| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Medicine, Arthritis Center, and
Department of Pathology, Boston University School of Medicine, Boston, MA 02118
l
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B and Egr and the AP-1 site identified in this study. We
found that TCR signaling induces AP-1 binding to this site and
regulates the fasl promoter function in a fashion
dependent on NF-B binding. However, mutation in the AP-1 site alone
did not show a significant effect on the promoter function. The data
suggest that the minimal promoter required at least two transcription
factors to function. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The transcription of the fasl gene is initiated following T
cell activation (12, 13, 14). Upon stimulation, several
kinases are activated at the proximal end of the TCR/CD3 complex to
transduce the activation signal (22). Gonzalez-Garcia et
al. showed that p56lck, but not
p59fyn, is essential for FasL expression
(23), and Eichen et al. showed that Zap-70 tyrosine kinase
is required for the expression of the fasl gene
(24). These secondary messengers activate various
transcription factors, some of which may directly act on FasL promoter
to induce fasl gene expression. For example, NF-AT, Egr,
and NF-B have been implicated in TCR-mediated fasl gene
induction (25, 26, 27, 28, 29, 30). These studies have also identified
binding sites for the respective transcription factors on the
fasl promoter (25, 26, 27, 28, 29, 30).
It has been reported that stress-causing agents such as UV radiation,
-irradiation, and DNA-damaging agents can induce fasl
gene expression (31, 32). Two AP-1-binding elements were
found to be important in stress-induced expression of fasl
gene (31, 32). In the present study, we examined the role
of TCR-induced AP-1 in fasl gene expression by using the
murine fasl promoter. We present evidence indicating that
TCR-induced AP-1 binds to the newly identified AP-1 site in the
fasl promoter and functionally cooperates with NF-B (that
binds to the FasL-B1 site (29) previously identified)
to activate transcription.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The generation, characterization, and maintenance of 5D5
hybridoma T cell, and the preparation of purified anti-CD3 Ab
(145-2C11) have been described previously (29, 33). T4
polynucleotide kinase and restriction enzymes KpnI and
HindIII were purchased from New England BioLabs (Beverly,
MA). Bacterial alkaline phosphatase was obtained from Life Technologies
(Gaithersburg, MA). Radioactive [-32P]ATP
(3000Ci/mmol) was purchased from New England Nuclear (Boston, MA).
Affinity-purified rabbit anti-mouse Abs against Egr-1 (catalogue
no. SC-189X), Egr-3 (catalogue no. SC-191X), c-Jun (catalogue no.
SC-822X), and c-Fos (catalogue no. SC-253X) were purchased from Santa
Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-Egr-2 Abs
(catalogue no. PRB-236P) were obtained from Covance (Richmond, CA).
Quick-Change Site-Directed Mutagenesis kit (catalogue no. 200518) was
purchased from Stratagene (La Jolla, CA), and Qiagen Plasmid Maxiprep
Kit (catalogue no. 12162) was obtained from Qiagen (Santa Clarita, CA).
Luciferase Detection Kit (catalogue no. E1500) and 5x reporter lysis
buffer (catalogue no. E3971) were purchased from Promega (Madison, WI).
Luminescent ß-gal Detection Kit (catalogue no. K2048-1) was purchased
from Clontech (Palo Alto, CA). Poly(dI-dC) (catalogue no. 27-7880) was
purchased from Amersham Pharmacia Biotech (Piscataway, NJ).
