Dual Function of Interleukin-1 (cid:1) for the Regulation of Interleukin-6-induced Suppressor of Cytokine Signaling 3 Expression*

Interleukin-6 (IL-6) exerts pro- as well as anti-inflam-matory activities in response to infection, injury, or other stimuli that affect the homeostasis of the orga-nism. IL-6-induced expression of acute-phase protein genes in the liver is tightly regulated through both IL-6-induced feedback inhibitors and the activity of pro-inflammatory cytokines such as tumor necrosis factor (cid:2) and interleukin-1 (cid:1) . In previous studies mechanisms for how IL-1 (cid:1) counteracts IL-6-dependent acute-phase protein gene induction have been proposed. Herein we analyzed IL-1 (cid:1) -mediated regulation of IL-6-induced expression of the feedback inhibitor SOCS3. In hepatocytes IL-1 (cid:1) alone does not induce SOCS3 expression, but it counteracts SOCS3-promoter activation in long term studies. Surprisingly, short term stimulation revealed IL-1 (cid:1) to be a potent enhancer of

The acute-phase reaction in response to inflammation, infection, or injury starts with the release of pro-inflammatory cytokines such as interleukin (IL) 1 -1␤ or tumor necrosis factor (TNF)-␣. In the later phase, IL-6, the major mediator of acute-phase protein induction in the liver, is expressed and secreted into the circulation. IL-6 exerts pro-as well as anti-inflammatory activities. IL-1␤ has a co-regulatory function on IL-6induced gene expression: Several IL-6-induced acute-phase proteins such as serum amyloid A, C-reactive protein, and ␣ 1 -acid glycoprotein are positively regulated by IL-1␤. On the other hand, fibrinogen, ␣ 1 -antichymotrypsin, ␣ 2 -macroglobulin, and thiostatin expression are suppressed by IL-1␤ (for review see Refs. 1 and 2).
IL-6 signaling occurs via JAK/STAT and MAPK cascades (for review see Refs. 3 and 4) and can be negatively regulated by SH2-containing protein-tyrosine phosphatase, by the protein inhibitor of activated STAT3 (PIAS3), or by the suppressors of cytokine signaling SOCS1 and SOCS3 (5). SOCS3 is supposed to act as a negative regulator of JAK/STAT-dependent IL-6signaling by binding to tyrosine 759 of gp130, resulting in reduced activation/phosphorylation of JAKs, gp130, and STATs (6,7). Inhibition of signaling by SOCS3 was found to be independent of SH2-containing protein-tyrosine phosphatase, although both proteins bind to the same site within the cytoplasmic part of the signal transducer gp130 (8). Besides being induced by IL-6, SOCS3 expression is also mediated by proinflammatory mediators such as lipopolysaccharide, TNF␣, and CpG-DNA (9 -11). Recent findings demonstrate that in SOCS3-deficient macrophages IL-6 induces an anti-inflammatory response. Thus, the rather restricted view of SOCS3 as an inhibitor of IL-6-induced JAK/STAT signaling should be extended to the role as a modulator of biological functions of IL-6 (12)(13)(14).
Recently, it has been shown in hepatocytes that inhibition of IL-6-induced expression of ␣ 2 -macroglobulin and fibrinogen by IL-1␤ involves the activation of NF-B and competition of NF-B and STAT3 for overlapping binding sites (20 -22). It has been suggested that other IL-6-induced proteins such as the ␣ 1 -antichymotrypsin and SOCS3 may be regulated similarly (22).
With respect to the importance of SOCS3 for the regulation of IL-6 signaling (12)(13)(14), we studied the concerted regulation of SOCS3 expression by IL-6 and IL-1␤ in more detail. Here we show that IL-1␤ does not induce SOCS3 in hepatocytes. However, a time-dependent dual regulatory function of IL-1␤ on IL-6-induced SOCS3 expression was found: we discovered an enhanced SOCS3 expression by IL-1␤ immediately after stimulation. In contrast, SOCS3-promoter activation was suppressed by IL-1␤ at later time points. Both functions of IL-1␤ were dependent on the activation of NF-B. Surprisingly, in the absence of NF-B activation early SOCS3 expression even decreased in response to IL-1␤. This inhibitory function seems to be compensated by the strong NF-B-mediated positive regulatory activity in the initial phase of IL-1␤/IL-6 co-stimulation and IL-1␤-dependent stabilization of SOCS3 mRNA.

EXPERIMENTAL PROCEDURES
Materials-Restriction enzymes were purchased from Roche Molecular Biochemicals (Mannheim, Germany); Taq polymerase was from Hybaid (Heidelberg, Germany); oligonucleotides were obtained from MWG-Biotech (Ebersberg, Germany); Dulbecco's modified Eagle's medium (DMEM) and DMEM/nutrient mix F-12 were from Invitrogen; fetal calf serum was from Seromed (Berlin, Germany); recombinant human IL-6 was prepared as described (23); recombinant human IL-1␤ was from Roche Applied Science. Recombinant erythropoietin (Epo) was a generous gift of Drs. B. Hilger and K. H. Sellinger (Roche Applied Science). The internal control plasmid DNA pCH110 was from Amersham Biosciences, cycloheximide from Sigma-Aldrich, and actinomycin D from Calbiochem. Antibodies for detection of SOCS3 (SOCS3-M20) in Western blots and for precipitation of p65 (p65-C20) in ChIP assays were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The SOCS3-M20 antibodies were biotinylated as described by Pierce (Rockford, IL). Antibodies for immunoprecipitation of SOCS3 were kindly supplied by J. Johnston (Belfast, Northern Ireland). Antibodies to the Tyr-701-phosphorylated STAT1 were obtained from Cell Signaling Technology (Frankfurt/Main, Germany), and antibodies to STAT1 and Ser-727-phosphorylated STAT1 were from Upstate Laboratories Inc. (Lake Placid, NY). HRP-coupled secondary antibodies were form DAKO (Hamburg, Germany).
