The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Goenka, S.
Right arrow Articles by Boothby, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Goenka, S.
Right arrow Articles by Boothby, M.
The Journal of Immunology, 1999, 163: 4663-4672.
Copyright © 1999 by The American Association of Immunologists

Paired Stat6 C-Terminal Transcription Activation Domains Required Both for Inhibition of an IFN-Responsive Promoter and Trans-Activation1

Shreevrat Goenka*, Jeehee Youn*, Linda M. Dzurek*, Ulrike Schindler{ddagger}, Li-yuan Yu-Lee§,¶ and Mark Boothby2,*,{dagger}

Departments of * Microbiology and Immunology and {dagger} Medicine (Rheumatology), Vanderbilt University, Nashville, TN 37232; {ddagger} Tularik, South San Francisco, CA 94080; and Departments of § Cell Biology and Medicine (Rheumatology), Baylor College of Medicine, Houston, TX 77030


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The cytokines IL-4 and IFN-{gamma} exert biologically antagonistic effects that in part reflect opposing influences on gene transcription. While the molecular mechanisms for IL-4-mediated transcription activation have been extensively studied, little is known about molecular mechanisms required for IL-4 inhibition of IFN-{gamma} signaling. We have investigated IL-4 inhibition of the IFN-{gamma}-inducible promoter for IFN regulatory factor-1 (IRF-1). In a cell line with low endogenous Stat6, increasing levels of activated Stat6 at constant doses of IFN-{gamma} and IL-4 leads to inhibition of the IRF-1 promoter. The Stat1-dependent IFN-{gamma} activation sequence element of the IRF-1 promoter is a target for Stat6-mediated inhibition despite apparently normal Stat1 DNA binding. However, our data are inconsistent with competition between Stat1 and Stat6 for access to the IRF-1 IFN-{gamma} activation sequence or for an essential coactivator as a mechanism for this Stat6-mediated inhibition. Instead, the data demonstrate that a threshold of Stat6 transcription activation domains is required for IL-4-dependent inhibition. The findings provide evidence of a novel mechanism in which the Stat6 transcription activation domains play a critical role in the IL-4-mediated inhibition of an IFN-{gamma}-inducible promoter.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cytokines exert crucial regulatory influences on immune responses and other aspects of biological function (reviewed in Refs. 1, 2, 3, 4). One role for these intercellular signaling molecules is to activate key developmental events, typically through induction of gene transcription. For example, IFN-{gamma} promotes the Th1 subset of effector T cells in a process dependent on the IFN-{gamma} target gene IFN regulatory factor 1 (IRF-1)3 as well as on other transcription factors (1, 5, 6, 7, 8, 9). In addition to these activating influences, normal biological regulation requires the integration of a multiplicity of inter- and intracellular signaling pathways. Thus, certain cytokines exert inhibitory influences on the actions of other cytokines. For instance, IL-4 inhibits the development of the Th1 subset of effector T cells (2, 5). Although, the basis for these reciprocal effects of cytokines is unclear, transcriptional antagonism between IL-4 and IFN-{gamma} has been amply documented (Refs. 10, 11, 12, 13, 14, 15, 16 ; reviewed in Ref. 17). IL-4-induced expression of class II MHC and CD23 Ags, as well as Ig class switching to the {epsilon} H chain, are repressed by IFN-{gamma} (10, 11, 12, 14, 18, 19). Similar transcriptional antagonism is mediated by IL-4 against IFN-{gamma}. Thus, IL-4 inhibits the induction of the IL-12Rß2 chain, IFN-{gamma}-induced high-affinity IgG receptor (Fc{gamma}RI), and IFN-{gamma}-enhanced Ig class switching to the C{gamma}2a H chain locus (13, 14, 15, 16, 17). Taken together, these findings suggest the existence of mechanisms by which IL-4 inhibits IFN-{gamma}-inducible genes.

Remarkably little is known about mechanisms by which one cytokine inhibits another or about the integration of cytokine signaling pathways such as IL-4 inhibition of IFN-{gamma}-induced transcription. In contrast, a considerable amount of information has emerged on the mechanisms by which cytokines activate gene transcription (3, 4, 20, 21, 22). IL-4 activates at least four distinct signaling pathways, each of which has the potential to influence transcription regulation. First, the Janus kinase (Jak)-Stat pathway is used by IL-4 for transcription activation (23, 24). IL-4 induces the Jak-mediated tyrosine phosphorylation of latent Stat6, which then dimerizes, translocates to the nucleus, and binds to specific DNA sequences so as to activate gene transcription. Second, the insulin-IL-4 receptor motif of IL-4 receptor {alpha}-chain (IL-4R{alpha}) activates a signaling pathway via phosphorylated insulin receptor substrate 1 and 2 (25, 26). These molecules signal a proliferative response through the activation of the phosphatidylinositol-3 kinase pathway and also may modulate IL-4-induced transcription activity (25, 26, 27). A second mechanism by which IL-4 can activate phosphatidylinositol-3 kinase is through tyrosine phosphorylation of the protein tyrosine kinase c-Fes, as c-Fes is tyrosine phosphorylated upon IL-4 engagement with its cognate receptor (28, 29). Two additional molecules, SH2 containing sequence (SHC) and IL-4 receptor interacting protein (FRIP), associate with the insulin-IL-4 receptor motif of IL-4R{alpha} and are thought to function as adaptors linking IL-4 signaling to the Ras/mitogen-activated protein kinase pathway (30). It is unclear which of these IL-4-activated pathways provides a basis for the cross-regulation of IFN-activated genes. Similarly, IFN-{gamma} activation of Jak tyrosine kinases and the Stat1 protein play central roles in IFN-induced gene transcription, but additional signaling pathways are activated by IFN-{gamma}, perhaps accounting for the inhibition of IFN-{gamma} by TGF-ß through a Jak-Stat-independent mechanism (31).

To investigate the mechanism by which IL-4 inhibits IFN-{gamma}-induced gene transcription, we have focussed on the promoter for IRF-1, a gene whose induction by IFN-{gamma} is reported to be inhibited by IL-4 (32). On the basis of titration experiments with IFN-{gamma} and in vitro DNA binding studies, it has been proposed that this IL-4-dependent inhibition is mediated by Stat6, and Stat6 competes for the Stat1-dependent IFN-{gamma}-activated sequence (GAS) element of the IRF-1 promoter. Using constant concentrations of IL-4 and IFN-{gamma} so as to avoid potential changes in the activation other signaling pathways, we present direct evidence that IL-4-induced inhibition of the IRF-1 promoter requires increasing levels of activated Stat6. Our data are inconsistent with the previous proposal that competition for DNA binding between Stat1 and Stat6 is the mechanism of inhibition. In contrast, we have found that a threshold of paired Stat6 transcription activation domains (TAD) needs to be recruited for IL-4 to exert an inhibitory effect. Thus, these findings represent evidence of a novel Stat6-mediated mechanism for IL-4 inhibition of IFN-{gamma}-inducible gene expression.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture

HepG2 and 293T cells were cultured in DMEM supplemented with 10% heat-inactivated FBS, 50 U/ml penicillin, 50 µg/ml streptomycin, 3 mM glutamine, and 1x MEM nonessential amino acids (Life Technologies) (DMEM/10F) at 37°C in a humidified CO2 incubator. HepG2 variants stably overexpressing Stat6 were isolated after transfection with linearized Stat6-pcDNA3 (33) and selection in G418 (1 mg/ml) (Life Technologies) containing DMEM/10F.

Generation of Stat6{Delta}C chimeric constructs

The Stat6{Delta}C-SD1, Stat6{Delta}C-SD2, Stat6{Delta}C-SD[1+1], Stat6{Delta}C-SD[2+2] and Stat6{Delta}C-SD[1+2] expression constructs were generated by subcloning PCR products corresponding to amino acids 661–715 (domain 1) and 753–810 (domain 2) of Stat6 into Stat6{Delta}C-pcDNA3 (33). The inframe insertion of the PCR products downstream of Stat6{Delta}C was confirmed by DNA sequencing.

