Syk and Hrs Regulate TLR3-Mediated Antiviral Response in Murine Astrocytes

Toll-like receptors (TLRs) sense the presence of pathogen-associated molecular patterns. Nevertheless, the mechanisms modulating TLR-triggered innate immune responses are not yet fully understood. Complex regulatory systems exist to appropriately direct immune responses against foreign or self-nucleic acids, and a critical role of hepatocyte growth factor-regulated tyrosine kinase substrate (HRS), endosomal sorting complex required for transportation-0 (ESCRT-0) subunit, has recently been implicated in the endolysosomal transportation of TLR7 and TLR9. We investigated the involvement of Syk, Hrs, and STAM in the regulation of the TLR3 signaling pathway in a murine astrocyte cell line C8-D1A following cell stimulation with a viral dsRNA mimetic. Our data uncover a relationship between TLR3 and ESCRT-0, point out Syk as dsRNA-activated kinase, and suggest the role for Syk in mediating TLR3 signaling in murine astrocytes. We show molecular events that occur shortly after dsRNA stimulation of astrocytes and result in Syk Tyr-342 phosphorylation. Further, TLR3 undergoes proteolytic processing; the resulting TLR3 N-terminal form interacts with Hrs. The knockdown of Syk and Hrs enhances TLR3-mediated antiviral response in the form of IFN-β, IL-6, and CXCL8 secretion. Understanding the role of Syk and Hrs in TLR3 immune responses is of high importance since activation and precise execution of the TLR3 signaling pathway in the brain seem to be particularly significant in mounting an effective antiviral defense. Infection of the brain with herpes simplex type 1 virus may increase the secretion of amyloid-β by neurons and astrocytes and be a causal factor in degenerative diseases such as Alzheimer's disease. Errors in TLR3 signaling, especially related to the precise regulation of the receptor transportation and degradation, need careful observation as they may disclose foundations to identify novel or sustain known therapeutic targets.


Introduction
Astrocytes constitute 19-40% of brain glial cells and are the key component responsible for homeostasis and immune and oxidative stress defense in the CNS [1,2]. By participating in the biogenesis and transport of a wide range of neuroactive substances, they affect neurons and other glial cells and thus regulate many physiological and pathophysiological processes [3]. In neuropathological conditions, immunologically silent astrocytes often undergo reactive reprogramming [4].
Recent reports reveal astrocytes as cells that play a substantial role in the pathogenesis of Alzheimer's disease (AD), the most frequent type of brain amyloidosis and the most common type of dementia in humans [5]. Amyloid-β (Aβ) plaques, AD's hallmark, activate cerebral glial cells and cause neuroinflammation resulting in neuronal cell death [6]. Such a process is decisive for the progression of AD. An increase in the number of reactive astrocytes that surround and may phagocytose Aβ plaques is observed in the neuronal vicinity [7]. However, following reactive reprogramming, astrocytes exhibit high concentrations of the amyloid precursor protein (APP) and βand γ-secretase that enable formulation of the Aβ plaques [6]. Interestingly, during reactive gliosis, about 50% of the altered gene expression in astrocytes is significantly dependent on the initiating brain injury [8]. One of the factors leading to neuroinflammation, which may contribute to astrocyte reprogramming and enhanced astrocytic secretion of Aβ, is herpes simplex type 1 virus (HSV-1) infection in the brain. Aβ accumulates in HSV-1-infected cell cultures, while viral particles, as well as viral nucleic acid were found in the vicinity of amyloid plaques in the brains of mice and humans [9]. Toll-like receptor 3 (TLR3) plays an essential role in the innate immune control of cerebral HSV-1 infection. Therefore, it is likely that infected astrocytes detect the virus through TLR3, thus activating them and contributing to production of Aβ.
Substantial expression of TLR3 occurs in neurons and astrocytes, oligodendrocytes, and microglia, of which expression in astrocytes is the most abundant [10,11]. An increase in the TLR3 expression in CNS-resident cells is usually associated with the development of neuroinflammation [12]. Stimulated glial cells and macrophages are responsible for the removal of microorganisms and injured cells. In addition to the production of various growth factors, chemokines and cytokines such as IL-1β, TNF-α, IL-6, and glial cells secrete oxides that may become neurotoxic during brain injuries or neurological diseases and exacerbate CNS dysfunction [13,14].
Toll-like receptors (TLRs) are important contributors to activation of the innate immune response in the brain during infection, injury, or degeneration [15][16][17][18]. TLR3 is an evolutionarily conserved protein which recognizes doublestranded RNA (dsRNA) in endosomes. Double-stranded RNA may constitute a viral nucleic acid or an intermediate product formed during replication of viruses such as HSV-1 and HSV-2. Patients, especially children with deficiencies in TLR3 or single-gene errors in components of the TLR3 signaling pathway, are more susceptible to HSV-1 and HSV-2 infections, which may be the cause of a devastating human disease-herpes simplex encephalitis (HSE) [19][20][21][22]. Being one of the most common viral brain diseases in the world, HSE entails many deleterious outcomes. The accession of external factors into the central nervous system (CNS) and the CNS immune response are precisely controlled; however, due to the neuronal latency of HSV, inflammation in the brain may last for years and have a recurrent character [23]. The length of inflammatory response in CNS and disease progression is affected by the balance between pro-and anti-inflammatory signals in the neuronal environment [24]. Sustainability of the inflammatory process or deficiencies in distribution of suppressive mechanisms may lead to pathological repercussions and influence the outcome of disease.