Construction of Expression Vectors-The SOCS3-reporter construct contains the region Ϫ511 to ϩ929 of the murine SOCS3 promoter (kindly provided by Shlomo Melmed, Los Angeles, CA). The IB␣(S/A) expression vector encodes for an IB␣ mutant, where serines 32 and 36 have been mutated to alanine to avoid phosphorylation and subsequent degradation of IB␣ (Upstate Laboratories Inc.). The SIE-tk-Luc construct, containing two copies of the STAT consensus binding sequence from the c-fos promoter upstream of a thymidine kinase minimal promoter, was kindly provided by Hugues Gascan (Angers, France) (24). pSuper-SOCS3 was cloned according to SOCS3-specific siRNA sequence described by Leung et al. (25). The following double-stranded oligonucleotide has been inserted into the BglII/HindIII opened pSuper: forward primer, 5Ј-GATCCCCGACCCAGTCTCGGGACCAAGAATTC-AAGAGATTCTTGGTCCCAGACTGGGTCTTTTTGGAAA-3Ј and reverse primer, 5Ј-AGCTTTTCCAAAAAGACCCAGTCTGGGACCA-AGAATCTCTTGAATTCTTGGTCCCAGACTGGGTCGGG-3Ј.
For transfection of HepG2 cells, cells were grown on 6-cm plates to 30% confluency and transfected in DMEM supplemented with 10% fetal calf serum. Calcium phosphate precipitation was performed with 3 g of reporter construct, 2 g of ␤-galactosidase expression vector (pCR3lacZ; Amersham Biosciences). Transfections were adjusted with control vectors to equal amounts of DNA. Cells were incubated with the precipitate for 16 h, washed twice with phosphate-buffered saline, and let for additional 8 -10 h in fresh medium. For reporter gene assays, including SOCS3 siRNA, HepG2 cells were grown in 6-well plates and transfected with 1.2 g of pSuper-SOCS3 or control vector, 1.4 g of pRcCMV-(EpoR/gp130) (26), 0.8 g of SIE-tk-luciferase-reporter gene, and 0.6 g of ␤-galactosidase expression vector (pCR3lacZ) using Fu-GENE 6 reagent (Roche Applied Science) according to the manufacturer's instructions. Cells were stimulated with cytokines for the times indicated in the figures. Cell lysis and luciferase assays were conducted using the luciferase kit (Promega, Madison, WI), as described by the manufacturer's instructions. All expression experiments were done at least in triplicate. Luciferase activity values were normalized to transfection efficiency monitored by the co-transfected ␤-galactosidase expression vector. Error bars represent ϮS.D.
Retroviral Gene Transfer/Generation of HepG2 Cells Stably Expressing IBa(S/A)-The pCFG5-IEGZ retroviral vector allowing expression of GFP from the same mRNA as the gene of interest has been described earlier (27). NX producer cells plated at a density of 1 ϫ 10 6 per 10-cm plate were transfected using the calcium phosphate precipitation method with 10 g of vector or pCFG5-IEGZ IB␣(S/A) plasmid DNA (28). Twenty-four hours later transfection efficiencies were determined by monitoring GFP expression. Subsequently, 1 mg/ml Zeocin (Invitrogen) was added to the cells, which were then grown in the presence of this agent for another 2 weeks until all the cells were positive for GFP. Retroviral infection of HepG2 cells was performed with supernatants from NX producer cells essentially as described for other cells (28). Briefly, supernatants were filtered through a 0.45-m filter, and 5 g/ml Polybrene (Sigma) was added to the filtrate. Thereafter, medium of HepG2 cells plated in 6-well plates was replaced by NX cell supernatant containing the IB␣(S/A)-cDNA containing retrovirus (giving HepG2-IB␣(S/A)) or control virus (giving HepG2-mock). Culture plates were centrifuged at 1000 ϫ g for 3 h. Medium was then replaced, and HepG2 cells were subsequently cultured in the presence of the selection marker Zeocin (Invitrogen) until all cells were positive for GFP.
Nuclear Extract Preparation and Electrophoretic Mobility Shift Assays-Preparation of nuclear extracts and EMSAs were performed as described previously (29). The protein concentration of the nuclear extracts was determined with a Bio-Rad reagent. 8 g of protein was incubated with a surplus of double-stranded 32 P-labeled B site (5Ј-GATCCATGGGGAATTCCCCATG-3Ј) for binding NF-B. The pro-tein⅐DNA complexes were separated on a 4.5% polyacrylamide gel containing 7.5% glycerol in 0.25-fold TBE (20 mM Tris base, 20 mM boric acid, 0.5 mM EDTA, pH 8) at 20 V/cm for 4 h. Gels were fixed in 10% methanol, 10% acetic acid, and 80% water for 1 h, dried, and autoradiographed.
Immunoprecipitation and Immunoblot Analyses-For immunoprecipitation, cells from a 60% confluent 10-cm plate were lysed in 500 l of radioimmune precipitation assay buffer (50 mM Tris/HCl, pH 7.5; 150 mM NaCl; 0.5% Nonidet P-40; 15% glycerol; 1 mM EDTA; 1 mM NaF; 1 mM Na 3 VO 4 ). Before adding antibodies protein amounts were adjusted to be equal. Buffers were supplemented with 10 g/ml each of aprotinin, pepstatin, and leupeptin. Immune complexes were separated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane. Antigens were detected by incubation with specific primary antibodies and HRP-coupled secondary antibodies (DAKO). SOCS3 detection was performed with biotinylated SOCS3-specific primary antibodies and HRPcoupled streptavidin (1:5000) (Pierce). The membranes were developed with an enhanced chemiluminescence kit from Amersham Biosciences. For immunoprecipitation and Western blot analyses of SOCS3 siRNAtreated cells, HepG2 cells were cultivated on 10-cm dishes and transfected with 2.5 g of pRcCMV-(EpoR/gp130), and 3.5 g of pSuper-SOCS3 or control vector using FuGENE 6 reagent. Stimulation was performed with 7 units/ml Epo and/or 100 units/ml IL-1␤.
