Chapter I Introduction
1.1 Interferons and their receptors
Chapter I Introduction
1.1 Interferons and their receptors
Type I interferons (IFNs) were discovered 50 years ago by Isaacs and
Lindenmann (Isaacs and Lindenmann, 1957; Isaacs et al., 1957), and has been well
recognized as cytokines for inducing cellular resistance to virus infection. Moreover,
IFNs also regulate cell growth (Grander et al., 1997) and possess immunomoduatory
activities (Biron, 2001). IFNs were classified into three families: type I, type II, and
type III IFN. Type I IFN comprises many subtypes of IFN-α and one IFN-β.
Furthermore, human IFN-ε, IFN-κ, IFN-ω, and limitins (IFN-δ in pigs and cattle,
IFN-τ in ruminants, IFN-ζ in mice) also belong to type I IFNs (Pestka et al., 2004;
Vilcek, 2003). All type I IFNs have structural homology and bind to a common
receptor, namely type I IFN receptor (consists of two chains IFNAR1 and IFNAR2)
(Uze et al., 2007). In contrast, type II IFN has only IFN-γ in this family, and bind to
type II IFN receptor (consists of two chains IFNGR1 and IFNGR2) (Bach et al., 1997).
Type III IFNs contain IFN-λ1 (IL-29), IFN-λ2 (IL-28A), and IFN-λ3 (IL-28B), which
are recognized by IFN-λ receptor (IFNLR1) and the IL-10Rβ subunit (IL-10Rβ)
(Takaoka and Yanai, 2006)
2 1.2 Type I IFN signaling pathway
Type I IFNs, IFN-α/β, are secreted in most cell types upon viral or other
microbial infection, and play an essential role in innate and adaptive immune response.
Binding of type I IFNs to IFNAR1 and IFNAR2, expressed in most cell types,
initiates several signaling cascades. The intracellular domain of IFNAR1 is
constitutively associated with TYK2, one of the Janus Kinases, whereas IFNAR2 with
JAK1. Phosphorylation of JAK1 and TYK2 is triggered by the interaction of type I
IFN and receptor. Activated JAKs further phosphorylate tyrosine residues in the
receptor for recruiting src-homology 2 (SH2) domain-containing proteins including
signal transducer and activator of transcription (STAT) 1, STAT2, and STAT3 in most
cell types (de Weerd et al., 2007). Other STATs, such as STAT4, STAT5, and STAT6,
seem to be strictly activated in limited cell types like endothelial or lymphoid cells.
Furthermore, JAKs also phosphorylate STATs on the tyrosine residue. Homodimers
(STAT1/1 and STAT3/3) or heterodimers (STAT1/2, STAT1/3, and STAT2/3) of
activated STATs translocate into nucleus, bind interferon-γ activated site (GAS)
elements ,and drive gene expression. STAT1 and STAT2 interact with IRF9 (p48) to
form interferon stimulated gene factor 3 (ISGF3) and target to the promoter
containing IFN-stimulated response elements (ISREs).
3 1.3 Effects of type I IFNs
Type I IFNs regulate several biological responses, such as induction of major
histocompatibility complex (MHC) class I expression, activation of natural killer (NK)
cell cytotoxicity, maturation of dendritic cells (DCs), and cancer immunoediting.
Notably, the most well known effect of type I IFNs is to establish an antiviral state
against virus infection by inducing ISG expression (Stark, 2007).
There are more than three hundred ISGs being induced following type I IFN
stimulation (Bowie and Unterholzner, 2008; Sadler and Williams, 2008). Some well-
characterized ISGs, including protein kinase R (PKR), 2’,5’-oligoadenylate-
synthetase (2’,5’-OAS), RNase L, Viperin (cig5) (Chin and Cresswell, 2001),
inducible nitric oxide synthetase (iNOS), and Mx, directly affect virus by suppressing
transcription and translocation, interfering RNA stability, blocking protein assembly,
or inducing cell apoptosis (Liu et al., 2011).
1.4 STATs
STATs were first discovered in the early 1990s. Mammalian STAT proteins
contain seven members, including STAT1, STAT2, STAT3, STAT4, STAT5a, STAT5b,
and STAT6 (Levy and Darnell, 2002). All of STATs have seven conserved domains,
such as N-terminal domain (NTD), coiled-coil domain (CCD), DNA binding domain
4
(DBD), linker domain, SH2 domain, and transactivation domain (TAD). The tyrosine
residue that undergoes phosphorylation upon activation, is especially highly
conserved in all STATs. Most activated STAT dimers bind to the GAS element
(TTCNNNGAA)(Hutchins et al., 2012), a palindromic sequence, whereas STAT1/2
and IRF9 heterotrimers bind to the ISRE (TTTCN2TTTC), a direct tandem repeat
(Robertson et al., 2007; Schindler et al., 2007).
