| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Tumor Immunology Program,
Tumorvirus-Immunology, German Cancer Research Center, Heidelberg, Germany;
Department of Neuroimmunology, Max Planck Institute of Neurobiology, Martinsried, Germany;
Institute for Clinical Neuroimmunology, Ludwig Maximilians University, Munich, Germany; and
¶
Institute for Clinical and Molecular Virology, University Erlangen-Nürnberg, Erlangen-Nürnberg, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and TCR/CD3 stimulation. Here we report the
identification of two positive regulatory IFN-regulatory
factor-dependent domains (PRIDDs) in the CD95L promoter and its 5'
untranslated region, respectively. EMSAs demonstrate specific binding
of IFN--induced IFN-regulatory factor 1 (IRF-1) to the PRIDD
sequences. Ectopic IRF-1 expression induces CD95L promoter activity.
Furthermore, we demonstrate that PRIDDs play an important role in
TCR/CD3-mediated CD95L induction. Most interestingly, viral IRFs of
human herpes virus 8 (HHV8) totally abolish IRF-1-mediated and strongly
reduce TCR/CD3-mediated CD95L induction. We demonstrate here for the
first time that viral IRFs inhibit activation-induced cell death. Thus,
these results demonstrate an important mechanism of HHV8 to modulate
the immune response by down-regulation of CD95L expression. Inhibition
of CD95-dependent T cell function might contribute to the immune escape
of HHV8. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The transcriptional regulation of CD95L upon T cell activation has been
studied over the last couple of years. Several binding sites for NF-AT
(12, 13, 14, 15), NF-
B (16, 17), and Egr (early
growth response protein) (18, 19) have been identified in
the CD95L promoter. TGF-
was described to inhibit CD95L expression
via down-regulation of c-Myc and, thereby, to block AICD
(20). Accordingly, overexpression of c-Myc results in
induction of CD95L (21). The forkhead transcription factor
was described as a silencer of CD95L expression (22).
Recently, the contribution of IRF-1 in CD95L induction upon TCR/CD3
triggering has also been demonstrated (23).
In addition to TCR/CD3 stimulation of T cells, apoptosis is induced
upon IFN- treatment in certain cell types. In microglia CD95 and
CD95L are simultaneously up-regulated by IFN- (24). In
neuroblastoma cells IFN- induces apoptosis via induction of the
CD95L (25). In addition, CD95L expression is up-regulated
in squamous cell carcinoma (26) and in Tera-2 embryonal
carcinoma cells (27). The molecular mechanisms of
IFN--mediated CD95L induction are still elusive. Stimulation of the
IFN- receptor leads to phosphorylation of STAT1 via activation of
the Janus family tyrosine kinases, JAK1 and JAK2.
Tyrosine-phosphorylated STAT1 assembles to form a homodimer, which then
translocates into the nucleus and binds to a specific DNA sequence
motif, termed the IFN- activation site to induce a variety of genes
(28, 29). Among these genes is IFN-regulatory factor 1
(IRF-1). IRF-1 belongs to a family of transcription factors termed the
IRF family. Originally identified as a transcription factor that binds
to the human IFN- gene promoter (30), it is now clear
that IRF-1 has a remarkable functional diversity; it is involved in
proliferation, blocks oncogenesis, and is able to induce apoptosis
(31, 32, 33, 34). Furthermore, IRF-1 is a regulator of the immune
response. It is involved in the function of multiple stages of the Th1
limb of the immune response (35, 36, 37); in addition, IRF-1
regulates the IL-15 gene and is therefore essential for the development
of NK cells (38). Studies in
IRF-1-/- mice demonstrated that IRF-1 plays a
crucial role in IFN-induced anti-viral and anti-bacterial
responses (35, 39). Also, DNA damage-induced apoptosis in
mitogenically activated mature T cells is dependent on IRF-1
(40).
Viral members of the IRF family (vIRFs) encoded by human herpesvirus 8 (HHV8) have also been identified (41). HHV8 is thought to be the viral agent involved in Kaposi’s sarcoma and primary effusion lymphoma. The genome of HHV8 contains a cluster of open reading frames encoding proteins with homology to the cellular transcription factors (41). Viral IRF1 and vIRF2 have been shown to interact with several IRFs, including IRF-1, and to inhibit IRF/IFN-mediated transcriptional activation (42, 43). At least two additional IRF-homologous genes were recently identified, namely K10.5 and K10, which are encoded by alternative splicing. Interestingly, according to two groups, K10.5/vIRF3 is expressed in viral latency (44, 45).
