SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

TGF-β1 is a well-known immunosuppressive cytokine; however, little is known of the effect of TGF-β1 on antigen-presenting cells (APCs). In this report, we investigated the molecular mechanisms of the suppressive effects of TGF-β1 on APCs including dendritic cells and macrophages. Although TGF-β1 did not greatly affect the activation of APCs, as assessed by the induction of IL-12 or the upregulation of CD40 in response to LPS, it strongly inhibited IFN-γ-induced nitric oxide (NO) production from macrophages and dendritic cells. Using murine macrophage-like cell line RAW 264.7, we demonstrated that TGF-β1 not only reduced the inducible NO synthase (iNOS) protein stability but also suppressed the iNOS gene transcription. We also found that TGF-β1 directly inhibited IFN-γ-induced STAT1 activation by reducing STAT1 tyrosine phosphorylation. The IFN-γ Type I receptor (IFNGR1) was found to be associated with the TGF-β1 Type I receptor (TGF-βRI) and was phosphorylated by the TGF-βRI. Reduced activation of STAT1 by TGF-β1 was abrogated by the mutation in the IFNGR1 in which the serine residues of potential sites of phosphorylation by TGF-βRI were replaced by alanine residues. Thus, multiple mechanisms are present for the TGF-β1-mediated reduction of iNOS production, and we propose a novel mechanism for regulating inflammatory cytokine by an anti-inflammatory cytokine, TGF-β1; i.e. suppression of IFN-γ-induced STAT1 activation by an association of the IFNGR1 with the TGF-βRI.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

TGF-β1 is a pleiotropic cytokine that regulates cell growth and differentiation, deposition of the extracellular matrix and immuno-modulation (Fortunel et al. 2000; Massaguéet al. 2000). Furthermore, TGF-β1 has been considered a major anti-inflammatory cytokine which suppresses T cell proliferation and macrophage activation. Receptor activation occurs upon binding of a ligand to the Type II receptor (TGF-βRII), which then recruits and phosphorylates TGF-βRI in its glycine- and serine-rich domain (GS domain). Once phosphorylated, TGF-βRI is activated and phosphorylates Smad2 and Smad3 which then are translocated into the nucleus after forming a trimer with Smad4 to induce expression of target genes (Feng & Derynck 2005).

On the other hand, IFN-γ is an important immuno-activating cytokine which activates macrophages and dendritic cells and promotes Th1 and cytotoxic T cell differentiation from immature CD4+ T and CD8+ T cells, respectively. IFN-γ augments the expression of class I and class II MHC molecules and costimulators on APCs. The IFN-γ receptor has two subunits: IFNGR1 and IFNGR2. IFNGR1 is the signal transduction subunit and recruits JAK1 and STAT1, while IFNGR2 has a short cytoplasmic domain composed of 66 amino acids for which the only known function is the recruitment of JAK2 (Kotenko et al. 1995; Bach et al. 1996). Upon ligand binding, there is a receptor rearrangement with auto- and trans-phosphorylation/activation of preassociated JAKs, phosphorylation of the IFNGR1 Y440 motif, and recruitment and phosphorylation of STAT1. The phosphorylated STAT1 dimer is released to migrate into the nucleus, where it activates transcription by binding to GAS (γ-activated sequence), with or without additional factors.

There is extensive crosstalk between TGF-β1/Smad signaling and the JAK-STAT pathway (Ulloa et al. 1999; Eickelberg et al. 2001). For example, IFN-γ suppresses TGF-β1 signaling through up-regulation of the inhibitory Smad7. IFN-γ also inhibits TGF-β1 responses via STAT1-mediated sequestration of the nuclear coactivator p300/CREB-binding protein, preventing its association with Smads and blocking Smad transcriptional activity (Ghosh et al. 2001). In contrast, little is known about the suppression of the JAK-STAT pathway by TGF-β1. However, there actually exists a suppression of inflammatory cytokine signaling by TGF-β1, which has been shown through the analysis of TGF-β1−/– mice. TGF-β1−/– mice develop severe inflammatory lesions with disregulated IFN-γ signaling (Shull et al. 1992; Kulkarni et al. 1993). In TGF-β1−/– mice, a high level of iNOS, which catalyzes the production of NO from l-arginine, was observed (Vodovotz et al. 1996; McCartney-Francis & Wahl 2002). TGF-β1 suppresses NO production from macrophages stimulated with LPS or IFN-γ, and TGF-β1 functions as a negative autocrine feedback regulator to prevent tissue injury caused by excessive NO (Ding et al. 1990; Nelson et al. 1991). Although previous reports have suggested that TGF-β1 reduces IFN-γ-induced iNOS mRNA and protein levels (Ding et al. 1990; Vodovotz et al. 1993; Mitani et al. 2005), the detailed molecular mechanisms of the suppression of iNOS expression by TGF-β1 remain to be clarified.

In this study, we investigated the mechanisms of anti-inflammatory effects of TGF-β1 using bone marrow-derived dendritic cells (BMDCs) and murine macrophage-like cell line RAW 264.7. We demonstrated that TGF-β1 suppresses IFN-γ-induced NO production by multiple mechanisms. We found that TGF-β1 not only accelerated proteosomal degradation of iNOS but also inhibited iNOS mRNA transcription by suppressing STAT1 activation. Additional analyses showed that TGF-βRI associated with and phosphorylated IFNGR1. We propose a novel mechanism of STAT1 repression by TGF-β1, which provides a new insight into the regulatory role of TGF-β1 in inflammatory diseases.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

TGF-β1 suppresses NO production but has little effect on cytokine and surface marker expression in APCs

