The FASEB Journal • Research Communication
Suppressor of cytokine signaling 1 inhibits IFN-␥ inflammatory signaling in human keratinocytes by sustaining ERK1/2 activation Stefania Madonna, Claudia Scarponi, Ornella De Pita`, and Cristina Albanesi1 Laboratory of Immunology, Istituto Dermopatico dell’Immacolata (IDI)-IRCCS, Rome, Italy IFN-␥ is a pleiotropic cytokine importantly involved in the development of skin inflammatory responses. Epidermal keratinocytes are extremely susceptible to IFN-␥ action, but, once transduced with the suppressors of cytokine signaling (SOCS)1 molecule, they can no longer express a number of IFN-␥inducible signal transducer and activator of transcription (STAT)1-dependent genes. Extracellular-signalregulated kinase (ERK)1/2 pathway is also involved in the protection of keratinocytes from the proinflammatory effect of IFN-␥. Here we show that, after IFN-␥ stimulation, SOCS1 inhibited IFN-␥ receptor and STAT1 phosphorylation but maintained ERK1/2 activation. SOCS1 was also necessary for the IFN-␥-induced RAS and Raf-1 activities in keratinocytes. The enhanced ERK1/2 pathway in SOCS1-overexpressing keratinocytes was in part responsible for their inability to respond to IFN-␥, in terms of CXCL10 and CCL2 production, and for the high production of CXCL8. Moreover, SOCS1 interacted with the RAS inhibitor p120 RasGAP and promoted its degradation after IFN-␥ stimulation. We hypothesize that SOCS1 functions as suppressor of IFN-␥ signaling, not only by inhibiting STAT1 activation but also by sustaining ERK1/2-dependent antiinflammatory pathways.— Madonna, S., Scarponi, C., De Pita`, O., Albanesi, C. Suppressor of cytokine signaling 1 inhibits IFN-␥ inflammatory signaling in human keratinocytes by sustaining ERK1/2 activation. FASEB J. 22, 3287–3297 (2008)
ABSTRACT
Key Words: skin inflammation 䡠 antiinflammatory pathways 䡠 skin immune responses 䡠 RAS 䡠 RAF-1 䡠 p120RasGAP IFN-␥ is a key cytokine that regulates a variety of cellular activities, including antiviral immunity, apoptosis, and cell cycle progression. In the skin, it plays a pivotal role in the development of inflammatory and immune responses. Keratinocytes, which constitutively bear the IFN-␥ receptor complex, are a primary target of IFN-␥ (1). During immune-mediated skin diseases, after exposure to IFN-␥, keratinocytes become a source of a plethora of inflammatory mediators involved in the initiation and amplification of pathogenetic processes (2– 4). In particular, IFN-␥ induces the expression of numerous chemokines, including CCL2 and CXCL10, which drive immigration of T cells, monocytes, and dendritic cells into inflamed skin. IFN-␥-treated kera0892-6638/08/0022-3287 © FASEB
tinocytes are also a source of CXCL8, a pleiotropic chemokine showing a strong chemoattractant activity on neutrophils as well as tissue-protective and prosurvival functions (5–7). IFN-␥ intracellular signaling activates a number of molecular cascades initiated by Janus-activated kinases 1 (Jak1) and Jak2 phosphorylation and culminates in the activation of transcription factors, mainly signal transducer and activator of transcription (STAT)1 and interferon regulatory factor-1, which can induce the IFN-stimulated gene (ISG) expression (8). In parallel, other signaling pathways functioning simultaneously with Jak/STAT1 are required for a complete IFN-␥ response, as IFN-␥ can still induce part of ISGs, mediate antiviral responses, and regulate cell proliferation in absence of STAT1. Aside from Jak1 and 2, other kinases are activated by IFN-␥, including extracellular signalregulated kinase (ERK)1/2, phosphatidylinositol 3-kinase and Akt, Pyk2, Ca/calmodulin-dependent protein kinase II, and protein kinase-C-␦ (9 –13). Among these enzymes, ERK1/2 kinases have been shown to activate the transcription factors CCAAAT/enhancer-binding protein- and activator protein-1, which are important for the induction of a subset of ISGs (14, 15). ERK1/2 pathway has also been found to regulate the serine phosphorylation of STAT1 and its transactivation potential (8, 16, 17). In human keratinocytes, ERK1/2 activation by IFN-␥ can involve the autocrine activation of epidermal growth factor receptor (EGFR) signaling and requires RAS and Raf-1 molecules (18). In these cells, ERK1/2 activity has an antiinflammatory and self-protective function by down-regulating CCL2, CCL5, and CXCL10 and enhancing CXCL8 expression (19). Accordingly, the impairment of ERK1/2 signaling in the epidermis led to enhanced inflammatory responses in mouse skin models of irritative dermatitis and T cell-mediated contact hypersensitivity (19). Keratinocytes can avoid the detrimental consequences of an excessive stimulation by IFN-␥ by expressing the suppressor of cytokine signaling (SOCS)1 and, to a lesser extent, SOCS3 proteins that complete a negative feedback loop attenuating IFN-␥ signal trans1
Correspondence: Laboratory of Immunology, IDI-IRCCS, Via Monti di Creta, 104, 00167 Rome, Italy. E-mail:
[email protected] doi: 10.1096/fj.08-106831 3287
duction (20, 21). The ability of SOCS1 to inhibit IFN-␥ signaling resides in its capacity to bind and inactivate Jak1 and Jak2 proteins. In particular, SOCS1 functions as a pseudosubstrate inhibitor of Jak1 and Jak2 and impedes the IFN-␥-induced tyrosine phosphorylation of IFN-␥ receptor (IFN-␥ R)␣ subunit and STAT1 activation (20). As a direct consequence of the lack of STAT1 function, keratinocytes overexpressing SOCS1 cannot express the ICAM-1 and MHC class II membrane molecules, as well as CXCL10, CXCL9, and CCL2 in response to IFN-␥ (20, 22). Moreover, SOCS1-expressing keratinocytes showed constitutively higher, but not IFN-␥-inducible, CXCL8 levels and were resistant to IFN-␥-mediated growth inhibition. A growing number of studies indicate that SOCS proteins not only suppress cytokine-mediated Jak/STAT signaling, but, in parallel, sustain other pathways that are triggered by the same receptor. In particular, IL-2-induced SOCS3 strongly inhibits STAT5 activation, but, by sequestering the RAS inhibitor p120 RasGAP, can induce RAS/ ERK1/2 pathways (23). Similarly, the SOCS box-containing protein SSB-1 enhances hepatocyte growth factor-induced ERK1/2 phosphorylation following its interaction with p120 RasGAP (24). Recently, SOCS1 has been found to promote simultaneously STAT1 inhibition and MAPK activation, following its interaction with fibroblast growth factor receptor 3 (25). In this study, we provide experimental evidence that SOCS1 is required to sustain the IFN-␥-induced RAS/ ERK1/2 signaling, which in turn is involved in the control of chemokine expression in epidermal keratinocytes. This event could be in part primed by SOCS1mediated degradation of p120 RasGAP. Our data indicate that SOCS1 suppresses IFN-␥ signaling in keratinocytes, not only by repressing STAT1 activation at the IFN-␥ R complex but also by inducing antiinflammatory and self-protective pathways triggered by RAS/ ERK1/2 molecules.
