Article15 January 1999free access Mutational analyses of the SOCS proteins suggest a dual domain requirement but distinct mechanisms for inhibition of LIF and IL-6 signal transduction Sandra E. Nicholson Corresponding Author Sandra E. Nicholson The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Tracy A. Willson Tracy A. Willson The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Alison Farley Alison Farley The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Robyn Starr Robyn Starr The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Jian-Guo Zhang Jian-Guo Zhang The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Manuel Baca Manuel Baca The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Warren S. Alexander Warren S. Alexander The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Donald Metcalf Donald Metcalf The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Douglas J. Hilton Douglas J. Hilton The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Nicos A. Nicola Nicos A. Nicola The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Sandra E. Nicholson Corresponding Author Sandra E. Nicholson The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Tracy A. Willson Tracy A. Willson The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Alison Farley Alison Farley The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Robyn Starr Robyn Starr The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Jian-Guo Zhang Jian-Guo Zhang The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Manuel Baca Manuel Baca The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Warren S. Alexander Warren S. Alexander The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Donald Metcalf Donald Metcalf The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Douglas J. Hilton Douglas J. Hilton The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Nicos A. Nicola Nicos A. Nicola The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia Search for more papers by this author Author Information Sandra E. Nicholson 1, Tracy A. Willson1, Alison Farley1, Robyn Starr1, Jian-Guo Zhang1, Manuel Baca1, Warren S. Alexander1, Donald Metcalf1, Douglas J. Hilton1 and Nicos A. Nicola1 1The Walter and Eliza Hall Institute of Medical Research and the Cooperative Research Center for Cellular Growth Factors, Parkville, Victoria, 3050 Australia *Corresponding author. E-mail: [email protected] The EMBO Journal (1999)18:375-385https://doi.org/10.1093/emboj/18.2.375 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info SOCS-1 (suppressor of cytokine signaling-1) is a representative of a family of negative regulators of cytokine signaling (SOCS-1 to SOCS-7 and CIS) characterized by a highly conserved C-terminal SOCS box preceded by an SH2 domain. This study comprehensively examined the ability of several SOCS family members to negatively regulate the gp130 signaling pathway. SOCS-1 and SOCS-3 inhibited both interleukin-6 (IL-6)- and leukemia inhibitory factor (LIF)-induced macrophage differentiation of murine monocytic leukemic M1 cells and LIF induction of a Stat3-responsive reporter construct in 293T fibroblasts. Deletion of amino acids 51–78 in the N-terminal region of SOCS-1 prevented inhibition of LIF signaling. The SOCS-1 and SOCS-3 N-terminal regions were functionally interchangeable, but this did not extend to other SOCS family members. Mutation of SH2 domains abrogated the ability of both SOCS-1 and SOCS-3 to inhibit LIF signal transduction. Unlike SOCS-1, SOCS-3 was unable to inhibit JAK kinase activity in vitro, suggesting that SOCS-1 and SOCS-3 act on the JAK–STAT pathway in different ways. Thus, although inhibition of signaling by SOCS-1 and SOCS-3 requires both the SH2 and N-terminal domains, their mechanisms of action appear to be biochemically different. Introduction Cytokine stimulation involves specific recognition of receptors expressed at the cell surface, leading to receptor oligomerization and activation of an intracellular cascade of signaling molecules. The JAK–STAT pathway is activated in a rapid, transient manner and has been shown to be critical in many biological responses to cytokines. Negative regulation of this response is paramount to maintaining appropriate control of cytokine responses and, with the exception of the phosphatase SHP-1, is poorly understood. Recent studies have identified a new family of negative regulators of cytokine signaling called SOCS (Starr et al., 1997). The prototype of this family, SOCS-1 or suppressor of cytokine signaling-1, was identified by retroviral expression of a cDNA library derived from the factor-dependent hemopoietic cell line, FDCP-1 (Rayner and Gonda, 1994), in the murine monocytic leukemic cell line, M1. Selection for cells which had lost the ability to differentiate in response to interleukin-6 (IL-6) resulted in the recovery of a cDNA encoding a novel 212 amino acid protein (Starr et al., 1997). SOCS-1 was found to be distantly related to CIS (cytokine-inducible SH2-containing protein; Yoshimura et al., 1995), the most striking homology being in a 40 amino acid domain located in the C-terminal region of both proteins. This novel domain has been termed the 'SOCS box', and database searches have since identified a large number of additional proteins containing a C-terminal SOCS box (Starr et al., 1997; Hilton et al., 1998). These proteins have been classified further on the basis of the structural domains located N-terminal to the SOCS box. Six other proteins (SOCS-2, -3, -4, -5, -6 and -7), like CIS and SOCS-1, contain an SH2 (src homology-2) domain preceded by an N-terminal region of variable length and limited homology (Hilton et al., 1998). To date, little work has been done to investigate the functional relationships within this group of SH2-containing SOCS proteins. SOCS-1 was discovered independently on the basis of its ability to interact with the kinase domain of JAK2 (Janus kinase 2) and was named JAB (JAK-binding protein; Endo et al., 1997). It was also discovered because of antigenic similarity of the SOCS-1 SH2 domain to a sequence motif in the Stat3 SH2 domain, and was referred to as SSI-1 or STAT-induced STAT inhibitor-1 (Naka et al., 1997). Transcription of the SOCS-1 gene is induced in response to a number of cytokines (Starr et al., 1997), and the protein has been shown to interact with and inhibit the tyrosine phosphorylation of each member of the JAK family of protein tyrosine kinases (Endo et al., 1997; Naka et al., 1997; Ohya et al., 1997). It has also been demonstrated to inhibit TEC tyrosine kinase activity (Ohya et al., 1997), suggesting that its spectrum of activity may extend beyond that of the JAK family. There is some evidence that transcription of the SOCS-1 gene is regulated by Stat3 (Naka et al., 1997) in a manner similar to the transcriptional regulation of CIS by Stat5 (Matsumoto et al., 1997). The initial data therefore suggest that SOCS-1, at least, is part of a classical negative feedback loop. In contrast to SOCS-1, CIS appears to negatively regulate cytokine signaling by competing with Stat-5 for binding to phosphotyrosine residues within the erythropoietin and IL-3 receptor cytoplasmic domains (Yoshimura et al., 1995). It is therefore of interest to determine whether other SOCS proteins (SOCS-2 to SOCS-7) can negatively regulate cytokine signal transduction, and if so, to characterize their mechanism of action. This study examines the ability of SOCS family members to negatively regulate IL-6 signal transduction in M1 cells and further delineates the functional domains within SOCS proteins required for inhibition of signaling by IL-6 and leukemia inhibitory factor (LIF). Results Assays of cytokine signaling To assess the ability of SOCS proteins to inhibit signal transduction we have utilized two systems. (i) In soft agar, unstimulated M1 cells form large compact colonies of undifferentiated blast cells. When cells are incubated in the presence of IL-6 or LIF, the colonies are dispersed, with a halo of differentiating macrophages migrating out from the central core. At high concentrations of cytokine, the number of colonies observed is markedly reduced, a phenomenon referred to as clonal suppression (Metcalf, 1989). As described previously, when assessed by migration in agar, clonal suppression or morphological differentiation, M1 cells that constitutively express SOCS-1 are unable to respond to either IL-6 or LIF (Starr et al., 1997). These cells retain the ability to differentiate in response to dexamethasone, indicating that SOCS-1 inhibition is specific to cytokine signaling (Starr et al., 1997). (ii) 293T is a human fibroblast line expressing endogenous LIF receptors. Cells were transiently transfected with a LIF-responsive reporter construct, in which an element of the α-2 macroglobulin acute phase protein promoter was placed upstream of the