1.4.1 STAT1
STAT1 was the first identified in STAT family, which is important in both IFN-α
and IFN-γ signaling pathway. STAT1-defecient mice are susceptible to microbial and
viral infections and tumor formation because of impaired type I and II IFN-mediated
responses (Durbin et al., 1996; Meraz et al., 1996). However, STAT1 also regulates
inflammation as well as antagonizes cell proliferation, indicating that STAT1 affects
diverse biological responses (Mui, 1999).
There are two isoforms of STAT1, including STAT1α and STAT1β, which are
resulting from alternative mRNA splicing. STAT1β lacks 38 amino acids at TAD of
STAT1α, but is still efficiently phosphorylate a at Y701 residue, forms dimers with
STAT1α isoform, and binds DNA. Nevertheless, overexpression of STAT1β does not
activate transcription, suggesting that STAT1β plays a dominant negative role (Lim
5 and Cao, 2006; Shuai et al., 1993).
1.4.1.1 Post-translational modification of STAT1
Post-translational modification of STAT1 includes the phosphorylation,
acetylation, ISGylation (Malakhova et al., 2003), SUMOylation (Rogers et al., 2003),
and ubiquitination (Kim and Maniatis, 1996). All the modifications can modulate
transcriptional or non-transcriptional activity of STAT1. However, there is still a
debate on whether activation of STAT1 is regulated by methylation (Komyod et al.,
2005; Meissner et al., 2004).
Recently, it has been reported that activation of STAT1 needs not only
phosphorylation but also deacetylation. STAT1 was found to interact with
CREB-binding protein (CBP) or p300 (Horvai et al., 1997), and downregulate
STAT1-mediated gene expression. Mutation of STAT1 at Lys410 and Lys413, two
acetylation sites of STAT1, attenuates the expression of STAT1-downstream gene
(Kramer et al., 2009). Besides, cells stimulated with HDAC inhibitor, TSA,
suppresses ISG expression, suggesting that deacetylation of STAT1 at Lys410 and
Lys413 is required for STAT1-dependent gene expression. The expression of ISGs
was suppressed by TSA, a HDAC inhibitor, and histone acetylation of ISRE
promoters decrease after IFN treatment (Chang et al., 2004; Nusinzon and Horvath,
6
2003). In these data, it suggested that promoter of ISGs are unlike other promoters,
histone deacetylation is required for its expression. However, these results are
controversial and are not consistent with others (Antunes et al., 2011). For example,
the interaction of ISGF3 complex with p300 can increase the DNA binding ability,
enhance downstream gene expression (Zhang et al., 2005b), and induce histone
acetylation (Varinou et al., 2003). Therefore, it is still unclear whether or not
acetylation of histone is required for IFN-α-mediated gene expression.
1.4.2 STAT3
STAT3 initially identified as acute phase response factor (APRF) due to its
ability to induce acute phase genes in the liver in response to IL-6 (Akira et al., 1994).
Biochemical and genetic studies demonstrate that STAT3 plays a crucial role in
transducing signal for IL-6 family, IL-10 family, granulocyte colony-stimulating
factor (G-CSF), Leptin, IL-21, IL-27, growth factor, oncogenes, and potentially IFNs
(Schindler and Plumlee, 2008). Unlike other STATs, ablation of STAT3 leads to
embryonic lethality at E6.5-7.5 (Takeda et al., 1997), suggesting that STAT3 is
important in development of various organs and cell proliferation.
STAT3, like STAT1, also has two isform, including STAT3α and STAT3β.
STAT3β missing the 55 C-terminal amino acids of STAT3α, and has 7 additional
7
amino acids residues at C terminus (Schaefer et al., 1995).
1.4.2.1 Post-translational modification of STAT3
Like STAT1, the activity of STAT3 is regulated by phosphorylation, acetylation,
and SUMOylation (Ohbayashi et al., 2008). Engagement of ligand and receptor
activates STAT3 by tyrosine phosphorylation at Y705. In addition, Serine
phosphorylation is required for full transactivation ability of STAT3 (Wen et al., 1995).
