Cytokines play an important role in the regulation of innate and adaptive immunity. In general, cytokines exert rapidly inducible but transient cellular responses. Whereas signal transduction by cytokines is nowadays known in molecular details, counteraction and termination by negative regulation are only partly understood. However, the integration of positive and negative signals is mandatory to ensure immunological homeostasis. Type I and II cytokine receptors (including receptors for interferons, interleukins, and hematopoietic growth factors) typically lack intrinsic kinase activity. Instead, cytokines that trigger those receptors activate receptor-associated cytoplasmic Janus kinases (JAKs) and signal transducer and activator of transcription (STAT) factors. The JAK/STAT signaling pathway regulates a variety of responses in immune cells, including development, activation, and differentiation. Negative regulation which terminates JAK/STAT signaling comprises internalization of cytokine receptors, secretion of decoy receptors that capture secreted cytokines, and expression of constitutively activated regulators including PIAS (protein inhibitors of activated STAT) proteins, protein tyrosine phosphatases (PTP) such as SHP-1 (SH2-containing phosphatase 1) and SHIP (SH2-containing inositol phosphatase), and protein modifications (ubiquitination and ISGylation) of signaling components. In 1995, CIS (cytokine-inducible SH2-containing protein) was discovered as the first member of the SOCS family during studies on the expression of immediate early cytokine-induced genes (Yoshimura et al. 1995). In 1997, suppressor of cytokine signaling (SOCS) proteins were subsequently identified independently by three groups as a protein family that regulates JAK/STAT signaling (Endo et al. 1997; Naka et al. 1997; Starr et al. 1997). Importantly, expression of SOCS proteins was induced by JAK/STAT signaling itself (explaining the synonyms STAT-induced STAT inhibitor, SSI and cytokine-inducible SH2-containing protein, CIS) and in turn SOCS proteins limited further signal transduction of this signaling pathway. Thus, SOCS proteins were identified and characterized as the first intracellular feedback inhibitors of type I and II cytokine receptors. They build up a negative regulatory pathway thereby limiting duration of JAK/STAT signaling. Since then, it has been discovered that the eight SOCS family members can be induced by a variety of cytokines that in turn are regulated in a (1) feedback inhibition mode. Moreover, it has turned out that besides this feedback loop, SOCS proteins can also be induced by other immune receptors and signals, including Toll-like receptors, TNF receptor, T-cell receptors, and cAMP, thereby regulating a cell’s sensitivity toward subsequent cytokine stimulation, referred to as (2) cross-talk inhibition.
The SOCS Family
Based upon structural similarities, the SOCS family consists of eight members: SOCS1–7 and CIS (Alexander 2002; Alexander and Hilton 2003; Yoshimura et al. 2007). Since SOCS proteins are negative feedback inhibitors that are induced through JAK/STAT signaling, they contain STAT binding regions within their promoters. Hence, SOCS expression is induced by activated STATs. However, SOCS proteins are pleiotropic; thus individual family members can be induced by and regulate different STAT signaling modules. SOCS genes have only few introns thus identifying them as rapidly inducible, immediate early genes. SOCS expression is regulated mostly at the transcriptional level: Promoters bear STAT factor binding elements but also additional transcription factors drive SOCS expression (e.g., NF-κB, egr-1, IRF1). Moreover, SOCS protein expression is also controlled by translational mechanisms. Thus, SOCS proteins feature a short half-life time (approximately 1–4 h) ensuring timely and rapid regulation of transient cytokine responses. SOCS1 and SOCS3 abundance is determined by protein stability: Pim-1-induced phosphorylation stabilizes SOCS1 which in turn increases half-life time of the latter. Protein stability of SOCS3 is regulated through a specific 35-residue PEST sequence (Pro, Glu, Ser, Thr) that enhances its degradation. Additionally, the SOCS box is discussed to influence the stability of SOCS proteins. Structures for CIS and SOCS2 in conjunction with Elongin B/C and SOCS3 together with a tyrosine-phosphorylated peptide from the IL-6 receptor have been solved.
