Role of SOCS1 in tumor progression and therapeutic application

Authors


Abstract

SOCS1, a prototype molecule of the SOCS family, was initially defined as a suppressor of cytokine signaling. The molecular mechanisms of SOCS1-mediated functions have been subsequently identified by studies using gene knockout mice and gene silencing technology. As part of a negative feedback regulation, SOCS1 downregulates cytokine signaling through direct inhibition of the JAK tyrosine kinase and the signaling cascade of activated cytokine receptors, thereby attenuating cytokine-initiated signal transduction. Moreover, other studies have demonstrated that SOCS1 also downregulates TLR signaling through direct and indirect mechanisms. Both cytokine receptor and TLR signaling pathways mediate important functions in survival, maturation and differentiation of various types of cells and in the regulation of immune function. Abnormal expression of SOCS1 in tumor cells has been detected in various human cancers, where it is associated with dysregulation of cytokine receptor and TLR signaling to promote cell transformation. Recent studies on the function of SOCS1 in tumor cells have revealed its novel role in carcinogenesis. In this review, we will focus on the mechanism of action of SOCS1 in both tumor cells and antigen-presenting cells in the tumor microenvironment. The potential of using SOCS1 as a diagnostic marker and therapeutic target in tumor diagnosis, prognosis and treatment is discussed.

Signaling mediated by toll-like receptor (TLR) and cytokine receptors are involved in innate and adaptive immunity. Aberrant or sustained activation of immune signaling has been associated with severe disorders such as septic shock, autoimmunity and cancer. An important mechanism to prevent overactivation of the immune system in a normal organism is the negative feedback regulation, which provides efficient spontaneous control for in vivo proliferative signals. Deficiency of a negative feedback regulation results in enhanced and prolonged activation of the proliferative signaling.1 Among the several critical signaling attenuators that have been identified in cells, suppressor of cytokine signaling 1 (SOCS1), a prototype molecule of the SOCS family, has been extensively investigated. SOCS1 was initially recognized as a negative feedback regulator of cytokine signaling, and its additional functions and mechanisms of action have been recently discovered. SOCS1 exhibits different effects on cells of different lineages. In antigen-presenting cells (APCs), SOCS1 is considered as a classic antigen presentation attenuator.2, 3 APC-secreted cytokines, whose signal transduction is negatively regulated by SOCS1, influence not only APCs but also the differentiation and function of T cells in microenvironment.4, 5 Meanwhile, SOCS1 also directly influences the differentiation, maturation and function of T cells through mechanisms that have been reviewed elsewhere.4, 5 Because of its critical role in immunomodulation, SOCS1 has been studied in the context of several immune-related diseases, including HIV infection,6 systemic lupus erythematosus,7 rheumatoid arthritis,7, 8 multiple sclerosis,9 immunological rejection10 and cancer, with the goal to identify its function in mediating these diseases and its potential value in novel therapies.

The excessive proliferative signaling resulted from deficiency of signaling attenuator can induce cell transformation. SOCS1, as a regulator in the negative feedback mechanism, can influence the transduction of proliferative signals and effect in the survival, differentiation and transformation of cells.5, 11 Indeed, alterations of SOCS1 expression in human cancers have been extensively reported and are associated with prolonged activation of JAK-STAT signaling pathway and insensitivity to IFNs.12–14 Therefore, the alterations of SOCS1 in tumor cells may be indicative in cancer diagnosis and prediction of clinical outcome in cancer patients.15–17 Based on the altered expression of SOCS1 in tumor cells and the critical role of SOCS1 in downregulating various signaling pathways, several studies have suggested that SOCS1 can be a potential target for antitumor therapy.13, 14, 18 However, a comprehensive understanding of the functions of SOCS1 in tumor cells was not available until recently. Here, we review the latest findings on the functions of SOCS1 in APCs and tumor cells, and discuss the potential value of SOCS1 in the diagnosis and prognosis of cancer patients and in antitumor therapy.