However, unphosphorylated STAT3 can still form dimers and induce transcription
(Yang and Stark, 2008). Acetylation of STAT3, unlike STAT1, plays a positive role in
gene transcription. Dimerization of STAT3 was inhibited by mutation of STAT3 at
K685, an acetylation site by p300, to arginine (Wang et al., 2005; Yuan et al., 2005).
STAT3 NTD (amino acids 1 to 130) alone can interact with p300, histone deacetylase
(HDAC) 1, and HDAC3 (Hou et al., 2008; Ray et al., 2008). Moreover, in addition to
K685, K49 and K87 at NTD of STAT3 can be acetylated by p300 in response to IL-6,
and the acetylation affects STAT3 downstream gene expression (Ray et al., 2005).
K49R and K87R, two mutations in STAT3, decrease the interaction of STAT3 with
p300 and HDAC1 (Hou et al., 2008; Ray et al., 2008). Other than acetylation, STAT3
can be methylated by SET9, when it binds to the promoter (Yang et al., 2010).
Moreover, STAT3 can form complex with DNA methyltransferase (DNMT) 1 and
8
HDAC 1 to silence SHP-1, an tumor suppressor gene in malignant T cells (Zhang et
al., 2005a). The acetylation of STAT3 at K685 is important for interaction with
DNMT1 to inhibit transcription of tumor suppressors (Lee et al., 2012). These results
indicated that post-translational modification of STAT3 may affect the gene
expression through the recruitment of histone modifying-enzymes.
1.5 Negative regulators of STAT signaling pathway
The signaling of STATs are not only regulated by post-translational
modifications, but also tightly controlled by several negative regulators, such as
protein tyrosine phosphotase (PTP), suppressors of cytokine signaling (SOCSs), and
protein inhibitor of activated STAT (PIASs) families. SOCS family is induced by
activated STATs, resulting in termination of STAT signals. IL-10 can inhibit the
activity of pro- inflammatory cytokines like IFN through induced SOCS3. PIAS can
interact with STATs, and inhibit STAT-mediated gene induction by a distinct
mechanism. For instance, DNA-binding activity of STAT1 and STAT3 is inhibited by
PIAS1 and PIAS3 (Chung et al., 1997; Liao et al., 2000). PTPs can also inactivated
STATs in either nucleus or cytoplasma by removing phosphate group from activated
STATs. It has been identified that TC45 and SHP2 (PTP) can inactivate STAT1
through dephosphorylation in nucleus (Shuai and Liu, 2003). However, a STAT
9
protein can also suppress other members of the STAT family. For example, activation
of STAT1 is prolonged in STAT3-deficient MEFs by IL-6 (Costa-Pereira et al., 2002).
STAT3 negatively regulates STAT1-dependent gene expression in IFN treatment (Ho
and Ivashkiv, 2006; Wang et al., 2011). Nevertheless, the detailed mechanism remains
unclear.
10 1.6 Rationales
Type I IFN-stimulation leads to activation of STAT1, STAT2, and STAT3, which
induces expression of different downstream genes. Besides, STAT3 not only regulates
gene expression, but also negatively regulates type I IFN-mediated response and
tumor suppressor genes (Ho and Ivashkiv, 2006; Lee et al., 2012; Wang et al., 2011).
In this study, we want to investigate the mechanism how STAT3 suppresses type I
IFN-mediated response. We have previously shown that STAT3 directly reduce type I
IFN-mediated gene expression. Hence, we want to address whether STAT3 may affect
phosphorylation, nuclear translocation, or DNA-binding ability of STAT1. Interstingly,
NTD of STAT3 is sufficient to suppress IFN-α-mediated gene induction. Acetylation
of STAT3 at NTD is critical for its function and interaction with proteins. Furthermore,
it has reported that STAT3 could silence gene expression through
acetylation-dependent mechanism. From these results, we hypothesize that acetylation
at NTD of STAT3 may influence the suppressive effect of type I IFN-mediated
response.
11
Chapter II Materials and methods
2.1 Cells
WT mouse embryonic fibroblast (MEF) and STAT3 knockout (STAT3KO) MEF
cell lines were obtained from Dr. Levy’s laboratory at New York University. STAT1
and STAT3 double knockout (DKO) MEF cell lines were generated previously by
using retroviral transduction of a vector encoding Cre recombinase into
STAT1-/-STAT3flox/flox cells. All cells were cultured in DMEM (Gibco) supplemented
with 10 % fetal bovine serum (Hyclone), and 10 ng/ml gentamicin (Gibco).