AP-1 and NF-B consensus oligonucleotides were purchased from Promega
(catalogue no. E3201, 5'-cgcttgatgagtcagccggaa-3'; catalogue no. E3291,
5'-agttgaggggactttcccaggc-3'). The WT, Egr mut, and AP-1 mut
oligonucleotides were synthesized by Integrated DNA Technologies
(Coralville, IA). The sequences are as follows. WT,
5'-tcagatgcaagtgagtgggtgtctcacaga-3'; Egr mut,
5'-tcagatgcaagtgagtttttgtctcacagagaagc-3'; AP-1 mut,
5'-gtgagtgggtgtctaaaagagaagcaaagagaaga-3'
(underlined nucleotides indicate the positions of the mutation). The
22-mer oligonucleotide
(5'-tgggtg
gagaagcaa-3')
bearing the putative AP-1 binding site (underlined and/or
strikethrough), but not the Egr binding site, was also synthesized (see
Results). SP-1 consensus oligonucleotide
(5'-attcgatcggggcggggcgagc-3') and Egr consensus (EBS-1,
34) oligonucleotide (5'-cgccctcgcccccgcgccggg-3') were
respectively provided by Dr. R. L. Widom and Dr. H. Cohen, both at
the Department of Medicine, Boston University School of Medicine
(Boston, MA). NF-AT consensus oligonucleotide was purchased from Santa
Cruz Biotechnology (catalogue no. SC-2577,
5'-cgcccaaagaggaaaatttgtttcata-3'). The rEgr-2 protein was produced
using the appropriate construct kindly provided by Dr. P. Charnay
(Laboratoire de Génétique Moléculaire, Paris, France)
(35).
Activation of 5D5 hybridoma T cells
5D5 cells were activated in wells (24-well or 6-well plates, catalogues nos. 3524 and 3516, respectively; Costar, Corning, NY) that had been coated with purified anti-CD3 mAb (145-2C11). Briefly, the purified mAb was prepared in 0.1 M NaHCO3 (pH 9) at 6 µg/ml. Then, 300 µl of this solution was used to coat the 24-well plates and 1.2 ml to coat the 6-well plates. Plates were kept in a 37°C incubator for 24–48 h and then washed three times with culture medium before use. 5D5 cells were activated at 0.4 x 106 cells/ml/well when 24-well plates were used. The electroporated 5D5 cells were activated at 2–3 x 106 cells/3 ml/well in the 6-well plates.
DNase I-footprinting assay
The antisense strand of the 333-bp fragment of the murine
fasl promoter was labeled with
[-32P]ATP and used as a probe. The p333
construct was digested with HindIII, dephosphorylated with
bacterial alkaline phosphatase, and labeled with
32P using T-4 polynucleotide kinase.
Subsequently, linearized 32P-labeled construct
was digested with KpnI to isolate the fasl
promoter fragment from the plasmid (pGL3). The labeled fasl
promoter fragment was gel purified.
Seventeen microliters of a mixture containing 1 µg of poly(dI-dC) (1 mg/ml) and 50 µg of the nuclear extract, either from unactivated or anti-CD3-activated (1.5 h) 5D5 cells, were kept on ice for 10 min. Subsequently, 3 µl of the solution containing the 32P-labeled fasl promoter fragment (10,000 cpm) and 33 mM MgCl2 were added to the mixture and incubated for another 90 min on ice. Then, the mixture was digested with 50 ng of DNase I for 5 min at room temperature. The reaction was terminated by adding 150 µl of the stop solution (100 mM Tris (pH 8), 100 mM NaCl, 1% SDS, 10 mM EDTA (pH 8), salmon sperm DNA (25 µg/ml), and proteinase K (100 µg/ml)). For the negative control, nuclear extract was replaced with the buffer used for nuclear extract preparation (36). The digested products were ethanol precipitated and suspended in the DNA loading dye (20 mM EDTA (pH 8), 500 µg/ml bromophenol blue, 500 µg/ml xylene cyanol, and 95% (v/v) formamide). Samples were heated at 95°C for 5 min before loading. The products were resolved by electrophoresis using a 7% polyacrylamide/8 M urea DNA-sequencing gel in 1x Tris-borate-EDTA buffer (pH 8) at a 30 mA constant current. After electrophoresis, the gel was vacuum dried at 80°C, and the products were visualized by autoradiography. As a marker, G+A sequence reaction was conducted and ran in parallel with the samples during electrophoresis. The Maxam-Gilbert G+A sequencing ladder was generated, as described previously (37).