Prior to launching the signaling cascade, nucleic acid-sensingTLRs such as TLR3, TLR7, and TLR9 enter the UNC93B1-dependent secretory pathway from the endoplasmic reticulum (ER) through the trans-Golgi network (TGN) to endosomes. However, receptors are subject to differential UNC93B1-related sorting mechanisms [25]. Furthermore, the endosomal sorting complex required for transport-0 (ESCRT-0), composed of hepatocyte growth factor-regulated tyrosine kinase substrate (Hrs) and signal transducing adaptor molecule (STAM), was recently implicated in post-Golgi trafficking by sorting ubiquitinated TLR7 and TLR9 to endosomes [26], and silencing of Hrs reduced signaling through TLR7 and TLR9 [27]. Chiang et al. [27] indicated that Hrs binds directly to particular TLRs and that the interaction of Hrs with TLR9 was much stronger than that with cell-surface-expressed TLR2. STAM, similar to Hrs, may demonstrate endosomal localization and display a potent sorting efficiency due to multiple ubiquitin-binding domains (UBDs). Moreover, it was demonstrated that STAM localizes prominently to early endocytic vesicles and decidedly regulates morphology of the Golgi apparatus [28], the site where TLRs are packaged en route to endosomes.
In addition to TLR trafficking aimed at ligand recognition, ESCRT-0-mediated sorting of receptors may direct them for recycling or degradation. Regardless of ligand stimulation, IL-2 receptor β and IL-4 receptor α were consistently internalized and delivered to late endosomes (LE) in an ESCRT-dependent manner by association with Hrs [29]. Following activation, receptors such as EGFR, PDGF, or TLR4 were endocytosed and targeted via the ESCRT pathway for lysosomal degradation [30][31][32]. The formation of the endosomal sorting machinery and its ability to target EGFR were regulated in this case by modulation of Hrs protein level, phosphorylation, and ubiquitination [33]. Furthermore, deficiencies and overexpression of ESCRT machinery components led to reduced EGFR degradation [34]. Because EGFR is responsible for TLR3 phosphorylation, posttranslational modifications, as well as alterations in the expression of ESCRT-0 subunits could affect TLR3 signaling.
During HSV infection, release of viral dsRNA from the cells at the site of brain injury entails TLR3 activation. Upon phosphorylation by Bruton tyrosine kinase (Btk), c-terminal Src kinase (c-Src), and epidermal growth factor receptor (EGFR), the receptor triggers signaling in a pathway that enrolls transcription factors such as nuclear factor kappa B (NF-κB), interferon regulatory factor 3 (IRF3), and interferon regulatory factor 7 (IRF7), responsible for the development of inflammatory response [35][36][37] (Figure 1). Compared to other endosomal TLRs, TLR3 engages a different adaptor protein (Toll/interleukin 1 receptor domain-containing adaptor protein inducing IFN-β (TRIF)) for initiation of IRF3 and NF-κB signaling. Spleen tyrosine kinase (Syk) has been shown to phosphorylate tyrosine residues of TRIF, the TLR3 adaptor protein. Such a process leads to the TRIF proteasomal degradation and entails downregulation of the TLR signaling [38]. Furthermore, Syk significantly regulates Hrs phosphorylation and ubiquitination, as well as its membrane/cytosol localization [39]. Syk is known to function at the plasma membrane, but also in cytoplasmic and nuclear compartments of the cells [40], while Hrs may interact with Golgi proteins or reside in membranes of early endosomes and mediate delivery of protein cargo to multivesicular bodies (MVBs) for the subsequent degradation [41]. Consequently, Hrs serves as an important trafficking regulator and both Syk and Hrs may be critical controllers of protein entry into lysosomes for degradation. However, influence of these proteins on TLR3 signaling in CNS cells remains largely unexplored. The impact of Syk and ESCRT-0 on activation of the immune response may be receptor-and pathway-specific. It may be distinctively regulated in various cellular compartments and types of cells or tissues, as the ESCRT-0 target proteins may fulfill their function in the endosomes or on the cell surface.
Progression of the antiviral defense is guided by increased levels of type I interferons (IFNs) (IFNα, IFNβ), cytokines (IL-1β, IL-6, and TNFα), chemokines (CXCL8, CCL5, and CXCL10), and other molecules, such as 2′5′OAS. The ground for the clinical signs of HSE is as follows: impaired but not abolished IFNα/β and IFNγ production in response to TLR3 stimulation [42]. Because HSE is manifested in CNS, attention should be paid to discovering and characterizing immunological engagement in HSV-1 and HSV-2 control, particularly related to TLR3 transportation, activation, and degradation in CNS-resident cells.