Transient transfection and promoter assays

HepG2 cells (106/60 mm dish) were transiently transfected using the SuperFect reagent (Qiagen, Chatsworth, CA) according to the manufacturer’s protocol. Briefly, DNA-SuperFect complexes were allowed to form in serum-free media for 10 min, then complete media was added to this transfection mixture and was then used to transfect cells for 3 h. The indicated expression constructs (0–8 µg), along with either the IRF-1-chloramphenicol acetyltransferase (CAT) (34), the (GAS)3-thymidine kinase (TK)-CAT (35), pBLCAT2 (36), or CCAAT/enhancer binding protein (C/EBP)-N4-TK-Luc (33) promoter-reporter chimeric plasmid constructs (2 µg), were used for each transfection. Expression vector DNAs (0–8 µg) were included so that the total DNA in each transfection was the same. After transfection, the cells were washed, incubated overnight in complete DMEM, then divided equally and stimulated overnight with recombinant human IL-4, (10 ng/ml), recombinant human IFN-{gamma} (Sigma, St. Louis, MO), (1 U/ml), or both IL-4 and IFN-{gamma} as indicated. The cells were then harvested and lysed by repeated freeze-thaw cycles in 0.1 M Tris, pH 7.5. For CAT assays the resultant lysate was incubated at 56°C for 10 min, clarified by centrifugation, and assayed using [3H]acetyl CoA (0.05 µCi) and chloramphenicol (1 mM) in 0.1 M Tris-Cl, pH 7.5. CAT activity was calculated by counting the tritiated acetylchloramphenicol product soluble in the organic phase created by an overlay of water-immiscible scintillation fluor (Econofluor-Packard, Meriden, CT) and expressing these counts as a percent of the total counts (39). For luciferase assays, the cell lysates were mixed with luciferase assay substrate (Promega, Madison, WI) and promoter activity was determined according to manufacturer’s protocol.

EMSA

Mobility shift assays were performed on whole-cell and nuclear extracts of HepG2 cells transfected with the indicated expression plasmids. Cells were transfected using the SuperFect reagent as described above, divided equally, and induced with IL-4 and IFN-{gamma} as indicated. For whole-cell extracts, the cells were lysed at 4°C using 0.5% Nonidet P-40 in 0.01 M Tris-Cl, pH 7.4, supplemented with 150 mM NaCl, 50 mM NaF, 1 mM DTT, 0.1 mM sodium-vanadate, 0.4 mM PMSF, 5 µg/ml aprotinin, and 1 µg/ml leupeptin, followed by pelleting insoluble materials (27, 40). For nuclear extracts, cells were lysed with 0.5% Nonidet P-40 in 0.01 M Tris-Cl, pH 7.4, the nuclei were collected by centrifugation, and nuclear proteins were extracted using 0.42 M NaCl in 0.01 M Tris-Cl, pH 7.4, supplemented with 50 mM NaF, 1 mM DTT, 0.1 mM sodium-vanadate, 0.4 mM PMSF, 5 µg/ml aprotinin, and 1 µg/ml leupeptin (27, 40). DNA binding reactions were performed using 5 µg of protein in 10-µl reactions containing 1 µg poly(dI-dC) competitor and 32P-labeled oligonucleotide, essentially as described (27, 37). The double-stranded oligonucleotide probe contained the "N3" GAS from the IRF-1 promoter, 5'-CCTGATTTCCCCGAAATGAT (stat recognition palindrome is underlined), and its complement (41). DNA-protein complexes were resolved on a 4.5% nondenaturing polyacrylamide gel in Tris-borate-EDTA buffer as described (37). Competition experiments with unlabeled oligonucleotides were performed by mixing nuclear extract with 10 ng of the indicated unlabeled double-stranded oligonucleotide followed by performance of mobility shift reactions as described above. The N4 binding site oligonucleotide, 5'-CAACTTCCCAAGAACAGA, is derived from the mouse Ig H chain germline {epsilon} promoter (37). To identify DNA-protein complexes as Stat1 or Stat6 by "supershift" analyses, nuclear extracts were incubated with 1 µg of Abs specific for Stat1 (Transduction Labs, Lexington, KY) or Stat6 (Santa Cruz Biotechnology, Santa Cruz, CA) for 30 min at 4°C followed by addition of radiolabeled IRF-1 GAS probe and nondenaturing gel electrophoresis.

Western blot analysis

HepG2 or 293T cells were transiently transfected with the indicated Stat6 expression plasmids. After treatment with IL-4, nuclear extracts were prepared as described above. Nuclear proteins (40 µg) were resolved on SDS-PAGE and blotted onto nitrocellulose membranes. These transferred proteins were detected using Abs against Stat6 (rabbit polyclonal against amino acids 280–480 of Stat6) (Santa Cruz Biotechnology) followed by anti-rabbit IgG conjugated with HRP, and then visualized using enhanced chemiluminescence reagent (Amersham, Arlington Heights, IL) (40).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
IL-4 inhibition of IRF-1 is dependent on Stat6 activation

Certain IFN-{gamma}-dependent responses are enhanced in IL-4-deficient mice, and a number of specific target genes induced by IFN-{gamma} or other stimuli have previously been shown to be subject to inhibition by IL-4 (13, 14, 15, 16, 42, 43, 44). The IRF-1 gene is IFN-{gamma} inducible, involved in the differentiation of Th function, and has a well-characterized promoter that mimics regulation of the endogenous gene (36, 41, 45, 46, 47). Moreover, IL-4 inhibits the induction of IRF-1 by IFN-{gamma}, a finding consistent with observations using primary spleen cells (Ref. 32 , and S. Goenka and M. Boothby, unpublished observations). Both IL-4 and IFN-{gamma} activate multiple signaling pathways. Of these, the Stat pathway is demonstrated to be important in transcription activation (21, 22), but its role in transcription inhibition is not clear. To dissect the molecular mechanism(s) by which IL-4 inhibits IFN-{gamma}-induced transcription, it was necessary to identify an IL-4, IFN-{gamma}-responsive cell line with relatively low levels of endogenous Stat6 so that Stat6 dose-response experiments could be performed at constant doses of IL-4 and IFN-{gamma} to maintain other signaling by these receptors constant.

The HepG2 cell line is responsive to both IL-4 and IFN-{gamma}, yet offered the advantage of low endogenous levels of Stat6 (33) and modest levels of Stat1 (as assessed by mobility shift assays and by virtue of a 4-fold increase in IFN-{gamma} inducibility in cells transiently transfected with a Stat1{alpha} expression construct (S. Goenka, unpublished observations)). Therefore, in transient transfection experiments to determine the effect of Stat6 on the IRF-1 promoter independent of other IL-4-activated signaling pathways, the level of Stat6 was varied with constant IL-4. A 1.7-kb IRF-1 promoter linked to the CAT reporter gene (34) was transfected into HepG2 cells along with increasing amounts of a Stat6 expression construct, and the resultant CAT activity was determined after stimulation with IL-4, IFN-{gamma}, and IL-4 with IFN-{gamma}. A 25-fold induction of the IRF-1 promoter was observed in the presence of IFN-{gamma}, and IL-4 treatment alone did not induce the promoter. Cotransfection of increasing amounts of a wild-type Stat6 expression vector had no effect on IFN-{gamma} induction of the IRF-1 promoter in the absence of IL-4R{alpha} engagement. Moreover, IL-4 exerted no inhibitory effect when Stat6 was present at the low endogenous levels characteristic of HepG2 cells. In sharp contrast, IFN-{gamma} inducibility of the IRF-1 promoter was progressively inhibited when these increasing amounts of cotransfected Stat6 were activated by IL-4. IRF-1 promoter induction was reduced to 0.25x control at the maximum level of inhibition that could be achieved without toxicity to the transfected cells (Fig. 1Go). These data indicate that activation of sufficient Stat6 is required for inhibition of the IRF-1 promoter at a constant concentration of IFN-{gamma} and IL-4. Taken together, these findings indicate that a threshold of Stat6 is required for it to mediate its inhibitory effect, as the endogenous level of Stat6 in HepG2 cells is not sufficient to decrease the IFN-{gamma} inducibility of the IRF-1 promoter maximally.