In this study we demonstrate the involvement of IRF-1 in CD95L expression. We identified two novel positive regulatory IRF-1-dependent domains (PRIDDs) in the human CD95L promoter. Our results show that ectopic expression of IRF-1 leads to induction of CD95L, which can be totally abolished in the presence of mutated PRIDDs. In addition, PRIDDs are necessary for TCR/CD3 induction of the CD95L. Moreover, we found that vIRF1 and vIRF2 of HHV8 inhibit IRF-1-induced CD95L induction and repress AICD via down-regulation of TCR/CD3-mediated CD95L expression. These data suggest that 1) IRF-1-mediated induction of the CD95L promoter might contribute to the activity of IRF-1 as a tumor suppressor-like protein; and 2) vIRFs of HHV8 act as modulators of the immune system by repressing AICD via modulation of TCR/CD3-mediated induction of the CD95L.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human T lymphoblastoid cell line Jurkat was cultured in
IMEM supplemented with 10% FCS, antibiotics (100 µg/ml), and
HEPES (10 mM). HeLa, LTK-, and C243
(46) cells were grown in DMEM supplemented with 10% FCS,
glutamine, and antibiotics. HaCaT, an immortalized human keratinocyte
cell line, was obtained from Prof. N. Fusenig (German Cancer Research
Center, Heidelberg, Germany) and cultured in DMEM/F-12 (Life
Technologies/BRL, Eggenstein, Germany). The hybridoma secreting
OKT3 (anti-CD3 mAb) was provided by G. Moldenhauer (German Cancer
Research Center) and cultured in RPMI supplemented with 10% FCS and
antibiotics. Human peripheral T cells were prepared as described
previously (7). For activation, resting T cells were
cultured at 2 x 106 cells/ml with 1 µg/ml
PHA (Sigma, Taufkirchen, Germany) for 16 h. T cells were then
washed three times and cultured in the presence of IL-2 (30 U/ml) for
additional 5 days. The anti-IRF-1 and IRF-2 polyclonal Abs were
purchased from Santa Cruz Biotechnology (Heidelberg, Germany). IFN-
was purchased from Biomol (Hamburg, Germany).
Plasmid constructs
Vector constructions were conducted by standard procedures
(48). The coding regions of vIRF1 and vIRF2 were isolated
by PCR amplification from 
phage DNA, clone CB47-1
(48), using specific primer pairs and subsequent insertion
in pcDNA3 (Invitrogen, Groningen, The Netherlands). Mutations in the
PRIDD sequences in the -1204/+100 construct were introduced using the
QuikChange mutagenesis kit (Stratagene, La Jolla, CA). The primers for
the mutagenesis reaction (MWG Biotech, Ebersberg, Germany) were
muPRIDDI (sense,
5'-gagagagagagatagagacagagacagacagaggtg-3';
antisense,
5'-cacctctgtctgtctctgtctctatctctctctctc-3') and
muPRIDDII (sense,
5'-cctctacaggactgagacgacgtaaaaccgtttgctgg-3';
antisense,
5'-ccagcaaacggttttacgtcgtctcagtcctgtagagg-3').
Underlined nucleotides represent the mutated sites compared with the
wild-type CD95L promoter sequence. The wild-type or mutated PRIDD
containing reporter plasmids were obtained through the integration of
oligomers of wild-type or mutated PRIDD sequences, which were used for
EMSAs (see below), into the BglII site of pGL2CVX (Promega,
Heidelberg, Germany). The sequences of all constructs were confirmed by
automated sequencing (Toplab, Martinsried, Germany).
Transient transfection
LTK, C243, and HeLa cells were transfected by DNA calcium phosphate coprecipitation (48, 50). Briefly, 1 x 105 cells were seeded in 9-cm2 plates. The medium was exchanged 4 h before transfection and was renewed 20 h post-transfection. One microgram of the indicated reporter plasmids, 0.5 µg IRF-1 expression plasmid (50) together with 3 µg vIRF expression plasmid or empty vector as a control, and 5 µg high m.w. DNA were isolated from LTK- cells. Transfection efficiency was determined by cotransfection of 0.3 µg myeloproliferative sarcoma virus-chloramphenicol acetyltransferase (CAT) (50). HaCaT cells were transfected using theFuGene6 transfection reagent obtained from Roche (Mannheim, Germany) following the manufacturer’s instructions.
Jurkat cells (2 x 106) were electroporated using a Bio-Rad Gene Pulser (Munich, Germany) set at 960 µF and 240 V. The transfected cells were allowed to recover for 36 h, then divided into aliquots for TCR/CD3 stimulation for additional 12 h. Five micrograms of the reporter plasmid together with 5 µg of the plasmids coding for the effector proteins or the empty vector were used for transfection. To compare transfection efficiency 3 µg of a plasmid coding for Renilla luciferase (Promega, Heidelberg, Germany) was added. Inductions took place during the last 12–16 h before cell harvest. Cells were always harvested 48 h after transfection. All transfections were performed in triplicate in one experiment. A representative experiment of three to five performed is always shown.