In this research, we investigated the involvement of Syk, Hrs, and STAM in the regulation of the TLR3 signaling pathway in the C8-D1A cell line. Our studies identify molecular events in murine astrocytes such as phosphorylation of Syk and Hrs and interaction of Syk and Hrs that occur shortly after TLR3 stimulation. We also show that the receptor undergoes ligand-induced proteolytic processing and that the N-terminal form of TLR3 exclusively interacts with Hrs. Finally, we demonstrate that silencing of Syk or Hrs in  Figure 1: TLR3 signaling in astrocytes. Upon dsRNA recognition in the endosomal compartment, TLR3 undergoes dimerization and interacts with the TRIF adaptor molecule. TRIF activation is followed by TRAF6 and TRAF3 recruitment. TRAF6 conducts the signal via RIP-1 and RIP-3 kinases which facilitate NEMO, IKK-α, and IKK-β complex formation, followed by NF-κB phosphorylation and translocation into the nucleus. TRAF3 engages TBK1 and IKK-i/Ɛ for IRF3 and IRF7 activation, followed by their dimerization and translocation into the nucleus. This leads to the induction of type I IFNs and proinflammatory cytokine gene expression. The dotted arrows highlight possible roles of ESCRT-0 in TLR3 transport from the ER to the endosome, as well as the role of Hrs and Syk in TLR3 degradation.

Western Blot
Analysis. At the indicated times or concentrations, astrocytes were processed for protein assays. Cells were lysed with radioimmunoprecipitation assay (RIPA) buffer (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 1% protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific), and protein concentration was determined with the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific) and spectrophotometry on an Epoch BioTek spectrophotometer. Proteins (20 μg/well) from the cells were separated by SDS-PAGE and electrotransferred onto PVDF membranes using the Bolt® System (Thermo Fisher Scientific). After blocking for 2 h in phosphate-buffered saline with Tween (PBST) containing 5% nonfat milk, the blots were incubated overnight at 4°C with primary antibodies. Subsequently, membranes were washed 3 times after which they were probed with secondary anti-goat or anti-mouse antibodies conjugated to horseradish peroxidase (HRP) (CST, 2.6. Immunostaining and Fluorescence Microscopy. Astrocytes were cultured on 22 mm glass cell culture coverslips in 24-well plates. Untreated or poly(I:C)-treated cells were fixed for 15 min in phosphate-buffered saline (PBS) with 4% paraformaldehyde (Sigma-Aldrich). After washing, cells were permeabilized with 0.5% Triton X-100 (Sigma-Aldrich) in PBS and blocked with 3% bovine serum albumin (Sigma-Aldrich) with 0.1% Triton X-100. Subsequently, cells were incubated for 1 h with anti-TLR3 or anti-STAM or anti-Syk or anti-Hrs antibodies (1 : 50, Thermo Fisher Scientific) and washed with 0.1% Triton X-100 in PBS. Then, astrocytes were incubated with secondary antibodies conjugated with rhodamine Red-X (1 : 100, Jackson ImmunoResearch Laboratories Inc., West Grove, PA, USA) for 1 h. In double immunofluorescence experiments, cells were incubated for 1 h with the mixture of anti-TLR3 and anti-PDI antibodies or anti-STAM and anti-PDI antibodies. Next, after washing, cells were incubated for 1 h with a mixture of secondary antibodies conjugated with rhodamine Red-X or FITC (Jackson ImmunoResearch Laboratories). The cells were then washed and stained with Hoechst 33342 (Sigma-Aldrich) for 10 min. Finally, after washing with PBS, the coverslips were mounted in ProLong Gold Antifade Reagent (Thermo Fisher Scientific). Fluorescence microscopy was performed with an Olympus BX60 fluorescence microscope and analyzed with Cell^F software (Soft Imaging System) (Olympus, Tokyo, Japan).

2.7.
Immunoprecipitation. C8-D1A cells were cultured in a 6-well plate to reach 80-100% confluence. Cells were stimulated with poly(I:C) or poly(I:C)/LyoVec at the indicated times. If the transfection with specific siRNA was required, astrocytes were pretransfected and 48 or 72 h posttransfection stimulated with poly(I:C) or poly(I:C)/LyoVec as described in Section 2.2. Subsequently, the cells were lysed with IP lysis buffer (Thermo Fisher Scientific) supplemented with protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific). Syk, Hrs, STAM, and TLR3 were immunoprecipitated using Catch and Release® v2.0 Reversible Immunoprecipitation System (Merck) according to the manufacturer's protocol and subjected for immunoblotting. Normal mouse IgG (Santa Cruz Biotechnology) was used as negative control for immunoprecipitation experiments. Quantification of each cytokine concentration in cell supernatants was determined by reading ODs on a linear calibration curve generated for each protein.
2.9. Cell Fractionation. C8-D1A cells were untreated or treated with poly(I:C) or poly(I:C)/LyoVec for 5, 8, 12, 15, 30, and 60 min. Subsequently, cells were collected and fractionated using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific) according to the manufacturer's protocol. Nuclear and cytoplasmic extracts' protein concentration was determined using BCA Protein Assay kit (Thermo Fisher Scientific). Next, nuclear and cytoplasmic extracts were subjected to western blot analysis.