lation with 200 U/ml human recombinant IFN-␥ (R&D Systems, Minneapolis, MN, USA) was performed in keratinocyte basal medium (KBM; Clonetics). The HaCaT human keratinocyte cell line was a gift from N. E. Fusenig (Deutsches Krebsforschungszentrum, Heidelberg, Germany) and was grown in Dulbecco modified Eagle medium (DMEM; Biochrom, Cambridge, UK) supplemented with 10% Fetalclone II serum (HyClone Laboratories, South Logan, UT, USA). Transient and permanent transfections Cultured keratinocytes were transiently transfected in triplicate using Lipofectin reagent (Invitrogen Life Technology, Carlsbad, CA, USA), as previously reported (20). HaCaT cells were stably transfected with myc/SOCS1 or empty pcDNA3 plasmids as previously reported (20). HaCaT clones expressing ectopically SOCS1 were screened by Western blot analysis using the anti-c-myc antibody (Ab) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). If necessary, HaCaT clones were treated with the chemical inhibitor PD98059 (Calbiochem, Gibbstown, NJ, USA) in DMEM, at concentrations ranging from 1 to 50 M. Transient RNA interference SOCS1 was knocked down in keratinocyte cultures by using a pool of four small interfering RNAs (siRNAs) provided by Dharmacon RNA Technology (Lafayette, CO, USA) (ONTARGETplus SMARTpool, L-011511– 00-0005), whose sequences (Sense Sequences: 5⬘-GAGCCCCGACUGCCUCUUCUU; 5⬘-UCCGUUCGCACGCCGAUUAUU; 5⬘-GCAUCCGCGUGCACUUUCAUU; 5⬘-CCAGGUGGCAGCCGACAAUUU) annealed on different regions of human SOCS1 mRNA (NM_003745). In parallel, a pool of four nontargeting siRNAs was used as negative control (ON-TARGET plus siCONTROL, D-001810 –10-05). Primary cells cultured in 10-cm dishes were transfected with SOCS1 or irrelevant siRNAs at a 30 nM final concentration using TransIT-TKO Trasfection Reagent (Mirus Bio Corporation, Madison, WI, USA), according to the manufacturer’s protocol. After 24 h of transfection, the cells were stimulated with IFN-␥. RNA isolation and real-time RT-PCR
MATERIALS AND METHODS Plasmids Myc-tagged full-length JAB/SOCS1 in pcDNA3 plasmid (myc/SOCS1) was a generous gift of Prof. A. Yoshimura (Department of Immunobiology and Neuroscience, Medical Institute of Bioregulation, Kyushu University, Fukuoka, Japan). His-tagged ubiquitin expression plasmid was a gift of Prof. S. Ferrari (Institute of Molecular Cancer Research, Zurich University, Zurich, Switzerland). HA-tagged RasGAP cDNA in pcDNA3 expression plasmid, kindly provided by Prof. C. Widmann (Dept. of Cellular Biology and Morphology, Faculty of Biology and Medicine, Lausanne University, Lausanne, Switzerland), was modified by adding three additional HA-tags at the 5⬘ terminus of cDNA.
Total RNA was extracted using the TRIzol reagent (Invitrogen). mRNA was reverse-transcribed into cDNA and analyzed by real-time RT-PCR. Real-time PCR was performed using SYBR Green PCR reagents (Applied Biosystems, Branchburg, NJ, USA) and primers specific for SOCS1 and -actin mRNAs. Fluorescence intensity was analyzed by the ABI PRISM SDS 7000 PCR Instrument (Applied Biosystems). The forward and reverse primers used for PCR were as follows: for SOCS1, 5⬘-TTTTTCGCCCTTAGCGTGA-3⬘ and 5⬘-AGCAGCTCGAAGAGGCAGTC-3⬘; for SOCS3, 5⬘-AAGGACGGAGACTTCGATTCG-3⬘ and 5⬘-AAACTTGCTGTGGGTGACCAT-3⬘; for -actin, 5⬘-CATCGAGCACGGCATCGTCA-3⬘ and 5⬘-TAGCACAGCCTGGATAGCAAC-3⬘. The levels of SOCS1 expression were determined by normalizing to -actin expression. The fold induction value for triplicate wells was averaged, and data were presented as the mean ⫾ se.
Keratinocyte cultures and treatments Immunoprecipitation and immunoblotting Normal human keratinocytes were obtained from skin biopsies of healthy volunteers (n⫽3), as previously reported (20). Briefly, cells were cultured in the serum-free medium, keratinocyte growth medium (KGM; Clonetics, Walkersville, MD, USA), for at least 3–5 days (at 60 – 80% confluence). Stimu3288
Vol. 22
September 2008
Protein extracts were prepared by solubilizing cells in RIPA buffer (1% Nonidet P-40, 0.5% sodium dehoxycholate, and 0.1% sodium dodecyl sulfate (SDS) in PBS containing a mixture of protease and phosphatase inhibitors. Proteins
The FASEB Journal
MADONNA ET AL.
were subjected to SDS-PAGE, and transferred to polyvinylidene difluoride (PVDF) membranes. The latter were blocked and probed with various primary Abs diluted in PBS containing 5% nonfat dried milk or 3% BSA. For immunoprecipitation, 500 g of protein extracts was incubated with protein A-Agarose beads (Amersham Pharmacia Biotech, Buckinghamshire, UK) and the specific Abs. Proteins eluted from the beads were resolved on 8% SDS-PAGE, transferred to PVDF filters, and probed with primary Abs. Filters were developed using the ECL-plus detection system (Amersham Pharmacia Biotech) or, otherwise, the SuperSignal West Femto kit (Pierce, Rockford, IL, USA). The Abs used for the study were as follows: anti-SOCS1 and anti-SOCS3 (MBL International Corporation, Nakaku Nagoya, Japan), anti-ERK1/2 (C16; Santa Cruz Biotechnology), anti-phospho-ERK1/2 (E4; Santa Cruz Biotechnology), anti-phosphotyrosine (clone 4G10; Upstate Biotechnologies, Temecula, CA), anti-phosphotyrosine (PY) STAT1 (Tyr701) (Santa Cruz Biotechnology), anti-IFN-␥ R␣ subunit (C-20; Santa Cruz Biotechnology), anti-p120 RasGAP (B4F8; Santa Cruz Biotechnology), anti-Sos1 (C-23; Santa Cruz Biotechnology), anti-NF1 (D; Santa Cruz Biotechnology), horseradish peroxidase (HRP) -conjugated anti-cmyc (9E10; Santa Cruz Biotechnology), anti-HA-probe (F-7; Santa Cruz Biotechnology), and anti-His-probe (H-3; Santa Cruz Biotechnology). In vitro kinase assays ERK1/2 and RAS activation assays were performed using commercial kits from Cell Signaling Technology (Boston, MA, USA) (p44/42 MAP kinase nonradioactive kit) and Upstate (RAS activation nonradioactive assay kit), respectively. To test Raf-1 activity, Raf-1 was previously immunoprecipitated from keratinocyte lysates and then incubated with the kinase assay mixture containing mitogen-activated ERKactivated kinase (MEK)1 as substrate (Raf-1 kinase assay, Upstate). Ubiquitin assay Ubiquitin assay was performed as follows: cultured keratinocytes were transfected with combinations of 1 g of HA-p120 RasGAP, 0.5 g of His-Ubiquitin, and 0.5 g of Myc-tagged SOCS1 plasmids. A plasmid containing the green fluorescence protein (GFP) gene was used to normalize the amounts of DNA transfected into the cells. After 48-h transfection, cells were treated with 20 M MG132 (Sigma-Aldrich) for 4 h and then lysed in RIPA buffer. Ubiquitinated HA-p120 RasGAP was detected by immunoprecipitation with anti-HA agarose beads (Santa Cruz Biotechnology) and Western blotting performed with an anti-His Ab. Carboxyfluorescein diacetate succinimidyl ester (CFSE) assay Cell suspensions of mock-transfected (n⫽8) or SOCS1 (n⫽8) clones were incubated with the fluorescent dye CFSE (5 M) (Sigma-Aldrich, St. Louis, MO, USA) for 4 min at 37°C in the dark and seeded in 6-well plates. After 24 h, an aliquot of CFSE-stained cells was subjected to FACScan analysis to set basal fluorescence intensity (day 0, 100% of positivity). The remaining cells were left untreated or stimulated with IFN-␥ for 2, 4, 7, 10, and 14 days and then analyzed for CFSE fluorescence. The proliferation rate of clones was expressed as a percentage of CFSE fluorescence reduction in cells collected at sequential time-points. Experiments were performed in triplicate for each condition. SOCS1 SUSTAINS ERK1/2 ACTIVATION IN KERATINOCYTES
ELISA CXCL10 was assayed using the purified 4D5/A7/C5 and the biotinylated 6D4/D6/G2 Abs (BD Pharmingen, Franklin Lakes, NJ, USA). CXCL8 and CCL2 were measured with OptEIA kits (BD Pharmingen) as reported in the manufacturer’s protocol. The plates were analyzed in an ELISA reader (model 3550 UV; Bio-Rad, Hercules, CA, USA). Cultures were conducted in triplicate for each condition. Results are given as ng/106 cells ⫾ sd. Densitometry and statistical analysis Immunoblots were subjected to densitometry using an Imaging Densitometer model GS-670 (Bio-Rad) supported by the Molecular Analyst software, and band intensities were evaluated in three independent experiments. Data are expressed as fold induction (FI) ⫾ sd in experimental IFN-␥ time course relative to untreated samples, to which were given a value of 1. Wilcoxon’s signed rank test (SigmaStat; Jandel, San Rafael, CA, USA) was used to compare data in P-ERK1/2 Western blot experiments, and in ERK1/2, RAS and Raf-1 assays. Significant differences were also determined in CFSE staining and ELISA experiments conducted in SOCS1 and control clones. Values of P ⱕ 0.05 were considered significant.