Northern Blot Analysis-Total RNA from 5 ϫ 10 6 cells was isolated with the total RNA kit from Qiagen (Hilden, Germany) as described in the manufacturer's instructions. 10 g of RNA was mixed with denaturing buffer (20 mM MOPS, 1 mM EDTA, 2.2 M formaldehyde, 50% formamide, bromphenol blue, 4 mM sodium acetate, pH 7.0) and denatured at 66°C for 5 min. RNA was subjected to electrophoresis on a 1% agarose gel containing 0.2 M formaldehyde, 20 mM MOPS, 1 mM EDTA, 5 mM sodium acetate, pH 7.0. The RNA was blotted by capillary blot method to NitroPlus transfer membrane (Micron Separations, Westboro, MA). RNA was immobilized at 80°C for 2 h. The membranes were blocked in pre-hybridization buffer (1 M NaCl; 10% dextran sulfate; 1% SDS) at 66°C for at least 2 h. Hybridization was performed at 66°C with labeled cDNAs overnight. The XbaI/Mlu1 fragment of pEF-FLAG-I/mSOCS3 (kindly provided by D. Hilton, Melbourne, Australia) was labeled with the Random Primed DNA Labeling kit (Roche Applied Science). The membranes were washed twice in 2ϫ NaCl/citrate buffer, 0.2% SDS at room temperature for 20 min. Then washing was performed in 0.2ϫ NaCl/citrate buffer, 0.2% SDS at 66°C for 20 min. The membranes were dried and subjected to autoradiography using an intensifying screen at Ϫ80°C. For counterstaining, membranes were blocked and re-probed with an EcoR1/XhoI DNA fragment of the GAPDH-cDNA.
Chromatin Immunoprecipitation-HepG2 cells were grown to 80% confluence and serum-starved for 24 h. Where indicated, cells were treated with 100 units/ml IL-6 or IL-1␤ for 20 min. For IL-1␤/IL-6 co-stimulation IL-1␤ was given 10 min prior to IL-6. After stimulation, cells were fixed by adding formaldehyde to a final concentration of 1% (HCHO, from a 37% HCHO-10% methanol stock, Merck, Darmstadt, Germany) at room temperature for 10 min. Cells were washed three times with ice-cold phosphate-buffered saline for 10 min, centrifuged, re-suspended in L1 buffer (50 mM Tris/HCl, pH 8.0; 2 mM EDTA; 0.1% Nonidet P-40; and 10% glycerol) supplemented with protease inhibitors for 5 min. Nuclei were centrifuged at 3000 rpm and re-suspended in L2 buffer (50 mM Tris/HCl, pH 8.0; 1% sodium dodecyl sulfate; and 5 mM EDTA). Chromatin was sheared by sonication (4 ϫ 16 s at about 50% output, Branson Sonifier). Debris was eliminated by centrifugation, and the supernatant was diluted (1:10) in dilution buffer (50 mM Tris/HCl, pH 8.0; 0.5% Nonidet P-40; 0.2 M NaCl; and 5 mM EDTA). Immunoprecipitation was performed with specific antibodies overnight. 20 l of Protein A-Sepharose (Amersham Biosciences) was added at 4°C for 45 min. Immune complexes were washed three times for 5 min with high salt buffer (20 mM Tris/HCl, pH 8.0; 0.1% SDS; 1% Nonidet P-40; 2 mM EDTA; and 0.5 M NaCl). Complexes were washed three times in 1ϫ TE buffer (10 mM Tris/HCl, pH 8.0, 1 mM EDTA) and finally extracted in 1ϫ TE buffer containing 2% SDS. Cross-linking of the protein with DNA was reverted by heating at 65°C overnight. Supernatants were then incubated with proteinase K (100 g at 50°C for 2 h). DNA was extracted with phenol-chloroform, precipitated in ethanol, and resolved in 50 l of TE buffer. 2.5 l of DNA was used for PCR, with a specific primer for the human IB␣ or SOCS3 promoter. The following promoter-specific primers were used: 5Ј-GACGACCCCAATTCAAATCG-3Ј and 5Ј-TCAGGCTCGGGGAATTTCC-3Ј for IB␣ promoter and 5Ј-GCT-CAGCCTTTCTCTGC-3Ј and 5Ј-CGAAGCGGCAGCAGC-3Ј for the Ϫ131 to ϩ32 region of the human SOCS3 promoter and 5Ј-CAGGGT-TGGCAAAGAAC-3Ј and 5Ј-ACCTGGAGAGCCTC-3Ј for the Ϫ722 to Ϫ420 region of the human SOCS3 promoter (numbering relative to the proposed transcription starting site).
The PCR reaction was performed for 35 cycles in a total volume of 25 l (1.25 unit of Taq DNA polymerase; 100 ng of each primer; 200 M dNTP; 2.5 l of 10ϫ Taq buffer). PCR conditions were as follows: the first denaturation was for 94°C for 180 s then 35 cycles consisting of denaturation (94°C for 45 s), annealing (60°C for 60 s), and elongation (72°C for 60 s); the final elongation was carried out at 72°C for 10 min. DNA fragments were separated by electrophoresis in a 2% agarose gel. DNA was visualized with ethidium bromide.