2.10. Statistical Analysis. Quantitative data are presented as mean ± standard deviation (SD) from at least three independent biological experiments (unless otherwise indicated). All data were analyzed in STATISTICA software (StatSoft, Poland). Comparisons were made using Student's t-test. A p value ≤0.05 ( * ) or ≤0.01 ( * * ) was considered statistically significant.  cleavage by pH-dependent cysteine proteases-cathepsins B, H, L, and/or S [45,46]. To evaluate whether astrocytic TLR3 responds to poly(I:C) treatment and undergoes proteolytic processing, we treated cells with poly(I:C) at different concentrations or with poly(I:C)/LyoVec (10 μg/ml) for various time intervals and performed western blot analysis with the antibodies directed against amino acid fragment localized at the N-terminus of the TLR3 protein.

Expression of Syk and Hrs Is Upregulated upon
Stimulation of Murine Astrocytes with Poly(I:C). Recent studies have reported that Syk may be an important controller of the immune receptor transportation and signaling [38,47,48] and that Hrs participates in the regulation of responses of multiple TLRs [27,49]. Importantly, Hrs may contribute to protein cargo sorting as the HRS-STAM heteromer, homotypic hexamer [50], or in combination with subunits from other ESCRT complexes, e.g., with tumor susceptibility gene 101 (TSG101), the ESCRT-I component [51]. Therefore, we investigated if stimulation of astrocytes with the TLR3 ligand entails changes in the level of Syk, Hrs, and STAM expression, which could point to the potential role of these proteins in TLR3 immune modulation. Following TLR3 stimulation, Syk and Hrs expression was upregulated in a time-dependent manner ( Figure 6(a)), and densitometric measurements showed highly statistically significant (p ≤ 0 01) difference for Hrs. Similarly, following poly(I:C)/LyoVec treatment, the Hrs expression level also increased, but not to such a high level as following poly(I:C) treatment ( Figure 6(b)). On the contrary, expression of Syk and STAM remained at a similar level throughout all stimulation time intervals with poly(I:C)/LyoVec ( Figure 6(b)). We also observed a significant increase in the expression of Syk in response to treatment with increasing concentrations of poly(I:C), in contrast to Hrs (Figure 6(c)). All the studied proteins exhibited reduced expression level when astrocytes were treated with poly(I:C)/LyoVec at concentrations of 2 and 5 μg/ml, which could indicate that these proteins may not participate in MDA5 signaling in murine C8-D1A cells ( Figure 6(d)). Interestingly, no significant differences in STAM expression were observed, either after poly(I:C) treatment of astrocytes with various concentrations or at different time courses (Figures 6(a) and 6(c)). Similar results were observed in STAM expression, after stimulation of cells with poly(I:C)/LyoVec (Figures 6(b) and 6(d)).

Distribution of TLR3 and STAM Is Altered upon Poly(I:C)
Stimulation of Murine Astrocytes. To study whether the observed changes have a reflection in protein distribution at the cellular level following stimulation and thus better understand the connection between TLR3, ESCRT-0, and Syk in astrocytes, we examined the localization of TLR3, Syk, Hrs, and STAM in C8-D1A cells stimulated with poly(I:C) (10 μg/ml) for 5 min, 4 h, and 24 h. We chose such a time scale to show that the posttranslational modifications of Syk and Hrs that occur within the first half hour after TLR3 stimulation are not significantly related to the change in their distribution. Consequently, noticeable modification in the arrangement of TLR3 and STAM was found at 4 h and 24 h after the addition of poly(I:C) (Figures 7(a) and 7(b)). Our findings indicate that TLR3, STAM, Syk, and Hrs are highly expressed in murine astrocytes (Figures 7(a)    Similarly, the majority of STAM was present in the perinuclear region in untreated cells, whereas in poly(I:C)-stimulated cells, the longer the duration of stimulation, the larger the dispersion and abundance of the protein vesicles was observed (Figures 7(b) and 7(f)). We observed a similar distribution of Syk and Hrs in untreated cells and cells stimulated with poly(I:C) (Figures 7(c) and 7(d)); most of the proteins were located near the nucleus, while part was dispersed in the cytoplasm.  [39]. Here, we demonstrate that poly(I:C) stimulation of astrocytes leads to rapid phosphorylation of Syk, which appears at 1 min and peaks at 2-5 min following TLR3 stimulation (Figure 2(a)). Western blot data corresponded with the densitometric analysis where the highest level of Syk phosphorylation was observed 1-2 min after the addition of poly(I:C) (Figure 2(a)). Such a result is consistent with the work of Cao et al. [53], where phosphorylation of the Tyr-346 residue of Syk was detected We have not observed Syk phosphorylation after the addition of poly(I:C)/LyoVec to the cells (Figure 2(a)). To investigate whether Syk could associate with Hrs, lysates from astrocytes stimulated or unstimulated with poly(I:C) for the indicated time points were subjected to immunoprecipitation with anti-Hrs antibody and probed with anti-Syk antibody. In the fifth minute after the addition of the TLR3 ligand, we detected an interaction of Syk and Hrs (Figure 2(b)), which peaked between 5 and 12 min and decreased within 15 min of stimulation, whereas such interaction did not occur after the addition poly(I:C)/LyoVec to the cells (Figure 2(b)). A similar result was observed after FcεRI activation, where the Syk-Hrs interaction was maximal at 5 min and decreased to near-baselinewithin 20 min of stimulation [39]. Next, we evaluated whether Hrs may serve as a substrate for Syk-mediated phosphorylation. Lysates from unstimulated or poly(I:C)-stimulated astrocytes were subjected to immunoprecipitation with anti-Hrs antibody and probed with anti-phosphotyrosine antibody. Stimulation of cells with poly(I:C) resulted in tyrosine phosphorylation of Hrs (Figure 2(b)). Importantly, Hrs phosphorylation peaked at 5-12 min following the addition of poly(I:C), which is consistent with the peak of the Syk-Hrs interaction. Hrs phosphorylation was undetectable in cells stimulated with poly(I:C)/LyoVec (Figure 2(b)). We did not observe phosphorylation of Hrs after the addition of poly(I:C) to the cells treated with siRNA against Syk (Figure 2(c)), indicating that Syk may be the crucial kinase responsible for Hrs modification. Furthermore, following immunoprecipitation of astrocytic lysates with anti-Hrs antibody and immunoblotting with antibody recognizing amino acids at the N-terminal part of TLR3, we observed that the TLR3 N-terminal progeny form interacts in increasing amounts with Hrs following poly(I:C) stimulation (Figure 2(b)).

Poly(I:C) Treatment of Murine Astrocytes Induces TLR3 Tyrosine Phosphorylation and Promotes Interaction with
Hrs. Previous studies have identified Hrs as the protein that directly interacts with TLRs such as TLR2, TLR4, TLR7, and TLR9 [27,32,49]. To investigate whether Hrs associates with TLR3, lysates from astrocytes unstimulated or stimulated with poly(I:C) were subjected to immunoprecipitation with anti-TLR3 antibody. Probing of the immunoblot with anti-Hrs antibody showed that Hrs bound to TLR3 in nonstimulated cells. However, an increasing proportion of Hrs associated with TLR3 in a time-dependent manner (Figure 8(a)) in poly(I:C) stimulated cells. A similar result was observed in the case of TLR4, the receptor constitutively associated with Hrs, and the interaction increased after TLR4 stimulation [32]. Following poly(I:C)/LyoVec cell stimulation, only a small amount of Hrs was associated with TLR3 ( Figure 8(a)). Similarly, following TLR3 immunoprecipitation, we probed immunoblots with anti-STAM antibody and determined whether the second ESCRT-0 subunit interacts with TLR3. Interestingly, we did not observe any interaction between TLR3 and STAM, either following poly(I:C) or poly(I:C)/LyoVec stimulation of astrocytes ( Figure 8(a)). Furthermore, we wanted to address if poly(I:C) induces TLR3 ubiquitination. By probing western blots of immunoprecipitated TLR3 with anti-ubiquitin antibody, we showed that TLR3 is similarly ubiquitinated in C8-D1A cells treated with poly(I:C) or poly(I:C)/LyoVec (Figure 8(b)). Such modification of TLR3 may condition the interaction with ubiquitin-binding molecules, e.g., Hrs. Because ubiquitin ligases are known to interact with tyrosinephosphorylated proteins, we examined the level of TLR3 phosphorylation in nonstimulated and poly(I:C)-stimulated cells. Phosphorylation of the receptor was observed 8 min after the addition of poly(I:C) and later (Figure 8(a)), which may indicate that TLR3 modification such as ubiquitination is not fully dependent on the receptor's phosphorylation. Finally, we investigated whether Hrs-STAM interaction may be promoted by stimulation of cells with poly(I:C). While the majority of the Hrs cellular pool remained unassociated with STAM, the greater part of STAM interacted with Hrs in a time-dependent manner following TLR3 stimulation (Figure 8(c)). Such a result indicates that Hrs may not reside in the same cellular localization as STAM; however, the addition of poly(I:C) may foster interaction between these proteins.