RESULTS IFN-␥ up-regulates SOCS1 and induces a biphasic activation of ERK1/2 signaling in human keratinocytes ERK1/2 induction by IFN-␥ in keratinocytes has been described as an early event (5 min after IFN-␥ administration) and can be mediated by EGFR signal transduction cascade. In particular, IFN-␥ promotes a rapid keratinocyte release of EGFR ligands, including transforming growth factor-␣, which induce EGFR autophosphorylation and, consequently, ERK1/2 activation (19). Since previous studies indicated an involvement of SOCS3 and SSB-1 proteins in sustaining ERK1/2 pathway, we asked whether SOCS1 could be responsible for the IFN-␥-induced ERK1/2 activation in keratinocytes. Thus, we firstly analyzed SOCS1 mRNA and protein expression in human keratinocytes at different time points after IFN-␥ treatment. As shown in Fig. 1, SOCS1 was up-regulated at significant levels only 1 h after IFN-␥ administration, reaching a peak at 3 h and slowly decreasing thereafter. These data excluded the possibility that the early ERK1/2 activation by IFN-␥ could depend on SOCS1 function. However, a more detailed kinetic analysis of ERK1/2 activation revealed that IFN-␥ had a biphasic effect on ERK1/2 phosphorylation and activity in cultured keratinocytes (Fig. 2). In particular, phosphorylation of ERK1/2 was up-regulated rapidly after treatment with IFN-␥ for 5 min (early activation), gradually declined to the basal level after 30 min, and returned to high at 3 h (late activation) (Fig. 2A). Rephosphorylation of ERK1/2 lasted for at least 3 h. In parallel, ERK1/2 function, as measured by its ability to phosphorylate the transcription factor Ets-like gene (Elk)-1, was increased by IFN-␥ following a bipha3289
Figure 1. SOCS1 expression is induced by IFN-␥ in human keratinocytes. Keratinocyte cultures prepared from skin biopsies of healthy volunteers were left unstimulated or treated with 200 U/ml IFN-␥ for the indicated time periods. A) SOCS1 mRNA expression analysis by real-time RT-PCR. The levels of SOCS1 mRNA are normalized to -actin transcripts. Data are representative of 3 independent experiments performed on 3 keratinocyte strains. B) Expression of SOCS1 protein as determined by immunoprecipitation (IP) and Western blotting (WB) analysis performed with an antiSOCS1 Ab on keratinocyte lysates. Data are representative of 3 independent experiments performed on 3 keratinocyte strains.
sic activation, with a first peak present at 5 min and a second at 3 h (Fig. 2B, top panel). This biphasic trend was absent in resting cells, which showed a constant ERK1/2 activity during the time-course experiment (Fig. 2B, bottom panel). These results prompted us to investigate on the possible role of SOCS1 in mediating the late activation of ERK1/2 pathway induced by IFN-␥ in cultured keratinocytes.
an enhancement of the phosphorylation state of IFN ␥R␣ and STAT1 in keratinocytes following their exposure to IFN-␥ (Fig. 3B, C). The increase started at 1 h after IFN-␥ administration and persisted for at least 5 h. As a consequence of the prolonged phosphorylation of IFN ␥R␣ and STAT1, SOCS1-depleted cells released higher levels of CXCL10 and CCL2 in response to IFN-␥ compared to controls (Fig. 3D). These data fit with our previous findings, which demonstrate the impairment of IFN-␥ R␣ and STAT1 phosphorylation and the reduction of CXCL10 or CCL2 release in keratinocyte clones permanently transfected with SOCS1 (20). Analysis of ERK1/2 phosphorylation and activity conducted on RNA interference experiments demonstrated that SOCS1 inhibition significantly diminished the late, but not early, ERK1/2 activation induced by IFN-␥ in keratinocytes (Fig. 4A, B, bottom panels). Differences between SOCS1-depleted and control keratinocytes in ERK1/2 activation kinetics could be clearly appreciated and quantified by densitometric analyses. The abrogation of ERK1/2 phosphorylation and activity was maximal at 3 h of IFN-␥ treatment in keratinocytes transduced with SOCS1, as compared to control siRNA. (P-ERK1/2 FI⫽13 in SOCS1-depleted keratinocytes vs. P-ERK1/2 FI⫽98 in control cells; P-Elk1 FI⫽14 in SOCS1-depleted keratinocytes vs. PElk1 FI⫽100 in control cells) (Fig. 4A, B). To further confirm the functional role of SOCS1 in sustaining the IFN-␥-dependent ERK1/2 pathway, the keratinocytelike cell line HaCaT was stably transfected with myc/ SOCS1 or control plasmids. However, before establishing clones, HaCaT cells were analyzed and compared to normal keratinocytes in IFN-␥-induced ERK1/2 signal-
SOCS1 sustains the late activation of ERK1/2 pathway induced by IFN-␥ in human keratinocytes To demonstrate the possible involvement of SOCS1 in sustaining the late activation of ERK1/2 by IFN-␥, we took advantage of RNA interference and stably transfectant technologies. In RNA interference experiments, we first transfected keratinocytes with a pool of four SOCS1 or irrelevant siRNA, and then treated cells with IFN-␥ in time-course experiments. As shown in Fig. 3A, keratinocytes transfected with SOCS1 siRNA, but not with scrambled (NC) siRNA, expressed reduced and negligible levels of endogenous SOCS1 mRNA and protein, respectively. The specificity of the SOCS1 mRNA down-regulation was validated by examining in parallel the IFN-␥-induced expression of endogenous SOCS3 (Fig. 3A). In addition, we evaluated the inhibitory role of SOCS1 in the IFN-␥-induced phosphorylation of the IFN-␥ R␣ subunit and STAT1. As compared to control, transfection with SOCS1 siRNA determined 3290
Vol. 22
September 2008
Figure 2. IFN-␥ induces ERK1/2 activation in keratinocytes in a biphasic fashion. A) ERK1/2 phosphorylation as monitored on IFN-␥-treated keratinocytes by WB with an anti-phospho ERK1/2 (P-ERK) Ab, specific for both p42 and p44 ERK phosphorylated at Tyr 204 residue. Filters were stripped and reprobed with an anti-ERK1/2 Ab. B) ERK1/2 activity assayed on lysates from IFN-␥-treated keratinocyte cultures (top panel) and from resting cells (bottom panel) and determined by IP of P-ERK1/2, followed by in vitro kinase assay with Elk-1 as substrate.
