Abstract
The goal of this study was to investigate how bacterial LPS affects macrophage responsiveness to the activating factor IFN-γ. Pretreatment of macrophages with LPS for <2 h increased the transcriptional response to IFN-γ. In contrast, simultaneous stimulation with IFN-γ and LPS, or pretreatment with LPS for >4 h, suppressed Stat1 tyrosine 701 phosphorylation, dimerization, and transcriptional activity in response to IFN-γ. Consistently, the induction of MHCII protein by IFN-γ was antagonized by LPS pretreatment. Neutralizing Abs to IL-10 were without effect on LPS-mediated suppression of Stat1 activation. Decreased IFN-γ signal transduction after LPS treatment corresponded to a direct induction of suppressor of cytokine signaling3 (SOCS3) mRNA and protein. Under the same conditions socs1, socs2, and cis genes were not transcribed. In transfection assays, SOCS3 was found to suppress the transcriptional response of macrophages to IFN-γ. A causal link of decreased IFN-γ signaling to SOCS3 induction was also suggested by the LPS-dependent reduction of IFN-γ-mediated Janus kinase 1 (JAK1) activation. Further consistent with inhibitory activity of SOCS3, LPS also inhibited the JAK2-dependent activation of Stat5 by GM-CSF. Our results thus link the deactivating effect of chronic LPS exposure on macrophages with its ability to induce SOCS3.
Maximal immunological activation of macrophages results from the effects caused by two independent stimuli. One signal is provided by the Th1 cytokine IFN-γ, whereas the second is most often provided by microbial products like LPS, the outer membrane constituent of Gram-negative bacteria (1, 2). IFN-γ influences nuclear gene expression by activating the IFN-γ receptor-associated Janus tyrosine kinases (JAK),3 JAK1 and JAK2. These phosphorylate transcription factor Stat1 on Y701 and cause its dimerization and translocation to the cell nucleus (3, 4). Stat1 dimers bind to the IFN-γ activation site (GAS) in the promoters of target genes and activate transcription (5). On the other hand, LPS stimulates gene expression predominantly through the activation of transcription factor NF-κB (6). However, LPS also exerts a direct and rapid effect on IFN-γ-mediated Stat1 activation by increasing the activity of a Stat1 S727 kinase (7). Phosphorylation of S727 strongly augments the transcription factor activity of Stat1 (7, 8).
In this study, we investigated the effect of prolonged LPS stimulation on the IFN-γ response of macrophages, a situation that may occur if bacteria are not rapidly cleared from a site of infection. We report the inhibition of signal transduction in response to IFN-γ through the LPS-induced synthesis of the suppressor molecule, suppressor of cytokine signaling 3 (SOCS3). The SOCS family of proteins, also known as cytokine-inducible SH2 domain-containing protein (CIS)/JAK-binding protein (JAB) or STAT-induced STAT inhibitors (SSI) (9, 10, 11), was recently described as feed-back inhibitors of cytokine signaling paths and appears to have a widespread role in their negative regulation (reviewed in Refs. 12, 13). The ability of socs genes to respond to cytokine-activated Stats explains how Jak-Stat signal transduction autoregulates in a negative feed-back loop (14). In addition, SOCS proteins can mediate negative crosstalk between distinct signaling avenues. Here, we establish socs3 as an LPS target gene, thus linking its regulation to a signaling path not employing Stats. We suggest that via SOCS3 the persistent presence of bacteria and their products exerts a direct suppressive effect on macrophage activation by IFN-γ. Our findings stress the importance of timing immunologically relevant extracellular stimuli with respect to each other.
Materials and Methods
Cells
Bac1.2F5 macrophages (15), or C11 cells, a clone of Bac1.2F5 transfected with a Stat-dependent luciferase gene (7), were maintained in MEM-α medium containing 10% FCS and 20% L cell-conditioned medium as a source of CSF-1. 293 kidney fibroblasts were cultured in DMEM containing 10% FCS.