View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 1. IRF-1 induction by IFN-{gamma} is inhibited by IL-4. IL-4-activated Stat6 inhibits IFN-{gamma} induction of the IRF-1 promoter. HepG2 cells were cotransfected with a construct containing a 1.7-kb IRF-1 promoter fragment linked to the CAT reporter gene (2 µg) (34 ), and increasing amounts of a Stat6 expression plasmid, pcDNA3-Stat6 (33 ). The total amount of DNA transfected was kept constant using appropriate amounts of empty expression vector, pcDNA3. After transfection, the cells were stimulated 16 h with nothing, IL-4 (10 ng/ml), IFN-{gamma} (1 U/ml), or both IL-4 and IFN-{gamma}, and CAT assays were performed using extracts of the treated cells. The data plotted represent the mean (±SEM) of three independent experiments.

 
Stat6 inhibition via the Stat1-binding GAS motif

To determine whether the Stat1 binding site in the IRF-1 promoter is a target of Stat6-mediated inhibition, we performed similar transient transfection experiments using a trimer of this Stat1 binding site (also termed GAS) linked to the basal TK promoter in the construct (GAS)3-TK-CAT (35). Cotransfection of this construct into HepG2 cells along with increasing amounts of a Stat6 expression construct led to Stat6 dose-dependent inhibition of (GAS)3-TK-CAT induction by IFN-{gamma}, but only when the Stat6 was activated by IL-4 (Fig. 2Go). Due to high basal activity of the (GAS)3-TK-CAT, the IFN-{gamma} inducibility was 5-fold over basal values for this promoter-reporter construct as compared with 25x basal for the full-length IRF-1 promoter. In contrast, the minimal TK promoter with no linked GAS was not IFN-{gamma} inducible, and no effect of activated Stat6 on this latter promoter was observed (data not shown). Taken together, these findings indicate that whereas Stat1 activated by IFN-{gamma} binds to the IRF-1 GAS to induce transcription (34, 45, 46), a GAS element is a sufficient target for Stat6-dependent inhibition by IL-4.



View larger version (24K):
[in this window]
[in a new window]
  		FIGURE 2. The Stat1-dependent GAS element of the IRF-1 promoter is a sufficient target for the Stat6-mediated inhibition. A plasmid containing three copies of the IRF-1 GAS element along with the TK promoter and the CAT reporter, (GAS)3-TK-CAT (35 ) (2 µg), was cotransfected into HepG2 cells with a constant mass of additional DNA (8 µg) representing a mixture of empty expression vector and a Stat6 expression vector (pcDNA3-Stat6, 0–8 µg). After stimulation with IL-4, IFN-{gamma}, or both IL-4 and IFN-{gamma} as indicated, the transfected cells were harvested and CAT assays were performed. The mean values (±SEM) from three independent experiments are plotted.

 
Normal nuclear induction of Stat1 DNA binding activity in IL-4-treated cells

The above results indicate that IL-4 inhibition of IFN-{gamma} responsiveness can be mediated through a Stat1 binding site. (Assay of a 1.7-kb IRF-1 promoter mutant lacking the GAS site detected no IFN inducibility, so the effect of IL-4 on this mutant promoter could not be determined.) One mechanism by which such inhibition could be achieved in HepG2 cells would be through decreased Stat1 activation and DNA binding. Therefore, we performed EMSA to determine whether IL-4 blockade of Stat1 DNA binding activated by IFN-{gamma} is the molecular mechanism for inhibition of the IRF-1 promoter. To mimic the transcription assays, nuclear extracts from cells transiently transfected with the Stat6 expression construct and treated with IL-4, IFN-{gamma}, or both IL-4 and IFN-{gamma} were analyzed in mobility shift experiments using the GAS element of the IRF-1 promoter. No difference in Stat1 DNA binding activity was observed when extracts from IFN-{gamma}-induced cells were compared with those from portions of the same population treated with IL-4 and IFN-{gamma} (Fig. 3GoA). IFN-{gamma}-induced levels of Stat1 were unaffected by IL-4 treatment even when transfection of increasing amounts of the Stat6 expression construct led to higher levels of nuclear Stat6 (data not shown). Consistent with published results in a different cell line (32), these findings indicate that IL-4 and Stat6 activation did not inhibit nuclear translocation of Stat1 or its GAS DNA binding activity.



View larger version (48K):
[in this window]
[in a new window]
  		FIGURE 3. Stat1 DNA binding in the presence of activated Stat6. HepG2 cells were cotransfected with expression vectors for overexpression of Stat6 (pcDNA3-Stat6, 2.5 µg) (33 ), then stimulated as indicated with IL-4 (10 ng/ml), IFN-{gamma} (1 U/ml), or both. A, Nuclear extracts prepared from these cells were subjected to mobility shift analyses using a 32P-labeled IRF-1 GAS probe (41 ). B, Where indicated, Abs against Stat1 (1 µg) or Stat6 (1 µg), or unlabeled double-stranded oligonucleotides (10 ng) containing the IRF-1 GAS or an "N4" Stat binding site (SBS) from the Ig H chain germline {epsilon} (G{epsilon}) promoter, were added to the reactions. (Some preparations of extracts generated a single Stat6 mobility shift band on initial assay but, when reassayed after several months storage, later generated a doublet (e.g., A vs B).) C, Nuclear extracts from two independent Stat6-overexpressing HepG2 clones were prepared after treatment with cytokine as indicated and subjected to EMSA with the IRF-1 GAS probe.

 
Both Stat1 and Stat6 bind efficiently to the GAS binding site but differ in their migration under nondenaturing conditions, such that Stat1 migrates faster than Stat6 when complexed with DNA. Thus, IL-4-treated cells generated increased levels of Stat6 DNA binding activity whereas extracts from IFN-{gamma}-induced cells led to a Stat1 band (Fig. 3GoA). The identity of these bands was confirmed by using oligonucleotide competitions and through Ab supershifting (Fig. 3GoB). Abs specific for Stat1 eliminated the IFN-{gamma}-induced DNA-protein complex through formation of a supershifted band comigrating with Stat6-DNA complexes (open circle). Similarly, Abs specific for Stat6 eliminated the weaker Stat6 mobility shifts and created a slower mobility band (closed circle). Consistent with these findings, unlabeled IRF-1 GAS competed for formation of each of the indicated complexes ("1" and "6"), whereas an oligonucleotide with a 4-bp spacer between its dyad repeats competed for the slower-migrating complexes (Stat6) but not for the Stat1 complex because only Stat6 binds to such N4 binding sites (48). Also consistent with these findings, we have observed that Stat1 DNA binding activity was normal in HepG2 variants stably transfected with Stat6 so as to express amounts of Stat6 DNA binding activity >=20-fold higher than the low endogenous level in HepG2 (Fig. 3GoC), and IL-4 inhibited the IRF-1 promoter in transfection experiments using these subcloned variants (data not shown). It was conceivable that these results were due to examination of cells that had only undergone short-term stimulation. However, we observed in time-course experiments that Stat1 activation wanes after several hours, consistent with previous observations in other cell lines (49). Periods of IL-4 pretreatment of up to 16 h led to a similar time-dependent decay in Stat6 activity but did not affect Stat1 DNA binding activity (data not shown). Therefore, no substantive decrease in Stat1 mobilization was apparent despite substantial inhibition of IRF-1 promoter activity. We conclude that when HepG2 cells are treated with IL-4 in addition to IFN-{gamma}, IRF-1 promoter inhibition occurs under conditions where Stat1 binding activity is normal.

