Signal transducer and activator of transcription (STAT) family of transcription factors convert extracellular cytokine signals into diverse biological responses through modulation of gene transcription (Levy and Darnell, 2002). Binding of the cytokine to its receptor at the cell surface induces receptor dimerization and activation of the cytoplasmic receptor associated Janus kinases (JAK) (Kotanides and Reich, 1993). The JAK then phosphorylate a specific tyrosine residue in STAT (Darnell et al., 1994), causing STAT dimerization, translocation to the nucleus and leading to specific DNA-binding to promoter sequences. This results in transactivation of the target genes (Ihle, 1996; Leonard and Oshea, 1998). Seven mammalian STAT proteins have been identified, and they all share a similar overall structure with central DNA binding domain, Src homology 2 (SH2) domains, and the C-terminal transactivation domain (TAD), which is the most divergent part among the STATs.
STATs play prominent roles in both pro- and anti-inflammatory processes, including cell proliferation, apoptosis, and differentiation (Rawlings et al., 2004). The different STAT molecules share significant homology in structure and function. STAT6 is activated in response to IL-4 and IL-13 (Hou et al., 1994; Quelle et al., 1995). DNA-binding sites of STAT6 have been identified in the promoter regions of several IL-4- inducible genes, including CD23, major histocompatibility complex class, Ig class ε and γ1, and the 3ε-hydroxysteroid dehydrogenase (3β-HSD)type1 (Moriggl and Berchtold, 1997; Gingras et al., 1999). Transcriptional responses to IFN-γ are largely dependent on STAT1. The phosphorylated STAT1 dimers translocate to the nucleus where they bind to specific regions (gamma-activated sequence, GAS) on IFN-γ responsive promoters and modulate their transcription (Darnell et al., 1994).
p100, encoded by SND1, is a ubiquitous, multifunctional protein that can interact with and modulate a broad spectrum of proteins involved in transcription (Leverson et al., 1998; Valineva et al., 2006), for example the transcription of EBNA-2–activated gene expression (Tong et al., 1995). In addition to binding to the acidic transactivation domain of EBNA-2, p100 has been found to interact with general transcription factor TFIIE, homeodomain-containing transcription factor c-Myb. Recently, p100 has been implicated in the pathogenesis of autosomaldominant polycystic kidney disease (ADPKD) (Low et al., 2006). Notably, p100 is also a known component of the RNA-induced silencing complex (RISC), promoting cleavage of double-stranded RNA and hyperedited double stranded RNA substrates (Caudy et al., 2003). These studies suggest that p100 may have several distinct roles. Therefore, the functional understanding of the human p100 protein requires full characterization of the complex networks of protein-protein interactions. In our previous study, we discovered that p100 protein functions as a transcriptional co-activator for STAT5-dependent gene regulation and the existence of a positive regulatory loop in PRL-induced transcription (Paukku et al., 2003). Meanwhile, p100 protein was also found as a key coactivator in the STAT6 enhanceosome (Yang et al., 2002). In this regard, our study aims at investigating the potential role of p100 in STAT1-dependent gene transcription.
MATERIALS AND METHODS
Cell Culture, Transfections, and Plasmids
HeLa cells and COS-7 cells were maintained in DMEM, supplemented with 10% FBS. Transfection of HeLa cells was performed using the calcium phosphate precipitation method for luciferase assay and using lipofectamin2000 (invitrogen) for confocal microscopy. COS-7 cells were transfected by electroporation with a Bio-Rad gene pulser at 260 V/960 microfarads.
pSG5-p100-Flag, pSTAT1-HA, pSTAT6-HA, pEGFP-CI-STAT6, pERFP-p100, pERFP-p100-SN, pERFP-p100-TSN were constructed as previously described (Yang et al., 2002).
GST Pull-Down Assays
GST (glutathione S-transferase) pulldown experiments were performed as previously described (Valineva et al., 2006). GST and GST-p100-SN, GST-p100-TSN fusion proteins were produced in BL21 bacteria and purified with glutathione-Sepharose 4B beads (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK) according to the manufacturer's instructions, and then incubated with total cell lysates of transfected COS7 cells. After washing, the bound proteins were eluted from beads, separated by SDS-PAGE and analyzed by silver staining or immunoblotting with mouse monoclonal anti-Flag M2 (Sigma) or mouse monoclonal anti-HA antibodies.
COS7 cells were transfected with pSTAT1-HA (hemagglutinin epitope-tagged) or pSTAT6-HA. Forty-eight hours after transfection, the total cell lysates of COS-7 cells were prepared as previously described (Valineva et al., 2006).