Luciferase and CAT determination
C243, HeLa, and LTK- cells were disrupted by the freeze-thaw method. HaCaT and Jurkat cells were lysed using the lysis buffer of the dual luciferase assay system (Promega). Luciferase assays were performed with this assay system using the DuoLumat LB9507 (Berthold, Wildbach, Germany). CAT amounts were determined using the CAT ELISA kit following the manufacturer’s instructions (Roche). Protein determination was performed using the Bio-Rad protein assay (Bio-Rad) to normalize the protein content of the transfected cells.
Determination of apoptotic cells
Jurkat cells were transiently transfected with a green fluorescence protein (GFP) expression plasmid together with an expression vector coding for vIRF1 or vIRF2 or an empty vector as a control. Living cells were isolated 24 h after transfection by Ficoll gradients, and TCR/CD3 stimulation was performed with plate-bound anti-CD3 Abs for additional 36 h, or cells were left untreated. Cell death was quantified by FACS analysis (forward/side scatter), gating on GFP-positive cells.
EMSA and supershift analysis
Nuclear extracts were prepared according to the method of
Andrews and Faller (51). HeLa, HaCaT, and Jurkat cells
were stimulated at 2 x 106/ml with IFN-
(1000 U/ml) for 16 h. PHA-activated primary T cells (2 x
107) were cultured in IL-2 medium for 6 days. In
vitro translated proteins were generated using the TNT Quick Coupled
Transcription/Translation System (Promega). Double-stranded
oligonucleotides (2.5 pmol) were end labeled with T4 polynucleotide
kinase (Roche) using 2.5000 Ci/mmol
[-32P]ATP (Amersham, Braunschweig, Germany).
Free nucleotides were removed using microspin G-25 columns. Sequences
of the single-stranded oligonucleotides were: PRIDDI (-55/-77) sense,
5'-agagaaagagaaagacagagg-3'; PRIDDII (+51/+71) sense,
5'-gagaagaagtaaaaccgtttg-3'; muPRIDDI sense,
5'-agagacagagacagacagagg-3'; muPRIDDII sense,
5'-gagacgacgtaaaaccgtttg-3'; +26/+45
oligonucleotide of the 5' untranslated region (5'UTR) of the CD95L
promoter sense, 5'-ttgacacctcagcctctaca-3'; and IFN-stimulated
responsive element (ISRE) sense,
5'-tcgggaaagggaaaccgaaactgaagcc-3'. The oligonucleotides contain
additional linker sequences for cloning into the BglII site
of pGL2CVX. Binding reactions were conducted at 4°C for 30 min using
2.5–5 µg nuclear proteins or 0.5 µl of in vitro translated
proteins in a buffer containing 10 mM HEPES (pH 7.5), 0.5 mM EDTA, 100
mM KCl, 2 mM DTT, 2% glycerol, 4% Ficoll 400, 0.25% Nonidet P-40,
and 0.1 µg/µl poly(dI/dC) in a total of 30 µl containing 50 fmol
labeled oligonucleotides. Competition was performed with the indicated
oligonucleotides at room temperature for 15 min. For supershift
analysis, 1 µg Ab was added to the binding reaction. Samples were
analyzed on a 6% nondenaturing polyacrylamide gel.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
induces CD95L promoter activity
To test whether IFN- induces CD95L expression via modulation of
its promoter activity, we transfected a reporter construct containing a
1204-bp fragment 5' of the transcription start site and the 100 bp of
the 5'UTR (-1204/+100) of the CD95L in front of the firefly luciferase
gene into different cell lines. IFN- stimulation of mouse
(LTK-, C243) and human cell lines (HeLa, HaCat,
Jurkat) led to up to a 5-fold induction compared with the untreated
controls (Fig. 1
). In contrast, the empty
vector was not induced by IFN- (data not shown). These results
indicate that IFN- can induce CD95L expression by modulation of
CD95L promoter activity.
|
As IRF-1 is induced and activated by IFN-, we investigated
whether IRF-1 is involved in the induction of the CD95L promoter. In
the murine cell lines LTK- and C243 as well as
in the human cell lines HeLa, HaCaT, and Jurkat the -1204/+100
reporter construct was induced up to 6-fold after coexpression of IRF-1
(Fig. 2A). To determine the
IRF-1-responsive domain in the CD95L promoter we tested serial 5'
deletions of the CD95L promoter for IRF-1 inducibility. Deletion of the
promoter sequence to -165 did not significantly influence
IRF-1-mediated induction. Deletion of the promoter sequence up to -88
partially decreased IRF-1-mediated CD95L promoter induction in HeLa and
Jurkat cells (Fig. 2B). Therefore, the sequence between
-165 to -88 contains an IRF-1-responsive element, which is in
agreement with the recently published data of Chow et al.