Inhibition requires Stat6 TAD(s) of the C terminus

The ability of Stat6 both to bind to and exert an inhibitory effect at the IRF-1 GAS suggested that IL-4 might inhibit the IRF-1 promoter because Stat1 and Stat6 compete for access to the same DNA binding site. Consistent with this model, Stat6 inhibits NF-{kappa}B-mediated induction of the E-selectin promoter due to overlap of a Stat binding site and an NF-{kappa}B/Rel DNA element (50). To investigate if this mechanism accounts for IL-4 inhibition of the IRF-1 promoter, we compared the inhibitory potency of wild-type Stat6 with that of a Stat6 mutant (Stat6{Delta}C) that retains normal DNA binding activity but lacks a TAD. Each of these expression constructs was cotransfected into cells along with the full-length IRF-1 promoter CAT construct. Although wild-type Stat6 inhibited IFN-{gamma} inducibility of the IRF-1 promoter construct (Fig. 4GoB), activated Stat6{Delta}C was unable to mediate an inhibitory effect (Fig. 4GoA). Similar results were observed with the (GAS)3-TK-CAT plasmid (data not shown). However, the Stat6{Delta}C construct directed similar steady-state protein levels and binding activity specific for the IRF-1 GAS compared with wild-type Stat6 under these conditions (Fig. 5Go). Moreover, in vitro competition experiments using increasing concentrations of nonradiolabeled oligonucleotide containing the IRF-1 GAS showed that the relative affinities of Stat6 and Stat6{Delta}C for the IRF-1 GAS are similar (Fig. 5GoC), consistent with previously published data using the germline {epsilon} Stat6 binding site (33). We conclude from these data that although Stat6 can bind to the IRF-1 GAS in vitro and has the potential to interfere with Stat1 access to this critical cis-acting element, such a competition mechanism did not account for the observed inhibition of the IRF-1 promoter by IL-4 in HepG2 cells.



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 4. Failure of Stat6 competition with Stat1 at the IRF-1 GAS to provide an inhibitory effect in the absence of a transcription activation domain. Pools of HepG2 cells were cotransfected with either IRF-1-CAT along with increasing amounts of plasmid constructs leading to overexpression of a Stat6 mutant lacking the C-terminal TAD (pcDNA3-Stat6{Delta}C) (33 ) (A) or wild-type Stat6 (pcDNA3-Stat6) (B) and divided in four equal portions. Cells were then treated 16 h with IL-4, IFN-{gamma}, or IL-4 with IFN-{gamma} as indicated at t = 24 h posttransfection and assayed for CAT activity. The values plotted represent the mean (±SEM) from three independent experiments. (CAT activities in these experiments reflected lower apparent promoter activity than in the earlier experiments of Fig. 1Go without affecting the underlying effect of Stat6 on IRF-1 inducibility; hence, all comparisons to wild-type Stat6 represent assays performed at the same time on the same population of HepG2 cells.)

 


View larger version (25K):
[in this window]
[in a new window]
  		FIGURE 5. A Stat6 mutant lacking the C-terminal activation domain retains normal IRF-1 GAS binding activity. Pools of HepG2 cells were transfected with either a plasmid construct leading to overexpression of a Stat6 mutant lacking the C-terminal TAD (pcDNA3-Stat6({Delta}C)) or wild-type Stat6 (pcDNA3-Stat6) and divided into four equal portions. Cells were then treated 30 min with IL-4, IFN-{gamma}, or IL-4 with IFN-{gamma} as indicated, and mobility shift assays were performed with the resultant extracts (A). Nuclear extracts from untreated and IL-4-treated cells were subjected to immunoblotting and probed with Abs directed against Stat6 (aa 280–480) (B). Nuclear extracts from HepG2 cells transfected with Stat6 or Stat6{Delta}C and treated with IL-4 were incubated with 0.25 ng of 32P-labeled oligonucleotide containing the IRF-1 GAS and increasing amounts of cold competitor containing the same sequence. The protein-DNA complexes were resolved on nondenaturing PAGE (C). The relative binding of Stat6 and Stat6{Delta}C was calculated and plotted against the molar excess of cold competitor after the radioactive complexes were quantified by a PhosphorImager (D).

 
Paired C-terminal domains mediate transactivation and Stat6-dependent inhibition

To dissect further the role of the Stat6 in inhibition of IFN-{gamma} inducibility by IL-4, we determined the role of the Stat6 TAD. We identified two distinct TADs within the Stat6 C-terminal amino acids using screening in yeast. Fragments of Stat6 spanning amino acid residues 661–715 (domain 1) and 753–810 (domain 2) induced strong GAL4-dependent lacZ reporter activity when fused with the GAL4 DNA binding domain (data not shown). To investigate the IL-4-mediated transcriptional activity of these two regions, we created translational fusions of Stat6{Delta}C with one or two copies of each domain (Fig. 6GoA). An additional chimera, Stat6{Delta}C-SD[1+2], was created in which both domains were fused to Stat6{Delta}C. Each of these constructs was cotransfected into HepG2 cells along with reporter plasmid that consisted of a minimal TK promoter linked to four copies of the Stat6 and C/EBP binding sites from the Ig H chain germline {epsilon} locus and luciferase (C/EBP-N4-TK-Luc) (Fig. 6GoB). The transfectants were cultured with or without IL-4 followed by measurement of promoter activity. As previously reported, synergism between the C/EBP and Stat6 binding sites allowed transactivation by the low level of endogenous Stat6 (33). Overexpression of wild-type Stat6 provided further trans-activation, while Stat6{Delta}C served as a trans-dominant inhibitor of the wild-type Stat6. In contrast to their potent trans-activation at GAL4 binding sites, a single copy of neither domain 1 nor 2 was able to restore significant trans-activation function to Stat6{Delta}C; instead, these proteins inhibited the function of endogenous wild-type Stat6. However, almost complete recovery of wild-type Stat6-mediated transcription was observed with Stat6{Delta}C when domains 1 and 2 were linked in cis (Fig. 6GoB). To test if this result implied a requirement for functional cooperation of unique sequences in domain 1 and 2, we tested the transactivation function of Stat6{Delta}C constructs with two copies of domain 1 or of domain 2. A pair of either domains 1 or 2 was able to restore wild-type transactivation function to Stat6{Delta}C (Fig. 6GoB). To test if each construct was expressed and could be activated by IL-4 so as to enter the nucleus, nuclear extracts of cells transfected with the various Stat6{Delta}C chimeric constructs and treated with IL-4 were analyzed by Western blotting and mobility shift assays (Fig. 7Go, A and B). These experiments detected Stat DNA binding activity in cells transfected with the different Stat6 constructs and activated by IL-4 (Fig. 7GoB). Taken together, our data show that Stat6 transactivation is mediated by two distinct domains. A single iteration of either domain was insufficient to mediate Stat6-dependent transcription activation. In contrast, two copies of any combination of two activation sequences were necessary and sufficient for transactivation function.



View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 6. The Stat6 C-terminal TAD consists of two distinct domains that contribute to IL-4-induced transcription. Stat6, Stat6{Delta}C, and the indicated Stat6{Delta}C chimeras with domains 1 and 2 were cloned in pcDNA3 as diagrammed (A). These plasmid constructs were then cotransfected into HepG2 cells with C/EBP-N4-TK-Luc (33 ). IL-4 inducibility of the promoters mediated by the Stat6 constructs was determined from the ratio of reporter enzyme activities in extracts from IL-4-treated and control cells. The values plotted represent the mean (±SEM) from three independent experiments.