IL-1␤
Counteracts IL-6-dependent SOCS3-promoter Activation-Our own initial experiments suggested that SOCS3 may be negatively regulated by IL-1␤ (22). To analyze the collabo- were transfected with a reporter gene construct containing the SOCS3 promoter fused to the firefly luciferase gene. An expression vector for ␤-galactosidase was co-transfected for monitoring transfection efficiency. After transfection, cells were stimulated with 100 units/ml IL-1␤ or IL-6 (100 units/ml) or both for 16 h. Luciferase activity in cellular extracts of these cells was determined and normalized to ␤-galactosidase activity, as outlined under "Experimental Procedures." B, HepG2-mock (left panel) and HepG2-IB␣(S/A) cells (right panel) were stimulated with 100 units/ml IL-1␤ for the times indicated. Cellular extracts were performed, and the same amounts of proteins were subjected to SDS-PAGE. Expression of endogenous IB␣ and exogenous IB␣(S/A) was monitored by Western blotting with an antibody specific for IB␣. C, HepG2-mock (left panel) and HepG2-IB␣(S/A) cells (right panel) were stimulated with 100 units/ml IL-1␤ for the times indicated. Nuclear extracts were prepared, and the same amounts of protein were subjected to EMSA with the B-site-containing DNA fragment. rative regulation of SOCS3 by IL-6 and IL-1␤, we performed promoter/reporter gene assays with an SOCS3-luciferase reporter construct in human HepG2 hepatoma cells. Fig. 1A (left part) shows that the SOCS3 promoter was responsive to stimulation with IL-6 (second bar), whereas the promoter was not inducible by IL-1␤ (third bar). Instead, IL-1␤ counteracted IL-6-dependent promoter activation, because co-incubation with IL-1␤ significantly reduced IL-6-induced reporter activity (compare bars 2 and 4).
Because NF-B is a major mediator of IL-1␤ signaling, we asked whether IL-1␤ may act through NF-B. Therefore, we tested whether expression of a non-degradable IB␣(S/A) inhibitor could overcome the effect of IL-1␤. As shown in the right part of Fig. 1A, stable expression of IB␣(S/A) strongly affected the inhibitory action of IL-1␤ on SOCS3-promoter-driven reporter expression.
Expression of IB was monitored by Western blotting. Fig.  1B shows that degradation of endogenous IB␣ occurred 2 min after IL-1␤ treatment of HepG2-mock cells. IL-1␤-induced IB␣ re-expression was detectable after 20 min (Fig. 1B, left  panel). In contrast, expression of the endogenous IB␣ as well as of the exogenous IB␣(S/A) mutant was not affected by IL-1␤ treatment in HepG2-IB␣(S/A) cells (right panel). IL-1␤-induced NF-B DNA-binding activity was monitored by EMSA with a B-site-containing DNA-probe (Fig. 1C). IL-1␤-induced NF-B binding to the B site in HepG2-mock cells was detectable already after 1 min and further increased during the following 40 min (left panel). In contrast, no NF-B binding in response to IL-1␤ was detectable in HepG2-IB␣(S/A) cells (right panel) demonstrating an efficient blockade of NF-B activation in these cells. These data suggest that IL-1␤ acts negatively on IL-6-mediated SOCS3 promoter activation through an NF-B-dependent pathway.
The Inhibitory Activity of IL-1␤ on SOCS3-promoter Activation Is Time-dependent-SOCS3 is an immediate early IL-6 response gene. Therefore, we decided to analyze the kinetics of IL-1␤-induced inhibition of the SOCS3-reporter in the initial phase of expression. SOCS3-promoter-driven reporter gene expression was assessed 50, 80, 180, and 240 min after stimulation of HepG2 cells with IL-6. Reporter gene activity increased within the first 3 h but did not rise further up to 4 h of IL-6 stimulation. To our surprise, gene induction was most efficiently inhibited by IL-1␤ in cells stimulated for more than 3 h with IL-6 ( Fig. 2, panels 3 and 4), whereas cells stimulated for 80 min were less sensitive to IL-1␤ (Fig. 2,  panel 2). A reproducible slight increase of reporter gene induction was observed in cells stimulated with IL-6 for only 50 min (Fig. 2, panel 1).
IL-1␤ Mediates Enhanced SOCS3 Expression in Response to IL-6 -To confirm the significance of the IL-1␤-dependent increase of SOCS3-promoter activation observed in Fig. 2, we extended the duration of IL-1␤ pre-stimulation up to 40 min prior to addition of IL-6 ( Fig. 3A). Stimulation of HepG2 cells with IL-6 for 50 min led to a 2-fold induction of reporter gene expression (left panel). As described in Fig. 2 a small but significant (p ϭ 0.007) increase of reporter gene expression was detectable when IL-1␤ treatment started 10 min prior to IL-6 stimulation (bars 2 and 4). This increase was more obvious when IL-1␤ was given 20, 30, or 40 min before addition of IL-6 ( bars 6, 8, and 10). These data suggest that IL-1␤ acts negatively on SOCS3 expression only in response to long term IL-6 stimulation, whereas a positive regulatory function of IL-1␤ is obvious early after stimulation.
To inspect a putative positive effect of IL-1␤ on early IL-6induced SOCS3 induction, we analyzed expression of IL-6induced endogenous SOCS3 in response to IL-1␤. Intentionally, in the following experiments only the initial phase of SOCS3 induction has been analyzed to exclude putative secondary effects mediated by SOCS3 itself.
We investigated whether the supporting activity of IL-1␤ on IL-6-induced SOCS3 expression also affects SOCS3 mRNA levels in the initial phase of IL-6 stimulation. SOCS3 mRNA was isolated from HepG2 cells stimulated for up to 70 min with IL-6 in the presence or absence of IL-1␤ and analyzed by Northern

IL-6-mediated SOCS3 Expression in Concert with IL-1␤
blotting (Fig. 3B). IL-6 induced remarkable amounts of SOCS3 mRNA already after 30 min with a peak at 45 min after IL-6 stimulation (left part). In the presence of IL-1␤ SOCS3 mRNA levels were significantly higher and therefore detectable already 15 min post stimulation.