Nuclear extraction and EMSA
The preparation of nuclear extract and the oligonucleotide probes has been described previously (36). A typical binding reaction mixture contained the labeled oligonucleotide (0.2 ng/µl), 5 µg of nuclear proteins, and 1 µg of poly(dI-dC) (1 mg/ml) in a final volume of 15 µl in a 0.6-ml microcentrifuge tube. The mixture was incubated for 20 min at room temperature and then electrophoresed through a 5% polyacrylamide gel under nondenaturing condition in 0.5x Tris-borate-EDTA (pH 8) at 200 V. After electrophoresis, the gel was fixed in 10% (v/v) glacial acetic acid and 10% (v/v) methanol and vacuum dried at 80°C. The bands were visualized by autoradiography. For the cold target competition analyses, 100-fold molar excess of competitors was incubated with the nuclear extract for 10 min at room temperature before the addition of the labeled oligonucleotide. The mixtures were incubated for 20 min at room temperature before electrophoresis. For the supershift analyses, 0.5 µl of the appropriate Abs was added to the reaction mixtures, and the samples were incubated at room temperature for 40 min before electrophoresis.
Transfection and luciferase assay
The cloning of the murine fasl promoter (GenBank accession AF045739) and the procedures for generating various mutant constructs have been described previously (29). DNA sequencing confirmed the mutations (data not shown). The transcriptional start site was determined and it was assigned +1 (38). The nucleotide residue immediately upstream of this transcriptional start site was assigned -1.
Hybridoma 5D5 cells (3 x 107), suspended in 300 µl of DMEM, were transfected by electroporation. The cells were mixed with 8 µg of the promoter-reporter construct and 8 µg of the pCMV-ß-gal plasmid. The pCMV-ß-gal construct was used to normalize the transfection efficiency. The mixtures were incubated on ice for 10 min before electroporation, which was conducted using a Bio-Rad Gene Pulser (Richmond, CA) at 240 V and 960 µF. Afterward, the cells were incubated at room temperature for 10 min and washed twice with culture medium. Each sample was suspended in 6 ml of culture medium. One-half of the cell suspension was put into 6-well plates that had been coated with anti-CD3 mAb. The other half was kept in untreated wells. The cells were incubated for 16 h at 37°C. Cells were harvested, washed three times with 1x PBS, and lysed with 110 µl of 1x reporter lysis buffer. Luminescent ß-gal Detection Kit was used to detect ß-gal activity. Briefly, 50 µl of the reaction buffer was incubated with 10 µl of the cell lysate for 1 h at room temperature and the samples were measured by a luminometer (Turner TD-20e model; Promega). To determine luciferase activity, 50 µl of the substrate provided in Luciferase Detection Kit were mixed with 40 µl of cell lysate before the measurement by the luminometer. After normalizing for transfection efficiency, the increase in luciferase activity in samples cultured in anti-CD3-coated wells was calculated as fold induction according to the following formula: Luciferase value from anti-CD3 activated cells/luciferase value from unactivated cells.
Preparation of rEgr-2 protein
The characterization and the induction of rEgr-2 protein in bacteria have been described previously (35).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previously, we showed the functional importance of a
NF-B-binding element, FasL-B1 (located -71 to -61 with respect
to the transcriptional start site), in the expression of murine
fasl promoter activity (29). Transient
transfection assays showed that mutation of this element reduces
fasl promoter activity by only 50%. This result suggests
that there are additional regulatory element(s) in the murine
fasl promoter. To identify these elements, a DNase
I-footprinting assay was performed. Nuclear extracts from both
unactivated and anti-CD3-activated 5D5 cells were tested (Fig. 1
). Three protected regions were
identified and designated as protected region A, protected region B,
and protected region C. Protected region A corresponds to the NF-B
binding site FasL-B1. This region was protected by nuclear extracts
prepared from both unactivated and anti-CD3-activated cells, as
shown by the presence of hypersensitive bands (indicated by arrows).