The FASEB Journal
MADONNA ET AL.
Figure 3. Inhibition of SOCS1 expression determines a long-lasting phosphorylation of IFN-␥R␣ and STAT1 in IFN-␥-treated keratinocytes. Transient RNA interference was performed by transfecting cultured keratinocytes with a pool of four SOCS1 or irrelevant (NC) siRNA and then by treating the cells with IFN-␥ for the indicated time periods. A) SOCS1 mRNA levels in siRNA-transfected cells measured by real-time PCR and compared to those present in cells transfected with NC siRNA. SOCS3 mRNA levels were also monitored to validate the specificity of SOCS1 siRNA. SOCS1 and SOCS3 protein levels in siRNA-transfected cells were determined by SOCS1 or SOCS3 IP from total lysates, followed by WB analysis with anti-SOCS1 and anti-SOCS3 Abs. B) IFN-␥ R␣ phosphorylation analysis by IP of keratinocyte lysates with the IFN-␥R␣ chain Ab, followed by WB analysis with anti-PY Ab. C) STAT1 phosphorylation as detected by WB with an Ab specific for STAT1 phosphorylated at Tyr 701 residue. D) Keratinocyte cultures transfected with 10 and 50 nM NC siRNA (white bars) or SOCS1 siRNA (black bars) and then stimulated with IFN-␥ for 24 h. Supernatants analyzed for CXCL10 and CCL2 content by ELISA. Data are expressed as mean ⫾ sd ng/106 cells of triplicate cultures.
ing pathways. IFN-␥ promoted the same profile and kinetics of ERK1/2 activation in HaCaT cells and normal keratinocytes (data not shown). HaCaT clones permanently expressing SOCS1 were screened for ectopic SOCS1, and 10 SOCS1 clones were obtained and included in this study. To test the effectiveness of SOCS1 overexpression, SOCS1 clones were tested for their inability to phosphorylate STAT1 and to express proinflammatory chemokines in response to IFN-␥ (data not shown). All the SOCS1 stable clones showed an enhanced basal ERK1/2 activity compared to control cells (Fig. 4C). The enhancement of ERK1/2 activity was proportional to the amounts of SOCS1 gene expressed by the keratinocyte clones. As a whole, these data demonstrate an involvement of SOCS1 in supporting ERK1/2 activation induced by IFN-␥ in human keratinocytes. SOCS1 participates in the induction of RAS pathway by IFN-␥ In many signaling pathways, ERK1/2 function is regulated by other upstream kinases, known as MEK1/2 and Raf-1, whose phosphorylation is, in turn, preceded by RAS activation. Therefore, we next investigated whether SOCS1 can also modulate the IFN-␥-induced RAS cascade in keratinocytes. To this end, RAS and Raf-1 activation were analyzed in SOCS1-interfered SOCS1 SUSTAINS ERK1/2 ACTIVATION IN KERATINOCYTES
keratinocytes and in SOCS1-overexpressing clones. Measurement of RAS activation was performed using an affinity precipitation assay based on the specific interaction between RAS-GTP and the RAS binding domain of Raf-1 (RAS pull-down). Phosphotransferase ability of active Raf-1 was assessed by an immunoprecipitation-kinase assay using MEK1 as substrate for Raf-1. As shown in Fig. 5A, RAS activity was rather low in untreated keratinocytes but could be substantially upregulated after 1 h of IFN-␥ treatment, reaching a maximum at 3 h (FI⫽100, P⬍0.01) and declining slowly thereafter. In contrast, IFN-␥-treated keratinocytes transfected with SOCS1 siRNA showed only a slight RAS activation (FI⫽3) (Fig. 5A). Vice versa, when SOCS1 was overexpressed in keratinocytes, the basal levels of activated RAS significantly augmented compared to untreated mock-transfected cells (Fig. 5B). RAS activation was low in untreated HaCaT cells compared to keratinocytes cultures but could be induced after 3 h of IFN-␥ treatment. GTP-RAS levels in SOCS1 clones were proportional to the amount of ectopically expressed SOCS1 (Fig. 5B and data not shown). Consistent with RAS, Raf-1 kinase also reached a maximal activity after 3 h of IFN-␥ stimulation (FI⫽74, P⬍0.01). This up-regulation was suppressed by transfecting keratinocyte cultures with SOCS1 siRNA (FI⫽2.2) (Fig. 5C). Together, our results suggest that SOCS1 is fun3291
Figure 4. SOCS1 sustains the late activation of ERK1/2 pathway induced by IFN-␥ in human keratinocytes. Protein lysates from cultured keratinocytes subjected to transient RNA interference experiments analyzed for both P-ERK1/2 and ERK1/2 expression by WB (A) or ERK1/2 activity by assaying Elk-1 phosphorylation (B). Bar graphs represent densitometric analyses of P-ERK1/2 (A) and P-Elk1 (B) in keratinocytes transfected with NC or SOCS1 siRNA. Data are expressed as FI ⫾ sd in experimental IFN-␥ time course relative to untreated samples (time 0), which are given a value of 1. C) HaCaT cells stably transfected with myc/SOCS1 or empty pcDNA3 plasmids (Mock-myc). Lysates from untreated clones assayed for ERK1/2 in vitro activity. Ectopical SOCS1 expression in different clones was determined by WB with an anti-c-myc Ab.
damental for the IFN-␥-induced RAS-Raf-1 signaling in human keratinocytes. As a consequence of activation of RAS/Raf-1/ERK1/2 signaling pathway, profound changes in the proliferation rate of keratinocytes could occur. We therefore evaluated whether overexpression
of SOCS1 in keratinocytes influenced cell division and viability using the CFSE dye, as described in Material and Methods. CFSE-labeled SOCS1 or control keratinocyte clones were kept in culture for different time periods and then analyzed by flow cytometry for CFSE
Figure 5. SOCS1 participates in the induction of RAS pathway by IFN-␥. RAS activity was measured on lysates obtained from keratinocytes transfected with SOCS1 or NC siRNA (A) and myc-mock control or SOCS1 HaCaT clones (B). GTP-bound RAS was pulled down using Raf-1 RBD agarose as substrate and detected by WB with an anti-RAS Ab. In parallel, lysate obtained from Mock-myc 2.1 clone was loaded with GTP␥S (first lane) or with GDP (second lane) before incubating with Raf-1 RBD agarose, as positive and negative controls, respectively. C) NC- or SOCS1-interfered keratinocytes are also analyzed for Raf-1 activity, as determined by IP of Raf-1, followed by in vitro kinase assay with MEK-1 as substrate. Data are expressed as FI ⫾ sd in experimental IFN-␥ time course relative to untreated samples (time 0), which are given a value of 1.