 


View larger version (36K):
[in this window]
[in a new window]
  		FIGURE 7. Paired Stat6 TADs are necessary and sufficient to restore its inhibitory effect on the IRF-1 promoter. Nuclear extracts from cells transfected with the Stat6 chimeras were used in immunoblotting experiments with anti-Stat6 (against aa 280–480) (A) and in mobility shift reactions with the IRF-1 GAS probe (B). The identity of these complexes as Stat6-related proteins was confirmed as in Fig. 3Go (data not shown). C, IRF-1-CAT was cotransfected into HepG2 cells with expression constructs containing the indicated Stat6 chimeras. After treatment of the transfectants with cytokines, CAT assays were performed to determine promoter activity. In each independent experiment, IL-4-dependent inhibition of the IRF-1 promoter was calculated based on IRF-1 inducibility in IL-4/IFN-{gamma}-cotreated cells as compared with cell treated with IFN-{gamma} alone. The bars represent mean (±SEM) data from three independent experiments.

 
To investigate if either of these two Stat6-derived TADs played a role in inhibition of the IRF-1 promoter, IL-4-induced antagonism mediated by full-length Stat6 was compared with that mediated by Stat6{Delta}C, Stat6{Delta}C-SD1, Stat6{Delta}C-SD2, Stat6{Delta}C-SD[1+1], Stat6{Delta}C-SD[2+2], or Stat6{Delta}C-SD[1+2]. Stat6{Delta}C failed to inhibit the IRF-1 promoter, and the fusions with either domain 1 or domain 2 alone were unable to inhibit the IRF-1 promoter (Fig. 7GoC). In contrast, when the two domains were linked in cis, or when either domain was dimerized, the wild-type activity of Stat6 was nearly restored so that transfection of the resultant molecule significantly decreased IRF-1 promoter induction (Fig. 7GoC). Taken together, these observations indicate that Stat6-mediated inhibition is not due to simple competition with Stat1 for binding at the GAS element and also confirm that the low level of endogenous Stat6 in HepG2 does not lead to inhibition. Instead, a threshold level of paired Stat6 TADs plays a critical role in the inhibitory process required for IL-4-induced antagonism of this IFN-{gamma}-dependent promoter.

Inhibition mediated by transcriptional activity rather than competition for a coactivator

The observed requirement for the transactivation domain of Stat6 to mediate its inhibitory effect suggests two distinct models. First, TADs are known to interact with essential coactivators that play a critical role in the integration of signaling pathways. Thus, Stat6 TADs might sequester limiting pools of a coactivator essential for Stat1 function. In this regard, cAMP response element binding protein (CREB) binding protein (CBP) is a nuclear coactivator essential for Stat1 activation of target promoters (51, 52). Based on a physical interaction between CBP and Stat6, it has been hypothesized recently that Stat6 may compete with Stat1 for CBP to inhibit IFN-{gamma}-induced expression of monokine induced by IFN-{gamma} (53). By analogous reasoning, Stat6 might mediate the inhibitory effect of IL-4 by sequestration of CBP, thereby inhibiting Stat1 activation of the IRF-1 GAS element. To test this model, we investigated whether overexpression of CBP was able to reverse Stat6-mediated inhibition of the IRF-1 promoter (Fig. 8Go). To confirm that the reported coactivation of Stat1 activity by CBP applies to a naturally occurring promoter, we transiently transfected cells with the IRF-1-CAT promoter-reporter construct and increasing amounts of a hemagglutinin-tagged CBP expression construct (54). A 2-fold increase in IFN-{gamma} inducibility of the IRF-1 promoter was observed when only endogenous Stat6 was present (Fig. 8GoA). In the same experiments, we cotransfected a Stat6 expression vector and increasing amounts of CBP to determine whether CBP would reverse Stat6-mediated inhibition of the IRF-1 promoter. In the absence of IL-4, the CBP-mediated coactivation of IFN-{gamma} inducibility was comparable in cells cotransfected with Stat6 to that obtained when only endogenous Stat6 was present (Fig. 8GoB). However, increasing amounts of CBP were unable to reverse IL-4-induced inhibition of IFN-{gamma} induction of the IRF-1 promoter when Stat6 was activated by IL-4 (Fig. 8GoB). The expression of CBP was confirmed by immunoblot analysis using an anti-hemagglutinin Ab (Fig. 8GoB). Moreover, CBP did not enhance inducibility of a Stat6-dependent promoter (data not shown) These findings indicate that CBP is not a sufficient target to explain the Stat6-mediated inhibition of the IRF-1 promoter.



View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 8. CBP, an essential coactivator for Stat1-dependent transactivation, is not a sufficient target for Stat6-mediated inhibition of the IRF-1 promoter. The 1.7-kb IRF-1-CAT promoter-reporter plasmid was cotransfected into HepG2 cells along with a constant mass (8 µg) of pRc/RSV plus pRc/RSV-CBP (54 ) (0–8 µg) and (A) pcDNA3 (2 µg) or (B) pcDNA3-Stat6 (2 µg). After overnight induction with IL-4, IFN-{gamma}, or IL-4 plus IFN-{gamma} as indicated, cell-free extracts were prepared and used in CAT assays. The mean values (±SEM) from three independent experiments are plotted.

 
Apart from CBP, one could hypothesize that Stat6 inhibits the IRF-1 promoter via the GAS element by sequestering some other coactivator of Stat1. Thus, a Stat6 DNA binding mutant with an intact TAD would be expected to retain the ability to compete for coactivators. Substitution of arginine 562 to leucine has been shown to generate a mutant form of Stat6 that is expressed normally in HepG2 cells but fails to bind DNA (33). Accordingly, wild-type Stat6 or the Stat6 DNA binding mutant, Stat6R->L (33), were cotransfected with the IRF-1-CAT construct. Cells transfected with Stat6R->L and induced with IL-4 and IFN-{gamma} showed no inhibition of IRF-1 promoter activity as compared with cells transfected with wild-type Stat6 (Fig. 9GoA). A similar result was obtained (S. Goenka and M. Boothby, unpublished observations) using a second DNA binding mutant, Stat6VVI (33). To test the nuclear localization of Stat6R->L, nuclear extracts from cells transfected with Stat6R->L and treated with IL-4 were immunoblotted with a Stat6-specific Ab. These experiments showed significant nuclear concentrations of the Stat6R->L mutant protein, which has a normal C terminus containing both TADs 1 and 2 (Fig. 9GoB). Taken together, these observations are inconsistent with the models in which inhibition of IFN-{gamma} transcription activation is due to Stat6 competing for a coactivator such as CBP.



View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 9. Stat6 DNA binding competence is required for inhibition of the IRF-1 promoter. Plasmids leading to overexpression of Stat6 or a Stat6 DNA binding mutant with amino acid substitution converting arginine 562 to leucine (Stat6R->L) (33 ) were cotransfected with the IRF-1-CAT construct into HepG2 cells; empty expression vector (pcDNA3) was transfected as a negative control. After overnight induction with IL-4, IFN-{gamma}, or IL-4 plus IFN-{gamma} as indicated, cell-free extracts were prepared and used in CAT assays; the mean values (±SEM) of three independent experiments are plotted (A). Nuclear translocation of Stat6 and Stat6R->L was assayed by immunoblotting experiments using nuclear extracts of IL-4-treated transfectants and an anti-Stat6 Ab (B).