2.2 Plasmids
pLPC-FH2-mSTAT3 was contructed using the following primers containg
BamHI and EcoRI site. Mouse STAT1 was PCR amplified using the following
primers. pLPC- FH2-mSTAT1 was subcloned by XhoI site.
(1) STAT3
Forward: 5'- CGGGATCCGCTCAGTGGAACCAG -3'
Reverse: 5'- GCGAATTCCCATGGGGGAGGTAGC-3'
12 (2) STAT1
Forward: 5'- CCGCTCGAGATGTCACAGTGGTTCGAG -3'
Reverse: 5'- CCGCTCGAGTACTGTGCTCATCATACTGTC -3'
2.2 Calcium phosphate precipitation transfection
Plasmid DNA was mixing with 250 mM CaCl2, followed by adding 2X BBS
drop-wise and then transferred into medium. After incubation for 6 hours at 37 , ℃
medium was refreshed.
2.3 Retroviral transduction
The retroviral bicistronic vector pLPC-FH2 plasmid encoding WT STAT3 or WT
STAT1 and puromycin resistant gene, respectively was cotransfected with a helper
plasmid (pCL-Eco) into Phoenix A, amphotropic packaging cell line, or HEK 293T
cells using calcium phosphate precipitation method. After transfection for two days,
the culture supernatant containing pseudo-typed virus was collected. MEFs were
incubated with viral supernatant in the presence of 8 μg/ml polybrene and spun at
1100 xg for 45 minutes at room temperature. Two days later, cells were treated with
puromycin to select the drug-resistant transfetants .
13 2.4 RT-QPCR
Total RNA was prepared from MEFs using TRIzol (Invitrogen) or TRIsure
(Bioline #Bio-38032) reagent. 1-3 μg of RNA was subjected to reverse transcription
(RT) with oligo dT and then cDNA prepared from the reaction was then subjected to
QPCR by iCycler IQ (Bio-rad) using the following primer sets.
(1) PKR
Forward: 5'- TGCGCAGACAATGAATGGTA -3'
Reverse: 5'- ATGTGACAACGATAGAGGAT-3'
(2) IP-10
Forward: 5'- TGAGCAGAGATGTCTGAATCCG -3'
Reverse: 5'- TGTCCATCCATCGCAGCA -3'
(3) IRF-7
Forward: 5'- AGCAAGACCGTGTTTACGAC -3'
Reverse: 5'- AGTGCTGAAGTCGAAGATGG -3'
(4) IFIT1
Forward: 5'- AGAGCAGAGAGTCAAGGCAGGT -3'
Reverse: 5'- TGGTCACCATCAGCATTCTCTCCCA -3'
14 (5) IFIT2
Forward: 5'- ATTGCGAACTACCGTCTG -3'
Reverse: 5'- CTTCAGTGCTAAGAGGAC -3'
(6) Socs3
Forward: 5'- ATGGTCACCCACAGCAAGTTT -3'
Reverse: 5'- TCCAGTAGAATCCGCTCTCCT -3'
(7) JunB
Forward: 5'- TCACGACGACTCTTACGCAG -3'
Reverse: 5'- CCTTGAGACCCCGATAGGGA -3'
(8) β-actin
Forward: 5'-GTGGGGCGCCCCAGGCACCA -3'
Reverse: 5'-CTCCTTACCGTCACGCACGATTT -3'
2.5 Western blot
Whole cell lysates are extracted by lysis buffer at 4 for 15 minutes. Lysates℃
were cleared of debris by centrifugation at 14,000 rpm (Eppendorf) for 15 minutes.
Equal amounts of samples were resolved in 7 % or 10 % sodium dodecyl
sulfate-polyacrylamide gels (SDS-PAGE), followed by transfering to polyvinylidene
difluoride membranes (PVDF) and immunoblotting with the indicated antibodies.
15 2.6 Cytosolic and nuclear extracts
Cells were washed twice with 1XPBS, and then scraped off of the dish and the
cell pellets were obtained by centrifugation (300 xg, 5 minutes, 4℃). Cells was
resuspended in RSB-G40 buffer and nuclei were centrifugated at 10,000 xg to obtain
the cytosolic supernatant. Nuclear extraction was obtained from resuspend nuclei by
using nuclear extraction buffer.
2.7 Co-immunopreciptation (CoIP)
Equal amount of whole cell lysetes were immunopreciptated using anti-HA
antibodies in cell lysis buffer, which was conjugated with protein G beads (Millipore).
Western blotting was performed as described in 2.5 using the indicated antibodies.