The SOCS Family

The SOCS family consists of eight members, namely SOCS1-7 and CIS.5, 19 Members of the SOCS family share similar structure, including a conserved Src homology (SH2) domain, a C-terminal SOCS box and an N-terminal region of variable length. The SH2 domain mainly mediates binding of SOCS to distinct phosphorylated tyrosine motifs on targeted proteins.5, 20 The conserved SOCS box in the C-terminus recruits Elongin B/C, Cullin 2 and Ring-box2 to form an ubiquitin E3 ligase complex, which promotes the degradation of SH2-binding proteins by proteasome.5, 21, 22 In addition, SOCS box also directly binds to other molecules, such as ATM and ATR, to confer additional regulatory functions. Recently, researchers found that a nuclear localization sequence (NLS) located between the SH2 domain and SOCS box mediates the translocation of SOCS1 from cytoplasm to the nucleus.23–25 Consistently, mutation of critical amino acid (aa) residues in the NLS results in predominant cytoplasmic localization of SOCS1.23–25 The N-terminal regions are variable in length and aa sequence. Among all members in the SOCS family, SOCS1 and SOCS3 contain a 12-aa region in the N-terminus, namely kinase inhibitory region (KIR), which includes a conserved tyrosine residue as a pseudosubstrate for JAK2 with the function to inhibit its kinase activity. The conserved tyrosine residue in KIR may also support the interaction between SOCS1/3 and target proteins, as its mutation abolishes the ability of SOCS1/3 to inhibit the JAK2 kinase without affecting binding of SOCS1/3 to JAK226, 27 (Fig. 1).

Figure 1.

The structure and functional domains of SOCS family. All eight members of SOSC family contain a central SH2 domain and C-terminal SOCS box. Both SOCS1 and SOCS3 harbor a kinase inhibitory region (KIR). The nuclear localization sequence (NLS) of SOCS1 is also shown. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

In this review, we focus on SOCS1, a gene initially cloned in 1997 through various approaches by three different groups.28–30 The SOCS1 gene is highly conserved among various species, including human, chimpanzee, dog, cow, rat and chicken. Human SOCS1, which contains a STAT-binding site within the promoter region, is located in the 16p13.31 This gene contains two exons and is transcribed into a 1215-nt mRNA encoding a 211-aa protein.

Mechanisms of SOCS1-Mediated Regulation

The cytoplasmic function of SOCS1

SCOS1 protein confers the function as signaling attenuator by localizing in the vicinity of the cell surface receptors they regulate. Since its discovery, the majority of SOCS1 studies focus on its cytoplasmic function in the regulation of signal transduction, as summarized below.

Inhibition of cytokine signaling by SOCS1.

The SOCS1 deficient (SOCS1−/−) mice die as neonates within 3 weeks after birth with systemic inflammatory responses that are similar to those observed in the wide-type (WT) mice administered with IFNs.32, 33 However, the SOCS1 and IFN-γ double knockout mice, or SOCS1−/− mice administered with IFN-γ neutralizing antibody after birth do not exhibit the phenotypes of SOCS1−/− mice.32, 33 These observations suggested the importance of SOCS1 in regulating IFN-γ signaling. Moreover, it has been shown that the promoter of SOCS1 gene has a STAT-binding site. As such, SOCS1 can be induced by several cytokines through the JAK-STAT signaling pathway, such as IFN-γ,12, 28, 32–34 IL-6,28, 29, 35 IL-2,36 IL-437 and IL-7.38 Induced expression of SOCS1, in turn, downregulates JAK-STAT signaling primarily through the SH2 domain-mediated binding to JAK and inhibition of STAT activation by JAK via the function of KIR as a pseudosubstrate that competes STAT binding to the JAK catalytic domain. This results in the termination or attenuation of JAK-mediated signaling, which is crucial in the proliferation, differentiation and survival of cells. Moreover, SOCS1 can mediate the ubiquitination and proteasome-dependent degradation of binding proteins via the SOCS box. Additionally, increasing evidence suggest that SOCS1 can also directly bind to the cytokine receptor to trigger its degradation or mask the docking sites for adaptor molecules.39

SOCS1 downregulates TLR signaling.