The protection by the unactivated nuclear extract can be explained by
the binding of the p50/p50 homodimer (29). Protected
region B (-125 to -95 from the transcriptional start site) was
strongly protected by nuclear extract from anti-CD3-activated
cells. This region corresponds to the reported Egr binding site in the
human fasl promoter (27). Finally, the
protected region C corresponds to the reported NF-AT binding site in
the human fasl promoter (25, 26, 39). Nuclear
extracts from both the unactivated and anti-CD3-activated cells
protected this region. This is perhaps due to the binding of the SP-1
family members. In fact, in addition to NF-AT, we can detect the
binding of SP-1 family members to this region by EMSA
(40). Although the DNase I-footprinting analysis weakly
indicated that the region between FasL-B1 and the protected region B
is protected, the region does not bind to any activation-induced
factors (unpublished observation).
|
To determine the role that each of the above DNA elements played
in murine fasl promoter regulation, we synthesized
5'-deletion mutant constructs and tested their transcriptional
activities in response to anti-CD3 Ab treatment. Fig. 2A shows the relative
locations and the presence of each binding element in the different
deletion mutant constructs. The transcriptional response of 5D5 cells
to anti-CD3 stimulation is shown in Fig. 2B. All
5'-deletion mutant constructs, except the p82, responded to
anti-CD3 treatment. Of the binding elements identified by DNase I
footprinting, only FasL-B1 binding site is contained in p82. In
contrast, the next smallest promoter fragment was the p125 construct,
which responded strongly to the anti-CD3 treatment (5.2-fold
induction). The p125 contains the FasL-B1 site (29) and
the Egr binding site (protected region B) (27, 28).
However, the NF-AT binding site reported by Latinis et al.
(25) is not present. These findings indicate that in our
transient transfection assay system, NF-B alone is not sufficient
for an optimal induction of the fasl gene. Moreover, the
NF-AT binding site is dispensable for TCR-mediated induction of the
fasl gene in this system. This piece of data is consistent
with the result reported by Mittelstadt and Ashwell (27),
who showed that the human fasl promoter functions in the
absence of the NF-AT site. Based on the information, we decided to use
a minimal promoter fragment (i.e., p125 construct) that gave optimal
response to anti-CD3 treatment for the rest of our study.
|
We conducted a series of EMSA to examine how many different
proteins may bind to the protected region B. We synthesized a
double-stranded oligonucleotide probe spanning nucleotides -125 to
-95 (WT) and used it in EMSA with nuclear extracts from 5D5 cells that
had been activated by anti-CD3 for various times (Fig. 3A). In comparison with the
unactivated 5D5 cell nuclear extract (lane 1), the
1-h activated nuclear extract displayed two prominent
activation-induced bands (lane 2, indicated by arrows
1 and 2). After 2 h of activation, a faster migrating band was
detected (lane 3, indicated by bracket 3). However,
further stimulation did not produce any new bands (4 and 8 h,
lanes 4 and 5). Therefore, it appears that there
are three activation-induced factors that bind to the WT
oligonucleotide. Based on this result, the 4-h activated nuclear
extract was used for most of the subsequent EMSA analyses.
|
B). As seen in lane
3, a 100-fold excess of the self oligonucleotide competed all of
the activation-induced bands, indicating that these factors
specifically bound the probe. When a 100-fold excess of unlabeled
EBS-1, an oligonucleotide capable of binding Egr family member
(34), was used as a competitor, almost all of the
activation-induced bands disappeared (lane 4).
However, one band remained (indicated by arrowhead). In contrast,
excess NF-AT, SP-1, or NF-B consensus oligonucleotides had no
obvious effect on the binding pattern (lanes 5,
6, and 8). Although the SP-1 consensus
oligonucleotide blocked the formation of the slowest migrating band
(lane 6), the self oligonucleotide did not
(lane 3). This indicated that the binding was weak
and/or involved non-SP-1 factors. Interestingly, when an AP-1 consensus
oligonucleotide was used, a very distinct change was observed in band 1
(lane 7) such that the size of the band is
noticeably reduced. The upper portion of the band is clearly
missing (indicated by the arrowhead). The data indicate that band 1 is
composed of two different closely migrating bands, and the identities
of the bound factors are likely to be members of the Egr and AP-1
families.