(23). However, in the cell lines investigated here, this
sequence contributed only partially to IRF-1-mediated induction of the
promoter. Further deletion of the promoter construct from -88 to -62
completely inhibited IRF-1-mediated induction in all cell lines
investigated. From these findings we conclude that at least one
additional positive regulatory IRF-1-dependent domain (PRIDD) is
contained in the -88/+100 fragment of the CD95L promoter.
|
To identify the PRIDD sequence we performed EMSAs using putative
IRF-1 binding sites derived from the -88/+100 CD95L promoter construct
as probes and in vitro translated IRF-1. Luciferase was used as a
negative control. Two IRF-1 binding sites were found: one sequence
located between -77 and -55 upstream of the transcription start site
(PRIDDI), and an additional sequence in the 5'UTR between +51 and +71
(PRIDDII). An ISRE was used as a positive control for IRF-1 binding,
and the sequence located in the 5'UTR between +26/+45 was used as a
negative control. As indicated in Fig. 3A, IRF-1 was able to bind the
PRIDDI and PRIDDII sequences. IRF-1 recognition sequences were located
in the CD95L promoter and 5'UTR, respectively. Binding to the +51/+71
sequence was much weaker compared with that to the -77/-55 sequence.
The typical ISRE sequence showed a much stronger binding of IRF-1
compared with the PRIDDs of the CD95L promoter. The random sequence
spanning from +26/+45 of the CD95L 5'UTR was not recognized by IRF-1.
In contrast, two different complexes of IRF-1 and the ISRE were formed
(Fig. 3A, IRF-1 I and IRF-1 II), probably due to oligomers
of IRF-1. Only one IRF-1 complex was detectable with the PRIDD
sequences, which probably reflects binding of the IRF-1 monomer. Also
the supershift with an Ab directed against IRF-1 was different. In case
of the typical ISRE sequence there were two supershifted complexes
(Fig. 3A, scI and scII) resulting in slower migration. In
contrast, binding of IRF-1 to the PRIDD sequences was blocked in the
presence of the anti-IRF-1 Ab (Fig. 3A). An
isotype-identical Ab directed against IRF-2 did not influence binding
of IRF-1; the labeled oligonucleotides with or without anti-IRF-1
Ab in the presence or absence of in vitro translated proteins did not
result in any signal (data not shown). We designated the two newly
identified IRF-1 recognition sequences of the CD95L promoter PRIDDI for
the -77/-55 sequence and PRIDDII for the +51/+71 sequence. Both
PRIDDs were able to competitively inhibit binding of IRF-1 to the ISRE.
Competition with PRIDDI was more efficient than competition with PRIDII
(Fig. 3B, compare lanes 6–8 with lanes
9–11), supporting the previous finding that binding of IRF-1 to
PRIDDII was not as strong as that to PRIDDI. Binding of IRF-1 to the
ISRE was much stronger than that to the PRIDDs (Fig. 3B,
compare lanes 3–5 and lanes 6–11). The
unlabeled control sequence containing the sequence (+26/+45) did not
compete for binding to the PRIDDs (Fig. 3B, lanes
12 and 13). To test whether binding of IRF-1 to the
PRIDDs occurs under physiological conditions we stimulated Jurkat and
HeLa cells with IFN- for 18 h and determined IRF-1 binding to
the PRIDDs. In both cell lines an induced band appeared upon IFN-
treatment using PRIDDI as a probe (compare Fig. 3C,
lanes 4 and 6, and lanes 8 and
9). In untreated Jurkat cells low amounts of complex
formation were detectable. Addition of anti-IRF-1 Abs interfered
with the IFN--induced complex (compare Fig. 3C,
lanes 6 and 7, and lanes 9 and
11). The complex consisting of cellular IRF-1 and PRIDDI
migrated as fast as the one obtained with the in vitro translated
material (compare Fig. 3C, lanes 2 and
6). IFN- treatment also induced IRF-1 binding to the
PRIDDII sequence in Jurkat and HeLa cells (Fig. 3D, compare lanes 3 and 4, and
lanes 5 and 6).