luciferase-coding region (Endo et al., 1997). As has been reported previously (Masuhara et al., 1997), LIF stimulation resulted in a clear increase in luciferase activity. Co-expression of a β-galactosidase reporter construct under a constitutive promoter (Srαβ-gal) was used to control for transfection efficiency, and luciferase activity was normalized against β-galactosidase activity. Effect of SOCS proteins on LIF and IL-6 signaling To determine whether other SOCS family members had similar activity to SOCS-1 in regulating IL-6 signaling, cDNAs encoding N-terminally Flag-tagged versions of SOCS-2, SOCS-3, CIS and SOCS-5 (Figure 1) were stably transfected into M1 cells, and protein expression was assessed by immunoprecipitation and Western blot analysis using anti-Flag antibodies. Several clones expressing each protein were isolated; expression of SOCS-1 was found to be lower than SOCS-3 and CIS, with SOCS-2 protein levels consistently high. SOCS-5 was also expressed, but at relatively low levels (Figure 2B). Figure 1.Schematic diagram of the SOCS proteins analysed in this study. The red flag indicates either an N- or C-terminal Flag epitope. The SH2 domains are cross-hatched, and SOCS boxes are stippled. Domain swaps are indicated by SOCS-specific colors: CIS, pink; SOCS-1, yellow; SOCS-2, blue; SOCS-3, green; SOCS-5, orange; SOCS-6, purple. Download figure Download PowerPoint Figure 2.(A) Effect of SOCS proteins on M1 cell differentiation. Semi-solid agar cultures of parental M1 cells (control) and M1 cells expressing SOCS-1, SOCS-2, SOCS-3, SOCS-5 or CIS, showing the percentage of colonies that differentiated (dispersed) in response to a titration of 1 μg/ml mIL-6. Two independent clones are shown for each construct, and a dashed line represents the control untransfected cell line. (B) SOCS protein expression levels. Equal numbers of parental M1 cells and M1 cells expressing SOCS-1, SOCS-2, SOCS-3, CIS or SOCS-5 were lysed and analyzed by immunoprecipitation and Western blot using anti-Flag antibody. Arrows indicate the migration of the Flag-tagged SOCS proteins. Download figure Download PowerPoint Constitutive expression of SOCS-3, like that of SOCS-1, inhibited all IL-6- and LIF-induced effects, including clonal suppression, changes in cell morphology and up-regulation of surface markers, whereas M1 cells expressing SOCS-2 or SOCS-5 were reduced 3- to 10-fold in their sensitivity to IL-6 and LIF (Figure 2A, data not shown). Expression of CIS had no effect on either IL-6- or LIF-induced differentiation or clonal suppression (Figure 2A, data not shown). Similar results were found using 293T cells. Again, SOCS-1 and SOCS-3 completely abolished the LIF-induced activation of luciferase; SOCS-5 partially inhibited the LIF response, while SOCS-2, CIS and SOCS-6 were unable to inhibit the response (Figure 3). Appropriate expression of the various SOCS proteins was confirmed by Western blot analysis using anti-Flag antibodies (Figure 3). Figure 3.Effect of SOCS proteins on LIF-induction of the Stat3 reporter gene. Upper panel: levels of transiently expressed SOCS-1 (S1), SOCS-2 (S2), SOCS-3 (S3), CIS, SOCS-5 (S5) and SOCS-6 (S6) proteins from a representative experiment were determined by Western blot with anti-Flag antibody. Lower panel: 293T cells were transiently transfected with either vector alone or cDNAs expressing the various SOCS proteins in the presence of the APRE-luc and Srα-β-gal reporter genes. Cells were incubated in the presence (+) or absence (−) of 10 ng/ml hLIF overnight and cell extracts prepared. Luciferase activity from triplicate samples was determined and normalized against β-galactosidase activity. Download figure Download PowerPoint SOCS proteins that inhibit M1 cell differentiation block STAT3 tyrosine phosphorylation To understand further the mechanism by which SOCS-1 and SOCS-3 were able to inhibit M1 cell differentiation, we examined LIF-induced Stat3 tyrosine phosphorylation in M1 cells expressing the various SOCS proteins. Stat3 tyrosine phosphorylation has been implicated previously in IL-6-induced differentiation of M1 cells, both by the use of dominant-negative Stat3 constructs and by specific tyrosine mutations within the IL-6 signaling chain gp130, which block recruitment of Stat3 to the receptor complex (Nakajima et al., 1996; Yamanaka et al., 1996). In each instance, Stat3 tyrosine phosphorylation correlated inversely with the ability of the expressed