In addition to the JAK-STAT signaling pathway, SOCS1 can also be induced by TLR signaling, either directly through activation of early growth responses-1 or indirectly through cytokines induced by initial TLR activation, such as IL-6 and INF-β.40 The induced SOCS1 then initiates a negative feedback regulation by antagonizing TLR signaling, as described in the following studies. It has been observed that WT mice survive longer than SOCS1+/− mice under the challenge of LPS. However, in SOCS1 and IFN-γdouble knockout mice, an increased sensitivity to LPS is observed, compared to the IFN-γ−/− mice, suggesting that SOCS1 can define the LPS sensitivity independent of IFN-γ.21, 41SOCS1−/− mice and the macrophages isolated from these mice are hyper-responsive to LPS and CpG ODN as a result of high levels of nitric oxide and proinflammatory cytokine production. Moreover, they are also sensitized to secondary TLR stimuli, whereas the wild-type cells are tolerant to the secondary stimulation as a result of the negative feedback regulation. These indicate that SOCS1 has essential role in the regulation of TLR signaling.21, 41 Although it is generally accepted that SOCS1 downregulates TLR signaling, the mechanisms underlying this regulation remain controversial. As we have summarized above, the potential mechanisms of SOCS1 action can be divided into two categories, i.e., directly through inhibiting TLR signaling and indirectly through downregulating JAK-STAT cascade activated by the initial TLR activation. Several studies have found that SOCS1 can target and inhibit molecules in TLR signaling pathway, including IRAK41 and Mal42 (Fig. 2). However, whether SOCS1 can influence TLR signaling initiated by other stimuli in addition to LPS, CpG ODN and poly(I:C), and whether similar mechanisms of SOCS1-mediated regulation of TLR signaling occur in different cell lineages are still unclear and need further investigations.21, 41, 43

Figure 2.

Mechanisms underlying the effects of SOCS1 on TLR and cytokine signaling. The induced SOCS1 could feedback regulate JAK-STAT signaling by directly inhibiting cytokine receptor and attenuating the signaling transduction through targeting JAK for degradation and avoiding the interaction between STAT and JAK. Simultaneously, SOCS1 can inhibit TLR signaling by targeting the molecules in TLR pathway, such as Mal, IRAK, JAK2 and p65, for degradation, as well as by attenuating cytokine signaling activated by initial TLR stimulation. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

The Nuclear function of SOCS1

Although the studies of SOCS1 initially focused on its role in the vicinity of receptors on the cell surface membrane, recent evidences suggest the localization and function of SOCS1 in the nucleus. One of its nuclear functions is the regulation of NF-κB. SOCS1 not only regulates NF-κb signaling through the mechanisms described above via its cytoplasmic function but also limit the duration of NF-κB signaling by decreasing the stability of p65 in nucleus.44, 45 In resting cells, NF-κB is inhibited by IκB. Upon stimulation, IκB becomes phosphorylated and degraded, liberating NF-κB, which then translocates into the nucleus and initiates the transcription of target genes. Previous studies indicate that the NF-κB signaling is terminated by IκB, which enters into the nucleus and exports NF-κB. However, it was recently reported that termination of NF-κB signaling also occurs in the absence of IκB, through a variety of other inhibitory circuits that have been discovered, including SOCS1. Mutation studies of SOCS1 indicate that the SH2 domain and SOCS box of SOCS1, respectively, mediate its binding to p65 and the ubiquitination of the latter. In addition, it is shown that SOCS1 can form a ternary complex with ATM and ATR in the nucleus, and contribute to p53 phosphorylation on serine 15 and its activation, thereby to promote the p53-dependent process of oncogene-induced cell senescence.46–48 This mechanism possibly explains the spontaneous occurrence of colorectal cancer in SOCS1−/− mice.49 Although the aa residuals responsible for nuclear translocation of SOCS1 have been determined, the mechanisms controlling the intracellular translocation of SOCS1 from cytoplasm to the nucleus remain unclear and are crucial for the understanding of the tumor suppressive activity of SOCS1.