COS-7 cells transfected with pSGF-p100-flag together with pSTAT6-HA or pSTAT1-HA. After 48 hr incubation, cells were collected in Nonidet P-40 lysis buffer (50 mM Tris-HCl, pH 7.6, 150 mM NaCl, 0.1mM EDTA, 0.5% Nonidet P-40, 20% glycerol, 0.1 mM sodium orthovanadate,1 mM sodium butyrate). The cell lysates were immunoprecipitated with mouse monoclonal anti-Flag M2 (Sigma) or rabbit polyclonal IgG antibody (Santa Cruz Biotechnology) as a control. The immunoprecipitated proteins were separated by SDS-PAGE and detected by blotting with anti-HA antibody.
HeLa cells were plated in 12-well plates at a density of 3 × 104 cells per well. The cells were transfected with pSTAT6-HA plasmid, Igε-reporter luciferase construct, β-galactosidase vector, and different amount of pSG5-p100-Flag. Other cells were transfected with pSTAT1-HA plasmid, Igε-reporter luciferase construct, β-galactosidase vector, and different amount of pSG5-p100-Flag. After incubation for 24 hr, cells were starved overnight and treated or left untreated with 30 ng/mL of recombinant human IL-4 or IFN-γ for 6 hr. Cells were lysed and luciferase acitivity was measured. The luciferase values were normalized to β-galactosidase activity of the lysates.
HeLa cells were seeded on glass coverslips. HeLa cells were transfected with different expression plasmids as indicated in the figures by lipofectamin2000 when the cells were grown to approximately 80% confluence. After incubation for 48 hr, the cells were starved in serum-free medium for 8 hr and treated with IL-4 (100 ng/mL) for 30 min. Confocal images were collected using LSM510 program and Zeiss confocal microscope, equipped with an Argon laser (488 nm) and HeNe laser (543 nm), and a ×63 objective. Green emission was detected using a 505-nm low pass filter and red emission using a 630-nm low pass filter.
p100 Protein Interacts with STAT6 but Not STAT1
We performed GST pulldown assay to analyze the interaction of p100 protein with STAT1, and STAT6 as positive control. GST, GST-SN, and GST-TSN were expressed in bacteria and bound to glutathione-Sepharose beads. COS-7 cells were transfected with STAT6-HA or STAT1-HA.Then the cell lysates were incubated with beads bound GST fusion protein. As shown in Fig. 1, STAT6 interacted efficiently with p100-SN domain, but not the p100-TSN domain or GST alone. However STAT1 could not interact with p100-SN, p100-TSN, or GST alone. This result indicates that SN domain of p100 protein specifically interacts with STAT6, but neither p100-SN, nor p100-TSN, could associate with STAT1 in vitro.
As we reported earlier that p100 protein can recruit CBP to STAT6 (Valineva et al., 2005), and STAT1 could interact with CBP (Valineva et al., 2005), therefore, CBP may also act as bridge factor which recruit p100 to STAT1. To analyze whether STAT1-p100 complex formation occurs in vivo, the protein interaction between STAT1 and p100 was investigated by co-immunoprecipitation assays. COS-7 cells were cotransfected with STAT1-HA and p100-Flag. p100 was immunoprecipitated from the total cell lysates with anti-Flag antibody, and the presence of STAT1 was detected by anti-HA immunoblotting. Likewise, the same experiment was done to detect in vivo complex formation between STAT6 and p100 as positive control. As shown in Fig. 2A, STAT6 was found to coprecipitate with p100. As a control, STAT6 was not detected in anti-IgG immunoprecipitations from cells transfected with STAT6-HA and p100-Flag. However, STAT1 was not detected to bind with p100.
p100 and p100-SN Enhance STAT6-Mediated Gene Transcriptional Activation
It has been reported that CBP [cAMP response element binding protein, (CREB)-binding protein] has the capacity to facilitate the gene transcription of STAT1, STAT3, and STAT5, p100 protein was identified to interact with the HAT domain of CBP and enhance the histone acetylation of STAT6-mediated gene transcription (Valineva et al., 2005). Although we characterized that p100 could not interact with STAT1 directly, to investigate whether p100 could act as coregulator to regulate the STAT1-dependent gene transcription, we carried out luciferase reporter assay. HeLa cells were transfected with STAT1-dependent GAS luciferase reporter gene, together with increasing amounts of p100 or CBP plasmids. As a control, HeLa cells were transfected with STAT6-dependent Igε luciferase reporter gene, together with increasing amounts of p100, or CBP plasmids.