To determine the identities of the activation-induced factors bound to
the WT oligonucleotide, an Ab-mediated supershift analysis was
performed. The results are shown in Fig. 3C. In the presence
of anti-Egr-1 Ab, most of the band 1 binding was blocked or shifted
(lane 3, arrow 1). However, a slow migrating portion
of the band 1 remained (indicated by arrowhead). When anti-Egr-2
and anti-Egr-3 were tested, the slow migrating portion of band 1
remained, although the formation of bands 2 and 3 was respectively
inhibited (lanes 4 and 5). Based on the
cold target competition analysis (Fig. 3B), which suggested
that the slow migrating portion of band 1 could be AP-1, we tested the
effect of anti-c-Jun and anti-c-Fos on the binding pattern
(lanes 6 and 8). Both Abs inhibited the
formation of the slow migrating portion of band 1. Anti-c-Jun clearly
shifted the binding, whereas anti-c-Fos did not. When
anti-c-Jun plus anti-Egr-1 or anti-c-Fos plus
anti-Egr-1 were used (lanes 7 and 9),
the entire band 1 formation was blocked.
To more precisely define the AP-1 binding site, we have conducted
binding experiments with a 22-mer oligonucleotide
(5'-tgggtgtctcacagagaagcaa-3') probe (Fig. 3D). We observed a specific binding with nuclear extract
from anti-CD3-activated 5D5 cells, and the binding band can be
shifted with anti-c-Jun and inhibited with anti-c-Fos. Although
the probe contains the critical sequence (gggtg) for Egr binding, the
binding by Egr proteins was either extremely weak or undetectable,
suggesting its location on the extreme 5' end of this probe is
inadequate for Egr binding. It is possible that additional bases on its
5' side are required to form an appropriate Egr-binding motif. The
result indicates that the absence of the Egr binding site does not
cause conformational changes in the AP-1 site. Taken together, the data
provide convincing evidence that AP-1 (c-Jun/c-Fos heterodimer) is the
transcription factor that binds to the -125 to -95 region of the
murine fasl promoter.
Because fasl mRNA in 5D5 cells was detected at 2 h
after anti-CD3 activation (unpublished observation) and the
induction of Egr-3 required 2 h of activation or more (Fig. 3A), we asked whether Egr-3 might be present in the nuclear
extract after 1.5 h of activation. Ab-mediated supershift analyses
showed that while both Egr-1 and Egr-2 proteins were present at this
time, Egr-3 was not (Fig. 3E, lanes 3–5).
Interestingly, the slow migrating portion of band 1, which was not
shifted by anti-Egr-1, was also detected in the extract. The
results suggest that AP-1 and Egr-2, but not Egr-3, may function in the
early phase of FasL transcription.
AP-1 cooperatively regulates the fasl promoter with
NF-B
Upon examination of the promoter sequence, we have identified a
potential AP-1-binding sequence that is located immediately downstream
of the Egr binding site. Based on the AP-1 consensus sequence
(TGACTCA), we predicted that the potential AP-1 binding site might
reside between nucleotide positions -108 to -102 (TGTCTCA) or between
positions -106 to -100 (TCTCACA). To determine the functional
significance of these sequences, we have mutated the nucleotides at
-103 and -101
(5'-gtgagtgggtgtctaaaagagaagcaaagagaaga-3'
(bold and underlined nucleotides indicate the positions of the
mutation). We synthesized the mutant construct (AP-1 Mut) using the
p125 construct as the wild-type control. For comparison, we have also
tested a construct that was mutated in the FasL-B1 site, i.e., the
same mutation we used in our previous work was introduced to the p125
construct (29). The result of the transient transfection
assay is shown in Fig. 4. In agreement
with our previous finding, the FasL-B1 mutant construct (B1 Mut)
showed a 50% reduction in the reporter assay. In contrast, mutation in
the AP-1 binding site did not show a significant effect (AP-1 Mut).
However, when both the FasL-B1 and the AP-1 binding sites were
mutated (B1/AP-1 Mut), the promoter-reporter was rendered
nonfunctional. The data suggest that AP-1 can cooperatively regulate
the fasl promoter with NF-B.