SOCS1-mediated regulation in APCs: Effects on antigen-presenting capacity in the tumor microenvironment

Mice with SOCS1 deficiency only in T cells and NKT cells do not develop the severe diseases as in the SOCS1−/− mice. Moreover, transplantation of bone marrow (BM) cells from SOCS1−/− donors into irradiated, WT recipient mice results in the development of symptoms that resemble the graft-versus-host disease, instead of the typical phenotypes associated with systemic SOCS1 deficiency.50 These results indicate that T cells in SOCS1−/− mice are necessary but not sufficient to cause symptoms and neonatal death, and that other cell types, in particular APCs, are likely required for the SOCS1−/− disease.51 As the most potent APCs in vivo, DCs have been considered as intriguing targets in antitumor immunity.2, 3 DCs from SOCS1−/−mice exhibited a more mature phenotype and were hyper-responsive to LPS.41 Consistently, immunization with SOCS1−/− DCs induces a hyper Th1 immune responses and antitumor activities.52 Macrophages, like DCs, play an essential role in activating innate immune response and adaptive immune response. In addition, macrophages are also the effector cells in antitumor immunity.53 In mice with SOCS1−/− background, Hashimoto et al. demonstrated that SOCS1−/− monocytic cells, including macrophages, cause suppression of tumor development.53 Simultaneously, the tumor-killing activity of macrophages can be enhanced by SOCS1 deficiency. Based on the SOCS−/− mice studies, SOCS1 is, therefore, considered as a constitutive antigen presentation attenuator in APCs.

SOCS1 in tumor cells: The controversy

The function of SOCS1 in tumor cells is controversial. Compared to primary melanoma specimens, the expression of SOCS1 was lower in the brain metastases of melanoma.54 This difference is also found between a brain-homing metastastic melanoma cell line and its parent cell line. Using melanoma cells that have metastasized to the brain and the parent melanoma cells, it has been shown that the brain-homing melanoma cells express higher levels of MMP-2, bFGF and VEGF both in vitro and in vivo as a result of reduced SOCS1 expression, and that this is associated with tumor invasion and angiogenesis to promote metastasis to the brain.54 Furthermore, SOCS1 is found to inhibit the growth of prostate cancer cells by downregulating the expression of cyclins and cyclin-dependent kinases.55 Decreased expression of SOCS1 is also observed in other tumors, including prostate cancer,54 hepatocellular carcinoma,56 laryngeal carcinoma,57 multiple myeloma (MM),18 pancreatic cancer58 and acute myeloid leukemia (AML).59 Consistent to these observations, increased expression of SOCS1 in cells leads to resistance to oncogene-induced transformation through inhibiting proliferation and inducing cell death.11, 60 All these reported SOCS1 functions indicate its role as a tumor suppressor in carcinogenesis. On the other hand, it has also been reported that the expression of SOCS1 in breast cancer tissue is higher than that in corresponding normal tissue.61 Similarly, melanoma cancer cells express higher levels of SOCS1 than their normal courterparts.62, 63 The high tumor-specific expression of SOCS1 observed in these studies suggests a tumor-promoting function of SOCS1, instead of its long-known function as a tumor suppressor (Tables 1 and 2).

Table 1. Summary of clinical studies that have examined the expression of SOCS1 in tumor tissue and corresponding normal tissue
inline image
Table 2. Summary of published studies related to the effects of SOCS1on tumor cells lines
inline image

The reasons of these discrepancies on SOCS1 expression in different cancers are still unknown. The higher levels of SOCS1 in breast tumor tissues that are associated with the inflammatory stroma but not in breast cancer cell lines may be a result of the induction of SOCS1 expression by inflammatory cytokines, such as GH and PRL through autocrine/paracrine in the tumor microenvironment.61 Similar to breast cancer, expression of SOCS1 is also decreased after androgen ablation and is elevated in recurrent patients with prostate cancer.55 It is, therefore, possible that factors in the tumor microenvironment, such as hormonal regulators, can affect the expression of SOCS1 in tumor cells.61, 64, 65