Antibodies
The antiserum to the Stat1 C terminus has recently been described (7). Rabbit antiserum to Y701-phosphorylated Stat1 was purchased from New England Biolabs (Beverly, MA) and used at a dilution of 1:1000. A mAb recognizing the α-chain of the murine IFN-γ receptor was obtained from PBL Biomedical Laboratories (New Jersey, NJ) and used at a 1:50 dilution for flow cytometry. To visualize binding of the anti-IFN-γ receptor Ab, PE-coupled rabbit anti-rat antiserum was used. FITC-conjugated Ab (clone AMS32.1) to the MHC class II protein IAd was purchased from PharMingen (San Diego, CA). The Py20 Ab to phosphotyrosine was purchased from Transduction Laboratories (Lexington, KY) and used at 3 μg/ml. Rabbit antisera to JAK1 and JAK2 kinases were a kind gift from Dr. Andrew Zimiecki (Laboratory for Clinical and Experimental Cancer Research, Bern, Switzerland). They were used at 1:100 dilution for immunoprecipitation and 1:1000 dilution for Western blots. Rabbit anti-SOCS3 antiserum was produced by immunization with a peptide corresponding to amino acids 5–21 of human SOCS3 protein. It was used at a 1:100 dilution in immunoprecipitation and a 1:500 dilution in Western blots. A polyclonal Ab specific for the two Stat5 isoforms, Stat5a and Stat5b, was recently described (16). It was used for EMSA supershift analysis at a final dilution of 1:100. A mAb recognizing Stat3 was kindly provided by David Levy (Department of Pathology, New York University School of Medicine, New York, NY). It was used for EMSA supershift at a final concentration of 10 ng/ml. Neutralizing Abs to IL-10 were purchased from Strathmann Biotech (Hannover, Germany). They were used at a dilution of 1:200, which neutralizes 2 ng/ml of IL-10. The amount secreted by LPS-treated monocytes or macrophages into the culture medium was reported to be at least 4-fold lower (17, 18). As control Abs, an equal dilution of an antiserum raised against human IFN-α was used.
Immunoprecipitation and Western blot
A protocol for these procedures has recently been described (7). Abs were used as indicated in the figure legends.
EMSA
An end-labeled, double-stranded oligonucleotide corresponding to the GAS sequence of the β-Casein promoter was used in most experiments. To determine Stat3 binding, an acute phase response element (APRE) oligonucleotide corresponding to the sequence in the rat α2 macroglobulin promoter was used (19). The assay was performed with whole cell extracts, as recently described (20, 21).
Transfection assays
293 cells and Bac1.2F5 macrophages were transfected using Superfect reagent according to the manufacturer’s instructions (Quiagen, Munich, Germany). Transfected plasmids contained a luciferase reporter gene under control of the IFN-γ-responsive IFP53 promoter (22) or, in cotransfections, the IFN-γ-responsive reporter gene and expression plasmids for either SOCS1, -2, and -3 or CIS. Luciferase assays were performed according to standard procedures. Results are stated as inducibility, the ratio obtained by dividing cpm light emission from IFN-γ and/or LPS-treated and unstimulated cells.
Northern blot
An amount of 15 μg of total RNA from Bac1.2F5 cells was separated on agarose gels and blotted to membrane using standard procedures. The blots were probed using SOCS or CIS cDNAs labeled by random priming.
Results
Time-dependent effect of LPS on IFN-γ-stimulated gene expression
Transcriptional activity of Stat1 was determined in Bac1.2F5 macrophages stably transfected with a Stat1-dependent reporter gene. In this experimental situation, we have previously shown that a short pretreatment with LPS causes an increase in Stat1-dependent transcription, due to the stimulation of a Stat1 S727 kinase (7). Consistent with our previous observation, a brief costimulation (up to 3 h) of the cells with both IFN-γ and LPS resulted in increased Stat1 transcriptional activity, compared with stimulation with IFN-γ alone. By contrast, treatment beyond 4 h converted the stimulatory effect of LPS on the IFN-γ response into suppressive activity (Fig. 1⇓A). This suppressive effect of LPS was confirmed when the cells were first pretreated with LPS for various periods, followed by 2 h of stimulation with IFN-γ (Fig. 1⇓B).
Effect of LPS on the transcriptional response to IFN-γ. A, Bac1.2F5 macrophages, stably transfected with a Stat-dependent luciferase reporter gene (clone C11) were stimulated with IFN-γ or LPS alone, or a combination of both agents for the indicated periods and luciferase activity determined in cell extracts. IFN-γ and LPS were simultaneously present for the indicated periods. B, The cells were pretreated with LPS for the indicated periods, followed by stimulation with IFN-γ for 2 h. C, Macrophages were treated with LPS for 12 h and left without further treatment for 48 h (dotted line). Further culture dishes were treated with IFN-γ for 48 h (solid line) or with IFN-γ for 48 h after 12 h of LPS pretreatment (dashed line). Subsequently, the expression of the MHCII (IAd) proteins on the cell surface was determined by flow cytometry.