SOCS3 protein expression was determined by immunoprecipitation and subsequent Western blotting immediately after stimulation with IL-6 ( Fig. 3, C and D). SOCS3 protein was detectable after IL-6 stimulation for 40 min and further increased during the following 20 min. Indeed, IL-1␤ treatment led to enhanced SOCS3-protein expression (compare lanes with and without IL-1␤ in Fig. 3C). The activity of IL-1␤ on IL-6induced SOCS3 expression was further analyzed dose dependently. Fig. 3D shows an IL-1␤ dose-dependent increase in SOCS3 expression in the early stage of IL-6-induced SOCS3 expression. From these data we conclude that IL-1␤ enhances time-and dose-dependent IL-6-induced endogenous SOCS3 expression.
NF-B Is Crucial for the Enhancing Effect of IL-1␤-on IL-6dependent Expression of Endogenous SOCS3-To clarify whether NF-B is crucial for IL-1␤-dependent enhancement of IL-6-induced SOCS3 expression, we monitored the amount of endogenous SOCS3 protein in HepG2 cells stably expressing the non-degradable IB␣(S/A) mutant, which represents an inhibitor of NF-B activation (Fig. 4A). Consistent with the data in Fig. 3, IL-1␤ enhanced IL-6-induced SOCS3 expression in HepG2 cells. Stimulation with IL-1␤ alone did not lead to SOCS3 expression. Surprisingly, in cells expressing IB␣(S/A), IL-1␤ was not simply ineffective to enhance IL-6-dependent SOCS3 expression but, rather, reduced SOCS3 protein amounts (compare lanes 6 and 8). This finding is not due to an effect of the retroviral gene transfer, because cells infected with control virus behave similar to wild-type cells (lower panel of Fig. 4A).
These observations were also confirmed for SOCS3 mRNA expression (Fig. 4B). IL-6 alone induced SOCS3 mRNA expression in cells expressing the IB␣(S/A) inhibitor 30 min post stimulation (lane 5). After 60 min of IL-6 stimulation SOCS3 mRNA levels further increased (lane 6). In line with SOCS3 protein shown in Fig. 4A, IL-1␤ reduced SOCS3 mRNA in cells expressing the IB␣(S/A) inhibitor. These data indicate that SOCS3 expression is positively regulated by IL-1␤ through an NF-B-dependent mechanism; however, in the absence of NF-B activation, IL-1␤ rather represses SOCS3 expression indicating a dual regulatory role of IL-1␤ on IL-6-induced SOCS3 expression. Fig. 2, but pre-stimulation with IL-1␤ was extended to up to 40 min prior to stimulation with IL-6 for 50 min. B-D, induction of IL-6-dependent expression of endogenous SOCS3 mRNA (B) and SOCS3 protein (C and D) was analyzed in the presence or absence of IL-1␤ in HepG2 cells. To exclude putative secondary effects mediated by SOCS3 itself, only the initial phase of SOCS3 induction was considered in these experiments. B, kinetics of SOCS3 mRNA expression. HepG2 cells were stimulated with 100 units/ml IL-6 for 15-70 min (left part). For co-stimulation, IL-1␤ (100 units/ml) was given 10 min prior to IL-6 (right part). Total mRNA was isolated and analyzed by Northern blotting with a specific probe for SOCS3 (upper panel). For loading controls, the membrane was re-probed with a GAPDH-specific probe (lower panel). C, kinetics of SOCS3 protein induction. Cells were treated with 100 units/ml IL-6 or IL-1␤ for the times indicated. For IL-1␤/IL-6 co-stimulation IL-1␤ was given 10 min prior to IL-6. Cellular extracts were prepared and lysates were incubated with antibodies against SOCS3 (IP: SOCS3). Protein/antibody complexes were separated by SDS-PAGE and analyzed by Western blotting with a second type of SOCS3 antibody (Santa Cruz Biotechnology), which was biotinylated before usage (IB: SOCS3). Staining of SOCS3 proteins was performed with streptavidin-coupled HRP. D, dose-dependent effect of IL-1␤ on IL-6-induced SOCS3 expression. HepG2 cells were stimulated with IL-6 (100 units/ml) for 60 min. IL-1␤ (0, 100, 400, or 800 units/ml) was given 10 min prior to IL-6. SOCS3 protein was measured as described above. (30) described an IL-1␤induced STAT1 serine phosphorylation that in turn leads to enhanced transcriptional activity of STAT1. We asked whether IL-1␤ exerts enhanced SOCS3 expression through STAT1-serine phosphorylation. Fig. 5A shows that IL-1␤ is a strong mediator of STAT1-serine phosphorylation in HepG2 cells (left panel, lane 3), whereas IL-6 mediates STAT1-tyrosine phosphorylation but was only a comparable weak inductor of STAT1-serine phosphorylation (right panel, lane 2). Being aware of the fact that NF-B activation is crucial for the enhancing effect of IL-1␤ on SOCS3 expression, we analyzed STAT1-serine phosphorylation in HepG2 cells expressing the non-degradable IB␣(S/A) inhibitor (Fig. 5B). The right part of the figure shows that IL-1␤-dependent serine 727 phosphorylation of STAT1 did not depend to NF-B activation. These data suggest that STAT1-serine phosphorylation is not responsible for enhanced SOCS3 expression after treatment with IL-1␤, because the IL-1␤-dependent increase in SOCS3 expression, but not the IL-1␤-induced STAT1-serine phosphorylation, requires NF-B activation.