Structure and Mode of Action
The SOCS box has been shown to operate as E3 ubiquitin ligase by interacting with Elongin B and C, cullin-5, and RING box 2 (Kile et al. 2002). In concert with two other enzymes of the ubiquitin system, E1 (ubiquitin activating) and E2 (ubiquitin conjugating), the E3 ubiquitin ligase induces ubiquitination of a specifically bound protein thus leading to the proteasomal degradation of the latter. SOCS proteins thus operate by inducing proteasomal decay of substrates which they bind by an additional N-terminal protein interaction domain, the SH2 domain. Inhibition of proteasomal degradation dampens the inhibitory effects of SOCS on cytokine signaling.
To access their substrates, SOCS1–7 and CIS bear a central SH2 domain which mediates binding to phosphorylated tyrosine residues within target proteins. SOCS1 binds to activated, phosphorylated Y1007 in JAK2 thereby inhibiting JAK2 signaling. It also interferes with JAK1-3 and Tyk2 activation. SOCS1 also binds to the type I and II IFN receptor chains IFNAR1 and IFNGR1. SOCS3 is also capable of binding to JAK2-pY1007 but with lower affinity as compared to SOCS1. Moreover, SOCS3, SOCS2, and CIS bind to phosphorylated residues exposed at cytoplasmic sites of various cytokine receptors. SOCS3 has a specific affinity to bind to gp130, thus inhibiting IL-6 and other gp130-dependent cytokines. In turn, STAT3 activity is controlled. Importantly, IL-10-induced STAT3 activity is not limited by SOCS3. SOCS3 inhibits JAK activity, whereas CIS and SOCS2 do not interfere with JAKs but rather compete with STAT recruitment to the receptor. A consensus sequence to which the SH2 domain of SOCS3 binds has been identified as (S/A/V/Y/F)-hydrophobic-(V/I/L)-hydrophobic-(H/V/I/Y).
Interestingly, the SH2 domain is also involved in binding proteins distinct from JAK/STAT, e.g., NF-κB p65, c-kit, insulin receptor, and tumor suppressor p53. Moreover, SOCS3 interferes also with MAP kinase activation by blocking the SHP2 binding site in gp130.
Additional small extended SH2 subdomains (ESS) N-terminal (SOCS1 and 3) of the SH2 domain essentially contribute to the binding of phosphorylated tyrosine residues. In case of SOCS1 and SOCS3, the SH2 domain mediates binding to substrates; however, inhibition of JAKs is achieved by a kinase inhibitory region (KIR) which is N-terminally attached to the SH2 domain and can only be found in SOCS1 and SOCS3. The KIR domain blocks the catalytic cleft of JAKs thereby contributing to termination of JAK/STAT signaling.
The further N-terminus is quite diverse among the SOCS family members and varies in length and sequence. In contrast to the specific function of SOCS proteins located at the cytoplasmic site of the cytokine receptor, SOCS1 revealed an unexpected subcellular localization pattern with mainly nuclear occurrence. This was attributed to a specific sequence composed of two clusters of basic amino acid residues which function as classical bipartite nuclear localization signal (NLS) (Baetz et al. 2008). Furthermore, SOCS3 and 6 are found ubiquitously expressed throughout the whole cell by immunofluorescence, and SOCS7 also has the capacity to translocate to the nucleus where it contributes in regulating the DNA damage response. The N-terminus of SOCS7 was shown to bear a basic region with homology to a classical monopartite NLS. The findings indicate that SOCS proteins might exert additional functions beyond inhibiting JAK/STAT signaling at the receptor level.