 
The other model conforming with the requirement for a threshold of Stat6 TAD recruitment to mediate an inhibitory effect is that IL-4 activates the transcription of an inhibitor of the Stat1 pathway. In principle, experiments using the protein synthesis inhibitor cycloheximide might permit one to test the hypothesis that Stat6 induces the transcription of an inhibitor. In practice, we have found that the IRF-1 gene is de-repressed after cycloheximide treatment of HepG2 cells, so that IRF-1 transcript levels after cycloheximide equaled those induced by IFN-{gamma} (unpublished observations). This de-repression of IRF-1 by cycloheximide precluded the use of protein synthesis inhibition to test whether IL-4 induces new gene expression so as to inhibit IFN-{gamma}. As one alternate means of testing this model, a trans-dominant inhibitor of Stat6 was titrated against repressive amounts of Stat6 in transfection experiments with the IRF-1-CAT construct. Increasing amounts of Stat6{Delta}C attenuated the Stat6-mediated inhibition of the IRF-1 promoter (Fig. 10GoA), indicating that Stat6{Delta}C can inhibit the Stat6-mediated trans-activation of an inhibitor of the Stat1 pathway. If indeed IL-4 induces an inhibitor of IRF-1 promoter activation by a Stat6-dependent mechanism, then pretreatment of cells with IL-4 could decrease promoter activity at a time point after Stat6 activation has ended. Therefore, HepG2 cells were cotransfected with IRF-1-CAT and pcDNA3-Stat6, split, and pretreated overnight with nothing (control cells) or with IL-4. Control and IL-4 pretreated cells were then each split and further incubated with nothing, IL-4, IFN-{gamma}, or both for an additional 24 h. These experiments showed that IL-4 pretreatment decreased IRF-1 promoter activation (Fig. 10GoB). Thus, promoter activity induced in IL-4 pretreated cells by IFN-{gamma} alone was similar to that in cells that were not pretreated and were then exposed to both IFN-{gamma} and IL-4. Importantly, control experiments indicated that IL-4 induction of Stat6 DNA binding activity had ended at the time that pretreated cells were split and restimulated with IFN-{gamma} (Fig. 10GoC), and this inhibitory effect was not apparent in measurements of the competence of IFN-{gamma} to induce Stat1 (Fig. 10GoD; also cf. Ref. 32). Taken together, these results provide evidence of a mechanism in which IL-4 induces the transcription of a gene product that mediates inhibition of the IRF-1 promoter. In contrast, the data are inconsistent with Stat6 competition for binding to GAS DNA in the IRF-1 promoter (Fig. 4Go and 5Go) and with saturation of a limiting pool of coactivator protein such as CBP (Fig. 8Go and 9Go).



View larger version (24K):
[in this window]
[in a new window]
  		FIGURE 10. Role of trans-activation in IRF-1 promoter inhibition by IL-4. A, Attenuation of Stat6-mediated inhibition of the IRF-1 promoter using the trans-dominant inhibitor Stat6{Delta}C. HepG2 cells were cotransfected with the IRF-1-CAT plasmid, a constant repressive amount of pcDNA3-Stat6 (2 µg), and increasing amounts of a Stat6{Delta}C expression plasmid as indicated. Transfected cells were treated with cytokines as indicated, and promoter activity was determined by CAT assays. The values plotted represent the mean (±SEM) from three independent experiments. An inset panel plots the same data as the inhibitory effect of IL-4 on IFN-{gamma}-induced IRF-1 promoter activity, where the percent inhibition is calculated as [IFN alone - (IL4 with IFN)]/IFN alone at each indicated concentration of Stat6{Delta}C. The asterisks indicate that p < 0.01 for the difference between inhibition with no Stat6{Delta}C and that with 2 µg or 8 µg of Stat6{Delta}C cotransfected. B, Pretreatment of HepG2 with IL-4 decreases IRF-1 promoter activity. HepG2 cells cotransfected with IRF-1-CAT and pcDNA3-Stat6 were cultured overnight in either complete medium or media supplemented with IL-4. The cells were then split and treated 24 h with IL-4 and IFN-{gamma} as indicated. The mean results (±SEM) of CAT assays using cell extracts from three independent transfection experiments are plotted. C, IL-4-mediated activation of Stat6 in HepG2 is transient and resolves after 8 h. Nuclear extracts of HepG2 cells transfected with Stat6-pcDNA3 and treated with IL-4 for different time intervals were subjected to EMSA with a Stat6 binding probe. D, Stat1 DNA binding activity is not diminished by IL-4 pretreatment. HepG2 cells transfected with Stat6-pcDNA3 were pretreated overnight with IL-4 as indicated and then incubated for 30 min with IFN-{gamma}. Extracts from these cells were then used for DNA binding assay with the IRF-1 GAS probe.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
During differentiation and homeostasis, cells are exposed concurrently to a plethora of molecules effecting intercellular communication. Accordingly, it is important to understand the mechanisms by which different signals are integrated. IL-4 frequently antagonizes the actions of IFN-{gamma}, and this reciprocal antagonism extends to transcription regulation (13, 14, 15, 16, 17). The molecular mechanisms which account for IL-4 inhibition of the transcriptional activity induced by IFN-{gamma}, and for the integration of IFN-{gamma}- and IL-4-activated signaling pathways, are not well understood. A number of signaling pathways are thought to play distinct roles in IL-4-mediated responses, of which the Jak-Stat, insulin receptor substrate 1/2, and mitogen-activated protein kinase pathways have all been implicated in IL-4-mediated gene activation. The central findings of this study are that 1) a threshold of Stat6 is required for the IL-4-mediated inhibition of an IFN-{gamma}-responsive promoter, which was analyzed using fixed concentrations of cytokine; 2) paired TADs of Stat6 are essential for both the trans-activation function and the inhibition mediated by Stat6; 3) our data are inconsistent with previously hypothesized models of competition between Stat1 and Stat6 for either DNA binding or a coactivator as mechanisms for Stat6 mediated inhibition; however, 4) a requirement for paired TADs to mediate inhibition, and additional evidence with a trans-dominant inhibitor of Stat6, suggest that Stat6 activates the transcription of an inhibitor affecting the Stat1-dependent induction of IRF-1.

The data provide direct evidence for the involvement of Stat transcriptional activation in the integration of IL-4- and IFN-{gamma}-activated signaling pathways and for a novel inhibitory mechanism that involves the Stat6 TADs and has the potential to influence any promoter dependent on an N3 Stat binding site for its activation. Importantly, an additional level (threshold) of Stat6 C-terminal TADs was essential for inhibition, for which the endogenous Stat6 pool in HepG2 was insufficient. Because the IRF-1 transcription factor is essential for Th1 development, an attractive possibility is that IL-4-activated Stat6 may inhibit IFN-{gamma}-induced IRF-1 gene transcription and thus contribute to the regulated development of Th1 and Th2 subsets of effector T cells (5, 6, 7). The observed increases in development of Th1 cells in Stat6-deficient mice, and increased Th2 cells in IRF-1-deficient mice, support the importance of this Stat6-mediated inhibition of IFN-{gamma} signaling (6, 7, 55, 56). Of note, IL-4 also inhibits the IL-12 (Stat4)-dependent induction of another cytokine receptor, IL-12Rß2, thereby reinforcing the polarization of function during development of Th subsets (16, 57).