|
To analyze the importance of the PRIDD sequences in the CD95L
promoter we introduced point mutations in PRIDDI and PRIDDII to
generate nonfunctional IRF-1 binding motifs. We first tested the
mutated sequences in competition experiments for IRF-1 binding using
the respective wild-type oligonucleotide as the probe. The wild-type
sequences, but not the mutated PRIDD sequences, influenced binding of
IRF-1 to the wild-type oligonucleotide (Fig. 4A). To test whether the
mutated PRIDDs indeed resemble nonfunctional elements in vivo we
generated luciferase reporter constructs containing oligomers of
wild-type or mutated PRIDD sequences and compared IRF-1-mediated
induction after transient transfection. As indicated in Fig. 4B, oligomerized wild-type PRIDDI was induced by IRF-1
15-fold in HeLa and 4-fold in Jurkat cells, whereas oligomerized
PRIDDII was only induced 2-fold in HeLa and Jurkat cells. In contrast,
the mutated oligonucleotides were not induced by IRF-1 (Fig. 4B). Interestingly, both PRIDDs were also induced via
TCR/CD3 in Jurkat cells; PRIDDI was induced 15-fold, and PRIDDII 3-fold
via TCR/CD3 stimulation (Fig. 4C). To test the contribution
of the PRIDDs to IRF-1- and TCR/CD3-mediated CD95L induction, the two
PRIDD sequences were mutated either individually or in combination in
the context of the -1204/+100 CD95L promoter construct. Mutation of
PRIDDI alone resulted in 50% decrease in Jurkat cells and a 95%
decrease in HeLa cells of IRF-1-mediated CD95L promoter activity (Fig. 4D). In contrast, mutation of PRIDDII reduced IRF-1-mediated
induction only moderately (20% in Jurkat and 70% HeLa cells).
Disruption of both, PRIDDI and PRIDDII abolished IRF-1-mediated
induction in HeLa cells completely. However, in Jurkat cells,
IRF-1-mediated induction was still detectable, however at reduced
levels compared with the wild-type sequences. Mutation of PRIDDI or
PRIDDII also strongly inhibited TCR/CD3-mediated CD95L promoter
induction. Mutation of both PRIDDs decreased the induction of the CD95L
in an additive manner (Fig. 4E). These results demonstrate
that 1) both PRIDD sequences are involved in IRF-1 as well as in
TCR/CD3-mediated induction of the CD95L promoter; 2) PRIDDII
contributes to a much lesser extent to IRF-1 as well as
TCR/CD3-mediated CD95L induction; and 3) an additional PRIDD sequence
contributes to IRF-1-mediated CD95L induction in Jurkat cells. Since
the PRIDDs are involved in TCR/CD3-mediated CD95L induction in Jurkat
cells (23), we asked whether IRF-1 binds to the PRIDDs in
activated primary human T cells. We performed EMSAs with nuclear
extracts of primary 6-day-activated T cells with or without TCR/CD3
restimulation using the PRIDD sequences as a probe. As demonstrated in
Fig. 4F by a supershift with an Ab directed against IRF-1,
binding of IRF-1 was detectable in primary activated T cells. In
TCR/CD3-restimulated T cells, IRF-1 binding to the PRIDD sequences was
no more detectable, but another not yet identified band became visible
in the case of PRIDDI. These results suggest that PRIDDs are also
important for TCR/CD3-mediated CD95L expression in activated primary
human T cells. However, IRF-1 binding is only detectable in activated T
cells and is replaced by another as yet unidentified factor after
TCR/CD3 restimulation.
|
It was shown recently that vIRFs of HHV8 counteract
IFN/IRF-1-mediated gene activation (52, 53). Since we
could demonstrate that PRIDDs play an important role in IRF-1 as well
as in TCR/CD3-mediated induction of the CD95L promoter we wanted to
know whether vIRFs repress induction of the CD95L promoter. To
investigate this question we performed cotransfection experiments using
the -1204/+100 reporter construct together with an expression plasmid
encoding IRF-1 in the absence or the presence of expression vectors
coding for vIRF1 or vIRF2 of HHV8. As demonstrated in Fig. 5A cotransfection of either
vIRF1 or vIRF2 together with IRF-1 nearly completely inhibited
IRF-1-mediated induction of the CD95L promoter in HeLa as well as in
Jurkat cells. TCR/CD3-mediated CD95L induction in Jurkat cells was
decreased up to 7.5- and 10-fold in the presence of either vIRF1 or
vIRF2, respectively, compared with 17-fold in the vector control (Fig. 5B). To investigate whether vIRFs act via the PRIDDs in case
of TCR/CD3-mediated induction, we transfected the -1204/+100 reporter
construct containing either mutated PRIDDI or PRIDDII alone or in
combination in the absence or the presence of expression vectors coding
for vIRF1 or vIRF2. As mentioned above, mutation of either PRIDDI or
PRIDDII significantly reduced TCR/CD3-mediated induction of the CD95L.