Several Stat1 target genes relevant to macrophage activation like inos or icam-1 contain NF-κB sites and can be directly stim-ulated by LPS (23, 24, 25, 26). By contrast, MHCII protein induction is an exclusive property of IFN-γ. Since the crucial transcription factor for MHCII induction by IFN-γ, CIITA, is a Stat1 target gene, the expression of MHCII is indirectly dependent on Stat1 (27). To test whether enhanced MHCII expression, an important attribute of activated macrophages, is influenced by LPS pretreatment similar to our reporter gene, we analyzed MHCII expression by flow cytometry. Relative to untreated controls, Bac1.2F5 macrophages showed moderate but significant enhancement of MHCII expression by IFN-γ within 48 h. This increase was strongly inhibited by 12 h of pretreatment with LPS (Fig. 1⇑C). The same result was obtained in primary bone marrow macrophages (data not shown). It is consistent with reporter gene expression and confirms that the block in transcriptional responsiveness to Stat1 affects target genes relevant for macrophage activation in their proper genomic context.
Prolonged LPS treatment inhibits Stat1 activation and decreases JAK1 tyrosine phosphorylation by IFN-γ
To determine whether decreased transcription was due to LPS-mediated down-regulation of Stat1 activation, we performed EMSA experiments. Consistent with results on Stat1-mediated transcription, costimulation with IFN-γ and LPS caused a more transient appearance of Stat1 DNA-binding activity, compared with IFN-γ alone (Fig. 2⇓A), and pretreatment with LPS for more than 4 h strongly reduced the dimerization of Stat1 in response to IFN-γ (Fig. 2⇓B). Consistent with the EMSA results, pretreatment with LPS also reduced the tyrosine phosphorylation of Stat1 in response to IFN-γ, as indicated by Western blot with an antiserum reactive with Stat1 phosphorylated on Y701 (Fig. 2⇓C). Control cells treated for 6 h with LPS alone (Fig. 2⇓A, lane 7) produced a Stat complex with lower mobility than Stat1. This complex is analyzed in detail below.
Effect of LPS on Stat1 tyrosine phosphorylation in Bac1.2F5 macrophages. Cells were stimulated with IFN-γ or LPS alone, or a combination of both agents for the indicated periods. Stat1 DNA-binding activity in cell extracts was determined in an EMSA with a labeled Stat binding site (GAS) from the rat β-Casein promoter. A, The cells were stimulated with IFN-γ alone (lanes 2–6) or simultaneously with LPS and IFN-γ for the indicated periods (lanes 8–12). Control cells were either left untreated (lane 1) or treated with LPS alone for 6 h (lane 7). B, Cells were pretreated with LPS for the indicated periods, followed by 10 min of stimulation with IFN-γ. C, Macrophage extracts after treatment with LPS and/or IFN-γ, as indicated, were analyzed for Stat tyrosine701 phosphorylation by Western blotting with a phosphospecific Stat1 antiserum. D, Macrophages, untreated (lanes 1 and 2) or pretreated with LPS for 12 h (lanes 3–8), were stimulated with IFN-γ for 30 min, and Stat1 activation was determined by EMSA with a β-Casein probe. During the LPS pretreatment period, the culture medium contained no Abs (lanes 3 and 4), neutralizing Abs to IL-10 (lanes 5 and 6), or an equal dilution of control Abs to human IFN-α. In lanes 9 and 10, a direct suppressive effect of the IL-10 Abs was tested. The corresponding cultures were treated with the Abs for 12 h before stimulation with IFN-γ.
IL-10 is a LPS-induced cytokine with deactivating activity toward macrophages. To determine the relative contributions of LPS and IL-10 to the suppression of IFN-γ-mediated Stat1 activation, we treated macrophages with LPS in the presence of control Abs, or of neutralizing Abs to IL-10, followed by a brief pulse of IFN-γ. Stat1 activation was assayed by EMSA (Fig. 2⇑D). The reduction of Stat1 activation by a 12-h LPS treatment was found similarly strong in samples without Ab (lanes 3 and 4), the control Abs (lanes 7 and 8), or IL-10 Abs (lanes 5 and 6). Treatment with IL-10 Abs for 12 h (lanes 9 and 10) did not per se reduce IFN-γ-mediated Stat1 activation, a possibility suggested by a recent report (28). Together, the data in Fig. 2⇑D demonstrate that the suppressive activity toward Stat1 activation by IFN-γ stems mostly from LPS directly, not from LPS-induced IL-10.