SOCS protein to inhibit M1 cell differentiation. Stat3 was phosphorylated rapidly in response to LIF in M1 cells which either did not express SOCS proteins or expressed SOCS-2 or CIS, while LIF-induced Stat3 tyrosine phosphorylation was inhibited in M1 cells expressing either SOCS-1 or SOCS-3 (Figure 4). This suggests that the ability of both SOCS-1 and SOCS-3 to inhibit M1 differentiation may be mediated through inhibition of the JAK–STAT pathway. Figure 4.Stat3 tyrosine phosphorylation in M1 cells expressing the various SOCS proteins. (A) Upper panel: untransfected M1 cells and M1 cells stably expressing SOCS-1 or SOCS-2 were incubated with 5×105 U/ml mLIF for the times indicated, lysed, and Stat3 proteins immunoprecipitated with anti-Stat3 antibody. Phosphorylated Stat3 protein was detected by Western blot with antibody specific to tyrosine-phosphorylated Stat3 (anti-phosphoSTAT3; New England Biolabs Inc., Beverly, MA). Lower panel: blots were stripped and re-probed with anti-Stat3 antibody. (B) Untransfected M1 cells and M1 cells stably expressing SOCS-3 or CIS were analyzed as in (A). Download figure Download PowerPoint JAK1 activity is critical for LIF signal transduction in 293T cells and is blocked by SOCS-1 but not SOCS-3 Increasing evidence suggests that JAK1 is the critical JAK kinase involved in signaling through the IL-6 family of receptors. JAK1 knockout mice exhibit defective responses to IL-6 and LIF (Rodig et al., 1998), as do somatic cell mutants lacking functional JAK1 (Guschin et al., 1995). To address which JAK kinases were required for LIF signaling in 293T cells, we examined the ability of mutant JAK proteins lacking a functional kinase domain to inhibit LIF-induced activation of the APRE-luc reporter gene (Figure 5). Transient expression of mutant JAK1 (kinase dead) resulted in a dominant-negative effect, markedly reducing LIF-induced luciferase activity (to 3% of control values). Expression of mutant JAK2 (kinase dead) partially blocked signaling (to 69% of control values), whilst expression of mutant JAK3 (kinase dead) had no effect on LIF-induced luciferase activity (97% of control values). Expression of wild-type JAK1 and JAK2 enhanced LIF-induced transcription of the APRE, whilst expression of wild-type JAK3 had no effect (Figure 5). Comparable protein expression of the various JAK constructs was confirmed by Western analysis (Figure 5). These results suggest that in 293T cells, JAK1, and to a lesser extent, JAK2, is required for LIF-induced activation of Stat3. Figure 5.JAK1 is critical for LIF signal transduction in 293T cells. Upper panel: levels of transiently expressed JAK1, kinase-dead JAK1 (JAK1-KD), JAK2, kinase-dead JAK2 (JAK2-KD), JAK3 and kinase-dead JAK3 (JAK3-KD) proteins from a representative experiment were determined by Western blot with anti-JAK1 (Transduction Laboratories, Lexington, KY), JAK2 and JAK3 (Santa Cruz) antibody, respectively. Lower panel: 293T cells were transiently transfected with either vector alone or cDNAs expressing the various JAK proteins in the presence of the APRE-luc and Srα-β-gal reporter genes. Cells were incubated in the presence (+) or absence (−) of 10 ng/ml hLIF overnight and cell extracts prepared. Luciferase activity from triplicate samples was determined and normalized against β-galactosidase activity. Download figure Download PowerPoint While SOCS-1 has been implicated directly in inhibiting members of the JAK kinase family, the mechanism of SOCS-3 action is unknown. Given the importance of JAK1 and JAK2 in IL-6 and LIF signal transduction (Figure 5; Guschin et al., 1995; Rodig et al., 1998), the ability of SOCS-1 and SOCS-3 to inhibit JAK kinase activity directly was examined. Overexpression of JAK protein results in a constitutively active kinase, presumably because the high level of expression allows dimerization and activation to take place in the absence of ligand stimulation. JAK1 or JAK2 were transiently expressed with or without Flag-tagged SOCS-1 or SOCS-3. JAK proteins were immunoprecipitated using specific antibodies, and intrinsic kinase activity (or autophosphorylation) assessed using an in vitro kinase assay. Co-expression of SOCS-1 inhibited both JAK1 and JAK2 kinase activity (Figure 6). Interestingly, co-expression of SOCS-3 did not inhibit either JAK1 or JAK2 kinase activity and in fact enhanced JAK1 kinase activity (Figure 6). However, SOCS-3 enhancement of JAK1 activity appeared restricted to co-expression in 293T cells as, in COS cells, SOCS-3 neither inhibited nor enhanced JAK1 activity (data not shown). This suggests that although SOCS-1 and SOCS-3 act on the JAK–STAT pathway, they may do so through different mechanisms. Figure 6.JAK kinase autophosphorylation is inhibited by SOCS-1 but not SOCS-3. 