Transcriptional Upregulation of SOCS 1 and Suppressors of Cytokine Signaling 3 mRNA in the Absence of Suppressors of Cytokine Signaling 2 mRNA After Infection with West Nile Virus or Tick-Borne Encephalitis Virus

Introduction

W est Nile Virus (WNV) and tick-borne encephalitis virus (TBEV) are members of the genus Flavivirus, and are transmitted by mosquito and tick vectors, respectively. Both viruses are neurotropic, and can cause severe encephalitis, which on occasions is fatal.

In recent years, both of these viruses have extended their geographical range (Charrel et al. 2004, Tyler 2009), and due to a number of factors both viruses may potentially have the ability to emerge in new regions in the near future. TBEV exists in an ecological cycle involving Ixodid ticks, deer, and wild rodents, but causes a substantial number of cases of TBE in humans throughout Europe and Russia each year, some of which are fatal (Gritsun et al. 2003, Mansfield et al. 2009). In comparison, WNV has become established worldwide, and exists in an ecological cycle involving mosquitoes and birds, with humans and equines becoming occasional dead-end hosts (Gubler et al. 2007). Both of these positive-strand RNA viruses cause severe neurological disease, and there is a clear correlation between disease severity and the ability of pathogenic strains to modulate the innate immune response (Suthar et al. 2009).

WNV can cause a persistent infection in experimentally infected hamsters (Tesh et al. 2005) and in nonhuman primates (Pogodina et al. 1983). More recently, WNV shedding in the urine of WNV-positive hypertensive patients in the United States provides further evidence for the possibility that persistent renal infection may occur in humans (Murray et al. 2010). In this report, ∼60% of these patients remained symptomatic 1 year after infection, and chronic symptoms were associated with the persistence of detectable WNV-specific immunoglobulin M in serum and detection of the gene encoding the WNV envelope protein in urine by reverse transcription polymerase chain reaction (RT-PCR) in 20% of patients.

After viral recognition by host pathogen-recognition receptors in mammalian cells, a number of intracellular signaling cascades are initiated, leading to the activation of transcription factors, such as interferon (IFN)-regulatory factors and nuclear factor-κB, which in turn regulate expression of many genes such as IFNs, cytokines, and chemokines (Der et al. 1998, de Veer et al. 2001). Secreted type I IFNs bind to cell surface receptors, thus activating members of the Janus tyrosine kinase (JAK) family of proteins. Once activated, JAKs phosphorylate the signal transducers and activators or transcription (STAT) family of transcription factors. This leads to the STAT proteins forming complexes with other transcription factors, which in turn activate transcription of IFN-stimulated genes. These gene products are the main effectors of the IFN response, for example, 2′-5′ oligoadenylate synthetase and Mx proteins (Stark et al. 1998, de Veer et al. 2001).

Suppressors of cytokine signaling (SOCS) proteins are a family of proteins, of which there are currently eight recognized members: cytokine-inducible SH2-containing protein (CIS) and SOCS 1–7. This group of proteins share a central SH2 domain and a carboxy-terminal SOCS-box domain, although the amino-terminus varies in length between the group members (Yoshimura et al. 1995, Starr et al. 1997, Hilton et al. 1998). The SOCS family of proteins function as E3 ubiquitin ligases, mediating the degradation of proteins that associate with their amino-terminal regions (Yoshimura et al. 2007). Early work has suggested that SOCS proteins act in a classic negative feedback loop to regulate cytokine signal transduction, as transcription of CIS and SOCS 1–3 was increased rapidly in response to interleukin-6, both in vitro and in vivo (Starr et al. 1997). SOCS proteins are now known to be induced upon stimulation of type I and II cytokine receptors, and in turn inhibit signal transduction of type I and II cytokine receptors, by using their SH2 domain to bind to phosphorylated tyrosine residues (Fig. 1). Furthermore, SOCS 1 and SOCS 3 also have an additional domain in the N-terminal of the SH2 region, which is termed the kinase inhibitory region and inhibits JAK activity by attaching to the catalytic cleft and preventing further JAK enzymatic activity (Sasaki et al. 1999, Yasukawa et al. 1999). While also binding to receptor phosphotyrosines, SOCS 2 inhibits signaling by competing with STAT molecules for recruitment to the receptor complex (Croker et al. 2008). Recent studies have suggested that SOCS 1 may be detected within the nucleus where it might also interact with nuclear factor-κB (Maine et al. 2007), thus suggesting a broader range of inhibitory activity than previously suggested (Baetz et al. 2008).

OPEN IN VIEWER

FIG. 1.

Interference of the JAK/STAT signaling pathway by SOCS 1, 2, and 3. SOCS proteins are induced upon stimulation of type I and II cytokine receptors, and in turn inhibit signal transduction of type I and II cytokine receptors, through binding of their SH2 domain to phosphorylated tyrosine residues. SOCS 1 and SOCS 3 inhibit JAK activity by attaching to the catalytic cleft and preventing further JAK enzymatic activity. SOCS 2 inhibits signaling by competing with STAT molecules for recruitment to the receptor complex. JAK, Janus tyrosine kinase; SOCS, suppressors of cytokine signaling; STAT, signal transducers and activators or transcription.

The induction of SOCS proteins by pathogens may inhibit the immune response and contribute to persistent infection or even a fatal outcome. In this study, we have investigated the induction of SOCS gene expression after peripheral infection with WNV or TBEV in the murine model.