Physiological Functions of SOCS Proteins
Functions of SOCS proteins as determined from genetic models
Neonatal lethality, multiorgan inflammation, fatty liver degeneration, lymphocyte apoptosis, aberrant T-cell activation, increased IFNγ responsiveness, aberrant DC and B-cell activation, autoimmunity
Spontaneous T-cell activation, increase in CD4 + T cells, decrease in γδ T cells, disturbed T-cell development
Gigantism, dysregulated growth hormone, and IGF-I signaling
Gigantism, dysregulated growth hormone signaling
Embryonic lethality with placental defects and aberrant erythropoiesis, increased hepatocarcinogenesis, prolonged and aberrant IL-6/STAT3 signaling in macrophages
Embryonic lethality with anemia, increased T H2 differentiation
Impaired T H2 development, disturbed IL-4 signaling
Mild growth retardation
Growth retardation, increased lethality postnatal due to hydrocephalus
No obvious phenotype
IL-2/STAT5 signaling defects, lactation failure, reduced numbers of NK, NKT cells, reduced body weight
SOCS1−/− mice succumb to perinatal death due to multiorgan inflammation. These mice reveal enhanced peripheral T cell (TC) and natural killer (NK) cell activation, macrophage organ infiltration, and fatty liver degeneration. Furthermore, thymus size is decreased and B lymphocytes undergo increased apoptosis. A major underlying dysregulation is undampened IFNγ signaling. Backcrossing to Rag2 −/− or IFNγ −/− mice rescues SOCS1 −/− mice from early death, and the disease can be reinduced in Rag2 −/− mice by transfer of hematopoietic SOCS1 −/− cells confirming the importance of hyperactivated lymphocytes. Natural killer cell depletion reduces hepatotoxicity in SOCS1 −/− mice. SOCS1/IFNγ double knockout mice show a delayed phenotype with polycystic kidneys, pneumonia, chronic skin ulcers, and granulomas in various organs. Moreover, the lethal phenotype can also be ameliorated by crossing SOCS1 −/− mice with STAT4 −/− or STAT6 −/− mice, emphasizing that SOCS1 action is not limited to IFNγ signaling. Additionally, SOCS1 −/− macrophages show higher proinflammatory protein production in response to Toll-like receptor (TLR) stimulation by lipopolysaccharide (LPS) indicating a role for SOCS1 in termination of TLR signal transduction. In dendritic cells, SOCS1 was required to suppress induction of autoimmune responses and to restrict breakdown of tolerance. Mice with isolated knockout of the SOCS box in SOCS1 have been generated and also show an amelioration of the acute lethal phenotype with a delayed onset of the multi-inflammatory disease of SOCS1 knockouts.
SOCS2 −/− mice exhibit augmented growth which is supposed to occur due to altered growth hormone and insulin-like growth factor I signaling. Interestingly, SOCS2 transgenic mice also show increased growth. SOCS2 is also necessary to mediate the anti-inflammatory effects of lipoxin A4 in dendritic cells.
SOCS3 −/− mice are embryonic lethal, presumably due to placental defects and increased LIF receptor sensitivity. They also show defects in erythropoiesis, but SOCS3 −/− fetal liver stem cells can reconstitute erythropoiesis when transferred into lethally irradiated adults. SOCS3 deficiency in macrophages shows altered IL-6 signaling: On the one hand, IL-6 signal transduction is not limited anymore resulting in prolonged STAT3 activity with IL-10-like effects. On the other hand, in SOCS3-deficient macrophages, IL-6 adopts IFN-like activities due to unchecked STAT1 activity. SOCS3 is also crucially involved in the regulation of leptin and G-CSF signaling.
SOCS5 transgenic mice show diminished TH2 differentiation due to constrained IL-4 signaling. SOCS6 −/− and SOCS7 −/− mice are smaller than wild-type littermates, the latter showing increased lethality due to hydrocephalus development. Both SOCS6 and SOCS7 regulate insulin receptor signaling. However, SOCS4–7 mice are less studied so far. CIS −/− mice do not show a distinct phenotype, but CIS transgenic mice have defects in lactation, a disturbed IL-2/STAT5 signaling, and reduced numbers of γδ T cells, natural killer cells, and natural killer T cells.