3.6. Changes in the Intracellular Localization of NF-κB, IRF3, and IRF7 in TLR3-, Syk-, and Hrs-Depleted Astrocytes following TLR3 Stimulation. After describing molecular interactions between Syk and Hrs, as well as TLR3, Syk, and Hrs posttranslational modifications following poly(I:C) stimulation of C8-D1A cells, we investigated the role of Syk and Hrs in poly(I:C)-induced NF-κB, IRF3, and IRF7 nuclear translocation. Following the addition of poly(I:C) to the control cells, we observed translocation of NF-κB and IRF3 to the nucleus (Figure 3(a)). In contrast, NF-κB and IRF3 did not accumulate in the nucleus of TLR3-depleted cells stimulated with poly(I:C) (Figure 3(b)). Furthermore, NF-κB nuclear translocation was downregulated in poly(I:C)-treated astrocytes with knocked-down Syk, and shortage of Syk also appeared to minimally affect nuclear translocation of IRF3 in C8-D1A cells (Figure 3(c)). In Hrs-depleted cells NF-κB activation was reduced to a certain extent after stimulation with poly(I:C), whereas translocation of IRF3 to the nucleus remained intact in comparison to the control cells (Figure 3(d)). Interestingly, the amount of IRF7 localized in the nucleus was similar throughout the duration of poly(I:C) or poly(I:C)/LyoVec stimulation, and its physiological level was also observed in the nontreated cells (0 min time point, Figures 3(a)-3(d)).

A Role for Syk, Hrs, and STAM in the Regulation of TLR3
Signaling. To verify the influence of Syk, Hrs, and STAM knockdown on the response of C8-D1A cells to TLR3 ligand stimulation, we transfected astrocytes with the specific siR-NAs and subsequently activated the cells with poly(I:C). Following cell stimulation, secreted IFNβ, IL-6, and CXCL8 were quantified by ELISA. In Syk-knockdown cells, poly(I:C) significantly increased secretion of IFNβ (p ≤ 0 05), IL-6 (p ≤ 0 01), and CXCL8 (p ≤ 0 05) compared to poly(I:C)-stimulated cells without Syk depletion (Figures 4(a)-4(c)).  (Figures 4(a)-4(c)). Collectively, these results indicate that Syk and Hrs are involved in TLR3mediated signaling events. Because Syk and Hrs knockdown results in the increase of IFN-β, IL-6, and CXCL8 secretion, Syk and Hrs may serve to regulate TLR3-mediated immune response in intact astrocytes.

Discussion
In our study, we reveal that astrocytic TLR3 undergoes proteolytic processing and that the expression of TLR3 N increases in proportion to the poly(I:C) dose and length of exposure. Presentation of TLR3 in such a manner may modulate the level of response to viral dsRNA, especially in CNS cells. For example, such TLR3 form may constitute a negative regulator of signaling, as found for TLR9 [54], but this requires further investigation. TLR3 present in mammalian cells has a size of approximately 110 kDa; however, cleaved TLR3 molecules have been shown to remain associated and are also capable of binding the TLR3 ligand [55,56]. Such TLR3 configuration may represent the primary form of the receptor, and it is possible that the cleavage is aimed at actuating the novel receptor attributes, other than separation of the two progeny fragments. Conclusions from previous studies on the role of TLR3 cleavage are ambiguous. Proteolytic processing of TLR7 or 9 by cathepsins is required for signaling; however, TLR3 cleavage may not determine the activation of the immune response [57]. The addition of the cathepsin inhibitor or mutation at the TLR3 cleavage site did not influence TLR3 response to poly(I:C) in 4 cell lines, although inhibition of cleavage decreased the abundance of the receptor to be degraded in lysosomes [46]. On the other hand, TLR3 cleavage was indispensable for the receptor activation in murine RAW cells [58]. Zhang et al. [59] linked mutation situated in the region of the TLR3 cleavage and critical for dsRNA binding (P554S) in a patient suffering from HSE with the loss of TLR3 function in CNS cells and increased penetrance of the disease through insufficient antiviral response. This highlights the importance of proper TLR3 cleavage and its possible influence on ligand recognition and activation of the signaling pathway. Hrs and STAM are components of the ESCRT-0 transportation complex; however, they also function as separate proteins, e.g., Hrs bound in the membrane occurs partly as a monomer and partly as a form associated with STAM [50]. Hrs appears to be only partially involved in cooperation with STAM in C8-D1A cells not stimulated or stimulated with poly(I:C) (Figure 8(c)), and we did not observe any significant differences in the TLR3-mediated immune response level following STAM depletion (Figures 4(a)-4(c)). This indicates that after TLR3 activation in astrocytes, Hrs may interact with proteins from complexes other than ESCRT-0. ESCRT-0 subunits, alone or in combination with other proteins, may moonlight in manifold activities; e.g., Hrs in cooperation with the product of tumor susceptibility gene 101 (TSG101), an ESCRT-I subunit, leads to the endocytic downregulation of EGFR [51]. This protein is a tyrosine kinase responsible for the phosphorylation of Tyr858 of TLR3, a modification indispensable for the recruitment of TRIF  [36]. Silencing of Hrs, resulting in the reduced degradation of EGFR and concomitantly increased activatory influence on TLR3, could manifest in an increase of TLR3-mediated innate response to poly(I:C) and possibly viral dsRNA. Furthermore, we have demonstrated that TLR3 interacts with Hrs, which may have a direct influence on the fate of the receptor. Hrs may be involved in directing TLR3 N to the degradative pathway, as is the case with various cell membrane receptors. This, particularly, appears likely