IL-1␤-induced Serine Phosphorylation of STAT1 Does Not Depend on NF-B Activation and Thus Is Not Crucial for the Enhancing Effect of IL-1␤ on IL-6-induced SOCS3 Expression-Recently, Stark and colleagues
NF-B Activated by IL-1␤ Does Not Act through Binding to SOCS3 Promoter-Because NF-B plays a crucial role for IL-1␤-dependent enhancement of IL-6-induced SOCS3 expression, it is reasonable to speculate that NF-B may act directly on the SOCS3 promoter. Two regions within the SOCS3 promoter were found by the MatInspector (Genomatix Software, Mü nchen, Germany) to harbor putative B sites. To monitor IL-1␤-dependent NF-B binding to these SOCS3 promoter regions we performed chromatin immunoprecipitation assays with p65-specific antibodies (Fig. 6A). As a positive control, binding of NF-B to the IB␣ promoter was determined (upper panel, left four bands). The applicability of the selected PCR primers was confirmed by the input control (four right bands in both panels). Unspecific binding of promoter fragments was controlled by a ChIP assay without antibodies (w/o antibodies). Fig. 6A shows that no NF-B binding to the SOCS3 promoter was detectable in response to IL-1␤, whereas NF-B binding was confirmed for the IB␣-promoter. These results suggest that NF-B does not act directly through binding to the SOCS3 promoter but more likely indirectly by the induction of secondary effector proteins.
The Enhancing Effect of IL-1␤ on IL-6-dependent SOCS3 Expression Requires Protein Synthesis-To further verify the idea that IL-1␤ does not act directly on the SOCS3 promoter but rather through the induction of downstream effector proteins, we checked whether the enhancing activity of IL-1␤ on IL-6-dependent SOCS3 expression requires newly synthesized proteins. We analyzed the effect of cycloheximide (CHX) on IL-6-induced SOCS3 mRNA expression in presence of IL-1␤. Fig. 6B shows that CHX treatment enhanced IL-6-induced SOCS3 expression probably due to the blockade of SOCS3 feedback inhibition (compare lanes 3 and 7). Furthermore, co-stimulation with IL-1␤ enhanced IL-6-dependent SOCS3 expression in Me 2 SO-treated cells (compare lanes 3 and 4) but not in CHX-treated cells (compare lanes 7 and 8). CHX treatment rather reduced SOCS3 mRNA levels in response to IL-1␤, similar as shown in response to the block of NF-B activation in Fig. 4. To exclude any effect of CHX treatment on IL-1␤-induced NF-B activation, we monitored NF-B DNA-binding activity in the presence and absence of CHX in EMSA. As shown in Fig. 6C CHX treatment did not affect NF-B activation. These data support the idea that IL-1␤ acts through NF-B to induce the synthesis of one or more regulatory proteins that finally act positively on the SOCS3 expression.
IL-1␤ Does Not Affect SOCS3 Protein Degradation but Stabilizes SOCS3 mRNA-Instead of increasing IL-6-dependent induction of SOCS3, IL-1␤ could also affect the stability of SOCS3 protein or mRNA to increase SOCS3 levels in the cell. To test whether IL-1␤ stabilizes SOCS3 protein we compared the time-dependent loss of SOCS3 in the presence or absence of

IL-6-mediated SOCS3 Expression in Concert with IL-1␤
IL-1␤ (Fig. 7A). SOCS3 protein expression in HepG2 cells was induced by stimulation with IL-6 for 40 min. Afterward, IL-6 was removed and cycloheximide was added to the medium to avoid ongoing protein synthesis. The loss of SOCS3 protein within the following 90 min was compared in the absence (left part) or presence (right part) of IL-1␤. No obvious effect of IL-1␤ treatment on the degradation of SOCS3 was observed, suggesting that IL-1␤ does not act by stabilizing SOCS3 protein.
Another possibility would be that IL-1␤ stabilizes SOCS3 mRNA. Therefore, we performed analogous experiments as described for Fig. 7A to analyze the decrease of SOCS3 mRNA in the absence and presence of IL-1␤ (Fig. 7B). SOCS3 mRNA was induced by stimulation of HepG2 cells with IL-6 for 30 min. Ongoing mRNA synthesis was blocked by eliminating IL-6 and adding actinomycin D into the medium. The blockage of SOCS3 transcription in the presence of actinomycin D is shown in the right part of Fig. 7B. The decrease of SOCS3 mRNA within the following 90 min was monitored in the absence (left part) or presence (right part) of IL-1␤ by Northern blotting (upper panel). Whereas SOCS3 mRNA is rapidly degraded in the absence of IL-1␤, SOCS3 mRNA was stabilized in cells treated with IL-1␤. These results suggest that IL-1␤ is able to increase SOCS3 expression by stabilizing its mRNA.
SOCS3 Expression Contributes to the Inhibitory Activity of IL-1␤ on gp130-mediated Signaling-The role of SOCS3 expression in the context of cross-talk between IL-1␤ and IL-6 signaling was studied by using SOCS3 siRNA. We transiently transfected HepG2 cells with vectors for SOCS3 siRNA. To be able to specifically stimulate the population of transfected cells, we co-transfected expression vectors for chimeric receptors composed of the extracellular part of the EpoR and the transmembrane and cytoplasmic part of gp130. These well established receptors allowed us to analyze gp130-dependent signaling in response to Epo (26,31). In Fig. 8A we first confirmed the efficiency of the SOCS3 siRNA construct generated. Stimulation of the transfected HepG2 cells with Epo led to SOCS3 expression (lane 2). In contrast, cells additionally expressing SOCS3 siRNA showed drastically reduced SOCS3 levels in response to Epo, indicating that the SOCS3 siRNA is functional. We then studied the effect of IL-1␤ on Epo-induced reporter expression in the absence or presence of SOCS3 siRNA (Fig. 8B). Epo stimulation led to a significant induction of the STAT3-driven reporter construct (second bar), whereas the expression was inhibited to 50% the presence of IL-1␤ (compare second and fourth bars). These results are quite similar to those already presented in Fig.  1A. Cells additionally expressing SOCS3 siRNA showed a drastically increased reporter expression due to the lack of SOCS3-mediated feedback inhibition (compare bars 2 and 6). Interestingly, IL-1␤ was hardly potent to inhibit expression of the reporter gene in these cells (compare bars 6 and 8) suggesting SOCS3 plays an important role in the IL-1␤-dependent inhibition of IL-6-induced gene expression.  Fig. 6. Chromatin immunoprecipitation was performed with antibodies specific for the p65 subunit of NF-B. Binding of NF-B to the SOCS3 promoter was tested by PCR with two different pairs of SOCS3 promoter-specific primers. PCR conditions for these primers were tested for amplification of SOCS3 promoter fragments out of the input chromatin without immunoprecipitation. Efficiency of immunoprecipitation of NF-B⅐DNA complexes was monitored with PCR primers specific for the IB promoter. B, after pretreatment with Me 2 SO (DMSO) or cycloheximide (CHX, 25 M) for 20 min, HepG2 cells were stimulated with IL-6 (100 units/ml), IL-1␤ (100 units/ml), or both for an additional 60 min. For IL-1␤/IL-6 co-stimulation, IL-1␤ was given 10 min prior to IL-6. Total mRNA was isolated and analyzed by Northern blotting with a specific probe for SOCS3 (upper panel). For loading controls, the membrane was re-stained with a GAPDH-specific probe (lower panel). C, HepG2 cells were stimulated with 100 units/ml IL-1␤ in the presence or absence of cycloheximide. Nuclear extracts were prepared and tested for NF-B DNA binding to the B-site by EMSA.