First, we examined whether TGF-β1 affects APC activation. Expression of activation markers, such as CD40, CD80, and CD86, was induced by either LPS or IFN-γ in BMDCs. Preincubation of BMDCs with TGF-β1 had little effect on CD40, CD80, and CD86 induction by LPS or IFN-γ (data for CD40 are shown in Fig. 1A; those for CD80 and CD86 are not shown). We then examined the effect of TGF-β1 on cytokine production from BMDCs. The cells were preincubated with or without TGF-β1 and stimulated with various doses of LPS. There were no significant differences in the amounts of IL-12 and TNFα produced from TGF-β1-pretreated cells and untreated cells in response to LPS (Fig. 1B). These data suggest that the anti-inflammatory effect of TGF-β1 on APCs is not dependent on the modulation of cell-surface molecules and cytokines.

image

Figure 1. Effect of TGF-β1 on the expression of CD40 and the production of IL-12 and TNFα induced by LPS or IFN-γ. (A,B) BMDCs were pretreated with or without TGF-β1 for more than 16 h and then stimulated with 10 ng/mL LPS or 20 U/mL IFN-γ for 16 h. Cells were washed and stained with FITC-conjugated anti-CD40 antibody (A), or cytokines in the culture supernatant were measured by ELISA (B, open lozenge; LPS, closed square; LPS+TGF-β1). (C) BMDCs were incubated with or without TGF-β1 for 16 h and then stimulated with 20 U/mL IFN-γ for 24 h. NO in the culture supernatant was measured. Statistical analyses were performed using Student's t-test. *P < 0.05, IFN-γ plus TGF-β1 vs. IFN-γ.

Download figure to PowerPoint

Next, we examined the effect of TGF-β1 on NO production. As shown in Fig. 1C, TGF-β1 has a strong suppressive effect on NO production from BMDCs stimulated with IFN-γ. Since we could not find any other profound effect by TGF-β1 on BMDC activation, reduced NO production may be a major anti-inflammatory effect of TGF-β1 on APCs. Therefore, we focused on the effect of TGF-β1 on IFN-γ-induced iNOS induction.

TGF-β1 suppresses IFN-γ-induced iNOS expression

To determine the molecular mechanism of the suppression of NO synthesis by TGF-β1, we used a macrophage-like cell line, RAW 264.7. TGF-β1 has also been shown to suppress IFN-γ-induced NO production in mouse peritoneal macrophages (Ding et al. 1990). Following stimulation with LPS or IFN-γ for 24 h, a large amount of NO was detected in the culture medium of RAW 264.7 cells (Fig. 2A). However, when the cells were pretreated with TGF-β1 for 24 h, the NO induction by LPS or IFN-γ was dramatically inhibited (Fig. 2A). This inhibition was not due to the reduction in the expression of the IFNGR1 by TGF-β1 because surface IFNGR1 expression level was not affected by TGF-β1 treatment (Fig. 2B). As shown in Fig. 2C, iNOS protein expression was strongly suppressed by TGF-β1: the protein level of IFN-γ-induced iNOS in the presence of TGF-β1 was about 40% of that in the absence of TGF-β1.

image

Figure 2. TGF-β1 suppresses IFN-γ-induced iNOS protein synthesis in RAW cells. (A) RAW cells were preincubated with or without 10 ng/mL TGF-β1 for 24 h and then were stimulated with 20 U/mL IFN-γ or 100 ng/mL LPS for 36 h. The stable NO metabolite nitrite present in the culture medium was determined. Statistical analyses were performed using Student's t-test. *P < 0.05, IFN-γ plus TGF-β1 vs. IFN-γ, LPS plus TGF-β1 vs. LPS. (B) RAW cells were cultured with or without 10 ng/mL TGF-β1 for 1 h and stained with biotin-conjugated rat anti-mouse CD119 monoclonal antibody followed by SA-PerCPCy5.5. Data are representative of at least two separate experiments. (C) RAW cells were preincubated with or without 10 ng/mL TGF-β1 for 24 h and then stimulated with 20 U/mL IFN-γ for 3 h. iNOS protein levels were determined by immunoblotting. (D) RAW cells were preincubated with 20 U/mL IFN-γ for 4 h to induce iNOS protein and then incubated with or without TGF-β1 for the indicated time. iNOS protein levels were determined by immunoblotting, and the band intensity was quantified. (E) RAW cells were preincubated with 20 U/mL IFN-γ for 11 h and then treated with or without 10 ng/mL TGF-β1 and MG132 for 3 h. iNOS protein levels were determined by immunoblotting. The intensity of the bands was quantified with Image Gauge (FUJIFILM) and ratios of the indicated band intensities are shown.

Download figure to PowerPoint

Next, we tested whether TGF-β1 would affect iNOS protein stability. RAW 264.7 cells were pretreated with IFN-γ for 4 h to accumulate the iNOS protein. Then the cells were washed and further incubated with or without TGF-β1. As shown in Fig. 2D, TGF-β1 reduced the iNOS protein levels, suggesting that TGF-β1 promotes iNOS protein degradation. TGF-β1-mediated degradation of iNOS was partially blocked by treating cells with MG132, a proteasome inhibitor (Fig. 2E). These results are consistent with the recent report that TGF-β1 enhances the degradation of the iNOS protein through the ubiquitin-proteasome pathway (Mitani et al. 2005).