While there are few data on IL-4 inhibition of IFN-{gamma} signaling, some molecular mechanisms for negative regulation of Jak-Stat pathways or inhibitory influences of a Stat protein have been proposed. Competition between two transcription factors for the IRF-1 GAS represents a hypothetical mechanism for Stat6-mediated inhibition of Stat1 function. Stat6 has been implicated in the inhibition of TNF-{alpha}-stimulated, NF-{kappa}B-dependent transactivation of the E-selectin gene. Thus, Stat6 competes with NF-{kappa}B for access to a region within the E-selectin promoter containing overlapping Stat6 and NF-{kappa}B binding sites (50). Stat proteins bind to similar palindromic DNA (TTC-Nx-GAA) sequences with N2-N4 spacing. Thus, although Stat6 exclusively is able to bind to an N4 site, it also binds to the IRF-1 GAS, an N3 site at which Stat1 binding is essential for promoter function (34, 41). This competition for DNA binding between Stat1 and Stat6 led to the interpretation that it is the mechanism for the observed inhibition of IFN-{gamma} by IL-4 (32). In sharp contrast with previous conclusions, results with the Stat6 C-terminal deletion mutant Stat6{Delta}C demonstrate that a form of Stat6 equally able to compete for binding to the IRF-1 GAS was nonetheless incompetent to inhibit Stat-1-mediated promoter function (Fig. 4Go and 5Go above; Ref. 33). Moreover, Stat6 and Stat6{Delta}C have similar affinity for the IRF-1 GAS as determined by cold oligo competition experiments (Fig. 5GoC). It is known that C/EBP-ß enhances the affinity of Stat6 for DNA binding (58). Thus, it might be speculated that Stat6{Delta}C may have lower affinity for the IRF-1 GAS compared with Stat6, as the C terminal of Stat6 might be required for C/EBPß-mediated DNA binding. However, the presence of C/EBP sites in the IRF-1 promoter has not been demonstrated. Moreover, Stat6 and Stat6{Delta}C were equally able to cooperate with C/EBP in assays of the enhancement of binding to the composite C/EBP-Stat6 site of the {epsilon} H chain germline promoter (58). Thus, any effects of binding cooperativity in this system remain speculative. We have further shown that restoration of the two Stat6 TADs either as homo- or heterodimers to Stat6{Delta}C is essential for IL-4-mediated trans-activation and is also necessary and sufficient to restore the inhibitory activity to this mutant whereas a single domain is not. Taken together, these data are inconsistent with a model of direct competition for access to DNA to explain the observed Stat6-dependent inhibition through the IRF-1 GAS.

Direct inhibition at the level of DNA binding had been proposed but is rendered unlikely by the evidence as discussed above. An alternative model for IL-4 inhibition of monokine induced by IFN-{gamma} was later proposed based on the coimmunoprecipitation of Stat6 and CBP (53). It was hypothesized that Stat6 TADs may sequester limiting pools of a coactivator essential for Stat1 function. Indeed, Stat1 activated by IFN-{gamma} appears to decrease M-CSF-induced trans-activation of an AP-1 element through titration of a limiting pool of CBP (51, 52). Similarly, sequestration of CBP or the homologous protein p300 has been proposed as the mechanism by which the adenovirus E1A protein inhibits transactivation by IFN-{alpha}-dependent Stat2 or IFN-{gamma}-activated Stat1 (59). The failure of a Stat6 mutant, lacking the TADs necessary for coactivator interaction, to inhibit the IRF-1 promoter and the restoration of the inhibitory effect by the addition of the TADs are consistent with the possibility that a coactivator essential for Stat1 activation may play a role in the integration of the Stat1 and Stat6 pathways. However, despite the essential role of p300/CBP as a coactivator of Stat1 (51, 52), we find that cotransfection of CBP is insufficient to reverse the Stat6-dependent inhibitory activity of IL-4. Moreover, the inability of Stat6 DNA binding mutants (with intact TADs) to inhibit the IRF-1 promoter is inconsistent with competition for an additional coactivator essential for Stat1 function.

The final model conforming with the requirement for the Stat6 TADs for an inhibitory effect is that IL-4 activates the transcription of a specific inhibitor of the Stat1 pathway. In support of this model, we have observed that Stat6{Delta}C, a trans-dominant inhibitor of Stat6, can significantly attenuate Stat6-mediated inhibition of the IRF-1 promoter when both Stat6{Delta}C and wild-type Stat6 are transfected. Moreover, we have demonstrated that a pair of Stat6 transcription activation domains is essential for both Stat6-mediated transcription activation and inhibition of IFN-{gamma} inducibility, a result consistent with the model of Stat6-mediated activation of a specific inhibitor. While there is no evidence that such specific inhibitors suppress cytokine signaling under physiological conditions and it is unclear whether the kinetics of induction could be sufficiently rapid to repress Stat-regulated transcription, there are specific inhibitors of Stat DNA binding activity (60, 61). Protein inhibitor of activated Stat1 associates specifically with Stat1 and inhibits its DNA binding activity when added to mobility shift reactions at supraphysiologic concentrations (60, 61). Although mobility shift data did not detect an inhibitory effect of IL-4 on the DNA binding activity of Stat1 (Fig. 3Go and 5Go; Ref. 32), it remains conceivable that a physiologically important Stat1 inhibitor induced by IL-4 dissociates during nuclear extraction or mobility shift reactions. Although there is no precedent for such mechanisms, these in vitro reactions would not detect IL-4-inducible inhibitors that block the Stat1{alpha} TAD or alter its phosphorylation status (62). Other cytokine-induced inhibitors of Jak like the suppressors of cytokine signaling (SOCS) proteins regulate Stat phosphorylation (38, 63, 64) and may be implicated in IL-4-mediated inhibition of Stat1 activity. Indeed IL-4 does induce certain SOCS transcripts in a mixed population of bone marrow cells (63, 64). However, our data are inconsistent with the involvement of SOCS in IL-4-mediated inhibition of Stat1 activity, as comparable levels of Stat1-DNA complexes are observed with or without IL-4. Moreover, other investigators have observed no inhibition by IL-4 of IFN-induced Jak activity. Thus, while it is likely that SOCS-like inhibitors are induced by IL-4, these findings suggest that they are not sufficient to account for the observed Stat6-mediated inhibition of the IRF-1 promoter in the systems analyzed. This study has demonstrated that IL-4 inhibits an IFN-induced promoter by a Stat6-dependent mechanism, and the GAS element is a sufficient target for this Stat6-mediated inhibition. The data are inconsistent with mechanisms involving competition between Stat1 and Stat6 for either DNA binding or a coactivator, whereas the results support a model in which Stat6 induces the transcription of a specific inhibitory gene product.


   Acknowledgments
 
We gratefully acknowledge the technical assistance of W. Armistead; generous gifts of purified recombinant human IL-4 from J. de Vries, and the expression constructs from R. Goodman and from C. Horvath and J. Darnell; helpful discussions with R. Stein, T. Aune, and J. Darnell; and a critical reading of the manuscript by A. Richmond and G. Oltz.


   Footnotes
 
1 This work was supported by National Institutes of Health Grant R 01 GM42550 (to M.B., S.G., and J.Y.), Core Functions and Pilot Project support funded through the Vanderbilt Diabetes Research and Training Center (P60 DK20593), and the Scholars Program of the Leukemia Society of America (to M.B.). Back

2 Address correspondence and reprint requests to Dr. Mark Boothby, Department of Microbiology and Immunology, AA-4214B Medical Center North, Vanderbilt University Medical School, Nashville, TN 37232-2363. E-mail address: Back

3 Abbreviations used in this paper: IRF-1, IFN-regulated factor 1; Jak, Janus kinase; GAS, IFN-{gamma}-activated sequence; CREB, cAMP response element binding (protein); CBP, CREB binding protein; IL-4R{alpha}, IL-4 receptor {alpha}-chain; TK, thymidine kinase; SOCS, suppressor of cytokine signaling; TAD, transcription activation domain; CAT, chloramphenicol acetyltransferase; C/EBP, CCAAT/enhancer binding protein. Back