To study possible causes of the LPS effect on Stat1 activation, we first determined IFN-γ receptor expression by flow cytometry. Expression in normal or LPS-pretreated macrophages was virtually identical (Fig. 3⇓). By contrast, LPS affected an early step in IFN-γ signaling, the activation of the receptor-associated tyrosine kinase JAK1, as indicated by diminished tyrosine phosphorylation in response to IFN-γ (Fig. 4⇓). Densitometric analysis after normalization to expression levels indicated a reduction of JAK1 tyrosine phosphorylation after 8 h of LPS pretreatment of ∼70%. Inhibition of JAK1 activation was always observed. In a total of six independent experiments, four showed no effect of LPS on JAK2 tyrosine phosphorylation, while two showed a somewhat reduced activation of JAK2. We have at present no explanation for this variation because there were no apparent differences in the experimental protocol. Importantly, normal JAK2 activation in response to IFN-γ, as shown in Fig. 4⇓, confirms that LPS did not affect the expression of the IFN-γ receptor. Moreover, the control lanes of the blots shown in Figs. 2⇑C and 4 indicate that LPS caused no significant reduction of either JAK or Stat1 protein.
LPS effect on IFN-γ receptor expression. Untreated or LPS (13.5 h)-pretreated Bac1.2F5 macrophages were labeled with a mAb to the IFN-γ receptor α-chain and PE-labeled second Ab. A total of 10,000 cells in each sample was analyzed by flow cytometry.
LPS effect on IFN-γ-mediated JAK activation in macrophages. Bac1.2F5 cells were pretreated with LPS for the indicated periods, followed by either no further treatment or stimulation with IFN-γ for 20 min. JAK activation was determined by immunoprecipitation with specific Abs, followed by Western blot with the phosphotyrosine-specific mAb Py20.
Direct effect of LPS on the expression of the socs3 gene
The inhibition of JAK1 phosphorylation prompted us to investigate the possible role of the previously described inhibitors of JAK activity, the SOCS/CIS/SSI proteins. Particularly one member of this family, SOCS1/JAB/SSI1, has been associated with inhibition of the IFN-γ response, and another one, SOCS3, was similarly shown to inhibit Stat1 activation (29, 30). Consistent with these published results, SOCS1 and SOCS3, but not the family members SOCS2 or CIS, inhibited reporter gene expression in response to IFN-γ in transformed 293 kidney fibroblasts (Fig. 5⇓A) and in macrophages (Fig. 5⇓B). Northern blots with RNA from macrophages treated with LPS for various periods demonstrated increased expression of SOCS3 mRNA (Fig. 6⇓A), but not of SOCS1, SOCS2, or CIS mRNA (data not shown). Maximal levels of SOCS3 mRNA were reached after 4 h of LPS treatment, and induction was resistant to the inhibition of protein synthesis by cycloheximide. This establishes SOCS3 as a direct LPS target gene, induced by LPS-responsive transcription factors like NF-κB, and/or responsive to LPS-mediated mRNA stabilization as in the case of many mRNAs encoding proinflammatory cytokines. SOCS3 protein was induced by LPS with maximal levels after 6 h (Fig. 6⇓B). This result argues for a causal effect of SOCS3 on suppression of IFN-γ-induced signal transduction where a clear effect on Stat1 tyrosine phosphorylation is evident after 4 h and a maximal effect after 8–12 h (Fig. 2⇑).
Effect of SOCS/CIS expression on Stat-reporter gene activity. A, 293 cells were transiently transfected with a luciferase gene under control of the IFN-inducible IFP53 promoter. The cells were cotransfected with either empty vector or expression plasmids for the indicated SOCS/CIS family members. Luciferase activity was determined after 6 h of treatment with IFN-γ. B, Cotransfection of the IFN-sensitive reporter gene and SOCS1 or SOCS3 expression plasmids was performed with Bac1.2F5 macrophages. Luciferase activity was determined after 8 h of treatment with IFN-γ.
LPS induces in macrophages the expression of SOCS3 RNA and protein. A, Bac1.2F5 cells were stimulated with LPS as shown. Where indicated, cycloheximide (CHX) was present at 50 μg/ml. Total RNA was extracted and analyzed for SOCS/CIS RNA by Northern blotting. SOCS1, SOCS2, and CIS RNA did not respond to LPS in these experiments. B, Extracts from LPS and/or IFN-γ-treated macrophages were subjected to immunoprecipitation with SOCS3 antiserum, followed by Western blot with the same Abs.