TGF-β1 suppresses iNOS gene transcription

As shown in Fig. 2E, however, the proteasome inhibitor could not fully overcome the suppressive effect of TGF-β1. Therefore, we examined the iNOS mRNA levels by Northern blotting. As shown in Fig. 3A, TGF-β1 treatment suppressed IFN-γ-induced iNOS mRNA induction to about 50%. This result indicates that reduced protein levels of iNOS by TGF-β1 (Fig. 2C) are explained by two effects, reduced iNOS protein stability and reduced iNOS mRNA levels. Induction of iNOS mRNA by IFN-γ is largely dependent on STAT1 (Meraz et al. 1996). Therefore, we examined the effect of TGF-β1 on IFN-γ-induced STAT1 transcriptional activity assessed by GAS reporter gene activation. GAS reporter activity was increased 12-fold by IFN-γ. However, this activity was reduced to about 50% in the cells pretreated with TGF-β1 (Fig. 3B), suggesting that TGF-β1 suppresses IFN-γ-induced STAT1 activation.

image

Figure 3. Suppression of iNOS gene transcription and STAT1 activation by TGF-β1. (A) RAW cells were preincubated with or without 10 ng/mL TGF-β1 for 1 h and then incubated with 20 U/mL IFN-γ for 3 h. iNOS mRNA levels were measured by Northern blotting. The relative intensity shows the ratio of the intensities of iNOS and GAPDH. (B) RAW cells transfected with pGAS-Luc and β-galactosidase control plasmids were cultured in the presence or absence of 20 U/mL IFN-γ and 10 ng/mL TGF-β1 for 5 h, and then luciferase and β-galactosidase activities were measured. (C) RAW cells were preincubated with or without 10 ng/mL TGF-β1 for 1 h and stimulated with 20 U/mL IFN-γ, 100 U/mL IFN-α or 100 U/mL IFN-β for the indicated time. Whole-cell extracts were immunoblotted with the indicated antibodies. (D) RAW cells were treated with or without 10 ng/mL TGF-β1 and 20 U/mL IFN-γ in the presence or absence of 20 µg/mL cycloheximide (CHX). Then STAT1 phosphorylation and iNOS protein levels were detected by Western blotting. (E, F) RAW cells were treated with or without 10 ng/mL TGF-β1 and CHX for 1 h and stimulated with 20 U/mL IFN-γ for 3 h. The iNOS and SOCS1 mRNAs were detected by RT-PCR. The relative band intensities are shown on the panels.

Download figure to PowerPoint

Next, we examined whether TGF-β1 would inhibit the IFN-γ-induced tyrosine phosphorylation of JAK and STAT1, which is essential for STAT1 dimerization and nuclear translocation. In response to IFN-γ, JAK1 was phosphorylated; however, JAK1 phosphorylation was not much affected by TGF-β1 (Fig. 3C) at any time point we tested. However, STAT1 phosphorylation was suppressed by about 50% in the presence of TGF-β1 (Fig. 3C). This suppression was not due to the induction of an IFN-γ specific inhibitor such as SOCS1 since the TGF-β1-mediated suppression of IFN-γ-induced STAT1 phosphorylation as well as iNOS mRNA induction were observed even in the presence of cycloheximide (Fig. 3D,E). SOCS1 has been shown to be an important physiological negative regulator of IFN-γ signaling. Moreover, TGF-β1 neither induced SOCS1 expression nor modulated IFN-γ-induced SOCS1 mRNA expression (Fig. 3F). The TGF-β1-mediated suppression of the IFN-γ signaling was not a general cytotoxic effect of TGF-β1, because TGF-β1 did not inhibit IFN-α/β-induced STAT1 activation (Fig. 3C). These data suggest that TGF-β1 suppresses STAT1 activation at the level of the downstream of JAKs but upstream of STAT1, probably at the IFN-γ receptor level.

Activated TGF-β1 receptor suppresses IFN-γ-receptor-mediated STAT1 activation

Reduced tyrosine phosphorylation of STAT1 could be due to reduced STAT1 recruitment to IFNGR1. Since phosphorylation of Y440 of IFNGR1 is essential for STAT1 recruitment, we have tried to examine the phosphorylation of IFNGR1 upon ligand stimulation. Unfortunately, we could not detect tyrosine phosphorylation of IFNGR1 in response to IFN-γ even in the absence of TGF-β1, probably because of the low detection sensitivity of anti-phosphotyrosine immunoblotting after immunoprecipitation of IFNGR1. Thus, to define the effect of TGF-β1 receptor serine/threonine kinase on IFNGR1 activity, we examined GAS reporter assay in RAW cells transfected with either constitutively active TGF-βRI (CA-TβRI) or kinase negative-TGF-βRI (KN-TβRI) (Wrana et al. 1994). As shown in Fig. 4A, forced expression of CA-TβR1 but not KN-TβRI partially suppressed murine IFN-γ-mediated STAT1 reporter activation. Thus, the TGF-βRI may directly suppress IFN-γ receptor-mediated STAT1 activation, which depends on kinase activity of TGF-βRI.

image

Figure 4. Suppression of IFNGR1-mediated STAT1 activation by an active TGF-β1 receptor. (A) RAW cells were transfected with pGAS-Luc, β-galactosidase control plasmid and CA-TβRI or kinase negative TβRI (KN-TβRI). After transfection, the cells were incubated for 24 h prior to the addition of 20 U/mL IFN-γ for an additional 5 h, and then the luciferase and β-galactosidase activities were measured. (B) Schematic representations of IFNGR1 and the alignment of the SSXS motif in mouse and human IFNGR1 are shown. (C) COS-7 cells were co-transfected with murine IFNGR1 WT, IFNGR1/3SA, IFNGR2, STAT1, pGAS-Luc, β-galactosidase control plasmid, and CA-TβRI. After incubation for 12 h, the cells were treated with 20 U/mL IFN-γ for 6 h, and the luciferase and β-galactosidase activities were measured. The luciferase activity was normalized by β-galactosidase activity. Statistical analyses were performed using Student's t-test. *P < 0.05, IFN-γ plus TGF-β1 vs. IFN-γ. (D) COS-7 cells were co-transfected with IFNGR1 WT, IFNGR1/3SA, IFNGR2, STAT1, and CA-TβRI. After 12 h, following a medium change, the cells were treated with 20 U/mL IFN-γ for 30 min. Immunoblotting was performed with the indicated antibodies. The relative band intensities are shown on the panels. (E) RAW cells were transfected with myc-IFNGR1/3SA and pEGFP-C1 vector. After 24 h, the cells were treated with 10 ng/mL TGF-β1 and 20 U/mL IFN-γ for 5 h, then stained with anti-iNOS antibody and analyzed by flow cytometry. The percentage of cell population in the gates is shown.