3292
Vol. 22
September 2008
The FASEB Journal
MADONNA ET AL.
Inhibition of ERK1/2 signaling in SOCS1 clones partially reverts their inability to produce CXCL10 and CCL2 in response to IFN-␥ and constitutive release of CXCL8
Figure 6. SOCS1 does not influence basal growth of keratinocytes but prevents the antiproliferative effect of IFN-␥. Control (black squares) and SOCS1 (white squares) HaCaT clones were CFSE-labeled, kept in culture for the indicated days, and then analyzed by flow cytometry. The reduction of CFSE staining is followed in absence (A) or in presence (B) of IFN-␥. * P ⬍ 0.05 vs. IFN-␥-treated control clones.
fluorescence intensity. The decrement of CFSE positivity, which paralleled the number of cell divisions, progressively augmented with the duration of keratinocyte culture, which reached a maximum at 14 days after seeding (Fig. 6A). The reduction of CFSE staining was similar in SOCS1 and control keratinocytes clones, indicating that the constitutive expression of SOCS1 does not influence cell division. Also, basal apoptosis was comparable between SOCS1 and mock-transfected cells (data not shown). However, when keratinocyte clones were subjected to IFN-␥, only SOCS1 clones cells showed a marked resistance to its antiproliferative activity (Fig. 6B). In fact, mock-transfected cells, unlike SOCS1-expressing keratinocytes, ceased to divide at 4 –7 days of cultures in the presence of IFN-␥. Therefore, although SOCS1 constitutive expression does not influence basal keratinocyte growth it can efficiently prevent the IFN-␥-induced antiproliferative activity.
The main feature of keratinocytes overexpressing SOCS1 is their inability to express STAT1-dependent genes, including CXCL10 and CCL2, in response to IFN-␥. In addition, SOCS1-overexpressing keratinocytes produce constitutive higher, but not IFN-␥-inducible, levels of CXCL8 compared to control or SOCS2–3 clones (20). Since SOCS1 sustains ERK1/2 and ERK1/2 is known to suppress the IFN-␥-driven activation of keratinocytes (19), we asked whether the SOCS1-dependent ERK1/2 activation in keratinocyte clones could be, in part, responsible for their inability to produce CXCL10 and CCL2 and for the enhanced CXCL8 release. For these experiments, we used clones expressing intermediate levels of SOCS1 so that, in addition to ERK1/2, a partial STAT1 activation was possible on IFN-␥ stimulation (data not shown). IFN-␥activated SOCS1 clones expressed reduced levels of CXCL10 and CCL2 compared to control cells (Fig. 7A, B). However, incubation of SOCS1 clones with the chemical MEK1/2 inhibitor PD98059 dose-dependently augmented the IFN-␥ induced production of CXCL10 and CCL2. However, the chemokine content of SOCS1 clones never did achieve that of mocktransfected cells, also at the highest dose of PD98059 (Fig. 7A, B). These data suggest that the ERK1/2 pathway is in part responsible for the inability of SOCS1 clones to express adequate levels of CXCL10 and CCL2 in response to IFN-␥. In addition, the PD98059 effect on CXCL8 production by SOCS1 keratinocyte clones was investigated. Unstimulated SOCS1 clones secreted higher levels of CXCL8 compared to control cells, which produced detectable CXCL8 only on IFN-␥ treatment (Fig. 7C and data not shown). Addition of PD98059 to culture medium dose-dependently reduced CXCL8 constitutive production by SOCS1 clones. These findings indicate that ERK1/2 signaling is fundamental for the enhanced expression of CXCL8 in SOCS1-overexpressing keratinocytes. SOCS1 interacts with the RAS inhibitor p120 RasGAP and promotes its degradation RAS is activated by exchange of GDP for GTP and is turned off by hydrolysis of bound GTP to GDP. RAS
Figure 7. Inhibition of ERK1/2 signaling in SOCS1 clones reverts their inability to produce CXCL10 and CCL2 in response to IFN-␥ and constitutive release of CXCL8. SOCS1 1.4 (white bars) keratinocyte clone was left untreated or stimulated with IFN-␥ for 24 h, in the presence of increasing doses (0.2, 2, 10, 20 M final concentration) of the chemical MEK1/2 inhibitor PD98059. Myc-mock 2.1 clone (black bar) was stimulated with IFN-␥ for 24 h and used as control. Supernatants analyzed for CXCL10 (A), CCL2 (B), and CXCL8 (C) content by ELISA. Data are expressed as mean ⫾ sd ng/106 cells of triplicate cultures. SOCS1 SUSTAINS ERK1/2 ACTIVATION IN KERATINOCYTES
3293
Figure 8. SOCS1 binds to p120 RasGAP and induces its ubiquitination and degradation. A) IP with anti-p120 RasGAP or anti-SOCS1 followed by WB with anti-SOCS1 or anti-p120 RasGAP, respectively, performed on lysates from unstimulated or IFN-␥-treated keratinocytes. B) SOCS1-p120 RasGAP interaction is detected in HaCaT clones by IP of lysates with anti-c-myc or anti-p120 RasGAP. C–E) p120 RasGAP levels were detected in control and SOCS1 keratinocyte clones (C), in keratinocyte cultures transiently transfected with HA-tagged p120 RasGAP either in absence or presence of increasing myc/SOCS1 plasmid (D), or in NC- and SOCS1-interfered keratinocytes (E). F) Keratinocyte cultures transfected with expression vectors as indicated, and left untreated or treated with 20 M MG132 for 4 h. Lysates were immunoprecipitated with anti-HA-agarose beads and immunobloted with anti-His for the detection of polyubiquitinated p120 RASGAP.
inactivation can be induced by a variety of GTPaseactivating proteins (GAPs), including p120 RasGAP. Previous studies have shown that both SOCS3 and SSB-1 can bind and inhibit p120 RasGAP and thus promote activation of the RAS/ERK1/2 cascade (23, 24). We therefore sought to examine whether the IFN-␥-induced RAS activation in keratinocytes could depend on the capacity of SOCS1 to interact with p120 RasGAP and to regulate its accumulation in IFN-␥activated keratinocytes. Coimmunoprecipitation experiments were performed on lysates from keratinocyte cultures stimulated with IFN-␥ for different time periods and incubated with an Ab specific for p120 RasGAP or SOCS1 (Fig. 8A). An association of SOCS1 with p120 RasGAP was revealed in IFN-␥-treated keratinocytes using either anti-p120 RasGAP or anti-SOCS1 Abs. SOCS1 and p120 RasGAP interaction started at 30 min after IFN-␥ stimulation and lasted for at least 6 h. 3294
Vol. 22
September 2008
However, the amounts of SOCS1 and p120 RasGAP in the immunoprecipitated complexes varied in a timedependent manner. In fact, p120 RasGAP protein was represented more in the immunoprecipitates at 30 min and 1 h after IFN-␥ simulation, whereas SOCS1 was more abundant at 1–3 h (Fig. 8A). SOCS1 and p120 RasGAP interaction was further confirmed in keratinocyte clones overexpressing SOCS1. The protein product of ectopic SOCS1 (myc/SOCS1) coimmunoprecipitated with endogenous p120 RasGAP in the presence of anti-c-myc or anti-p120 RasGAP Abs (Fig. 8B). In contrast, no myc/SOCS1/p120 RasGAP interaction was observed in mock-transfected keratinocyte clones. The well-known role of SOCS1 in mediating the degradation of a number of target proteins has prompted us to examine the ability of SOCS1 to influence p120 RasGAP accumulation in keratinocytes. p120 RasGAP levels were significantly lower in SOCS1 overexpressing
The FASEB Journal
MADONNA ET AL.