because IFNβ, IL-6, and CXCL8 secretion following knockdown of Hrs was higher compared to cells with intact Hrs. In the noncanonical ESCRT-0 pathway, which does not engage MVBs or lysosomes, Hrs has been implicated in targeting of ubiquitinated TLR9 and TLR7 to endosomes for ligand recognition. Furthermore, Hrs silencing blocked the nuclear transportation of NF-κB p65 and reduced the level of TNFα and IFNα secretion following TLR9 stimulation [27]. We performed a similar experiment to examine NF-κB activation in the astrocytes with the knockdown of Hrs. Following TLR3 stimulation, NF-κB exhibited reduced but not abolished transportation to the cell nucleus. However, we did not observe a significant influence of Hrs knockdown on the nuclear accumulation of IRF-3 or IRF-7, suggesting that Hrs neither is a key moderator nor participates in signaling steps leading to the intracellular transfer of these transcription factors. Furthermore, knockdown of Syk and Hrs in C8-D1A cells increased TLR3-dependent IFNβ, IL-6, and CXCL8 secretion following stimulation with poly(I:C). This suggests that Syk and Hrs participate in TLR3-mediated innate response to dsRNA; however, it is likely that Hrs potentially engages in the signaling events not related to TLR3 endosomal trafficking aimed at ligand detection, as in the case of TLR7 and TLR9. Syk may play a dual role in TLR4 activation by promoting the signaling through mediation of endocytosis or inhibiting signaling from the plasma membrane [60], similar to Hrs, which supports contradictory cellular events leading to degradation or recycling of numerous receptors. Following LPS sensing, ubiquitinated TLR4 interacts with Hrs on the way to degradation in lysosomes and it is possible that other cell surface TLRs are trafficked in the analogous manner [32]. It should be noted that although we were unable to acquire complete silencing of Syk, the knockdown achieved was adequate to affect Hrs phosphorylation, NF-κB nuclear translocation, or IFNβ, IL-6, and CXCL8 secretion. We speculate that deprivation of the cells of the vast majority of Syk is sufficient to prevent Hrs phosphorylation, which may indicate Syk as the key protein responsible for Hrs activation. Therefore, the shortage in Syk followed by lack of Hrs phosphorylation may be reflected in subsequent cellular events associated with TLR3 signaling. Posttranslational modifications play important roles in the regulation of Hrs activity, and Hrs phosphorylation has been proven to correlate with EGFR degradation [33]. Interestingly, astrocytes exhibited phosphorylation of a small proportion of the cellular pool of Hrs (Figure 2(b)), which is consistent with the Hrs phosphorylation level observed downstream of EGFR activation by Stern et al. [33]. If Hrs was implicated in retaining of degradative transportation of TLR3, the depletion of Syk and Hrs would open the possibility of prolonged TLR3 recruitment in the signaling cycle. It is highly probable that following viral dsRNA-mediated stimulation of astrocytes, concurrent with endosomal trafficking, ligand recognition, and execution of the TLR3 signaling cascade, a process leading to ligand-induced degradation of the receptor is initiated with the aim of modulating and maintaining the proinflammatory response at an adequate level.
Indirect immunofluorescence analysis revealed that TLR3 and STAM exhibited distinct staining patterns at different time courses after poly(I:C) stimulation (Figures 7(a)  and 7(b)). STAM was found to colocalize with ER, Golgi, and endosomal markers and participate in the reconstruction and restoration of the Golgi structure [28]. Owing to the VHS domain, STAM may interact with the Golgi-localizing proteins and participate in protein sorting at the trans-Golgi network [61,62]. Thus, the distribution pattern of STAM observed after poly(I:C) stimulation may reflect its functions related to intracellular TLR3 trafficking; however, this requires further investigation. Another unique attribute of STAM is the diphosphorylated immune tyrosine activation motif (ITAM) [63], the distinctive sequence found in the cytoplasmic subunits of B and T cell receptors, as well as Fc receptors. Upon phosphorylation, ITAM serves as a ligand for SH2 domains of various cytoplasmic tyrosine kinases, e.g., Syk. Following docking at ITAM, kinase undergoes conformational changes, resulting in autophosphorylation of the Syk catalytic domain. Such an event increases Syk enzymatic activity and leads to propagation of downstream signaling [64,65]. We used specific anti-Syk mouse (pY342) antibodies to confirm that conserved tyrosine Y-342 in the Syk SH-2 linker region was phosphorylated following TLR3 activation (Figure 2(a)). Mutation of this particular tyrosine residue alone was found to diminish Syk ability to interact with other proteins, while a concomitant mutation in Y-346 induced a significant reduction in FcεRI-mediated signaling [66]. Lin et al. [52] confirmed that stimulation with poly(I:C) led to the activation of Syk in bone marrow-derived macrophages (BMDM) and RAW.264.7 macrophages; however, phosphorylation of the kinase was detected 15 min after stimulation of the cells. We have shown that Syk undergoes rapid phosphorylation and associates with Hrs after poly(I:C) stimulation of astrocytes (Figures 2(a) and 2(b)); however, it cannot be precluded that such interaction occurs with the participation of other proteins. Further, studies are necessary to confirm whether STAM may associate with the phosphorylated SH-2 domain of the Syk via its ITAM region, although the ITAM-independent pathway may underlay Syk-Hrs interaction.