Download figure to PowerPoint

TGF-βRI induces serine phosphorylation of the C-terminal SSXS motif of Smads (Massaguéet al. 2000). Therefore, we searched related SSXS motifs in mouse and human IFNGR1. We have found that an SSXS motif exists between the JAK-binding domain and the STAT-binding site in its cytoplasmic domain of both mouse and human IFNGR1 (Fig. 4B), whereas no such motif is present in mouse and human IFNGR2. Therefore, we hypothesized that the conserved SSXS motif of IFNGR1 was phosphorylated by activated TGF-βRI, resulting in reduced STAT1 recruitment. To test this hypothesis, we constructed a mutant IFNGR1 in which three serines at the SSXS motif were changed to alanines (IFNGR1/3SA). Since transfection efficiency in COS-7 cells is better than that in RAW cells, we used an over-expression system with the transfection of murine IFNGR1 and IFNGR2 and constitutively active TGF-βRI (CA-TβRI) in COS-7 cells. As shown in Fig. 4C, IFNGR1/3SA could activate STAT1 in response to IFN-γ, suggesting that these serine residues are not required for IFN-γ signaling. However, IFNGR1/3SA failed to respond to CA-TβRI although IFNGR1/WT-mediated STAT1 activation was suppressed (Fig. 4C). Therefore, these serine residues are essential for CA-TβRI-mediated suppression of STAT1 activity. We further investigated tyrosine phosphorylation of STAT1 by Western blotting (Fig. 4D). The tyrosine phosphorylation of STAT1 by IFNGR1 WT was suppressed by CA-TβRI. In contrast, the suppression of STAT1 phosphorylation was not observed when IFNGR1/3SA was expressed.

To examine the effect of this SA mutant IFNGR1 on TGF-β1 in RAW cells, we co-transfected IFNGR1/3SA cDNA with EGFP cDNA and measured IFN-γ-induced iNOS protein expression by flow cytometry (Fig. 4E). We could successfully detect iNOS protein by intracellular staining. TGF-β1 reduced iNOS protein in GFP-negative cells to about 50% levels (2.9% vs. 1.62%). However such reduction was not evident in GFP-positive cells which were supposed to express IFNGR1/3SA (1.23% vs. 1.28%). Therefore, the activity of IFNGR1/3SA was not modulated by TGF-β1 in RAW cells. These data suggest that activated TGF-βRI reduces STAT1 phosphorylation and iNOS production through the SSXS motif of IFNGR1.

IFNGR1 interacts with and is phosphorylated by TGF-βRI

Because TGF-βRI directly inhibited IFN-γ-induced receptor-mediated STAT1 activation, we assumed that TGF-βRI would directly interact with IFNGR1. Therefore, we performed co-immunoprecipitation experiment. As shown in Fig. 5A, HA-CA-TβRI was co-immunoprecipitated with myc-IFNGR1; reciprocally, myc-IFNGR1 was co-immunoprecipitated with HA-CA-TβRI. A similar interaction was observed between HA-WT-TGF-βRI and IFNGR1 (data not shown).

image

Figure 5. IFNGR1 was associated with and phosphorylated by CA-TβRI. (A) COS-7 cells were transfected with myc-IFNGR1 WT or myc-IFNGR1/3SA and with or without HA-CA-TβRI. Cell lysates were subjected to immunoprecipitation with anti-HA or anti-myc, and then receptor complexes were analyzed by immunoblotting with the indicated antibodies. (B) RAW cells lysates were subjected to immunoprecipitation with the indicated antibodies, then receptor complexes were analyzed by immunoblotting with anti-TβRI. Anti-myc is an unrelated antibody for a negative control. (C) COS-7 cells were transfected with myc-IFNGR1 WT or HA-CA-TβRI. Cell lysates were immunoprecipitated with an anti-HA antibody and then incubated with [γ-32P]-ATP. The phosphorylation reaction was stopped by adding SDS sample buffer and boiling. After dilution in the lysis buffer, samples were subjected to immunoprecipitation with an anti-myc antibody, SDS-PAGE, and autoradiography.

Download figure to PowerPoint

To measure interaction between endogenous TGF-βRI and IFNGR1, we performed immunoprecipitation-Western blot analysis (Fig. 5B). Endogenous TGF-βR1 was co-immunoprecipitated with endogenous IFNGR1 without IFN-γ/TGF-β stimulation (Fig. 5B). These data suggest that TGF-βR1 can associate with IFNGR1 before activation at physiological conditions.

Next, we examined whether IFNGR1 was phosphorylated by CA-TβRI. COS-7 cells were transfected with myc-tagged IFNGR1 WT and either HA-tagged CA-TβRI or HA-tagged KN-TβRI. The TGF-βRI-IFNGR1 complex was immunoprecipitated with an anti-HA antibody and then subjected to in vitro kinase reaction using [γ-32P]ATP. After in vitro phosphorylation, myc-IFNGR1 was immunoprecipitated, and phosphorylation was detected by autoradiography. Phosphorylation of IFNGR1 was observed when CA-TβRI was co-expressed, while no phosphorylation occurred when KN-TβRI was transfected (Fig. 5C). This result suggests that IFNGR1 is directly phosphorylated by activated TGF-βRI.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

In this report, we demonstrated that TGF-β1 suppresses IFN-γ-induced NO production by multiple mechanisms, including inhibition of iNOS mRNA transcription and enhancement of iNOS protein degradation through proteasome pathway. Furthermore, we found that TGF-β1 suppresses IFN-γ-induced STAT1 phosphorylation and activation. Suppression of STAT1 activation by TGF-β1 was about 50% at several conditions. Thus, major mechanism of TGF-β1 seems to be the degradation of iNOS protein. However, we consistently observe reduction of IFN-γ-induced STAT1 phosphorylation by TGF-β1 and this STAT1 repression by TGF-β1 was not observed in case to IFN-α and IFN-β (Fig. 3C). In addition, IFN-γ and TGF-β1 have been shown to counteract each other in many cases. Thus, we believe that suppression of IFN-γ-mediated STAT1 activation by TGF-β1 is physiologically important. We hypothesize that TGF-βRI negatively regulates STAT1 activation by association with and phosphorylation of IFNGR1. This novel mechanism is supported by the facts that: (i) IFNGR1 was co-immunoprecipitated with TGF-βRI, (ii) IFNGR1 was phosphorylated by CA-TβRI, (iii) IFNGR1/3SA mutant was resistant to TβRI-mediated inhibition of STAT1 phosphorylation and activation. We could not detect Smad7 protein in RAW cells treated with IFN-γ and TGF-β1. Therefore, possibility of the involvement of Smad7 in this system may be low.