SOCS Proteins in Disease
SOCS proteins also act as tumor suppressors (Rottapel et al. 2002): SOCS1 and SOCS3 are frequently downregulated in human cancer. SOCS1 is involved in the early stage of tumor development by limiting inflammation and tissue damage, whereas SOCS3 terminates STAT3-mediated proliferation and tissue remodeling. SOCS1 expression is lost in many tumors. Especially in liver fibrosis and hepatocellular carcinomas, the SOCS1 promoter is hypermethylated resulting in decreased SOCS1 expression, enhanced inflammation, and tumorigenesis. SOCS1 is also downregulated by hepatitis C virus core protein. SOCS1 has also been discovered as essential activator of the p53 tumor suppressor thus impeding with tumor development. Besides, SOCS1 has been identified to degrade HPV (human papillomavirus) E7 oncogene thus blocking the proliferation of infected cells. The SOCS3 gene is hypermethylated in lung cancer with restoration of SOCS3 expression resulting in STAT3 inhibition and concomitant growth inhibition. SOCS2 and SOCS6 have been shown to be reduced in breast cancer.
In addition, SOCS proteins play a role in chronic inflammatory diseases. SOCS1 is involved in the pathogenesis of Crohn’s disease, where it promotes apoptosis of intestinal epithelial cells via p53 signaling and in suppression of experimental colitis. Likewise, SOCS3 is elevated in human ulcerative colitis and regulates STAT3 activity in an animal model of inflammatory bowel disease. Differential regulation of SOCS proteins in inflammatory diseases is a common phenomenon. For example, SOCS2 and CIS are downregulated in osteoarthritis. MicroRNA-155 is suggested to modulate the inflammatory phenotype in both ulcerative colitis and atherosclerosis via targeting SOCS1. Patients with allergic contact dermatitis and psoriasis have dysregulated SOCS1–3 expression, and SOCS3 can also be found elevated in rheumatoid arthritis. Physiological expression of both SOCS3 and SOCS5 is mandatory for prevention of allergic conjunctivitis. Recent evidence suggests that SOCS1 plays an important role in asthma pathogenesis. It has been shown that SOCS1 gene expression is significantly lower in the airways of severe asthmatics as compared to mild or moderate asthmatics and inversely associated with airway eosinophilia. In line with this finding, serum IgE levels and infiltrating eosinophils are considerably increased in SOCS1-/- IFNγ-/- mice upon OVA immunization. Moreover, SOCS proteins have been suggested to be important for the pathophysiology of diabetes. SOCS2 was identified as a candidate disease gene for type-2 diabetes mellitus, and SOCS3 has been shown to inhibit insulin signaling. Also hematopoietic malignancies and loss of therapeutic IFN responsiveness might be associated with SOCS expression. The question remains to be solved whether disturbed SOCS expression is a consequence of altered cytokine networks or, at least in some conditions, might cause and contribute directly to pathogenesis of those diseases. Nevertheless, those observations implicate that SOCS manipulation is a promising approach to alter cytokine signaling.