The protein tyrosine phosphatase SHP-1 (also called PTP-1C or SH-PTP1) was shown to dephosphorylate and inactivate JAK2 in the course of a response to erythropoietin (31). We tested whether LPS affects SHP-1 activity by immunoprecipitation with specific Abs and phosphotyrosine blot. These experiments demonstrated a constant basal activity of SHP-1 throughout a 12-h period of LPS treatment (data not shown). A participation of the phosphatase in LPS-mediated down-regulation of IFN-γ signaling is therefore unlikely.
LPS inhibits GM-CSF-mediated JAK-Stat signal transduction
Recent publications suggest an ability of SOCS3 to inhibit prolactin, leptin, and growth hormone-induced signal transduction (32, 33, 34). All corresponding receptors signal via JAK2. Therefore, we tested whether SOCS3 induction by LPS in macrophages causes a general inhibition of cytokine responses employing JAK2. The GM-CSF receptor employs a JAK2-Stat5 signaling path to elicit immediate gene expression (20, 35). Rapid activation of Stat5 by GM-CSF in Bac1.2F5 macrophages was strongly suppressed after 5 h of LPS pretreatment (Fig. 7⇓A). This adds GM-CSF-induced signal transduction to the list of SOCS3-inhibitable cellular responses and further demonstrates the activity of the inhibitory protein in LPS-treated macrophages.
The effect of LPS pretreatment on Stat5 activation by GM-CSF. A, Macrophages without pretreatment or after pretreatment with LPS for the indicated periods were stimulated with GM-CSF for 20 min. Cellular extracts were analyzed by the EMSA with a β-Casein promoter GAS sequence for Stat activation. B, The extract derived from LPS-treated macrophages was subjected to EMSA supershift analysis with Abs specific for Stat3 and Stat5. APRE and β-Casein promoter probes were used to select for the preferential binding of Stat3 and Stat5 dimers, respectively.
As already noted in the EMSA experiments of Fig. 2⇑, LPS treatment for 4 h or longer caused the appearance of a Stat complex with lower mobility, compared with Stat1. The same observation is documented in Fig. 7⇑A. LPS does not directly activate Stats (Ref. 7 , and Refs. therein). Consistently, the slow appearance and cycloheximide-sensitivity (data not shown) of the LPS-induced complex indicated that it is activated indirectly by an LPS-induced cytokine. Careful analysis revealed the LPS-induced band to represent two closely spaced complexes. To identify the relevant Stats, we reacted the extracts from LPS-treated macrophages with an APRE probe that strongly binds Stat3, but associates only weakly with Stat5 and with the β-Casein promoter probe, which shows the opposite behavior. Coupled with EMSA supershift analysis using specific Abs to Stat3 and the Stat5 isoforms, we could demonstrate that the LPS-induced complexes consist of Stat3 and Stat5 dimers (Fig. 7⇑B). The transcription experiment in Fig. 1⇑B suggests that the LPS-induced complexes do not possess a strong potential to activate a Stat-dependent reporter gene. This is not surprising because Stat tyrosine phosphorylation was shown in a number of situations to produce dimers capable of binding to DNA, but without an inductive effect on target gene transcription (16, 20, 36). LPS-mediated generation of Stat3 and Stat5 complexes occurs in the presence of intracellular SOCS3. This indicates that signal transduction in response to the responsible LPS-induced cytokine(s) is either not inhibited by SOCS3, or that activation levels would be much higher if the same concentration of cytokine(s) would be acting on cells not treated with LPS.
Discussion
The major conclusions from the results in our study are that the socs3 gene is directly targeted by LPS-derived intracellular signals and that this results in negative crosstalk to cytokine signals through JAK-Stat pathways. Our particular focus was the activation of macrophages by IFN-γ. Stat1 knockout mice demonstrate the crucial importance of the protein for macrophage activation in vivo (37, 38). SOCS3 induction by LPS inhibits Stat1 activation, thereby interfering with the ability of macrophages to produce their complete arsenal of antibacterial responses.