TGF-β1 has been reported to inhibit several STAT pathways other than STAT1 in response to various cytokines, including IL-6 (Wierenga et al. 2002), IL-12 (Bright & Sriram 1998) and IL-2 (Bright et al. 1997). However, in all cases, the mechanisms by which TGF-β1 inhibits the JAK-STAT pathway activated by these cytokines have never been identified. We noticed that IFNGR1, but not IFNGR2, contains a conserved potential phosphorylation amino acid motif (SSXS) by TGF-β receptor serine/threonine kinase. TGF-β1 did not affect IFN-α- or IFN-β-induced STAT1 activation, probably because there are no SSXS motifs in IFN-α/β R1 and IFN-α/β R2. In addition, we found that a mutant IFNGR1 lacking the SSXS motif could not inhibit STAT1 phosphorylation. We demonstrated a constitutive association between TGF-βRI and IFNGR1 and the phosphorylation of IFNGR1 by activated TGF-βRI when they are over-expressed. These results strongly suggest that TGF-β1 inhibits STAT1 activation by the direct phosphorylation of IFNGR1 by TGF-βRI. This serine phosphorylation of IFNGR1 may reduce tyrosine phosphorylation of IFNGR1 by JAKs or may inhibit the binding of STAT1 to tyrosine-phosphorylated IFNGR1. Further study is necessary to determine the precise molecular mechanism of the suppression of STAT1 activation by the serine phosphorylation of IFNGR1.

Interplays between different cytokine and growth-factor receptors have been demonstrated for the IL-6 receptor and EGF receptor (Badache & Hynes 2001), c-kit and the EPO receptor (Wu et al. 1995) and the IL-6 receptor and IFN receptors (Mitani et al. 2001). Crosstalk between IFN-γ and IFN-α/β signaling components in the caveolar membrane domains has been shown (Takaoka et al. 2000). These associations usually facilitate receptor-signal transduction. However, interaction between IFN-γ receptors and TGF-β receptors is unique because this interaction may be a mechanism of the cross-suppression of inflammatory and anti-inflammatory cytokines, which raises the interesting possibility that the IFN-γ receptor may inhibit TGF-β1 signaling by the induction of tyrosine phosphorylation of the TGF-β1 receptor. However, this possibility has not been examined yet. The discovery of such an interaction between cytokine/growth factor receptors may be important for the modulation of cytokine networks. For example, a compound that disrupts the TGF-β1 and IFN-γ receptors may enforce immune reactions, and a compound that induces stronger interaction may be useful as an immune modulator.

Other mechanisms for the TGF-β1-mediated suppression of IFN-γ signaling have been reported. In astrocytes, TGF-β1 inhibits the induction of class II MHC mRNA not by inhibiting receptor-proximal signaling events but by suppressing the transcriptional induction of CIITA, which is essential for class II MHC expression (Nandan & Reiner 1997; Dong et al. 2001). Smads activated by TGF-β1 may function as a transcriptional repressor for CIITA (Dong et al. 2001). Another report suggests that TGF-β1 inhibits T-bet induction through the induction of protein tyrosine phosphatase Src homology region 2 containing phosphatase-1 (Shp-1) in murine CD4+ T cells (Park et al. 2005). Thus, there seems to be cell- and gene-specific mechanisms for the TGF-β1-mediated suppression of cytokine signaling. Such multiple, cell type-specific mechanisms may exist for TGF-β1-mediated suppression of IFN-γ signaling.

Negative regulation of inflammatory cytokines is essential to avoid potentially deleterious consequences of excessive immune cell activation. TGF-β1 has been generally considered a major anti-inflammatory cytokine; however, its suppressive mechanisms for cytokine signaling are not fully understood. In this report, we have proposed a TGF-β1 suppresses IFN-γ-induced STAT1 activation. This mechanism can represent a broad suppressive effect of TGF-β1 on IFN-γ. In addition, we have shown that TGF-β1 has an iNOS-specific effect: TGF-β1 induced destabilization of the iNOS protein. These effects were not observed for activation markers such as CD40 and cytokines in dendritic cells. Therefore, TGF-β1 has a gene-specific inhibitory effect in addition to a general STAT1 inhibitory effect. Induction of CD40 and inflammatory cytokines may not be strictly dependent on STAT1 since similar up-regulation of CD40 and IL-12 was observed in STAT1-deficient dendritic cells in response to LPS (Gautier et al. 2005). TGF-β1 may not affect the protein stability of these gene products.