IL-6-mediated SOCS3 Expression in Concert with IL-1␤ 45286 DISCUSSION IL-6 is the major mediator of acute-phase protein gene induction in the liver in response to inflammation, injury, and other stimuli (32,33). The inflammatory cascade underlying the acute-phase response is largely controlled by the action of different mediators (34,35). Monocytes and macrophages control hepatic IL-6 signaling through the release of pro-inflammatory cytokines such as IL-1␤ and TNF-␣. Several mechanisms for IL-1␤ and TNF-␣ action on IL-6 signaling have been proposed, and additional complexity stems from cell-specific differences. In macrophages, TNF␣ has been suggested to mediate repression of IL-6 signaling through the induction of SOCS3 protein expression (9). Additionally, de novo protein synthesis-independent mechanisms for inhibition of IL-6 signal transduction in macrophages have been suggested (15,16).
In contrast to macrophages, SOCS3 expression in hepatocytes is not induced by IL-1␤. Nevertheless, IL-1␤ affects IL-6-induced acute-phase gene expression. Several groups have proposed a crucial role of NF-B for IL-1␤-dependent repression of IL-6-induced gene induction in hepatocytes and suggest a competitive binding of NF-B and STAT3 for overlapping binding sites within the respective promoters (20 -22). Furthermore, IL-1␤ counteracts IL-6-dependent STAT3 activation through the activation of p38 MAPK (15). The induction of the SOCS3 feedback inhibitor by IL-6 is also affected by p38 (19). In the study presented here, we found that IL-1␤ increases IL-6-induced SOCS3 expression. Blocking SOCS3 expression impairs the inhibitory action of IL-1␤ on gp130-dependent gene expression (Fig. 8), indicating an important role of SOCS3 for IL-1␤-mediated repression of IL-6-induced gene expression.
SOCS3 plays a key role in the regulatory network of pro-and anti-inflammatory cytokines (12)(13)(14). On one hand SOCS3 is induced by IL-6, similar to type 2 acute-phase proteins. On the other hand SOCS3 suppresses IL-6 signaling and may thus be a mediator of IL-1␤-dependent repression of acute-phase protein gene induction in the liver. Therefore, it was important to analyze the concerted regulation of SOCS3 expression by IL-1␤ and IL-6 in detail.
IL-1␤ was found to inhibit IL-6-induced SOCS3 promoter activation by activating NF-B in human hepatoma cells (Fig.  1). A similar inhibition of IL-6-mediated expression of acutephase protein genes upon IL-1␤ stimulation has been reported (20 -22, 34). Surprisingly, IL-1␤ did not exert its inhibitory function on the SOCS3 promoter immediately after stimulation. Instead, IL-1␤ showed an NF-B-dependent co-stimulatory activity (Figs. 2 and 3A). Importantly, this IL-1␤/IL-6synergism could also be confirmed for endogenous SOCS3 expression in a time-and dose-dependent manner for IL-1␤ (Fig. 3). It is intriguing to speculate that the ratio of stimulation and suppression of SOCS3 gene induction may explain the differential regulation of type 1 and type 2 acute-phase protein genes by IL-1␤.
Knowing that IL-1␤ alone does not induce SOCS3 gene expression in hepatocytes, the key question of how IL-1␤ signal-FIG. 7. IL-1␤ does not affect SOCS3 protein degradation but stabilizes SOCS3 mRNA. A, HepG2 cells were stimulated with 100 units/ml IL-6 for 40 min. Subsequently, cells were washed and cultivated for an additional 20 min in IL-6-free medium containing cycloheximide (25 M) for 20 min. Afterward, 100 units/ml IL-1␤ was added to the medium for the times indicated. SOCS3 protein was monitored as described for Fig. 3. B, HepG2 cells were stimulated with 100 units/ml IL-6 for 30 min. Subsequently, cells were washed and cultivated for additional 20 min in IL-6-free medium containing actinomycin D (4 M) to block transcription. Afterward, 100 units/ml IL-1␤ was added to the medium for the times indicated. SOCS3 mRNA (upper left panel) and GAPDH mRNA (lower left panel) were monitored as described for Fig. 3. To show that actinomycin D blocks IL-6-induced SOCS3 mRNA synthesis, HepG2 cells were stimulated with IL-6 (middle lane) or preincubated with actinomycin D (4 M) for 20 min and afterward stimulated with 100 units/ml IL-6 in the presence of actinomycin D for 30 min.