2.8 Chromatin immunoprecipitation (ChIP)
MEFs were stimulated with IFN-α4 for 30 minutes, and then fixed in 1.4 %
formaldehyde for 15 minutes at room temperature. Cells were lysed with
immunoprecipitation buffer prior to sonication (Nelson et al., 2006). Chromatins were
sheared by sonication with a Vibra-Cell VCX 130 sonicator (Sonics & Materials).
Cells were sonicated using 1 sec on/ 1 sec off pulses for 5 min at 70% power output to
shear the DNA to ~200 to 500 base pairs. Protein G-Sepharose beads (Millipore) were
16
added to cell lysates preincubated with corresponding antibody overnight at 4°C.
After extensive washes, bead-bound DNA was reverse-crosslinked by incubation
overnight at 67°C. Protein were removed by incubating with 20 μg proteinase K in
proteinase K buffer at 55°C for 4 hours. Recovered DNA from ChIP was analyzed by
QPCR using primers specific for corresponding ISRE elements in the promoters of
the indicated genes. Primer sequences are as follows:
(1) ISRE of IFIT1 promoter
Forward: 5'- GTGGAGAATGCAGTAGGGCAAAC -3'
Reverse: 5'- GTCACACCAACTGGAAGCTCAGG -3'
(2) ISRE of MDA5 promoter
Forward: 5'-ACCAAAGTCCTCACCTAAC-3'
Reverse: 5'-TATTGCCTTCCACCCAC-3'
2.9 In vitro antiviral state assay
MEF cells were pretreated with or without 2-fold serial dilution of IFN-α4 from
240 IU/ml for 24 hours. Cells were infected with EMCV at am MOI of 0.1. After
infection for 18 hours, the medium was removed and cells were fixed with 10 %
formaldehyde for 20 minutes at RT. After fixation, cells were visualized with crystal
violet. The excessive dye was then removed by immersing the plate in water.
17
2.10 Single primer based site-directed mutagenesis
pLPC-FH2-mSTAT3 was subjected to site-directed mutagenesis using one
primer in a single cloning step (Makarova et al., 2000). After PCR, the construct was
digested by DpnI twice to remove the parental unmutated strand DNA and then
transformed into Ecoli.. DNA sequencing was performed to confirm the mutation
sites.
The primers used for mutagenesis are as follows
(1) K49Q: GGCATATGCAGCCAGCCAAGAGTCACATGCCAC Tm= 60 ℃
(2) K87Q: CAACCTTCGAAGAATCCAGCAGTTTCTGCAGAG Tm= 60 ℃
(3) K49R: GCATATGCAGCCAGCAGAGAGTCACATGCCACG Tm= 60 ℃
(4) K87R: AACCTTCGAAGAATCAGGCAGTTTCTGCAGAGC Tm= 58 ℃
(5) K685R: GAGGAGGCATTTGGAAGGTACTGTAGGCCCGAG Tm= 56 ℃
PCR conditions are as follows
95 , 5 min. 98 ℃ ℃, 20 sec. Tm ℃, 20 sec. 72 ℃ 4 min. 72 ℃, 10 min.
2.11 Statistics
A student’s T test (two-tailed) was performed for statistical analysis.
18
Chapter III Results
3.1 STAT3 negatively regulates type I IFN-mediated response
Using STAT3KO MEFs, we have previously reported that STAT3 could suppress
type I IFN-mediated antiviral response (Wang et al., 2011). Since WT MEFs and
STAT3KO MEFs were generated from different mice, it is likely that the epigenetic
modifications are different, which may have additional effects independent of STAT3.
In this study a different approach was taken, STAT1 and STAT3 double knockout
(DKO) MEFs were restored with STAT1, STAT3, or both molecules. In principle, the
genetic makeup of different STAT-restored DKO MEFs should be similar. DKO
MEFs restored with empty vector (EV), Flag- and HA-tagged STAT1 or STAT3, or
both molecular (Fig. 1) with retroviral transduction were stimulated with IFN-α4 for
30 minutes (Fig. 2). STAT1 or STAT3 was expressed in WT, STAT1-restored,
STAT3-restored, and STAT1/3-restored DKO MEFs, and all of their STAT1 or STAT3
were phosphorylated in response to IFN-α4. As expected, STAT1 and STAT3 can also
be detected by anti-Flag and anti-HA antibodies in the restored cells. More
importantly, the phosphorylation of STAT1 and STAT3 in the restored cells was
comparable to that in WT MEFs. Interestingly, phosphorylation of STAT2 was
increased in cells restored with STAT1. These results suggested that DKO MEFs had
19
been successfully restored with Flag- and HA-tagged STAT1 and/or STAT3. We next
examined the functions of the restored STAT1 and STAT3 in these cells. Expression
of STAT1- or STAT3-downstream genes was measured by RT-QPCR. After IFN-α4
stimulation for 1 hour and 2 hours, Socs3 and JunB, two STAT3-dependent genes,
were upregulated, respectively, in STAT3- and STAT1/3-restored DKO MEFs (Fig. 3).