The mechanisms involved in silenced SOCS1 expression include loss of heterozygosity (LOH) and DNA hypermethylation. Compared to LOH, hypermethylation has been more extensively explored in tumor samples. It is believed that LOH is not the major mechanism underlying gene silencing in carcinoma, although it may occur in combination with hypermethylation.66, 67 Initial reports in the field have investigated the CpG islands within the exon 2 of SOCS1 gene. More recent reports, however, focus on the CpG islands in the 5′ UTR of the gene. It is found the SOCS1 hypermethylation within exon 2 is unlikely to be involved in the regulation of gene transcription,58 whereas hypermethylation of SOCS1 promoter is detected in various cancers, including 40% of hepatoblastoma,68 39–60% of hepatocellular carcinoma,15, 16, 69, 70 50% of pancreatic cancers,71 75% of melanoma,72 60% of AML,73 23% of ovarian cancer,64 63.9–74.5% of MM18, 74, 75 and 44% of gastric cancer.76

In addition, SOCS3, which closely reassembles SOCS1 in structure and functions in SOCS family, was also supposed to be a tumor suppressor that was found in downregulation of tumor cells, such as lung cancer, resulted from the promoter hypermethylation and induced by the factors, such as androgen, estrogen and PRL, in the tumor microenvironment.77–81 However, tumor cells, such as melanoma cells, constitutively expressing high level of SOCS3 may rather be indicative of a tumor-protecting function.82–84 These are all similar to SOCS1 in tumor cells.

Potential Value in Clinical Applications

A potential biomarker for tumor diagnosis and prognosis

A good diagnostic biomarker of tumor is expected to have high sensitivity and specificity, and be readily detectable in the clinic. Assessments of alterations of substances in peripheral blood of patients have been considered as the most convenient and minimally invasive approach. It was found that SOCS1 expression is undetectable in normal individuals but overexpression of SOCS1 mRNA is detected in 65% ph-positive CML patients in total WBC and 60% in granulocytes at diagnosis. Overexpression of SOCS1 mRNA is associated with poor cytogenetic responses to IFN-α and shorter median PFS.85 As to the functions of SOCS1 in carcinogenesis, levels of SOCS1 in tumor cells show a potential in predicting the outcome of cancer patients. It has been demonstrated that higher expression of SOCS1 mRNA is associated with earlier tumor stage and better clinical outcome in breast cancer.17 However, other reports indicate that the positivity and intensity of SOCS1 staining were associated with tumor progression, as indicated by tumor invasion, tumor thickness and stage of disease.62

Hypermethylation of SOCS1 is common in tumors, and recent studies have devoted to identify the correlation between SOCS1 hypermethylation and the clinical characteristics of tumors. Evidence has been presented showing that the hypermethylation of pannel genes, including SOCS1, CCND2, THBS1, PLAU and VHL, in the plasma DNA in patients with pancreatic cancer is a good diagnostic biomarker with 76% sensitivity and 59% specificity for tumors, with better tumor-predicting potential than the CA199 and CA125 markers that are more closely associated with tumor burden, during early diagnosis.86 In contrast, for MM, the predicative value of SOCS1 hypermethylation seems to be different. High frequency of SCOS1 hypermethylation is found in the BM mononuclear cells of MM patients but is related to opposite prognostic results in different studies.18, 35, 74, 75 In solid tumors, although it has been reported that hypermethylation of SOCS1 is associated with lymph node metastasis and advanced stage of gastric and hepatocellular cancer patients,56, 76 this is not detected in other types of tumors. More studies focusing on the combination of SOCS1 hypermethylation and other gene markers, such as P16, CDH1, GSTP1 and DAPK, have been carried out to further define the prognostic value of SOCS1 in various tumors, but no definitive conclusion can be drawn at this point.16, 68, 87

Overall, for both plasma DNA and tumor samples, the role of SOCS1 as a tumor diagnostic and prognostic marker requires further verification using larger sample sizes and tumors of different types.