Further Functions of SOCS Within the Immune System
To address the role of SOCS within the immune system, regulation of both, innate and adaptive immunity, has been studied experimentally (Dalpke et al. 2008; Yoshimura 2009). It has been observed that SOCS2 is upregulated upon TLR or NOD activation and regulates IL-1ß and IL-10. In addition, SOCS1 has been shown to affect TLR responses, with increased secretion of proinflammatory cytokines once SOCS1 is missing. SOCS1 and 3 are also strongly induced by TLR triggering thus limiting the action of subsequent cytokine receptor triggering. Thus, SOCS1 limited GM-CSF (granulocyte–monocyte colony-stimulating factor) and IL-4-mediated differentiation of CD14 + monocytes to dendritic cells. SOCS1 also is important for STAT6 to STAT1 switching during dendritic cell maturation. Further experiments show that SOCS1 encounters activated NF-κB within the nucleus and inhibits its action as transcription factor via ubiquitination and subsequent degradation of the NF-κB subunit p65. Also, TLR-induced SOCS1 inhibits paracrine signal amplification by type I IFN, and this circuit is mainly responsible for increased LPS sensitivity of SOCS1 −/− mice because SOCS1 −/−/Tyk2 −/− mice which lack type I IFN signaling had no increased LPS sensitivity anymore. Furthermore, single reports claim interaction of SOCS1 with the TLR2/4 adaptor protein Mal through interaction with Bruton’s tyrosine kinase and with IRAKs as well as inhibition of TLR-induced JAK activity. In turn, SOCS1 −/− mice exhibit lack of endotoxin tolerance. SOCS1 is therefore involved in regulating innate immune responses; however, the distinct outcome of its action seems to be dependent on the respective gene analyzed. SOCS1 deficiency also leads to hyperactivated dendritic cells that supported autoimmune responses.
Concerning adaptive immunity, SOCS proteins play an important role in the differentiation of T-cell subsets. On the one hand, SOCS1 is upregulated during T-cell development in the thymus, and SOCS1 −/− mice show a defect in the development of single-positive thymocytes with a bias toward CD8 + T cells. On the other hand, SOCS proteins are induced during cytokine-mediated T helper (TH) development which in turn is regulated by SOCS proteins.
TH1 cells preferentially develop in an IL-12 environment: They have high levels of SOCS1 and SOCS5 which suppresses IL-4-induced STAT6 activation directing T cells into TH2 development. In contrast, TH2 cells mainly develop in a “non-IL-12” environment and express GATA-3 as transcription factor due to IL-4-mediated STAT6 activation. TH2 cells possess high levels of SOCS3 suppressing IL-12 signaling and thus inhibiting T-bet induction and TH1 differentiation. Indeed, patients with TH2-shifted diseases like asthma and atopic dermatitis show a correlation between SOCS3 levels and T-cell-mediated disease activity. TH17 cells for which IL-6, IL-21, and IL-23 with subsequent activation of STAT3 are important are also affected by SOCS3. Since SOCS3 blocks STAT3 action, it has been identified as negative regulator of TH17 differentiation. SOCS1 and 3 have also been implicated to play a role in development of regulatory T cells (Treg). Treg cells are CD4 + cells defined by high expression of CD25 and the transcription factor FOXP3. Expression of FOXP3 induces microRNA (miRNA) 155 which targets SOCS1. This miRNA-mediated SOCS1 downregulation enhances IL-2/STAT5 signaling which in turn increases the number of Treg cells. Furthermore, miRNA-155 has also been shown to target SOCS1 which is used by the host’s immune system upon viral infection to positively regulate type I IFN signaling. Indeed, SOCS1 suppresses rhinovirus induction of interferons (Gielen et al. 2015). It has been further shown that SOCS1 expression impairs viral clearance and exacerbates lung injury during influenza infection.
The role of SOCS proteins as negative regulators of the immune system is confirmed by pathogenic microbes that misuse cell-intrinsic negative regulation circuits to evade immune recognition. SOCS3 gets induced by both viruses and bacteria, and elevated SOCS3 expression has been linked to pathogenic immune evasion (Mahony et al. 2016). Murid herpesvirus-4 (MuHV-4) encodes for a SOCS box motif-containing protein (ORF73) that targets NF-κB for ubiquitination and proteasomal degradation thereby inhibiting NF-κB activity. Other viruses have evolved SOCS-dependent mechanisms to escape host immune system: HIV-1 Tat protein is a key regulator of viral replication; however, it additionally induces SOCS2 which in turn interferes with JAK/STAT signaling thus supporting virus replication. Likewise, RSV surface proteins G and F induce SOCS1 expression. For HSV-1 direct effects on the SOCS1 promoter, followed by increased SOCS1, have been shown. Not only viruses but also parasites take advantage of the SOCS negative regulatory system: Toxoplasma gondii replicates in macrophages and induces SOCS1, 3, and CIS which deactivate JAK/STAT signaling. Leishmania major infection was associated with lower pathogenicity in SOCS1-deficient cells, though at the cost of detrimental effects due to uncontrolled cytokine signaling. Also, intracellular mycobacteria are reported to interfere with macrophage activation by IFNγ through SOCS1 and 3 induction. Furthermore, SOCS3 induction by hepatitis C virus might be a mean to circumvent antiviral activities of type I interferons.