Previous reports have detailed the mechanisms through which TGF-β1 suppresses NO production in several cell types. These reports suggest multiple mechanisms for the suppression of iNOS expression. It was reported that TGF-β1 did not exert a transcriptional effect on the iNOS gene, whereas it reduced the stability and rate of translation of iNOS mRNA and increased the rate of degradation of the iNOS protein in mouse peritoneal macrophages (Vodovotz et al. 1993). Other reports indicated that TGF-β1 reduced the expression of iNOS at the transcriptional (Perrella et al. 1994; Perrella et al. 1996) as well as the post-transcriptional and post-translational levels (Finder et al. 1995) in smooth muscle cells. A recent study also showed that TGF-β1 enhanced the ubiquitination- and proteasome-dependent degradation of iNOS protein in IFN-γ treated RAW cells (Mitani et al. 2005). We have confirmed that TGF-β1 induces degradation of the iNOS protein. The molecular details of the iNOS-specific effect of TGF-β1 remain to be clarified.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Cell culture

RAW 264.7 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) (Gibco) containing 10% fetal bovine serum, antibiotics (Gibco), and nonessential amino acids (Gibco). COS-7 cells were cultured in DMEM containing 10% fetal bovine serum and antibiotics. BMDCs were generated from bone marrow according to the method described by Inaba et al. (1992), with modifications. Briefly, bone marrow samples from the femurs and tibiae of mice were cultured in RPMI 1640 (Sigma) with 10% heat-inactivated FBS containing 20 ng/mL GM-CSF (J558 supernatant) for 7 days with replenishment of the medium every other day.

Construction of expression vectors

Viral hemagglutinin (HA)-tagged constitutive activation TGF-βRI (CA-TGF-βRI) in pCMV5 was a gift from Dr J. Massagué (HHMI, Memorial Sloan-Kettering Cancer Center, New York, USA). TGF-βRI mutants were generated by the standard polymerase chain-reaction method as described by Kunkel (1985). The cDNAs encoding mouse IFNGR1 and IFNGR2 were cloned from the total RNA of RAW264.7 cells by RT-PCR and subcloned into pcDNA4 Myc-His A (Invitrogen) and pFLAG-myc-CMV21 (Sigma). Mutations were introduced by site direct mutagenesis using PCR. All constructs were confirmed by sequencing.

Transient transfection and reporter gene analysis

Cells were seeded on to 6-well plates one day before transfection. The GAS-luciferase plasmid (Mikita et al. 1996) was co-transfected with the β-galactosidase control plasmid using FuGENE 6 (Roche) according to the manufacturer's instructions. After 12 h, the cells were treated with 20 U/mL IFN-γ for 6 h, and the luciferase and β-galactosidase activities were measured.

ELISA and flow cytometric analysis

BMDCs were preincubated with or without TGF-β1 for more than 16 h and stimulated with 10 ng/mL LPS or 20 U/mL IFN-γ for 16 h. Culture supernatants of resting or stimulated Day 7 BMDCs were assayed for protein levels of TNFα (BD Bioscience) and IL-12p40 by ELISA (BD Biosciences) according to the manufacturer's instructions. CD40 and IFNGR1 expression was analyzed using FACScalibur (Becton Dickinson) with the FITC-conjugated anti-CD40 antibody (eBioscience) or biotin-conjugated rat anti-mouse CD119 (IFNGR1) monoclonal antibody (BDPharMingen).

Intracellular staining

RAW cells were transfected with myc-IFNGR1/3SA and pEGFP-C1 vector (Clontech). After for 24 h, cells were stimulated with 10 ng/mL of TGF-β or 20 U/mL IFN-γ for 12 h. Then, cells were collected and fixed with intracellular staining kit (eBioscience) according to the manufacturer's instructions. Then cell were stained with anti-iNOS antibody, and then analyzed using FACScalibur.

Measurement of nitrite concentration

RAW 264.7 cells (1 × 104 cells) were preincubated with or without 10 ng/mL recombinant human TGF-β1 (PeproTech), which can activate murine cells, for 24 h. Then they were incubated with 20 U/mL recombinant murine IFN-γ (PeproTech) for 36 h. NO synthesis (Nitrate) was spectrophotometrically determined by adding 50 µL Griess reagent (1% sulfanilamide/0.1%-N-(1-napthyl)-ethylenediamine dihydrochloride/2.5%) to 50 µL of culture medium as described (Aki et al. 2005). After incubation for 5 min at room temperature, the absorbance at 550 nm was measured and compared with standard NaNO2.

Immunoprecipitation and immunoblotting

Cells were solubilized in the NP-40-lysis buffer (50 mm Tris-HCl pH 7.4, 150 mm NaCl, 1% NP-40, protease inhibitor cocktail (Roche), phosphatase inhibitor cocktail I (Sigma), and phosphatase inhibitor cocktail II (Sigma) on ice for 30 min and then centrifuged at 15 000 g for 10 min at 4 °C. The supernatants were separated by SDS-PAGE, and the gel was transferred on to polyvinylidene difluoride membranes. The membranes were then blocked with Tris-HCl buffered saline (TBS) containing 5% skim milk, immunoblotted with specific antibodies, and visualized with the appropriate horseradish peroxidase-conjugated secondary antibodies using the Super Signal West Pico Chemiluminescent Substrate (PIERCE). For immunoprecipitation, cells were lyzed in the digitonin-lysis buffer (50 mm Tris-HCl pH 7.4, 150 mm NaCl, 1% digitonin, protease inhibitor cocktail, phosphatase inhibitor cocktail I, and phosphatase inhibitor cocktail II) and then centrifuged at 15 000 g for 10 min at 4 °C. The supernatants were incubated with anti-HA or anti-myc or anti-IFNGR1 antibodies and protein G-sepharose (Amersham Pharmacia) for 3 h at 4 °C. The immunoprecipitates were collected by centrifugation, washed 4 times in the lysis buffer, and then analyzed by SDS-PAGE.