clones compared to control cells (⬎3.5-fold) and inversely related to the amounts of myc/SOCS1 (Fig. 8C and data not shown). Consistently, by cotransfecting keratinocyte cultures with a constant amount of HAtagged p120 RasGAP either in absence or in presence of increasing amounts of myc/SOCS1, we found that SOCS1 could induce the degradation of HA-tagged RasGAP in a dose-dependent manner (Fig. 8D). In parallel, the treatment of cells with the proteasomal inhibitor MG132 for 4 h reverted SOCS1 effect on RasGAP accumulation (Fig. 8D), suggesting that p120 RasGAP degradation is mediated by SOCS1 through the 26S proteasome machinery. To further investigate SOCS1 function in regulating p120 RasGAP accumulation after IFN-␥ stimulation, we performed interference experiments on endogenous SOCS1 mRNA. The slight increase of p120 RasGAP observed in keratinocytes after 30 min of IFN-␥ treatment significantly declined at 1 h to reach minimal levels at the 3– 6 h (Fig. 8E). In contrast, p120 RasGAP reduction was no longer detected in keratinocyte cultures transfected with SOCS1 siRNA, confirming that SOCS1 can reduce the intracellular accumulation of p120 RasGAP. Because SOCS1 has E3 ligase activity that results in polyubiquitination of target proteins (22) and that we had demonstrated a role for SOCS1 in mediating p120 RasGAP degradation, we wanted to determine whether the mechanism of this degradation depended on the ability of SOCS1 to induce p120 RasGAP ubiquitination. To this end, HA-tagged p120 RasGAP was coexpressed with Histagged ubiquitin plasmids, either in presence or absence of SOCS1 in keratinocyte cultures. Transfected cells were then incubated with the proteasome inhibitor MG132 or were left untreated and analyzed by immunoprecipitation and immunoblotting. Significant amounts of ubiquitinated p120 RasGAP molecules could be detected in keratinocytes only in presence of SOCS1 and following their treatment with MG132 (Fig. 8F). As a whole, these data indicate that SOCS1 can reduce p120 RasGAP protein accumulation in IFN-␥treated keratinocytes by inducing its ubiquitination and degradation by proteasome machinery.
DISCUSSION The findings that ERK1/2 suppress a number of potent proinflammatory chemokines in epidermal keratinocytes and eventually in skin immune responses suggest that these mitogen-activated protein kinases are part of a homeostatic mechanism that tends to oppose skin inflammation (19). SOCS1 also is an important negative regulator of the proinflammatory functions of IFN-␥-activated keratinocytes and is known to act primarily through the inhibition of STAT1 at the IFN-␥ R complex (20). In this study, we provide evidence that SOCS1 suppresses IFN-␥ signaling in keratinocytes, not only by repressing IFN-␥ R␣ and STAT1 phosphorylation but also by sustaining antiinflammatory pathways triggered by ERK1/2. The latter were up-regulated in IFN-␥-treated keratinocytes in a biphasic fashion, with an early and late peak of induction occurring at 5 min and 3 h, respectively. Silencing of SOCS1 expression SOCS1 SUSTAINS ERK1/2 ACTIVATION IN KERATINOCYTES
significantly diminished the late, but not early, ERK1/2 activation in keratinocytes. The inhibition of both ERK1/2 phosphorylation and activity could be detected starting from 1 h after IFN-␥ exposure and was highest at 3 h. In addition, we found that keratinocytes permanently transfected with SOCS1 showed an enhanced basal ERK1/2 activity compared to control cells. As a whole, these data demonstrate the involvement of SOCS1 in supporting ERK1/2 activation, but it is clear that SOCS1 cannot be required for the early activation of ERK1/2. In fact, ERK1/2 early up-regulation occurs when endogenous SOCS1 is not yet expressed at adequate levels. Only later (1–3 h after IFN-␥ stimulation), when SOCS1 significantly accumulates, can it mediate and sustain the late activation of ERK1/2. Previous works have demonstrated extensively the biphasic pattern of ERK1/2 activity in response to various stimuli, such as stromal cell-derived factor-1 and retinods, and the molecular mechanisms underlying the induction of the first and second peak of ERK1/2 induction (26 – 28). The current study is the first demonstration that SOCS1 is required for the late activation of ERK1/2 in response to IFN-␥. It remains to comprehend the significance of ERK1/2 biphasic induction, although it can be argued that the early activation of ERK1/2 could somehow regulate SOCS1 expression and, ultimately, the late ERK1/2 induction. Similar to ERK1/2, RAS and Raf-1 activities were also found to be sustained by SOCS1. In fact, inhibition of SOCS1 expression in keratinocytes did not permit any RAS or Raf-1 activation after IFN-␥ stimulation. Vice versa, when SOCS1 was overexpressed by stable transfection, keratinocytes showed enhanced basal levels of activated RAS when compared with control clones. These findings are in line with a previous study showing a regulatory function of SOCS3 in sustaining the activation of RAS/ERK1/2 pathway triggered by IL-2 (23). This SOCS3-dependent regulation of RAS signaling depended on a direct interaction of phosphorylated SOCS3 with p120 RasGAP SH2 domains. Although we could not detect SOCS1 in a phosphorylated status after IFN-␥ treatment (data not shown), coimmunoprecipitation experiments conducted on keratinocyte cultures or SOCS1 clones demonstrated an association of SOCS1 with p120 RasGAP. The fact that SOCS1 cannot be phosphorylated after IFN-␥ treatment could depend on the lack in its C-terminus region of an YXXP motif, determinant for phosphorylation, which is instead present in SOCS3, but also in other SOCS proteins, such as CIS, SOCS2– 6, WSB, and ASB (29). Differently from SOCS3, unphosphorylated SOCS1 can interact with a number of protein partners, including apoptosis signal-regulating kinase (Ask)1 and Myd88-adaptor like, whose phosphorylation is, however, determinant for their binding to SOCS1 (30, 31). Indeed, SOCS1/p120 RasGAP interaction could be dependent on p120 RasGAP phosphorylation in tyrosine residues, which was detected in keratinocytes 1–3 h after IFN-␥ administration (data not shown). Unlike p120 RasGAP, neurofibromin 1, and the son-of-sevenless 1 protein, other negative or positive regulators of RAS activity, respectively, did not interact with SOCS1 in IFN-␥-activated keratinocytes (data not shown). We found that the reduced p120 3295