293T cells were transiently transfected with vector alone or cDNAs expressing either Flag-tagged SOCS-1 or SOCS-3 in the presence (+) or absence (−) of cDNA expressing JAK1 or Flag-tagged JAK2. (A) JAK kinase activity. After 48 h, cells were lysed and JAK1 and JAK2 proteins immunoprecipitated using specific anti-JAK antibodies. Immunoprecipitates were then subjected to an in vitro kinase assay. (B) Immunoprecipitates from (A) containing either JAK1 or JAK2 protein were split prior to kinase assay and analyzed by Western blot with anti-JAK1 and anti-Flag antibody, respectively. (C) SOCS protein expression levels. Lysates from (A) were analyzed by Western blot with anti-Flag antibody. Download figure Download PowerPoint Amino acids 51–80 are critical for SOCS-1 function Previous work has suggested that the interaction between SOCS-1 and the JAK family of protein tyrosine kinases is mediated through binding of the SOCS-1 SH2 domain to a phosphorylated tyrosine residue within the JAK kinase domain (JH1). In these studies, JAK2 was constitutively activated through overexpression in COS cells or yeast (Endo et al., 1997; Naka et al., 1997; Ohya et al., 1997). To investigate the structural basis of SOCS-1 action in a ligand-inducible system, a series of SOCS-1 deletion mutants was constructed, in which either the N-terminal region (S1Δ1-76), the C-terminal SOCS box (SΔ170–212) or both the N-terminal region and the SOCS box (S1Δ1-76&170–212) were deleted. In addition, a construct was created in which the SH2 domain and the SOCS box were both deleted, leaving just the N-terminal 81 amino acids (S1Δ82-212; Figure 1). The SOCS-1 deletion mutants were transiently expressed in the 293T reporter system and tested for their ability to inhibit LIF-induced luciferase activity. Expression of the mutant lacking the N-terminal region (S1Δ1-76), or lacking both the N-terminal region and the SOCS box (S1Δ1-76&170–212), did not block LIF induction of luciferase activity, whereas the mutant lacking the SOCS box (S1Δ170–212) was fully active and completely blocked luciferase activity (Figure 7A). In addition, a protein lacking both the SH2 domain and SOCS box (S1Δ82-212) was unable to inhibit LIF-induced luciferase activity, suggesting that the N-terminal region alone was insufficient to mediate this action. Protein expression of the various SOCS proteins was confirmed by Western blot using anti-Flag antibodies. The low-level expression of S1Δ1-76 and S1Δ1-76&170–212 was observed consistently in 293T cells, making assessment of the activity of these constructs difficult. Deletion mutants were therefore stably expressed in M1 cells and several independent transfectants obtained for each construct. Consistent with the 293T reporter assay, constitutive expression of the mutant either lacking the N-terminal region (S1Δ1-76) or lacking both the N-terminal region and the SOCS box (S1Δ1-76&170–212) did not block IL-6- or LIF-induced differentiation. In contrast, cells expressing either the mutant lacking the SOCS box (S1Δ170–212) or full-length SOCS-1 fully blocked IL-6- and LIF-induced differentiation (Figure 8A, data not shown). Analysis with anti-Flag antibodies confirmed expression of S1Δ1-76, S1Δ1-76&170–212 and S1Δ170–212 at equivalent levels to cells expressing full-length SOCS-1 (Figure 8B). These results indicated that although neither alone was sufficient, the combination of the SOCS-1 N-terminal region and SH2 domain was critical for SOCS-1-inhibition of LIF and IL-6 signaling. Figure 7.Mutational analysis of SOCS-1. Upper panels: levels of transiently expressed SOCS-1 mutant proteins from representative experiments were determined by Western blot with anti-Flag antibody. Lower panels: 293T cells were transiently transfected with vector alone or cDNAs expressing the various SOCS-1 mutant proteins in the presence of the APRE-luc and Srα-β-gal reporter genes. Cells were incubated in the presence (+) or absence (−) of 10 ng/ml hLIF overnight and cell extracts prepared. Luciferase activity from triplicate samples was determined and normalized against β-galactosidase activity. (A) 293T cells