The presence of vIRFs further reduced the induction; however, induction
of the construct containing both mutated PRIDD sequences was not
further repressed (Fig. 5C). The presence of vIRFs also
prevented IRF-1- as well as TCR/CD3-mediated induction of reporter
constructs containing oligomerized PRIDDs (data not shown). We conclude
from these results that vIRFs down-regulate CD95L expression via
modulation of PRIDD activity. To test the anti-apoptotic activity
of vIRFs we transfected Jurkat cells with expression constructs coding
for vIRFs together with a GFP expression plasmid as a transfection
marker. Twenty-four hours after transfection Jurkat cells were sorted
for GFP expression and stimulated via TCR/CD3 for an additional 36
h, and AICD was determined in GFP-positive cells. As demonstrated in
Fig. 5D, both vIRFs strongly inhibited TCR/CD3-mediated cell
death. The antagonistic CD95L Ab NOK1 inhibited TCR/CD3-mediated
apoptosis in control transfected cells, but no further inhibition was
observed in vIRF-transfected cells. Therefore, we conclude that vIRF1
and vIRF2 of HHV8 inhibit AICD by down-regulation of TCR/CD3-mediated
CD95L expression.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
induces IRF-1
binding to these sequences, as demonstrated by anti-IRF-1 Abs,
which specifically inhibit binding of IRF-1 to PRIDDs. It was shown
previously that IRF-1-/- cells can be
transformed by an activated c-Ha-ras oncogene. Furthermore,
ectopic expression of IRF-1 suppressed c-myc or
fosB-induced cell transformation (31, 32).
Therefore, IRF-1-mediated induction of apoptosis via CD95L expression
might contribute to its tumor suppressor function.
Mutation of either PRIDDI or PRIDDII in the context of the CD95L
promoter strongly inhibits IRF-1-mediated CD95L induction; PRIDDI
contributes significantly more to induction, which is due to most
likely stronger binding of IRF-1 to this sequence, as indicated by
EMSAs performed with the PRIDDI or PRIDDII sequence in competition
experiments and with in vitro translated IRF-1 (Fig. 3
). Binding of
IRF-1 to PRIDDs is different from IRF-1 binding to the ISRE, since
addition of anti-IRF-1 Abs results in stronger retardation of the
IRF-1/ISRE complex, whereas binding to the PRIDD sequences was
abolished under these conditions. These results indicate that binding
of IRF-1 to PRIDDs requires different domains or conformations of
IRF-1, which may be masked in the presence of anti-IRF-1 Abs.
Mutation of both PRIDDs prevents IRF-1-mediated CD95L induction. In
activated human primary T cells, IRF-1 binding to the PRIDD sequences
was also detectable. We detected a faint band of CD95L mRNA and low
amounts of CD95L protein in activated human primary T cells (data not
shown). The question arises of whether IRF-1 binding contributes to
this low CD95L expression. Alternatively, since IRF-1 contains a
repressor domain in the N-terminus, it might act as a silencer under
these conditions, as described previously for a forkhead transcription
factor (22, 54). Following TCR/CD3 restimulation, binding
of IRF-1 to PRIDDs is no more detectable, but is replaced by another as
yet uncharacterized protein (Fig. 4, PRIDDI). We tested possible
binding of other members of the IRF family (IRF-2, IRF-4, IRF-8) in
supershift experiments, but these IRF family members did not bind to
the PRIDD sequences. However, the PRIDD sequences play an important
role in TCR/CD3-mediated induction of the CD95L promoter, since
mutation of the PRIDD motifs strongly interfered with induction. The
contribution of IRF-1 to TCR/CD3-mediated induction in Jurkat cells is
in line with a recent report by Chow and co-workers (23)
showing that expression of antisense IRF-1 strongly reduced
PMA/ionomycin induction of CD95L in Jurkat cells. A putative IRF-1
binding site 5' of the NF-AT and Egr-3 recognition sequence, which is
located between 165 and 88 bp 5' of the transcription start site
according to our nomenclature, was described. In our hands deletion up
to -88 led to a slight decrease in IRF-1-mediated CD95L induction,
which is more prominent in Jurkat cells. However, deletion up to -62
completely inhibited IRF-1-mediated CD95L induction (Fig. 2).
Therefore, we conclude that PRIDDI, which is located between -88/-62
is an essential cis-acting element in IRF-1-mediated
promoter activation. A promoter construct containing only PRIDDII,
which is located in the 5'UTR (-62/+100, Fig. 2) was not sufficient
for IRF-1-mediated gene induction. Mutation of the PRIDDI sequence in
the presence of the wild-type PRIDDII sequence contained in the
-1204/+100 promoter construct still allowed IRF-1-mediated induction.