IL-6-mediated SOCS3 Expression in Concert with IL-1␤
ing affects IL-6-mediated SOCS3 expression needed to be answered. Blocking NF-B activation impaired the co-stimulatory function of IL-1␤ on SOCS3 expression (Fig. 4). Moreover, in hepatoma cells lacking NF-B activation, the response to IL-1␤ was completely reversed and IL-1␤ led to a reduced IL-6-dependent SOCS3 expression. These results demonstrate a crucial role of NF-B for the positive action of IL-1␤ on IL-6induced SOCS3 expression. Thus, prevention of NF-B activation revealed a so far hidden potential of IL-1␤ to inhibit SOCS3 expression (Fig. 4). The observed potential of IL-1␤ to inhibit SOCS3 expression in cells incapable of NF-B activation is likely due to the inhibition of STAT3 activation/tyrosine phosphorylation, because IL-1␤ seems to have a direct effect on STAT3 activation/tyrosine phosphorylation as well (15,17). However, despite these observations, we and others previously presented data suggesting that activated NF-B is at least partially able to inhibit IL-6-induced STAT3 binding to acutephase response elements within the promoters of the ␣ 2 -macroglobulin and the fibrinogen genes. This observation was at least partially explained by a competition of STAT3 and NF-B for binding to overlapping DNA-binding sites within the respective promoter elements (20 -22). Thus, IL-1␤ regulates the induction of IL-6-dependent genes (i) at the level of STAT3 activation by affecting all STAT3-dependent genes and (ii) at the level of promoter activation. The latter mechanism results in a specific down-regulation of genes containing overlapping NF-B/STAT3 binding sites within their promoters.
Enhanced SOCS3 expression could be a result of enhanced transcriptional activity of STAT1 and/or STAT3. Serine phosphorylation of both transcription factors has been shown to affect transcriptional of activity (36,37). Recently, Stark and collaborators described STAT1 serine phosphorylation in response to IL-1␤ (30). In respect to this study and the knowledge that NF-B activation is crucial for the enhancing effect of IL-1␤ on IL-6induced SOCS3 expression, we asked whether IL-1␤-induced serine phosphorylation of STAT1 is also dependent on NF-B (Fig. 5). We confirmed the results of Nguyen et al. (30) and additionally excluded a requirement of NF-B for STAT1 serine phosphorylation, suggesting that enhanced SOCS3 expression is not a result of STAT1 serine phosphorylation. The necessity of NF-B for the enhancing effect of IL-1␤ on SOCS3 induction and the computer-aided identification of two putative NF-B binding regions within the SOCS3 promoter led us to look for NF-B binding in this promoter area in vivo (Fig. 6A). Although we found IL-1␤-dependent NF-B binding to the IB␣ promoter, we did not detect any binding of NF-B to the SOCS3 promoter. These results argue against a direct function of NF-B on the SOCS3 promoter but rather for the induction of one or more unknown IL-1␤-induced factors that finally, in concert with IL-6, support SOCS3 expression. Indeed, we could show that de novo protein synthesis is required for IL-1␤ to exert its enhancing effect on SOCS3 gene induction (Fig. 6B). Inhibition of protein biosynthesis counteracted the enhancing activity of IL-1␤ on SOCS3 expression, similar as observed in response to impaired NF-B activation (Fig. 4).
Lipopolysaccharide and CpG activate TLR4 and TLR9, re-FIG. 8. Block of SOCS3 expression counteracts the inhibitory effect of IL-1␤. A, to monitor SOCS3 siRNA function HepG2 cells were co-transfected with vectors expressing SOCS3 siRNA or control vectors together with expression vectors of EpoR/gp130 chimeric receptors. After transfection, cells were stimulated with Epo (7 units/ml) for 50 min, and SOCS3 expression was analyzed by immunoprecipitation and Western blotting as described for Fig. 3. B, HepG2 cells were co-transfected with vectors expressing SOCS3 siRNA or control vector together with the STAT3-responsive SIE-luciferase reporter gene construct containing two STAT3 binding sites, expression vector of EpoR/gp130 chimeric receptors, and an expression vector for ␤-galactosidase. After transfection, cells were stimulated with 100 units/ml IL-1␤ or Epo (7 units/ ml) or both for 16 h. Luciferase activity in cellular extracts of these cells was determined and normalized to ␤-galactosidase activity, as outlined under "Experimental Procedures." spectively. Both pathogen patterns are known to induce SOCS3 in immune cells (9 -11). Viral double strand RNA binds to TLR3 and induces interferon-␤ gene expression by activation and subsequent nuclear translocation of NF-B and IRF-3 (38,39). Because IL-1␤ does not induce interferon-␤ expression in HepG2 cells (data not shown), we could exclude interferon-␤ to mediate the IL-1␤-dependent increase of SOCS3 expression. Instead, we found IL-1␤ to stabilize SOCS3 mRNA. From our knowledge, this is the first time that SOCS3 mRNA stabilization is shown as part of the regulation of SOCS3 expression.
To sum it up, IL-1␤ plays a positive regulatory role on IL-6induced SOCS3 expression, but enhanced immediate early SOCS3 expression in response to IL-1␤/NF-B results in enhanced feedback inhibition of IL-6 signaling and in turn, later, in reduced SOCS3 expression. What is the clue of such a complex regulatory network? IL-1␤ counteracts IL-6-induced acute-phase protein gene expression by inhibiting STAT3 activation or by competition of NF-B and STAT3 for overlapping binding sites in promoters of IL-6-inducible genes. Consequently, the induction of the SOCS3 feedback inhibitor would also be repressed by IL-1␤. The lack of feedback inhibition by SOCS3 would finally result in enhanced IL-6 signaling and thus reverse the initial pro-inflammatory function of IL-1␤. To overcome this problem enhanced SOCS3 expression in response to IL-1␤, especially when STAT3 activation is inhibited by IL-1␤, is reasonable. Our study provides evidence that enhanced SOCS3 expression contributes to the inhibitory effect of IL-1␤ on IL-6 signaling.