SOCS1 as a target for antitumor therapy

Based on the discoveries in basic research, researchers have developed ideas to regulate the state of DCs through modulating the expression of SOCS1. Silencing SOCS1 in DCs has been found effective in optimizing the antigen presentation capacity of DCs by abrogating self-tolerance at the cellular and host levels, to control the growth of tumor and promote the memory T cell responses at a higher efficiency.2, 3, 88 Similarly, researchers also obtained encouraging results by silencing SOCS1 in macrophages to prompt antitumor immunity.53

Because of the controversial function of SOCS1 in carcinogenesis as described above, modulation of SOCS1 expression in tumor cells for antitumor therapy is highly context-dependent. SOCS1 expression is higher in IFN-resistant tumor cells,12, 14, 34 and its overexpression is inversely associated with disease progression. The underlying mechanism of SOCS1 in mediating IFN resistance has, therefore, been investigated. It is found that in neuroendocrine cancer cells, SOCS1 abrogates apoptosis induced by IFN-α, which is commonly used in biotherapy for this type of carcinomas. Silencing SOCS1 in cancer cells, instead, improves the sensitivity of cancer cells to IFN-α, as indicated by the decreased cell viability and increased apoptosis.14 Meanwhile, silencing of SOCS1 inhibits cell proliferation by blocking cell cycle progression, resulting in accumulation of G0/G1 phase and reduction of S phase.14 Similar results are also obtained in melanoma cells.13, 63 On the other hand, elevated expression of SOCS1 in tumors that have reduced endogenous SOCS1 expression due to the hypermethylation of SOCS1 gene has also been explored. Restoration of SOCS1 in these tumor cells inhibits the colony growth in breast cancer and metastasis of melanoma cells to the brain. Demethylation drugs, such as DAC, can restore SOCS1 expression in tumor cells and may serve as a potential strategy for antitumor therapy.18 Additionally, modulation of SOCS3 in tumor cells obtains similar results. Overexpression of SOCS3 in tumor cells could induce apoptosis and G0/G1 arrest, as well as inhibit the proliferation of tumor cells.89 However, for some other tumor cells, silencing SOCS3 could enhance therapeutic efficacy of IFN-α.90, 91 Thus, these opposing observations prompt us to further clarify the function of SOCS1 in various tissues and at various stages during carcinogenesis, and to evaluate the overall clinical benefits of modulating SOCS1 expression in larger patient cohorts.

Concluding Remark

The mechanisms of SOCS1-mediated negative feedback regulation of various signaling pathways have started to be understood. Data from basic researches have prompted investigators to develop novel strategies to improve the antigen presenting capacity by modifying SOCS1 expression in APCs.2, 53 The functions of SOCS1 in tumor cells during carcinogenesis appear to be complex and controversial in previous studies. On one hand, elevated expression of SOCS1 has been observed in some tumor cells; on the other hand, silencing of SOCS1 gene expression by hypermethylation is found in other tumor cells. As a result, opposite strategies have been explored to, respectively, reduce or restore SOCS1 expression in these tumors. Although some reports suggest silencing SOCS1 in tumor cells can improve the sensitivity of tumor cells to IFNs and inhibit the proliferation of tumor cells, others show that by using demethylation drugs, such as DAC, to restore the expression of SOCS1 in tumor cells, one can restore the tumor-suppressive function of SOCS1. The same situation is also in SOCS3. The contradictory results in different tumors may be the results of the different tumor origins and/or the different microenvironments the tumor resides. Therefore, future in vivo studies of SOCS1 will be necessary for a comprehensive understanding of the function of this intriguing molecule. Further understanding of the mechanisms of action of SOCS1 in cancer will also facilitate its future applications as novel diagnostic and prognostic biomarker and therapeutic target.

Acknowledgements

The authors declare that they have no competing interests.

Ancillary