|
Because the AP-1 site is close to the Egr binding site, we
conducted an EMSA to determine whether the mutation introduced into the
AP-1 binding site affected Egr binding. Because the WT oligonucleotide
binds to multiple proteins, a bacterial lysate containing rEgr-2
protein (35) was used to analyze Egr protein binding. We
tested three oligonucleotides: WT, Egr mut, and AP-1 mut (see sequence
information in Materials and Methods). We mutated the
sequence gggtg to aaatg (-111 to -109) to eliminate the Egr binding
site. As shown in Fig. 5A, the
rEgr-2 protein bound to the WT and AP-1 mut oligonucleotides, but not
to the Egr mut oligonucleotide (indicated by arrow). Furthermore, the
rEgr-2 protein/DNA complex can be completely shifted by anti-Egr-2
Ab (data not shown), indicating that Egr-2 is responsible for the band
formation. This result shows that the mutation introduced into the AP-1
binding site does not prevent Egr proteins from binding to the
promoter.
|
B. The self (AP-1 consensus), WT, and Egr mut effectively
competed for AP-1 binding and inhibited the formation of the
AP-1-binding band (lanes 4, 5, and
7). However, AP-1 mut oligonucleotide was unable to inhibit
AP-1 binding (lane 6), indicating that the mutation
in AP-1 mut had effectively eliminated the AP-1 binding site. Moreover,
the mutation in Egr binding site did not affect the AP-1 binding
because the Egr mut oligonucleotide displayed a comparable level of
inhibition as the WT oligonucleotide. Collectively, these observations
provide convincing evidence that the mutation of the AP-1 site did not
affect Egr-1 binding, and conversely, the mutation of Egr-1 site did
not affect AP-1 binding either. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). This
region corresponds to the reported Egr binding site in the human
fasl promoter (27). We showed, in addition to
the Egr binding site, that there is an AP-1-binding element within this
region of the mouse fasl promoter. In addition, AP-1
cooperatively regulates the promoter activity with NF-B, which bound
to the nearby FasL-B1 site.
The result in Fig. 4 suggests that Egr proteins cannot act alone to
stimulate the murine fasl promoter since the B1/AP-1
double mutation completely abolishes the promoter function of the p125
reporter construct. We also showed that mutation of the AP-1 binding
site eliminated the ability of the oligonucleotide to compete for AP-1
(Fig. 5B). However, the same mutation affected neither the
binding of Egr-2 (Fig. 5A) nor the promoter activity of the
construct (Fig. 4). These results suggest that a cooperative
interaction between NF-B and AP-1 is important in regulating the
murine fasl promoter. A cooperative interaction between
NF-B and AP-1 has been documented (41, 42, 43, 44, 45, 46, 47). Of
interest, D’Adamio et al. (48) reported that
dexamethasone activates transcription of the glucocorticoid-induced
leucine zipper (GILZ) gene in T cells. GILZ is a leucine
zipper-containing factor expressed in the nucleus that can protect T
cells from activation-induced cell death by inhibiting FasL expression.
It may be possible that GILZ inhibits fasl gene expression
by interfering with AP-1 function. Perhaps GILZ recognizes the
AP-1-binding sequence and acts as a repressor. Alternatively, GILZ may
directly interact with the AP-1 family members, and this complex may be
blocking the sites that are required for the interactions among
different transcription factors.
The fact that mutating the AP-1 binding site does not significantly
affect the ability of the murine fasl promoter suggests that
NF-B and Egr can activate the minimal promoter in the absence of
AP-1. However, in our studies, neither NF-B nor Egr alone is
sufficient for promoter activation (p82 construct in Fig. 2B, and B1/AP-1 Mut construct in Fig. 4). This may be
explained if NF-B and Egr functionally interact. Hence, mutating
both the Egr and the AP-1 binding sites should eliminate murine
fasl promoter activity by preventing an interaction between
NF-B and either AP-1 or Egr. Indeed, the deletion mutant construct
p82, which contains only the FasL-B1 site, did not respond
significantly to anti-CD3 activation (Fig. 2B). In
support of this notion, NF-B and Egr-1 have been reported to
regulate the p105 gene in a cooperative fashion (49). Our
data suggest that such an interaction could be a possible mechanism
that regulates fasl gene expression.
In addition to the NF-B/AP-1 and the NF-B/Egr interactions, our
data also suggest that Egr and AP-1 can interact. Mutation at the
FasL-B1 site resulted in 50% of the promoter activity in comparison
with the WT promoter (Fig. 4). Because Egr or AP-1 alone is not
sufficient for promoter function, it suggests that they act
cooperatively to stimulate transcription.