Besides JAK/STAT regulation, SOCS proteins show further effects in a variety of additional signaling pathways: Thus SOCS1 has been reported to regulate the insulin receptor, TNF receptor, and Toll-like receptors; SOCS3 in vitro also inhibits insulin receptor, IL-1R, and CXCR4 signaling to name a few. Although the molecular details are not known, it seems probable that the function of SOCS proteins is not restricted to the inhibition of classical type I and II cytokine receptors.
SOCS proteins have been shown to be involved in a variety of cellular processes and could therefore present a potential target for the treatment of inflammatory diseases. To confirm the function of specific domains of SOCS and to imitate SOCS expression, SOCS mimetics have been produced. SOCS mimetics comprise peptides corresponding to functional domains coupled to protein translocation domains facilitating cell penetration. A peptide that resembles the KIR region has been described to mimic the inhibitory effects of SOCS on JAK activity. SOCS1-KIR mimetics are potent therapeutics in experimental allergic encephalomyelitis mouse model of multiple sclerosis and show promising results in a psoriasis model and a model of diabetes-associated cardiovascular diseases. Topical administration of this SOCS1-KIR mimetic has been shown to inhibit ocular inflammation in murine uveitis. Treatment of mice with SOCS mimetics prevented vaccinia virus transcription and replication. Regarding the role of SOCS1 in regulating NF-κB signaling, there is evidence that SOCS1 administration limits the TNF production in synovial fluid macrophages isolated from patients suffering from rheumatoid arthritis. Similarly, delivery of recombinant whole proteins has been proposed. Other techniques to control SOCS expression include siRNA or miRNA as well as CRISPR/Cas approaches which will lead to a downregulation of the respective protein.
SOCS proteins comprise a family of eight proteins (CIS, SOCS1–7) that have first been identified as intracellular negative regulators of JAK/STAT signaling upon type I and II cytokine receptor stimulation. Therefore, SOCS proteins contain an extended SH2 domain which allows them to bind to tyrosine-phosphorylated cytokine receptors or Janus kinases. Through a kinase inhibitory region (SOCS1 and 3) as well as a SOCS box domain which targets substrates for proteasomal degradation, SOCS proteins limit signal transduction. However, the activity of SOCS exceeds the control of JAK/STAT signal transduction: SOCS proteins are also involved in the termination of NF-κB responses, negatively influence insulin signaling, and function as tumor suppressors. Depending on the cytokine milieu, SOCS proteins shape the immune system by influencing macrophages, dendritic cells, and lymphocyte differentiation. Evidence for crucial roles of individual SOCS proteins within the immune system comes from mice with targeted disruption of SOCS genes that show major deregulated cytokine pathways. Since SOCS proteins are essential modulators of immunity, precise understanding of their defined mode of action could be beneficial to develop new therapeutic strategies to combat inflammatory and autoimmune diseases as well as cancer.
- Dalpke AH. Suppressor of cytokine signaling (SOCS) proteins in innate immunity. In: Pandalai SG, editor. Recent res develop. Immunology. 6 Research Signpost: Trivandrum; 2004. p. 317–39.Google Scholar
- Mahony, R., Ahmed, S., Diskin, C. & Stevenson, N. J. SOCS3 revisited: a broad regulator of disease, now ready for therapeutic use? Cell Mol Life Sci. 2016;73(17):3323–36. doi: 10.1007/s00018-016-2234-x.