ISGs, including PKR, IP-10, IRF7, IFIT1, and IFIT2, were induced in STAT1 and
STAT1/3-restored DKO MEFs. Consistent with previous studies, the expressions of
ISGs were decreased in the STAT1/3-restored DKO MEFs as compared to
STAT1-restored DKO MEFs (Fig. 4). Besides, we also perform microarray to do
whole gene profiling, and found that there were 137 type I IFN-induced genes, such
as STAT2, IRF1, OAS2, and MX1, displaying the same phenotype (Fig. 5). The
scatterplot, which showed relationship of gene expression, showed that it a good
approach to do the experiment (Fig.6 and Fig. 7A). It revealed that STAT3 could exert
suppressive effect in STAT1/3-induced DKO MEFs as compared with STAT1-restored
DKO MEFs (Fig. 7B). Furthermore, we used antiviral state assay to confirm negative
effect of STAT3 on type I IFN-mediated antiviral response. As shown in Fig. 8 EV
and STAT3-restored DKO MEFs were susceptible to EMCV infection, which was due
to the absence of STAT1. STAT1- and STAT1/3-restored DKO MEFs were resistant to
EMCV infection. Nevertheless, single restoration of STAT1 showed more resistant to
20
EMCV infection than did STAT1/3 double restoration in DKO MEFs. These results
indicate that the phenotype of STAT1- and STAT3-restored DKO MEFs indeed
mimicked the phenotypes seen in STAT3KO and WT MEFs, respectively.
3.2 Suppression of type I IFN response by STAT3 is independent of phosphorylation and nuclear translocation of STAT1 or STAT2
Negative regulation of type I IFN by STAT3 may due to competition for receptor
occupancy to decrease the phosphorylation of STAT1 or STAT2. To examine this
possibility, STAT-restored DKO MEFs were treated with IFN-α4 for different times
and activation of STAT1 and STAT2 was assessed. As shown in Fig. 9,
phosphorylation of STAT1 was transient and decreased in the time-dependent manner,
and the activation of STAT2 was prolonged and remained, at least, for 18 hours.
However, STAT1/3-restored DKO MEFs failed to alter IFN-α-induced activation of
STAT1 and STAT2 as compared to STAT1-restored DKO MEFs. These results
suggested that STAT3 did not affect kinetics of phosphorylation of STAT1 or STAT2,
and it also implied that STAT3 does not compete receptor occupancy with STAT1 or
STAT2. We next investigate whether STAT3 influences nuclear translocation of
STAT1 or STAT2. After IFN-α4 treatment, the level of nuclear STAT1 or STAT2 was
comparable in STAT1- and STAT1/3-restored DKO MEFs, suggesting that STAT3 did
21
not alter nuclear translocation of activated STAT1 and STAT2 following stimulation
(Fig. 10).
3.3 STAT3 suppresses type I IFN-mediated response through blocking the recruitment of ISGF3 complex to ISRE in the ISGs promoters
Since STAT3 did alter phosphorylation or nuclear translocation of STAT1 and
STAT2, we next examined whether STAT3 influenced the binding of ISGF3 complex
binding to ISRE containing promoter using ChIP assay. After IFN-α4 stimulation,
STAT1-containing ISGF3 complex was recruited to the ISRE of IFIT1 and MDA5 in
STAT1-restored DKO MEFs. However, the binding of STAT1 on ISRE site of IFIT1
and MDA5 was reduced in STAT1/3-restored DKO MEF as compared to
STAT1-restored DKO MEFs (Fig. 11), suggesting that STAT3 attenuated ISGF3
complex binding and/or recruitment to the promoter of ISRE of ISGs.
3.4 STAT3 negatively regulates type I IFN-induced gene expression through acetylation-dependent mechanism by HDAC inhibitor
It has been reported that the interaction of STAT1 with p300 can increase the
binding ISGF3 complex to ISRE (Zhang et al., 2005b). However, whether STAT1 and
binding ISGF3 complex to ISRE (Zhang et al., 2005b). However, whether STAT1 and