These results indicate that IRF-1 acts in cooperation with another
factor that binds a sequence located between -62 and -1204 of the
CD95L transcription start site. We observed that mutations of both
PRIDDI and PRIDDII totally inhibited IRF-1-mediated induction in HeLa
cells, but still allowed a weak induction in Jurkat cells (Fig. 2).
Most likely, the IRF-1 recognition site identified by Chow and
co-workers (23) contributes to the induction in Jurkat
cells. Interestingly, deletion of the CD95L promoter from -88 to -62
nearly completely inhibited PMA/ionomycin-mediated induction, which
also suggests the involvement of PRIDDI in TCR/CD3-mediated induction
(57). In summary, PRIDD sequences are involved in
IRF-1-mediated CD95L induction. Therefore, one can speculate that
pathways leading to IRF-1 binding to the PRIDDs also modulate CD95L
induction. However, IFN--mediated CD95L induction is hardly
detectable in Jurkat cells. Therefore, it most likely depends on the
cell type whether the induced amount of IRF-1 is sufficient for CD95L
expression. In addition, the composition of the transcription factors
in one particular cell type upon stimulation might also play a role in
the activity of IRF-1 on the CD95L promoter.
Many other transcription factors, such as NF-AT, Egr, NF-B, and
c-Myc have been found to be involved in CD95L induction. Mutation of
the recognition sequence of one of these transcription factors strongly
reduces CD95L induction (12, 13, 14, 15, 16, 17, 18, 19, 58). Therefore, it is
reasonable to assume that these transcription factors act
synergistically. These factors may form a large complex, an
enhanceosome, resembling the scenario described for the IFN-
promoter (56, 57, 58, 59). If one factor is missing, the
conformation of the enhanceosome is disturbed, and induction is
strongly reduced.
Most interestingly, vIRF1 and vIRF2 of HHV8 completely inhibited
IRF-1-mediated and reduced TCR/CD3-mediated induction of the CD95L,
which resulted in strong reduction of AICD (Fig. 5). One oncogenic
mechanism of the HHV8 virus might be to prevent T cell-mediated
apoptosis of tumor cells by inhibition of CD95L expression in activated
T cells. The inhibition of CD95L expression by HHV-8 might be linked to
a potentially proinflammatory effect of the CD95L, namely triggering of
neutrophil inflammation or induction of maturation of dendritic cells
(60, 61). Interestingly, HHV8 also expresses the
anti-apoptotic protein, vFLIP, which protects cells from
CD95-mediated apoptosis (62). Thus, several viral proteins
exist that inhibit the CD95 pathway and might contribute to immune
escape of HHV8. Our results strongly suggest that vIRFs act as
modulators of the immune response via CD95. The biological function of
vIRFs in the context of the immune response has to be worked out in
more detail. Also, the molecular functions of vIRF-mediated
transcriptional inhibition are not fully understood. It has been
reported that vIRF1 inhibits p300/CBP activity, which might be a
possible explanation for the inhibition of CD95L promoter activity
(63). In addition, in vitro translated vIRF1 reduced
protein binding to PRIDDs, but could not bind by itself (data not
shown). This result suggests that vIRF1 prevents CD95L expression by
inhibition of IRF-1 binding to PRIDDs. In contrast, vIRF2 did not
modulate protein binding to PRIDDs. Viral IRF2-mediated inhibition of
NF-B was described previously (43). Since NF-B is
involved in CD95L induction upon TCR/CD3 stimulation (17),
and IRF-1 can act synergistically with NF-B (64, 65),
this can be a possible explanation for vIRF2-mediated repression of
CD95L induction. In summary, the interplay between the transcription
factors that seem to be involved in the regulation of CD95L expression
under physiological conditions remains to be clarified.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Peter H. Krammer, German Cancer Research Center, Division of Immunogenetics (G0300), Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany. E-mail address: p.krammer{at}dkfz-heidelberg.de
3 Abbreviations used in this paper: CD95L, CD95 (APO-1/Fas) ligand; AICD, activation-induced cell death; CAT, chloramphenicol acetyltransferase; Egr, early growth response protein; GFP, green fluorescence protein; HHV, human herpesvirus; IRF, IFN-regulatory factor; ISRE, IFN-stimulated responsive element; PRIDD, positive, regulatory, IRF-dependent domain; UTR, untranslated region; vIRF, viral IRF.