As shown in Fig. 3, ectopic expression of p100 did not affect the basal activity of the Igε reporter, but enhanced the IL-4-stimulated reporter gene activity in a dose-dependent manner (Fig. 3A), but there is no obvious enhancement of IFN-γ-induced STAT1-mediated reporter gene activity (Fig. 3C). Interestingly, verse visa, the overexpression of CBP alone has no significantly effects on STAT6-mediated gene transcriptional activation (Fig. 3B), but dramatically enhanced the STAT1-mediated gene transcription (Fig. 3D). These data indicate that CBP may regulate the STAT1- and STAT6-mediated gene transcriptional activation via different mechanisms.
To verify the physical interaction of p100-SN domain and STAT6 in vivo, we have constructed the plasmids which express the RFP tagged p100-SN or p100-TSN fusion protein, and GFP tagged STAT6 fusion protein. To verify the proper functions of the plasmids, we performed luciferase reporter assay. HeLa cells were transfected with Igε luciferase reporter gene, with different amounts of RFP-p100-SN or RFP-p100-TSN plasmids. As shown in Fig. 3E,F, consistent with our earlier report (Yang et al., 2002), the RFP-p100-SN domain alone is sufficient to enhance the STAT6-mediated gene transcriptional activation, while RFP-p100-TSN could not. Thus, the RFP-p100-SN and RFP-p100-TSN protein has the same function as the original form.
Intracellular Localization of STAT6 and p100 Functional Domains
Our early study demonstrated that STAT6 interacts with the SN domain of p100, and SN domain is efficient to increase the IL-4/STAT6 regulated gene transcription. To further confirm the physical association of STAT6 and p100-SN domain, intracellular localization of overexpressed STAT6, p100, and its functional domains was investigated in HeLa cells by confocal microscopy. Cells were cotransfected with GFP-STAT6 and RFP-p100/RFP-p100-SN/RFP-p100-TSN, respectively. After overnight starvation, the cells were treated with IL-4 (100 ng/mL) for 30 min and visualized by confocal microscopy. As shown in Fig. 4, in unstimulated cells, STAT6 and p100 were distributed throughout the cells, full-length p100 protein (Fig. 4A-b) and p100-SN (Fig. 4B-b) are mainly localized in the cytoplasm, p100-TSN (Fig. 4C-b) is mainly in the nucleus. After IL-4 stimulation, more STAT6 protein translocated to the nucleus, both full-length p100 protein (Fig. 4A-e) and p100-SN (Fig. 4B-e) are also partially translocated into nucleus where also colocolized with STAT6 (Fig. 4A-f,B-f). The merged images demonstrate colocalization of STAT6 and p100/p100-SN. However, the RFP and GFP control did not show colocalization image. These results further confirmed the physical and functional interaction of STAT6 and p100-SN domain.
Transcriptional activation of eukaryotic genes is regulated by the mechanism termed combinatorial control, whereby it depends on the precise and ordered recruitment of transcriptional activators, coactivators, and general transcription factors to the promoters of target genes. The major emphasis in our laboratory is to understand the mechanism of STAT transcriptional activation. In our previous study, we identified that p100 can interact with STAT5 (Paukku et al., 2003) and STAT6 (Yang et al., 2002), and promote their transcriptional function. The different STAT proteins belong to the same family due to the similar structure and mode of action. Hereby, the aim of this study is to determine whether p100 is critical for STAT1-induced transcription. Our data show that neither p100, nor its domains can interact with STAT1, and over expression of p100 has no effect on the IFN-γ/STAT1-mediated gene transcriptional activation. Thus, it suggests that p100 is not an essential coactivator involved in STAT1 dependent transcriptional activation events. The confocal microscopy further confirmed that it is the SN domain of p100 which functionally interacts with STAT6 with IL-4 stimulation.