The possibilities where Syk may affect TLR responses are manifold; in this work, we addressed the ways in which Syk could influence TLR3 signaling and cytokine responses in cells of brain origin. Syk may play reverse roles in mediating TLR-dependent responses by opposingly regulating the ubiquitination of TRAF3 or TRAF6 [60]. Furthermore, Syk might inhibit MyD88-dependent production of proinflammatory cytokines and augment the TRIF-dependent expression of IFN-dependent genes. Despite the fact that Syk-mediated TRIF phosphorylation leads to TRIF proteasomal degradation resulting in downregulation of TLR signaling [38], phosphorylation of this adaptor protein is critical for the activation of the type I IFN pathway [67]. Thus, the observed restriction of NF-κB nuclear localization in the absence of Syk (Figure 3(c)) appears to correspond with insufficient activatory influence on the TRIF phosphorylation [67]. Most recently, preclinical and clinical studies highlight pharmacological inhibitors of Syk as promising drug targets, due to their inhibitory influence on the inflammatory responses [52,68,69]. Nevertheless, some studies present contradictory findings-Syk deficient cells exhibit higher proinflammatory response than do wild-type cells [70,71], and such a relationship has been pointed out for Syk-dependent inhibition of TLR signaling [38]. Our data distinguish Syk as the balancing component in TLR3-mediated immune response, intended to avoid unstrained production of inflammatory factors.
Importantly, recent reports explore and confirm the possible role of HSV-1 infection in the pathogenesis of the most common form of dementia-AD [78,79]. Therapies targeting glial cells might benefit the cells affected by neurodegenerative disorders. There is good support for the hypothesis that Aβ secreted by cells may constitute an antimicrobial protein (AMP) and astrocytes may produce it as an essential defensive component of the innate immunity [80]. Furthermore, inflammatory agents that appear during both acute and chronic brain infections may upregulate amyloid precursor protein levels in both astrocytes of murine and human brain [81]. During H3N2 and H1N1 influenza A viruses (IAV) or HSV-1 infections, Aβ may play a protective role against these pathogens and constitute a host response to infection, e.g., reduce virus replication in neurons or prevent viral entry into the cells [82,83]. On the other hand, HSE, and particularly recurrent or persistent HSV-1 infections in the brain, may be the determining factors that increase the risk of AD development. The extent to which HSV-1 infection may contribute to the deposition of Aβ in the brain was analyzed by Wozniak et al. [9]. In brains of people suffering from AD, HSV-1 nucleic acid was found in 90% of Aβ plaques, while over 70% of viral DNA was associated with the plaques. These data indicate HSV-1's presence in the brain as one of the initiating factors in the formation of Aβ plaques in the brain, as well as an important factor that may lead to the onset of AD. Knowledge regarding TLR3 biology in brain cells is significant since the receptor is crucial in combating HSV-1 infection. Results obtained in this work contribute to the understanding of TLR3 functioning in astrocytes.
A functional TLR3 reveals as an essential component of natural immunity to HSV-1 in the brain, while impaired innate immunity to HSV-1 may increase susceptibility to HSE in children and adults. Here, we point out Syk and Hrs as immune factors which influence TLR3 signaling that may affect inflammatory-mediated encephalitic responses during HSE. Details regarding how Syk and Hrs influence TLR3-mediated antiviral response call attention to novel elements which may require careful examination when analyzing the TLR3 activity in CNS, such as the role of posttranslational modifications of these proteins, or contribution of TLR3 N-terminal form in mounting the effective antiviral defense. Precise regulation of the TLR3 transportation and degradation, most likely related to Syk and Hrs, is essential for maintaining the adequate level of an active receptor and generating an effective immune response.

Conclusions
Endosomal TLR3 undergoes cleavage upon poly(I:C) stimulation of murine astrocytes in a dose-and time-dependent manner. Stimulation of murine astrocytes with poly(I:C) upregulates the expression of Syk and Hrs in a timedependent manner and additionally in a dose-dependent manner for Syk, while the expression of STAM is not affected. Distribution of TLR3 and STAM is altered, from a perinuclear location in nonstimulated cells to a much dispersed arrangement upon poly(I:C) stimulation of astrocytes. The increased expression of Syk appears to orchestrate its activation and eventual interaction with Hrs followed by tyrosine phosphorylation of Hrs, which in turn interacts with the N-terminal form of TLR3. Knockdown of TLR3, Syk, or Hrs followed by TLR3 stimulation of astrocytes leads to perturbations in nuclear translocation of NF-κB and IRF3, while IRF7 is not influenced. Moreover, Syk and Hrs knockdown results in the increase of IFNβ, IL-6, and CXCL8 secretion. These results suggest that Syk and Hrs have a regulatory role in signaling through TLR3 in murine astrocytes.

Data Availability
All data used to support the findings of this study are included within the article.

Conflicts of Interest
The authors declare that there is no conflict of interest regarding the publication of this paper.