Received for publication February 14, 2001. Accepted for publication November 15, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B. J. Biol. Chem. 274:987.B. Eur. J. Immunol. 30:661.[Medline]
1 inhibits Fas ligand expression and subsequent activation-induced cell death in T cells via downregulation of c-Myc. J. Exp. Med. 189:231. induces apoptosis and augments the expression of Fas and Fas ligand by microglia in vitro. Exp. Neurol. 162:290.[Medline]
in human neuroblastoma cell lines through the CD95/CD95L autocrine circuit. Cell Death Differ. 6:652.[Medline]
and cisplatin. Int. J. Cancer 80:564.[Medline]
and all-trans retinoic acid. Biol. Chem. 379:1083.[Medline]
-regulated DNA-binding factor. EMBO J. 10:927.[Medline]
gene regulatory elements. Cell 54:903.[Medline]
enhanceosome. Cold Spring Harbor Symp. Quant. Biol. 63:609.[Medline]
, and the production of interferon in the absence of IL-12 during DC-T cell cognate interaction: a new role for fas ligand in inflammatory responses. J. Exp. Med. 192:1661.B as a transcriptional activator of vascular cell adhesion molecule 1. Mol. Cell. Biol. 15:2558.[Abstract]
B activation is required for interferon regulatory factor-1-mediated interferon beta induction. Eur. J. Biochem. 261:546.[Medline]This article has been cited by other articles:
|
|
 |
|
  J. A. Fielhaber, Y.-S. Han, J. Tan, S. Xing, C. M. Biggs, K.-B. Joung, and A. S. Kristof Inactivation of Mammalian Target of Rapamycin Increases STAT1 Nuclear Content and Transcriptional Activity in {alpha}4- and Protein Phosphatase 2A-dependent Fashion J. Biol. Chem., September 4, 2009; 284(36): 24341 - 24353. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Areste, M. Mutocheluh, and D. J. Blackbourn Identification of Caspase-mediated Decay of Interferon Regulatory Factor-3, Exploited by a Kaposi Sarcoma-associated Herpesvirus Immunoregulatory Protein J. Biol. Chem., August 28, 2009; 284(35): 23272 - 23285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Y. Tan, J.-M. Li, R. Stocker, and L. M. Khachigian Angiotensin II-Inducible Smooth Muscle Cell Apoptosis Involves the Angiotensin II Type 2 Receptor, GATA-6 Activation, and FasL-Fas Engagement Circ. Res., August 28, 2009; 105(5): 422 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Conti, G. Regis, A. Longo, P. Bernabei, R. Chiarle, M. Giovarelli, and F. Novelli In the absence of IGF-1 signaling, IFN-{gamma} suppresses human malignant T-cell growth Blood, March 15, 2007; 109(6): 2496 - 2504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. R. Rezaee, C. Cunningham, A. J. Davison, and D. J. Blackbourn Kaposi's sarcoma-associated herpesvirus immune modulation: an overview J. Gen. Virol., July 1, 2006; 87(7): 1781 - 1804. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Fuld, C. Cunningham, K. Klucher, A. J. Davison, and D. J. Blackbourn Inhibition of Interferon Signaling by the Kaposi's Sarcoma-Associated Herpesvirus Full-Length Viral Interferon Regulatory Factor 2 Protein J. Virol., March 15, 2006; 80(6): 3092 - 3097. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Lubyova, M. J. Kellum, A. J. Frisancho, and P. M. Pitha Kaposi's Sarcoma-associated Herpesvirus-encoded vIRF-3 Stimulates the Transcriptional Activity of Cellular IRF-3 and IRF-7 J. Biol. Chem., February 27, 2004; 279(9): 7643 - 7654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Dourmishev, A. L. Dourmishev, D. Palmeri, R. A. Schwartz, and D. M. Lukac Molecular Genetics of Kaposi's Sarcoma-Associated Herpesvirus (Human Herpesvirus 8) Epidemiology and Pathogenesis Microbiol. Mol. Biol. Rev., June 1, 2003; 67(2): 175 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Cunningham, S. Barnard, D. J. Blackbourn, and A. J. Davison Transcription mapping of human herpesvirus 8 genes encoding viral interferon regulatory factors J. Gen. Virol., June 1, 2003; 84(6): 1471 - 1483. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
IRF), using a typical ISRE, the -55/-77
(PRIDDI) sequence of the human CD95L promotor, or the +51/+71 (PRIDDII)
or the +26/+45 sequence from the 5'UTR. The arrows indicate migration
of the two IRF-1 complexes, I and II. Asterisks indicate the
supershifted complexes (scI and scII). B, Competition
experiments with the indicated oligonucleotides using in vitro
translated IRF-1 or luciferase (Luc) as a control with an ISRE sequence
as a radiolabeled probe. Oligonucleotides were used in 20-, 50-, or
100-fold molar excess as competitors. The arrows indicate migration of
the two IRF-1 complexes, I and II. C, Nuclear extracts
of HeLa or Jurkat cells, which were treated with IFN-