Northern blot analysis

Poly(A)+ RNA of RAW cells was prepared using TRIzol Reagent (Invitogen) followed by Oligotex-dT30<Super> (Takara Bio). For Northern hybridization, poly(A)+ RNA (0.5 µg each) was separated on 1% agarose gel and transferred to a nylon membrane. Riboprobes were prepared by in vitro transcription with digoxigenin RNA Labeling Mix (Roche) using appropriate cDNA fragments: a 500 bp fragment as a probe for GAPDH and a 440 bp fragment as a probe for iNOS. Hybridization was performed overnight at 68 °C with probes (0.1 µg/mL) in solutions containing 50% formamide, 5 × SSC, 0.02% SDS, 0.1%N-lauroylsarcosine, and 2% blocking reagent (DIG Wash and Block Buffer Set, Boehringer Mannheim). After hybridization, filters were washed in 0.1 × SSC and 0.1% SDS twice for 10 min at room temperature and then twice for 20 min at 65 °C. The blot was visualized using the DIG Luminescent Detection Kit (Boehringer Mannheim) with CSPD (disodium 3-(4-methoxyspiro{1,2-dioxetane-3,29-(59-cholo)tricyclo[3.3.1.13,7]decan}-4-yl)phenylphosphate) as a substrate.

RT-PCR

RAW cells were pretreated for 1 h with or without TGF-β1 (10 ng/mL) and stimulated with IFN-γ for 3 h. Total RNA of RAW cells was prepared using TRIzol Reagent (Invitogen) following the manufacturer's instructions. RT-PCR was carried out using the one-step RT-PCR kit (Applied Biosystems) according to the manufacturer's instructions. The following oligonucleotides were used for mouse iNOS: 5′-CCCTTCCGACGTTTCTGGCAGCAGC-3′ and 5′-TCAGAGAGCCTCGTGGCTTTGG-3′; SOCS1: 5′-CACTCACTTCCGCACCTTCC-3′ and 5′-CAGCCGCTCAGATCTGGAAG-3′; and for G3PDH: 5′-ACCACAGTCCATGCCATCAC-3′ and 5′-TCCACCACCCTGTTG CTG TA-3′.