were transiently transfected with full-length SOCS-1 or SOCS-1 deletion mutants S1Δ82-212 (lacking both the SH2 domain and the SOCS box), S1Δ170–212 (lacking the SOCS box), S1Δ1-76&170–212 (lacking the N-terminal region and the SOCS box) and S1Δ1-76 (lacking the N-terminal region). (B) 293T cells were transiently transfected with full-length SOCS-1 or SOCS-1 deletion mutants S1Δ1-30, S1Δ1-40, S1Δ1-50, S1Δ1-60, S1Δ1-70, S1Δ50-60, S1Δ60-70 and S1Δ70-78. (C) 293T cells were transiently transfected with full-length SOCS-1 or SOCS-1 point mutants S1-D53R, S1-R57E, S1-D64R, S1-R70E, S1-P51A, S1-F56A, S1-F59A and S1-G52A. (D) 293T cells were transiently transfected with full-length SOCS-1 or SOCS-1 N-terminus or SH2 domain swap mutants S2/1/1, S3/1/1, C/1/1, S5/1/1, S6/1/1, S1/2/1, S1/3/1, S1/C/1, S1/5/1 and S1/6/1. Download figure Download PowerPoint Figure 8.Expression of SOCS-1 deletion mutants in M1 cells. (A) Effect of SOCS proteins on M1 cell differentiation. Semi-solid agar cultures of parental M1 cells (control) and M1 cells expressing SOCS-1, S1Δ1-76&170–212 (lacking both the N-terminal region and the SOCS box), S1Δ1-76 (lacking the N-terminal region) and S1Δ170–212 (lacking the SOCS box) showing the percentage of colonies that differentiated (dispersed) in response to a titration of 1 μg/ml mIL-6. Two independent clones are shown for each construct, and a dashed line represents the control untransfected cell line. (B) Protein expression levels. Equal numbers of parental M1 cells (M1) and M1 cells expressing SOCS-1 (S1), S1Δ1-76&170–212, S1Δ1-76 and S1Δ170–212 were lysed and analyzed by immunoprecipitation and Western blot using anti-Flag antibody. Download figure Download PowerPoint The luciferase reporter system was utilized further to analyze a series of smaller truncations within the N-terminal region of SOCS-1. Deletion of the N-terminal 10, 20, 30, 40 or 50 amino acids did not alter the ability of SOCS-1 to inhibit LIF-induced luciferase activity (Figure 7B and data not shown). However, deletion of the N-terminal 60 amino acids completely abolished SOCS-1 inhibitory action (Figure 7B). To narrow down the critical region, three additional mutants were generated in which amino acids 50–60, 60–70 or 70–78 were deleted. When tested in the 293T reporter assay, none of these constructs was able to inhibit LIF-induced luciferase activity (Figure 7B). The 30 amino acids directly N-terminal to the SOCS-1 SH2 domain therefore appear critical for inhibition of LIF signal transduction, whilst the first 50 amino acids are dispensable for SOCS-1 action. To define the critical residues within the N-terminal region required for SOCS-1 activity, a series of point mutations was assessed for their effect on SOCS-1 function. These mutations either replaced individual residues with alanine or resulted in a charge reversal. The following mutations had no effect when proteins were assayed using the 293T reporter system: P51A, G52A, D53A, D53R, T54A, H55A, R57A, R57E, T58A, R60A and R70E (data not shown; Figure 7C). Despite expression at equivalent levels to wild-type SOCS-1, mutation of F56A, F59A or D64R resulted in a non-functional SOCS-1 protein (Figure 7C). The SOCS-1 and SOCS-3 N-terminal domains are functionally interchangeable As described above, we have demonstrated functional differences between individual SOCS proteins in their ability to inhibit biological activity. Further, we have identified critical regions in the N-terminal region of SOCS-1 that are crucial for this response. To determine whether the SOCS protein domains are interchangeable, a series of chimeric proteins was created. The N-terminal region of SOCS-1 was replaced with the N-terminal domain of SOCS-2 (S2/1/1), SOCS-3 (S3/1/1), CIS (C/1/1), SOCS-5 (S5/1/1) or SOCS-6 (S6/1/1). These constructs were Flag-tagged, transiently expressed in 293T cells and LIF induction of luciferase activity assayed. The SOCS-1 chimeric protein containing the N-terminal region of SOCS-3 was able to inhibit LIF induction of luciferase activity to the same level as wild-type SOCS-1, whereas the N-terminal regions of SOCS-2 or CIS could not functionally replace the N-terminal region of SOCS-1 (Figure 7D). Consistent with the levels of wild-type SOCS-5 and SOCS-6 protein, chimeric proteins S5/1/1 and S6/1/1 were expressed at low levels, making assessment of the activity of these constructs difficult (Figure 7D). A series of chimeric proteins was