The induction of SOCS3 mRNA was previously reported to occur in response to IL-6 and IL-10 (9, 39). Particularly IL-10 is found in culture supernatants of LPS-activated macrophages. IL-10 is a potent activating stimulus for Stat3 (40), and this protein was part of the Stat complex found after long LPS stimulation in macrophage extracts (Fig. 7⇑). As shown for LPS in this study, SOCS3 produced in response to IL-10 was similarly suggested to down-regulate Stat1 activation by IFN-γ (39). For these reasons, it was important to neutralize IL-10 activity in culture supernatants and to assess the residual suppressive effect of LPS on the IFN-γ response. This experiment, documented in Fig. 2⇑D, indicates that IL-10 is not required for LPS to down-regulate Jak-Stat signal transduction in response to IFN-γ. Therefore, despite the ability of both agents to induce SOCS3 and interfere with an IFN-γ response, macrophage deactivation after long LPS treatment and after IL-10 stimulation is likely to be different. For example, IL-10 directly activates transcriptionally active Stat3 (40) and induces a set of genes that partially overlaps with that induced by IFN-γ-activated Stat1. This became apparent from our recent study on the expression of the FcγRI gene in macrophages. It demonstrated not only a direct induction of this gene by IL-10 and IFN-γ individually, but an additive effect on induction by the simultaneous presence of both cytokines (41). Therefore, both LPS- and IL-10-treated macrophages will be refractory to activation by IFN-γ, but the IL-10-treated cell will express part of the IFN-γ-responsive genes through the activity of Stat3, while the LPS-treated cell will not.
From our recent findings (7) and the results presented here, we propose a dual time-dependent effect of LPS on macrophage activation by IFN-γ. A brief exposure to LPS enhances IFN-γ-mediated macrophage activation through its specific target genes and through the activation of a Stat1 S727 kinase, which increases the transcriptional potential of Stat1 dimers. However, the continuous presence of LPS before the Th1 cytokine inhibits the responsiveness to IFN-γ and thus counteracts macrophage activation. Long exposure to LPS also causes macrophages to become tolerant to LPS itself (42). However, LPS-mediated LPS tolerance involves the formation of NF-κB p50 dimers (43) and differs in kinetics and dose-dependency from LPS-mediated tolerance to IFN-γ (our unpublished observations). Therefore, the two phenomena appear to be mechanistically unrelated.
The mechanism of SOCS3 action is currently unclear. While all reports agree with respect to the ability of the family member SOCS1 to bind to and inhibit JAK1 and JAK2 kinases (44), some studies suggest a similar ability of SOCS3, while others find a very limited potential of SOCS3 to directly inhibit JAKs, and therefore consider alternative ways by which SOCS proteins might interfere with cytokine signaling (13, 45). Assuming a direct SOCS3-JAK interaction, the predominant effect on JAK1 displayed in Fig. 4⇑ of this paper may indicate that SOCS3 binds to JAK1 and inhibits its activity. This would resemble the situation recently described for the IL-2 response of T lymphocytes (46). Alternatively, SOCS3 may bind to tyrosine-phosphorylated JAK2 and inhibit its kinase activity, with the possible consequence that JAK1 tyrosine phosphorylation and activation by cross-phosphorylation cannot occur. The latter assumption is consistent with our results, showing an inhibition of Jak-2-dependent Stat5 activation by GM-CSF and the ability of SOCS3 to inhibit Stat5 activation by IL-3 (46). Pull-down assays employing GST-SOCS3 have not been successful, either because of insufficient sensitivity of the assay or because SOCS3 does not stably associate with complexes containing JAKs (our unpublished observations). Recent publications suggest an ability of SOCS3 to inhibit prolactin, IL-6, LIF, leptin, growth hormone, IFN-γ, IFN-α, and IL-2-induced signal transduction (30, 32, 33, 34, 39, 45, 46). Therefore, SOCS3 interferes with both JAK2-dependent and JAK2-independent cytokine responses.
Thus, while the details of SOCS3 activity remain to be clarified, our studies emphasize the important role of this protein in mediating inhibitory crosstalk not only between signals from different cytokine receptors, but also between cytokine signals and those derived from bacteria during infection.
Footnotes
-
↵1 This work was supported by Grant P11530-MED from the Austrian Fonds zur Förderung der wissenschaftlichen Forschung (to T.D.).
-
↵2 Address correspondence and reprint requests to Dr. Thomas Decker, Vienna Biocenter, Institute of Microbiology and Genetics, Dr. Bohr-Gasse 9, A-1030 Vienna, Austria. E-mail address: decker{at}gem.univie.ac.at
-
↵3 Abbreviations used in this paper: JAK, Janus kinase; GAS, γ IFN activation site; SOCS, suppressor of cytokine signaling; APRE, acute phase response element; CIS, cytokine-inducible SH2 domain-containing protein; JAB, JAK-binding protein; SSI, STAT-induced STAT inhibitors.
- Received April 28, 1999.
- Accepted June 28, 1999.
- Copyright © 1999 by The American Association of Immunologists
References
In this issue