RasGAP protein accumulation in IFN-␥-treated keratinocytes could depend on the ability of SOCS1 to induce p120 RasGAP degradation via ubiquitination, similarly to what observed for other target proteins (31–33). The SOCS1-dependent p120 RasGAP degradation can be, in part, responsible for the IFN-␥-dependent activation of RAS/ERK cascade in human keratinocytes. RAS activation has an important role in epidermal homeostasis, as it maintains the proliferative capacity of epithelial progenitors and actively opposes the onset of differentiation (34, 35). However, biological effects of RAS signaling strongly depends on strength of inducing stimuli and on the activities of both exchange factors and GAPs involved in RAS-GTP induction. Studies performed on fibroblasts derived from mouse embryos with a null mutation in the p120 RasGAP gene showed that p120 RasGAP has specific functions in down-regulating the RAS pathway, but it is not required for cell growth and mitogenic signaling (35). These findings could explain why SOCS1 overexpression in keratinocyte clones did not show an altered proliferation rate despite of massive RAS activation. Moreover, the evidence that SOCS1 prevented the IFN-␥-induced inhibition of keratinocyte growth is in line with previous studies showing the ability of SOCS1 to inhibit the activation of STAT1, a molecule required for mediating the antiproliferative effect of IFN-␥ (20). An intriguing result of our study showed that inhibition of ERK1/2 signaling by PD98059 in SOCS1 clones to some extent reverted their inability to produce CXCL10 and CCL2 following IFN-␥ stimulation. The explanation for these results can reside in the ability of ERK1/2 to enhance the decay of CCL2 and CXCL10 transcripts (19). However, CXCL10 and CCL2 content of PD98059-treated SOCS1 clones never did achieve that of mock-transfected cells, indicating that both STAT1 and ERK1/2 activation concur to the regulation of these chemokines in IFN-␥-treated keratinocytes. Another feature of SOCS1 clones is their capacity to produce high constitutive amounts of CXCL8, which are remarkably reduced on ERK1/2 inhibition with PD98059. This reduction can be correlated with ERK1/2 ability to induce CXCL8 mRNA synthesis and stabilization (19). In conclusion, the negative regulation of IFN-␥ signaling by SOCS1 in human keratinocytes is executed through the direct inhibition of Jak1/2 and, in parallel, through an indirect activation of ERK1/2 (Fig. 9). The first mechanism has been amply described and implies the lack of recruitment and phosphorylation of STAT1, which can no longer mediate expression of proinflammatory genes. The second mechanism, proposed in this study, involves the negative regulation of p120 RasGAP by SOCS1 together with a sustained activation of RAS/ ERK1/2 pathway. In turn, induction of ERK1/2 signaling drives prosurvival and antiinflammatory programs, which counteract the deleterious effects of IFN-␥ on keratinocytes. The unforeseen function of SOCS1 in sustaining ERK1/2 pathway confirms its important role in down-regulating inflammatory responses elicited by IFN-␥ in keratinocytes. Down-regulation of proinflammatory events triggered and/or perpetuated by keratinocytes represents 3296
Vol. 22
September 2008
Figure 9. Schematic representation of SOCS1-dependent mechanisms inhibiting IFN-␥ signaling in human keratinocytes. IFN-␥ is responsible for the expression of proinflammatory genes in keratinocytes, mainly through the activation of STAT1 transcription factor. However, other than proinflammatory genes, STAT1 activates gene expression of SOCS1, which in turn dampens IFN-␥ signaling by using two mechanisms. SOCS1 inhibits Jak1/2, which in turn can no longer phosphorylate IFN-␥ R and STAT1. As a direct consequence of the lack of STAT1 function, keratinocytes do not express a number of proinflammatory genes (mechanism 1). In parallel, IFN-␥-induced SOCS1 sustains RAS/ERK1/2 activation, in part through a negative regulation of p120 RasGAP. Induction of ERK1/2 counteracts the deleterious effects of IFN-␥ by repressing proinflammatory gene expression (mechanism 2).
an important strategy for the therapy of many immunemediated skin diseases. Our findings confirm epidermal SOCS1 as a promising target for the modulation of skin inflammation. Strengthening of SOCS1 action in keratinocytes could represent a valid therapeutic approach for the treatment of IFN-␥-dependent skin diseases. Indeed, SOCS1 manipulation by synthetic mimetic peptides has been performed previously to suppress immune responses in other cell systems. In particular, SOCS1 analogues were found to inhibit Jak1/STAT1 activation in IFN-␥-activated macrophages and to prevent inflammatory processes in mice affected by allergic encephalomyelitis (36, 37). Future studies using SOCS1 mimetic peptides in keratinocyte cultures and in experimental models of IFN-␥-dependent skin inflammation (i.e., contact hypersensitivity reactions) should better define the therapeutic efficacy of SOCS1. This work was supported by the Italian Ministry of Health. We thank Dr. A. Cavani and Prof. G. Girolomoni for the critical reading of this manuscript.
REFERENCES 1.
Bos, J. D., de Rie, M. A., Teunissen, M. B., and Piskin, G. (2006) Psoriasis: dysregulation of innate immunity. Br. J. Dermatol. 152, 1098 –1107
The FASEB Journal
MADONNA ET AL.
2. 3.
4.
5. 6. 7.
8. 9.
10.
11.
12.
13.
14.
15.
16.
17. 18.
19. 20.
Albanesi, C., Scarponi, C., Giustizieri, M. L., and Girolomoni, G. (2005) Keratinocytes in inflammatory skin diseases. Curr. Drug Targets Inflamm. Allergy 4, 329 –334 Albanesi, C., Cavani, A., and Girolomoni, G. (1999) IL-17 is produced by nickel-specific T lymphocytes and regulates ICAM-1 expression and chemokine production in human keratinocytes: synergistic or antagonistic effects with IFN-␥ and TNF-␣. J. Immunol. 162, 494 –502 Albanesi, C., Scarponi, C., Sebastiani, S., Cavani, A., Federici, M., De Pita`, O., Puddu, P., and Girolomoni, G. (2000) IL-4 enhances keratinocyte expression of CXCR3 agonistic chemokines. J. Immunol. 165, 1395–1402 Zhang, X., Chen, X., Song, H., Chen, H. Z., and Rovin, B. H. (2005) Activation of the Nrf2/antioxidant response pathway increases IL-8 expression. Eur. J. Immunol. 35, 3258 –3267 Gillitzer, R., and Goebeler, M. (2001) Chemokines in cutaneous wound healing. J. Leukoc. Biol. 69, 513–521 Li, A., Dubey, S., Varney, M. L., Dave, B. J., and Singh, R. K. (2003) IL-8 directly enhanced endothelial cell survival, proliferation, and matrix metalloproteinases production and regulated angiogenesis. J. Immunol. 170, 3369 –3376 Platanias, L. C. (2005) Mechanisms of type-I- and type-IIinterferon-mediated signalling. Nat. Rev. Immunol. 