CBP is a multi-functional protein which plays important roles in transcriptional regulation. It acts as a protein bridge connecting sequence-specific transcription factors to the transcription apparatus, it also processes intrinsic histone acetyltransferase activity to remodel chromatin by acetylating nucleosomal histones and facilitate transcription. Therefore, recruitment of CBP may play a functional role in STATs transcriptional activation. The transactivation domain (TAD) of STATs, which is essential for gene transcription, is located in the C-terminal portion. The TAD domains are the most divergent domains in STATs, and it is reasonable to assume that STATs may utilize different coactivator proteins to regulate gene transcription. Interestingly, all TAD of STATs is capable of recruiting CBP, though share no sequence similarity. For example, STAT1 and STAT3 interacts with N-terminus of CBP, STAT2 associates CH1 domain of CBP, STAT5 is capable of binding KIX domain of CBP/p300. Comparing with other members of the STAT family, the TAD of STAT6 is rich in porline, serine, threonine, leucine, and glutamine residues but not acidic residues. The TAD of STAT1 has been shown to interact directly with CBP/p300 (Horvai et al., 1997) while STAT6 cannot associate with CBP directly, but need p100 protein recruiting CBP to the TAD of STAT6 (Valineva et al., 2005).
From the luciferase assay, CBP alone is sufficient to promote the STAT1-mediated gene transcription, but not STAT6-mediated transcription, while ectopic expression of p100 protein alone is sufficient to enhance the STAT6- mediated transcription, but not the STAT1-mediated transcription. This is consistent with our original report that a ternary complex of STAT6, p100, and CBP is formed in cells, and that p100 can bridge the HAT activity of CBP to STAT6, while STAT1 was found to recruit CBP and HAT activity directly and p100 did not affect the recruitment of HAT activity (Valineva et al., 2005).
The function of HAT domain is to transfer acetyl group from Acetyl-CoA to acceptor substrate. There are two hypothetic mechanisms on catalyzing acetyl transfer, one is called “ping-pong” mechanism, in which the acetyl group is transferred first to an enzyme nucleophile, and then the enzyme subsequently transfers the group to an acceptor substrate. The other one is called sequential mechanism, in which the acetyl group is directly transfered to the substrate acceptor. The interaction between p100 protein and CBP may provide some insight related to facilitate the transfer of acetyl group to histones, leading further modification of chromatin structure. Our findings support a hypothesis, that p100 brings CBP to STAT6 response elements causing the nucleosomes to unfold and facilitating the access of STAT6-p100 protein complex to the basal transcription machinery and formation of the preinitiation complex. However, the precise mechanisms by which these events are connected to activation of transcription are still unknown. Furthermore, our results also indicate that various STATs are utilizing different mechanisms to recruit CBP and HAT activity. While STAT6 connects to CBP through bridging factors like p100, STAT1 was found to recruit CBP and HAT activity directly and p100 did not affect the recruitment of HAT activity.
IL-4 is a pleiotropic cytokine that regulates a variety of cellular functions, and plays an important role in the pathogenesis of asthma and allergy. In addition, IL-13 induces phosphorylation of Stat6 in lung fibroblast (Ingram et al., 2004), and many of the characteristics of airway remodeling (eosinophilia, mucous cell metaplasia, and airway fibrogenesis) are absent in the model of allergic asthma in Stat6-deficient mice (Kuperman et al., 1998). Thus, STAT6 is required for many aspects of IL-13-induced airway remodeling. Stat1 is also activated by IL-13 in a variety of lung cell types (Wang et al., 2004) and yet negatively regulates IL-13-induced signaling in pulmonary cell types (Quarcoo et al., 2004; Yu et al., 2004; Ray et al., 2005). It was demonstrated that STAT1-deficient (STAT1−/−) mice exhibited more severe pulmonary fibrosis after bleomycin injury compared with wild-type STAT1 mice (Walters et al., 2005). Therefore, STAT6 and STAT1 are both important modulators of airway remodeling but appear to exert opposing biologic effects on cytokine signaling.
It was reported that Toxoplasma gondii infection triggers the autonomous STAT6 activation pathway to obstruct the innate immunity induced by IFN-γ pathway (Nam, 2009). On the other hand, IFN-γ stimulation directly restricts the IL-4 action via inhibition of STAT6 phosphorylation (Huang et al., 2005) therefore may inhibit the growth and replication of T. gondii within the host cells. However, the underline mechanism is unknown yet.
In conclusion, we have described the mechanism by which p100 protein coactivates STAT6. We also found that p100 cannot associate with STAT1. IFN-γ only triggers the phosphorylation of STAT1 but not STAT2. However STAT2 and STAT1 form heterodimers with INF-α/β stimulation and then induce the formation of the transcription factor IFN-stimulated gene factor 3 (ISGF3) which recognize the IFN-stimulated response elements (IRSEs). Further studies are required to test whether other molecules of STAT family could interact with p100, such as STAT2. These studies might provide more insights into interaction between STAT family and p100.
The authors thank O. Silvennoinen for the plasmids and H. Zhang for technical assistance.