In vitro kinase assay

COS-7 cells transiently transfected with myc-IFNGR1 WT (or deletion mutants) together with HA-tagged CA-TβRI or HA-tagged KN-TβRI. After being washed with PBS, collected cells were lyzed in the digitonin-lysis buffer and incubated for 30 min on ice. The whole extract was centrifuged at 15 000 g for 10 min at 4 °C, and the supernatants were collected. The supernatants were incubated with anti-HA and protein G-sepharose for 3 h at 4 °C. The immunoprecipitates were collected by centrifugation and then were washed 2 times in the digitonin-lysis buffer and 3 times in kinase reaction buffer (5 mm Tris-HCl pH 7.4, 1 mm MgCl2, 0.1 mm CaCl2). The beads were incubated with 5 µCi of [γ-32P]-ATP and 10 µm of cold ATP in 20 µL of reaction buffer for 30 min at room temperature. The reaction was stopped by adding 20 µL of 2% SDS with 10% 2-mercaptoethanol (2ME) and boiling for 5 min. To isolate IFNGR1, the supernatant was collected, and the beads were re-extracted in 50 µL of boiling water. The combined supernatants were mixed with 0.4 mL of lysis buffer containing 1% NP-40 and then immunoprecipitated with anti-myc antibody and protein G-Sepharose. The immunoprecipitates were boiled in SDS-sample buffer containing 5% 2ME for 5 min and subjected to 8.0% SDS-PAGE. The gel was dried and exposed to X-ray film with an intensifier screen at −80 °C for 2–14 days.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank T. Yoshioka, M. Othsu and Y. Yamada for technical assistance and Y. Nishi for manuscript preparation, Dr Joan Massagué for providing pCMV5-TbRI-HA (T204D). This work was supported by special grants-in-aid from the Ministry of Education, Science, Technology, Sports, and Culture of Japan for Y.A. and K.T.; the Haraguchi Memorial Foundation; the Yamanouchi Foundation for Research on Metabolic Disorders; the Takeda Science Foundation; the Mochida Memorial Foundation; the Kato Memorial Foundation; and the Uehara Memorial Foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Aki, D., Mashima, R., Saeki, K., Minoda, Y., Yamauchi, M. & Yoshimura, A. (2005) Modulation of TLR signalling by the C-terminal Src kinase (Csk) in macrophages. Genes Cells 10, 357368.
  • Bach, E.A., Tanner, J.W., Marsters, S., et al. (1996) Ligand-induced assembly and activation of the gamma interferon receptor in intact cells. Mol. Cell. Biol. 16, 32143221.
  • Badache, A. & Hynes, N.E. (2001) Interleukin 6 inhibits proliferation and, in cooperation with an epidermal growth factor receptor autocrine loop, increases migration of T47D breast cancer cells. Cancer Res. 61, 383391.
  • Bright, J.J., Kerr, L.D. & Sriram, S. (1997) TGF-β inhibits IL-2-induced tyrosine phosphorylation and activation of Jak-1 and Stat 5 in T lymphocytes. J. Immunol. 159, 175183.
  • Bright, J.J. & Sriram, S. (1998) TGF-β inhibits IL-12-induced activation of Jak-STAT pathway in T lymphocytes. J. Immunol. 161, 17721777.
  • Ding, A., Nathan, C.F., Graycar, J., Derynck, R., Stuehr, D.J. & Srimal, S. (1990) Macrophage deactivating factor and transforming growth factors-β1, -β2 and -β3 inhibit induction of macrophage nitrogen oxide synthesis by IFN-γ. J. Immunol. 145, 940944.
  • Dong, Y., Tang, L., Letterio, J.J. & Benveniste, E.N. (2001) The Smad3 protein is involved in TGF-β inhibition of class II transactivator and class II MHC expression. J. Immunol. 167, 311319.
  • Eickelberg, O., Pansky, A., Koehler, E., et al. (2001) Molecular mechanisms of TGF-β antagonism by interferon γ and cyclosporine A in lung fibroblasts. FASEB J. 15, 797806.
  • Feng, X.H. & Derynck, R. (2005) Specificity and versatility in TGF-β signaling through Smads. Annu. Rev. Cell Dev. Biol. 21, 659693.
  • Finder, J., Stark, W.W. Jr, Nakayama, D.K., et al. (1995) TGF-β regulates production of NO in pulmonary artery smooth muscle cells by inhibiting expression of NOS. Am. J. Physiol. 268, L862L867.
  • Fortunel, N.O., Hatzfeld, A. & Hatzfeld, J.A. (2000) Transforming growth factor-β: pleiotropic role in the regulation of hematopoiesis. Blood 96, 20222036.
  • Gautier, G., Humbert, M., Deauvieau, F., et al. (2005) A type I interferon autocrine-paracrine loop is involved in Toll-like receptor-induced interleukin-12p70 secretion by dendritic cells. J. Exp. Med. 201, 14351446.
  • Ghosh, A.K., Yuan, W., Mori, Y., Chen, S. & Varga, J. (2001) Antagonistic regulation of type I collagen gene expression by interferon-γ and transforming growth factor-β. Integration at the level of p300/CBP transcriptional coactivators. J. Biol. Chem. 276, 1104111048.
  • Inaba, K., Inaba, M., Romani, N., et al. (1992) Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J. Exp. Med. 176, 16931702.
  • Kotenko, S.V., Izotova, L.S., Pollack, B.P., et al. (1995) Interaction between the components of the interferon γ receptor complex. J. Biol. Chem. 270, 2091520921.
  • Kulkarni, A.B., Huh, C.G., Becker, D., et al. (1993) Transforming growth factor β1 null mutation in mice causes excessive inflammatory response and early death. Proc. Natl. Acad. Sci. USA 90, 770774.
  • Kunkel, T.A. (1985) Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. USA 82, 488492.
  • Massagué, J., Blain, S.W. & Lo, R.S. (2000) TGFβ signaling in growth control, cancer, and heritable disorders. Cell 103, 295309.
  • McCartney-Francis, N.L. & Wahl, S.M. (2002) Dysregulation of IFN-γ signaling pathways in the absence of TGF-β1. J. Immunol. 169, 59415947.
  • Meraz, M.A., White, J.M., Sheehan, K.C., et al. (1996) Targeted disruption of the Stat1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signaling pathway. Cell 84, 431442.
  • Mikita, T., Campbell, D., Wu, P., Williamson, K. & Schindler, U. (1996) Requirements for interleukin-4-induced gene expression and functional characterization of Stat6. Mol. Cell. Biol. 16, 58115820.
  • Mitani, Y., Takaoka, A., Kim, S.H., et al. (2001) Cross talk of the interferon-α/β signalling complex with gp130 for effective interleukin-6 signalling. Genes Cells 6, 631640.
  • Mitani, T., Terashima, M., Yoshimura, H., Nariai, Y. & Tanigawa, Y. (2005) TGF-β1 enhances degradation of IFN-γ-induced iNOS protein via proteasomes in RAW 264.7 cells. Nitric Oxide 13, 7887.
  • Nandan, D. & Reiner, N.E. (1997) TGF-β attenuates the class II transactivator and reveals an accessory pathway of IFN-γ action. J. Immunol. 158, 10951101.
  • Nelson, B.J., Ralph, P., Green, S.J. & Nacy, C.A. (1991) Differential susceptibility of activated macrophage cytotoxic effector reactions to the suppressive effects of transforming growth factor-β1. J. Immunol. 146, 18491857.
  • Park, I.K., Shultz, L.D., Letterio, J.J. & Gorham, J.D. (2005) TGF-beta1 inhibits T-bet induction by IFN-gamma in murine CD4+ T cells through the protein tyrosine phosphatase Src homology region 2 domain-containing phosphatase-1. J. Immunol. 175, 56665674.
  • Perrella, M.A., Patterson, C., Tan, L., et al. (1996) Suppression of interleukin-1β-induced nitric-oxide synthase promoter/enhancer activity by transforming growth factor-β1 in vascular smooth muscle cells. Evidence for mechanisms other than NF-κB. J. Biol. Chem. 271, 1377613780.
  • Perrella, M.A., Yoshizumi, M., Fen, Z., et al. (1994) Transforming growth factor-β1, but not dexamethasone, down-regulates nitric-oxide synthase mRNA after its induction by interleukin-1β in rat smooth muscle cells. J. Biol. Chem. 269, 1459514600.
  • Shull, M.M., Ormsby, I., Kier, A.B., et al. (1992) Targeted disruption of the mouse transforming growth factor-β 1 gene results in multifocal inflammatory disease. Nature 359, 693699.
  • Takaoka, A., Mitani, Y., Suemori, H., et al. (2000) Cross talk between interferon-γ and -α/β signaling components in caveolar membrane domains. Science 288, 23572360.
  • Ulloa, L., Doody, J. & Massagué, J. (1999) Inhibition of transforming growth factor-β/SMAD signalling by the interferon-γ/STAT pathway. Nature 397, 710713.
  • Vodovotz, Y., Bogdan, C., Paik, J., Xie, Q.W. & Nathan, C. (1993) Mechanisms of suppression of macrophage nitric oxide release by transforming growth factor β. J. Exp. Med. 178, 605613.
  • Vodovotz, Y., Geiser, A.G., Chesler, L., et al. (1996) Spontaneously increased production of nitric oxide and aberrant expression of the inducible nitric oxide synthase in vivo in the transforming growth factor β1 null mouse. J. Exp. Med. 183, 23372342.
  • Wierenga, A.T., Schuringa, J.J., Eggen, B.J., Kruijer, W. & Vellenga, E. (2002) Downregulation of IL-6-induced STAT3 tyrosine phosphorylation by TGF-β1 is mediated by caspase-dependent and -independent processes. Leukemia 16, 675682.
  • Wrana, J.L., Attisano, L., Wieser, R., Ventura, F. & Massagué, J. (1994) Mechanism of activation of the TGF-β receptor. Nature 370, 341347.
  • Wu, H., Klingmüller, U., Besmer, P. & Lodish, H.F. (1995) Interaction of the erythropoietin and stem-cell-factor receptors. Nature 377, 242246.