5, 375–385 Goh, K. C., Haque, S. J., and Williams, B. R. (1999) p38 MAP kinase is required for STAT1 serine phosphorylation and transcriptional activation induced by interferons. EMBO J. 18, 5601– 5608 Nguyen, V. A., Chen, J., Hong, F., Ishac, E. J., and Gao, B. (2000) Interferons activate the p42/44 mitogen-activated protein kinase and JAK-STAT (Janus kinase-signal transducer and activator transcription factor) signalling pathways in hepatocytes: differential regulation by acute ethanol via a protein kinase C-dependent mechanism. Biochem. J. 349, 427– 434 Takaoka, A., Tanaka, N., Mitani, Y., Miyazaki, T., Fujii, H., Sato, M., Kovarik, P., Decker, T., Schlessinger, J., and Taniguchi, T. (1999) Protein tyrosine kinase Pyk2 mediates the Jak-dependent activation of MAPK and Stat1 in IFN-gamma, but not IFN-alpha, signaling. EMBO J. 18, 2480 –2488 Nguyen, H., Ramana, C. V., Bayes, J., and Stark, G. R. (2001) Roles of phosphatidylinositol 3-kinase in interferon-␥-dependent phosphorylation of STAT1 on serine 727 and activation of gene expression J. Biol. Chem. 276, 33361–33368 Nair, J. S., DaFonseca, C. J., Tjernberg, A., Sun, W., Darnell, J. E. Jr., Chait, B. T., and Zhang, J. J. (2002) Requirement of Ca2⫹ and CaMKII for Stat1 Ser-727 phosphorylation in response to IFN-␥. Proc. Natl. Acad. Sci. U. S. A. 99, 5971–5976 Hu, J., Roy, S. K., Shapiro, P. S., Rodig, S. R., Reddy, S. P., Platanias, L. C., Schreiber, R. D., and Kalvakolanu, D. V. (2001) ERK1 and ERK2 activate CCAAAT/enhancer-binding proteinbeta-dependent gene transcription in response to interferongamma. J. Biol. Chem. 276, 287–297 Gough, D. J., Sabapathy, K., Ko, E. Y., Arthur, H. A., Schreiber, R. D., Trapani, J. A., Clarke, C. J., and Johnstone, R. W. (2007) A novel c-Jun-dependent signal transduction pathway necessary for the transcriptional activation of interferon-␥ response genes. J. Biol. Chem. 282, 938 –946 Ala-aho, R., Johansson, N., Gre´nman, R., Fusenig, N. E., Lo´pezOtı´n, C., and Ka¨ha¨ri, V. M. (2000) Inhibition of collagenase-3 (MMP-13) expression in transformed human keratinocytes by interferon-gamma is associated with activation of extracellular signal-regulated kinase-1,2 and STAT1. Oncogene 13, 248 –257 Ramana, C. V., Gil, M. P., Schreiber, R. D., and Stark, G. R. (2002) Stat1-dependent and -independent pathways in IFNgamma-dependent signaling. Trends Immunol. 23, 96 –101 Mascia, F., Mariani, V., Girolomoni, G., and Pastore, S. (2003) Blockade of the EGF receptor induces a deranged chemokine expression in keratinocytes leading to enhanced skin inflammation. Am. J. Pathol. 163, 303–312 Pastore, S., Mascia, F., Mariotti, F., Dattilo, C., Mariani, V., and Girolomoni, G. (2005) ERK1/2 regulates epidermal chemokine expression and skin inflammation. J. Immunol. 174, 5047–5056 Federici, M., Giustizieri, M. L., Scarponi, C., Girolomoni, G., and Albanesi, C. (2002) Impaired IFN-gamma-dependent inflammatory responses in human keratinocytes overexpressing the suppressor of cytokine signaling 1. J. Immunol. 169, 434 – 442
SOCS1 SUSTAINS ERK1/2 ACTIVATION IN KERATINOCYTES
21. 22. 23.
24.
25.
26.
27.
28.
29.
30.
31.
32. 33.
34.
35.
36.
37.
Knisz, J., and Rothman, P. B. (2007) Suppressor of cytokine signaling in allergic inflammation. J. Allergy Clin. Immunol. 119, 739 –745 Yoshimura, A., Naka, T., and Kubo, M. (2007) SOCS proteins, cytokine signalling and immune regulation. Nat. Rev. Immunol. 7, 454 – 465 Cacalano, N. A., Sanden, D., and Johnston, J. A. (2001) Tyrosine-phosphorylated SOCS-3 inhibits STAT activation but binds to p120 RasGAP and activates Ras. Nat. Cell Biol. 3, 460 – 465 Wang, D., Li, Z., Messing, E. M., and Wu, G. (2005) The SPRY domain-containing SOCS box protein 1 (SSB-1) interacts with MET and enhances the hepatocyte growth factor-induced ErkElk-1-serum response element pathway J. Biol. Chem. 280, 16393– 16401 Ben-Zvi, T., Yayon, A., Gertler, A., and Monsonego-Ornan, E. (2006) Suppressors of cytokine signaling (SOCS) 1 and SOCS3 interact with and modulate fibroblast growth factor receptor signalling 1. J. Cell Sci. 119, 380 –387 Han, Y., He, T., Huang, D., Pardo, C. A., and Ransohoff, R. M. (2001) TNF-␣ mediates SDF-1␣-induced NF-B activation and cytotoxic effects in primary astrocytes. J. Clin. Invest. 108, 425– 435 Sah, J. F., Eckert, R. L., Chandraratna, R. A. S., and Rorke, E. A. (2002) Retinoids suppress epidermal growth factor-associated cell proliferation by inhibiting epidermal growth factor receptor-dependent ERK1/2 activation. J. Biol. Chem. 277, 9728 –9735 Porcile, C., Bajetto, A., Barbieri, F., Barbero, S., Bonavia, R., Biglieri, M., Pirani, P., Florio, T., and Schettini, G. (2005) Stromal cell-derived factor-1a (SDF-1a/CXCL12) stimulates ovarian cancer cell growth through the EGF receptor transactivation. Exp. Cell Res. 308, 241–253 Hilton, D. J., Richardson, R. T., Alexander, W. S., Viney, E. M., Willson, T. A., Sprigg, N. S., Starr, R., Nicholson, S. E., Metcalf, D., and Nicola, N. A. (1998) Twenty proteins containing a C-terminal SOCS box form five structural classes. Proc. Natl. Acad. Sci. U. S. A. 95, 114 –119 He, Y., Zhang, W., Zhang, R., Zhang, H., and Min, W. (2006) SOCS1 inhibits tumor necrosis factor-induced activation of ASK1-JNK inflammatory signaling by mediating ASK1 degradation. J. Biol. Chem. 281, 5559 –5566 Mansell, A., Smith, R., Doyle, S. L., Gray, P., Fenner, J. E., Crack, P. J., Nicholson, S. E., Hilton, D. J., O’Neill, L. A., and Hertzog, P. J. (2006) Suppressor of cytokine signaling 1 negatively regulates Toll-like receptor signaling by mediating Mal degradation. Nat. Immunol. 7, 148 –155 De Sepulveda, P., Ilangumaran, S., and Rottapel, R. (2000) Suppressor of cytokine signaling-1 inhibits VAV function through protein degradation. J. Biol. Chem. 275, 14005–14008 Ungureanu, D., Saharinen, P., Junttila, I., Hilton, D. J., and Silvennoinen, O. (2002) Regulation of Jak2 through the ubiquitin-proteasome pathway involves phosphorylation of Jak2 on Y1007 and interaction with SOCS-1. Mol. Cell Biol. 22, 3316 –3326 Dajee, M., Tarutani, M., Deng, H., Cai, T., and Khavari, P. A. (2002) Epidermal Ras blockade demonstrates spatially localized Ras promotion of proliferation and inhibition of differentiation. Oncogene 21, 1527–1538 Van der Geer, P., Henkemeyer, M., Jacks, T., and Pawson, T. (1997) Aberrant Ras regulation and reduced p190 tyrosine phosphorylation in cells lacking p120-Gap. Mol. Cell Biol. 17, 1840 –1847 Waiboci, L. W., Ahmed, C. M., Mujtaba, M. G., Flowers, L. O., Martin, J. P., Haider, M. I., and Johnson, H. M. (2007) Both the suppressor of cytokine signaling 1 (SOCS-1) kinase inhibitory region and SOCS-1 mimetic bind to JAK2 autophosphorylation site: implications for the development of a SOCS-1 antagonist. J. Immunol. 178, 5058 –5068 Mujtaba, M. G., Flowers, L. O., Patel, C. B., Haider, M. I., and Johnson, H. M. (2005) Treatment of mice with the suppressor of cytokine signaling-1 mimetic peptide, tyrosine kinase inhibitor peptide, prevents development of the acute form of experimental allergic encephalomyelitis and induces stable remission in the chronic relapsing/remitting form. J. Immunol. 175, 5077– 5086 Received for publication March 3, 2008. Accepted for publication May 15, 2008.
3297