It has been reported that overexpression of either Egr-2 or Egr-3 can
induce fasl mRNA accumulation in HeLa cells (27, 28). However, which factor plays a more critical role in
fasl transcription is not known. The above studies suggest
that one of the Egr family members may be able to compensate for the
absence of the other member. It appears that Egr-2 is more important
than Egr-3 in our system. In 5D5 cells, fasl mRNA is first
detected 2 h after activation (unpublished observation). However,
the induction of Egr-3 requires 2 h of activation, whereas Egr-2
is detected 1 h after activation (Fig. 3, A and
D). Furthermore, the level of Egr-2 protein, but not Egr-3,
is maintained during activation. The Egr-3 level peaks at about 4
h after activation and then declines. At 16 h after activation,
Egr-3 is almost undetectable, but there is still a high level of
fasl mRNA (unpublished observation). Furthermore, He et al.
have recently reported that the ectopic expression of retinoic acid
receptor-related orphan receptor (RORt) inhibited the
activation-induced fasl mRNA expression in T cells
(50). However, egr-3 mRNA expression was not
inhibited, suggesting that Egr-3 may not be critical in regulating
fasl gene expression.
Mittelstadt and Ashwell (27) have shown that mutating the
Egr binding site in the human fasl promoter abolishes
promoter activity in a murine cell line. However, our preliminary data
using transient promoter-reporter assays suggest a different mechanism
for the murine fasl promoter. The mutation we made in the
murine Egr binding site, which completely abolished its Egr
protein-binding activity (Fig. 5A), did not cause a complete
inhibition of the function of the p125 promoter construct. In our
system, mutating the Egr binding site resulted in a 50–60% reduction
of the promoter activity. Moreover, our preliminary study showed that
mutating both the Egr and the FasL-B1 sites could eliminate 90–95%
of the reporter activity (data not shown). Hence, our findings suggest
a differential regulation of the fasl promoter between two
species, humans and mice. With respect to the AP-1 binding site, the
sequence immediately downstream of the murine Egr binding site
(GGGTG, the underlined letters indicate the possible
overlapping nucleotides shared by Egr- and AP-1-binding elements)
differs only by a single base (A
T) from the AP-1 consensus sequence
(TGTCTCA vs TGACTCA). However, the human sequence (TGTTTCT) is not
similar to the AP-1 consensus sequence. Furthermore, Li-Weber et al.
(51) reported that rAP-1 (c-Jun and c-Fos) did not
bind to an oligonucleotide that contains the TGTTTCT sequence of
the human fasl promoter. Also, Mittelstadt and Ashwell
(27) reported that AP-1 does not bind to an
oligonucleotide containing this region homologous to the human
fasl promoter. These studies further support the notion that
the fasl gene may be differentially regulated between humans
and mice.
We conclude that the p125 is a minimal murine fasl promoter that responds to TCR signal. Sequence downstream of nucleotide position -125 of the murine fasl promoter contains three defined transcription factor binding sites, and it is sufficient to trigger a response upon TCR signaling. This transcriptional response requires the interaction of at least two transcription factors induced by TCR cross-linking. Our results are consistent with the hypothesis that multiple transcription factor interactions are required for fasl gene expression.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. S.-T. Ju, K508, 71 East Concord Street, Boston University School of Medicine, Boston, MA 02118. E-mail address:
3 Abbreviations used in this paper: FasL, Fas ligand; Egr, early growth response; GILZ, glucocorticoid-induced leucine zipper; WT, wild type.