Materials and Methods

Results

The IN route of inoculation is well documented for flavivirus-inoculation experiments, as it allows rapid entry of the virus into the brain for the study of clinical disease (Sheahan et al. 2002). It has previously been demonstrated that IN inoculation of mice with WNV enabled the virus direct access to the rich olfactory neuroepithelium in the roof of the nasopharynx and leading to 100% mortality on day 7 p.i. (Wacher et al. 2007). This method was also used to demonstrate WNV-induced seizures in Balb C mice (Getts et al. 2007). However, this is not representative of the natural route of transmission and may be far more neuroinvasive than alternative peripheral routes, such as the intradermal route which may be more representative of a tick or mosquito bite.

After IN infection with TBEV or WNV, transcriptional responses of SOCS genes were measured within the brain at 6 days p.i., and compared to the uninfected controls (Fig. 2). There was an increase in the transcriptional response for SOCS 1 (Fig. 2), whereas 28-fold increase in mRNA transcripts was observed for WNV, whereas TBEV demonstrated a 17-fold increase, which was statistically significant (p = 0.00643). Similarly, a 28-fold increase in mRNA transcripts for SOCS 3 was observed for TBEV and a 33-fold increase for WNV, which were both statistically significant (p = 0.00019 and p = 0.00997, respectively). Changes in the transcriptional response for SOCS 2 for either virus were not observed. It is apparent that the fold change for WNV SOCS 1 is greater than that observed with TBEV; this was despite the fact that the mean fold change for WNV was not statistically significant, whereas the mean fold change for TBEV was statistically significant. It is possible that variation between mice could be due to variation in the level of uptake and subsequent replication of virus. Quantification of virus may enhance understanding of the relationship between viral replication and SOCS gene expression.

OPEN IN VIEWER

FIG. 2.

SOCS mRNA transcripts in mouse brain, after intranasal inoculation with WNV and TBEV. Statistical analysis by unpaired t-test; significant changes in comparison to negative controls (p < 0.05) denoted by asterisks (*). Error bars denote standard error of the mean. TBEV, tick-borne encephalitis virus; WNV, West Nile virus.

Immunoblotting for SOCS 3 confirmed that we were able to detect SOCS 3 protein in TBEV- and WNV-infected mouse brain samples (data not shown), demonstrating a translational response within the brain.

Discussion

The JAK-STAT pathway is a common signaling pathway used by many cytokines in the innate immune response to viral infection. We have shown that IN infection of mice with WNV or TBEV induced an upregulation of mRNA transcripts for SOCS 1 and SOCS 3 in the brain. This suggests that during flavivirus infection, SOCS proteins influence the innate immune response via JAK rather than STAT, since SOCS 3 inhibits JAK activity by binding to gp130, and SOCS 1 binds to and inhibits JAKs directly, whereas SOCS 2 (which was not upregulated) is known to interfere with STAT recruitment to the receptor complex.

We were able to confirm detection of SOCS 3 protein in TBEV- and WNV-infected brain samples (data not shown), although it is possible that changes in the stability of SOCS proteins may lead to downregulation (degradation) at the translational level, thus regulating feedback inhibition of cytokine signal transduction and enabling induction of SOCS by cytokines to continue in a feedback loop. Earlier work has suggested that elongins B and C, components of ubiquitin ligases, were able to link SOCS proteins to the proteasome, and target them for degradation (Zhang et al. 1999). However, later studies suggest that it is the stability of SOCS proteins which is mediated via this interaction with elongin C, since Jak-mediated phosphorylation of SOCS 3 at two tyrosine residues within the conserved SOCS box (Tyr²⁰⁴ and Tyr²²¹) has been shown to inhibit the interaction between SOCS 3 and elongin C, thus activating proteasome-mediated degradation of SOCS 3 (Haan et al. 2003).

Infection with WNV and TBEV induce an innate immune response that involves at least SOCS 1 and SOCS 3. However, it is known that flaviviruses such as TBEV and WNV also have a number of alternative mechanisms of regulating the host immune response, such as the interaction between TBEV nonstructural 5 protein and the PSD/Dlg/ZO-1 protein scribble (hScrib), which influences IFN type I- and II-mediated JAK-STAT signaling (Werme et al. 2008). Similarly, it has been demonstrated that WNV nonstructural proteins were responsible for preventing phosphorylation and activation of the Janus kinases JAK1 and Tyk2 (Guo et al. 2005).

For flaviviruses such as WNV and TBEV, the mechanisms by which they are able to cross the blood brain barrier are not completely understood, although certain cytokines such as tumor necrosis factor-α are known to alter the permeability of blood brain barrier endothelial cells, thus allowing viral entry into the brain (Diamond and Klein 2004, Wang et al. 2004). The subsequent innate immune response within the brain includes the release of a wide range of cytokines, in an attempt to control infection. However, recent studies have suggested that an increase in IFN-γ can also have an immunopathological effect within the central nervous system, characterized by seizures, after WNV infection in mice (Getts et al. 2007). Therefore, it is clear that SOCS proteins may play a role in limiting cytokine responses in the brain as a neuroprotective mechanism. However, the role of SOCS proteins in the innate immune response to viral infection is a clearly complex one, as this suppression of cytokine signaling may actually enhance the ability of neuroinvasive viruses such as WNV and TBEV to spread and cause disease via a mechanism of immunosilencing that may facilitate viral persistence.