Received for publication February 9, 1999. Accepted for publication August 10, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Mosmann, T. R., R. L. Coffman. 1989. Th1 and Th2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 7:145.[Medline]
  2. Sher, A., R. L. Coffman. 1992. Regulation of immunity to parasites by T cells and T cell-derived cytokines. Annu. Rev. Immunol. 10:385.[Medline]
  3. Miyajima, A., T. Kitamura, N. Harada, T. Yokota, K.-I. Arai. 1992. Cytokine receptors and signal transduction. Annu. Rev. Immunol. 10:295.[Medline]
  4. Kishimoto, T., T. Taga, S. Akira. 1994. Cytokine signal transduction. Cell 76:253.[Medline]
  5. Seder, R. A., and W. E. Paul. Acquisition of lymphokine-producing phenotype by CD4+ T cells. Annu. Rev. Immunol. 12:635.
  6. Lohoff, M., D. Ferrick, H.-W. Mittrucker, G. S. Duncan, S. Bischof, M. Rollinghoff, T. W. Mak. 1997. Interferon regulatory factor-1 is required for a T helper 1 immune response in vivo. Immunity 6:681.[Medline]
  7. Taki, S., T. Sato, K. Ogasawara, T. Fukuda, M. Sato, S. Hida, G. Suzuki, M. Mitsuyama, E.-H. Shin, S. Kojima, T. Taniguchi, Y. Asano. 1997. Multistage regulation of Th-1type immune responses by the transcription factor IRF-1. Immunity 6:673.[Medline]
  8. Thierfelder, W. E., J. M. van Deursen, K. Yamamoto, R. A. Tripp, S. R. Sarawar, R. T. Carson, M. Y. Sangster, D. A. Vignali, P. C. Doherty, G. C. Grosveld, J. N. Ihle. 1996. Requirement for Stat4 in interleukin-12-mediated responses of natural killer and T cells. Nature 382:171.[Medline]
  9. Kaplan, M. H., Y. L. Sun, T. Hoey, M. J. Grusby. 1996. Impaired IL-12 responses and enhanced development of Th2 cells in Stat4-deficient mice. Nature 382:174.[Medline]
  10. Mond, J. J., J. Carman, C. Sarma, J. O’Hara, and F. D. Finkelman. Interferon-{gamma} suppresses B cell stimulation factor (BSF-1) induction of class II determinants on B cells. J. Immunol. 137:3534.
  11. Lee, C. E., S. R. Yoon, K. H. Pyun. 1993. Mechanism of interferon-{gamma} down-regulation of the interleukin 4-induced CD23/Fc{epsilon} RII expression in human B cells: post-transcriptional modulation by interferon-{gamma}. Mol. Immunol. 30:301.[Medline]
  12. Rousset, F., R. de Waal Malefijt, B. Slierendredt, J.-P. Aubry, J.-Y. Bonnefoy, T. Defrance, J. Banchereau, J. E. de Vries. 1988. Regulation of Fc receptor for IgE (CD23) and class II MHC antigen expression on Burkitt’s lymphoma cell lines by human IL-4 and IFN-{gamma}. J. Immunol. 140:2625.[Abstract]
  13. Lehmann, J., D. Seegert, I. Strehlow, C. Schindler, M.-L. Lohmann-Matthes, T. Decker. 1994. IL-10-induced factors belonging to the p91 family of proteins bind to IFN-{gamma}-responsive promoter elements. J. Immunol. 153:165.[Abstract]
  14. Snapper, C. M., W. E. Paul. 1987. Interferon-{gamma} and B cell stimulatory factor-1 reciprocally regulate Ig isotype production. Science 236:944.[Abstract/Free Full Text]
  15. Lehn, M., W. Y. Weiser, S. Engelhorn, S. Gillis, H. G. Remold. 1989. IL-4 inhibits the H2O2 production and anti-Leishmanial capacity of human cultured monocytes mediated by IFN-{gamma}. J. Immunol. 143:3020.[Abstract]
  16. Szabo, S. J., A. S. Dighe, U. Gubler, K. M. Murphy. 1997. Regulation of the interleukin (IL)-12R ß2 subunit expression in developing T helper 1 (Th1) and Th2 cells. J. Exp. Med. 185:817.[Abstract/Free Full Text]
  17. Boehm, U., T. Klamp, M. Groot, J. C. Howard. 1997. Cellular responses to interferon-{gamma}. Annu. Rev. Immunol. 15:749.[Medline]
  18. Snapper, C. M., K. B. Marcu, P. Zelazowski. 1997. The immunoglobulin class switch: beyond "accessibility". Immunity 6:217.[Medline]
  19. Xu, L., P. Rothman. 1994. IFN-{gamma} represses {epsilon} germline transcription and subsequently down-regulates switch recombination to {epsilon}. Intl. Immunol. 6:515.[Abstract/Free Full Text]
  20. Ihle, J. N., B. A. Witthuhn, F. W. Quelle, K. Yamamoto, O. Silvennoinen. 1995. Signaling through the hematopoietic cytokine receptors. Annu. Rev. Immunol. 13:369.[Medline]
  21. O’Shea, J. J.. 1997. Jaks, STATs, cytokine signal tranduction, and immunoregulation: are we there yet?. Immunity 7:1.[Medline]
  22. Jr Darnell, J. E., I. M. Kerr, G. R. Stark. 1994. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264:1415.[Abstract/Free Full Text]
  23. Hou, J., U. Schindler, W. J. Henzel, T. Z. Ho, M. Brasseur, S. L. Mcknight. 1994. An interleukin-4 induced transcription factor: IL-4 Stat. Science 265:1701.[Abstract/Free Full Text]
  24. Kotanides, H., N. C. Reich. 1993. Requirement of tyrosine phophorylation for rapid activation of a DNA binding factor by IL-4. Science 262:1265.[Abstract/Free Full Text]
  25. Keegan, A. D., K. Nelms, M. White, L. M. Wang, J. H. Pierce, J. H. Paul, W. E. Paul. 1994. An IL-4 receptor region containing an insulin receptor motif is important for IL-4-mediated IRS-1 phosphorylation and cell growth. Cell 76:811.[Medline]
  26. Sun, X. J., L-M. Wang, Y. Zhang, L. Yenuch, Jr M. G. Meyers, E. Glasheen, W. S. Lane, J. H. Pierce, M. F. White. 1995. Role of IRS-2 in insulin and cytokine signalling. Nature 377:173.[Medline]
  27. Ryan, J. J., L. J. McReynolds, A. Keegan, L.-H. Wang, E. Garfein, P. Rothman, K. Nelms, W. E. Paul. 1996. Growth and gene expression are predominantly controlled by distinct regions of the human IL-4 receptor. Immunity 4:123.[Medline]
  28. Izuhara, K., R. A. Feldman, P. Greer, N. Harada. 1994. Interaction of the c-fes proto-oncogene product with IL-4R. J. Biol. Chem. 269:18623.[Abstract/Free Full Text]
  29. Izuhara, K., R. A. Feldman, P. Greer, N. Harada. 1996. Interleukin-4 induces association of the c-fes proto-oncogene product with phophatidylinositol-3 kinase. Blood 88:3910.[Abstract/Free Full Text]
  30. Nelms, K., A. L. Snow, J. Hu-Li, W. E. Paul. 1998. FRIP, a hematopoietic cell-specific rasGAP-interacting protein phosphorylated in response to cytokine stimulation. Immunity 9:13.[Medline]
  31. Nandan, D., N. E. Reiner. 1997. TGF-ß attenuates the class II transactivator and reveals an accessory pathway of interferon-{gamma} action. J. Immunol. 158:1095.[Abstract]
  32. Ohmori, Y., T. A. Hamilton. 1997. IL-4-induced Stat6 suppresses IFN-{gamma}-stimulated STAT1-dependent transcription in mouse macrophages. J. Immunol. 159:5474.[Abstract]
  33. Mikita, T., D. Campbell, P. Wu, K. Williamson, U. Schindler. 1996. Requirements for interleukin-4-induced gene expression and functional chracterization of Stat6. Mol. Cell. Biol. 16:5811.[Abstract]
  34. Stevens, A. M., L.-y. Yu-Lee. 1994. Multiple prolactin-responsive elements mediate G1 and S phase expression of the interferon regulatory factor-1 gene. Mol. Endocrinol. 8:345.[Abstract]
  35. Luo, G., L.-y. Yu-Lee. 1997. Transcriptional inhibition by Stat5: differential activities at growth-related versus differentiation-specific promoters. J. Biol. Chem. 272:26841.[Abstract/Free Full Text] </