Received for publication October 12, 1999. Accepted for publication January 6, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B sites in mouse CD95 ligand (Fas ligand) promoter: functional analysis in T cell hybridoma. J. Immunol. 161:3469.B. J. Biol. Chem. 274:987.B and AP-1. Mol. Cell 1:543.[Medline]
B and Fos/Jun transcription factors induces potentiated biological functions. EMBO J. 12:3879.[Medline]
B/AP-1 in T cells. Biochem. Biophys. Res. Commun. 199:1064.[Medline]
promoter in human monocytic cells. J. Biol. Chem. 272:17795.B and AP-1. Eur. J. Immunol. 27:1091.[Medline]
B and AP1 synergistically transactivate the human GM-CSF promoter. Oncogene 14:2845.[Medline]
-induced NF-B1 transcription. J. Exp. Med. 185:491.t, a novel isoform of an orphan receptor, negatively regulates Fas ligand expression and IL-2 production in T cells. Immunity 9:797.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. Morgan, P. M. Pirard, L. Shears, J. Sohal, R. Pettengell, and H. S. Pandha Antagonism of HOX/PBX Dimer Formation Blocks the In vivo Proliferation of Melanoma Cancer Res., June 15, 2007; 67(12): 5806 - 5813. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Villa-Morales, J. Santos, E. Perez-Gomez, M. Quintanilla, and J. Fernandez-Piqueras A Role for the Fas/FasL System in Modulating Genetic Susceptibility to T-Cell Lymphoblastic Lymphomas Cancer Res., June 1, 2007; 67(11): 5107 - 5116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hess, P. Angel, and M. Schorpp-Kistner AP-1 subunits: quarrel and harmony among siblings J. Cell Sci., December 1, 2004; 117(25): 5965 - 5973. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Xiao, U. S. Deshmukh, S. Jodo, T. Koike, R. Sharma, A. Furusaki, S.-s. J. Sung, and S.-T. Ju Novel Negative Regulator of Expression in Fas Ligand (CD178) Cytoplasmic Tail: Evidence for Translational Regulation and against Fas Ligand Retention in Secretory Lysosomes J. Immunol., October 15, 2004; 173(8): 5095 - 5102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. M. Droin, M. J. Pinkoski, E. Dejardin, and D. R. Green Egr Family Members Regulate Nonlymphoid Expression of Fas Ligand, TRAIL, and Tumor Necrosis Factor during Immune Responses Mol. Cell. Biol., November 1, 2003; 23(21): 7638 - 7647. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Crist, T. S. Griffith, and T. L. Ratliff Structure/Function Analysis of the Murine CD95L Promoter Reveals the Identification of a Novel Transcriptional Repressor and Functional CD28 Response Element J. Biol. Chem., September 19, 2003; 278(38): 35950 - 35958. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Yang, B. Dong, P. R. Mittelstadt, H. Xiao, and J. D. Ashwell HIV Tat Binds Egr Proteins and Enhances Egr-dependent Transactivation of the Fas Ligand Promoter J. Biol. Chem., May 24, 2002; 277(22): 19482 - 19487. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. T. Miller and L. J. Berg Defective Fas Ligand Expression and Activation-Induced Cell Death in the Absence of IL-2-Inducible T Cell Kinase J. Immunol., March 1, 2002; 168(5): 2163 - 2172. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Herr and K.-M. Debatin Cellular stress response and apoptosis in cancer therapy Blood, November 1, 2001; 98(9): 2603 - 2614. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-M. Steff, S. Trop, M. Maira, J. Drouin, and P. Hugo Opposite Ability of Pre-TCR and {{alpha}}{{beta}}TCR to Induce Apoptosis J. Immunol., April 15, 2001; 166(8): 5044 - 5050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Lindgren, K. Axcrona, and T. Leanderson Regulation of Transcriptional Activity of the Murine CD40 Ligand Promoter in Response to Signals Through TCR and the Costimulatory Molecules CD28 and CD2 J. Immunol., April 1, 2001; 166(7): 4578 - 4585. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. T. Eichhorst, M. Müller, M. Li-Weber, H. Schulze-Bergkamen, P. Angel, and P. H. Krammer A Novel AP-1 Element in the CD95 Ligand Promoter Is Required for Induction of Apoptosis in Hepatocellular Carcinoma Cells upon Treatment with Anticancer Drugs Mol. Cell. Biol., October 15, 2000; 20(20): 7826 - 7837. [Abstract] [Full Text]
|
|
|
|
|
|
X. Yu and S. M. Weissman Characterization of the Promoter of Human Leukocyte-specific Transcript 1. A SMALL GENE WITH A COMPLEX PATTERN OF ALTERNATIVE TRANSCRIPTS J. Biol. Chem., October 27, 2000; 275(44): 34597 - 34608. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
