Negative Regulation of RIG-I by Tim-3 Promotes H1N1 Infection

ABSTRACT The mechanisms by which retinoic acid-inducible gene I (RIG-I), a critical RNA virus sensor, is regulated in many biological and pathological processes remain to be determined. Here, we demonstrate that T cell immunoglobulin and mucin protein-3 (Tim-3), an immune checkpoint inhibitor, mediates infection tolerance by suppressing RIG-I-type I interferon pathway. Overexpression or blockade of Tim-3 affects type I interferon expression, virus replication, and tissue damage in mice following H1N1 infection. Tim-3 signaling decreases RIG-I transcription via STAT1 in macrophages and promotes the proteasomal dependent degradation of RIG-I by enhancing K-48-linked ubiquitination via the E3 ligase RNF-122. Silencing RIG-I reversed Tim-3 blockage-mediated upregulation of type I interferon in macrophages. We thus identified a new mechanism through which Tim-3 mediates the immune evasion of H1N1, which may have clinical implications for the treatment of viral diseases.


Introduction
Influenza virus infections kill approximately half a million people each year globally (Iuliano et al. 2018). The innate immune system is essential for controlling viral infections (Liu et al. 2018;Takeuchi and Akira 2009). Following an infection, cytoplasmic viral RNA is recognized by retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs), including RIG-I and MDA5 (Kato et al. 2006). RIG-I comprises N-terminal caspase activation and recruitment domains (CARDs), a central RNA helicase domain that is essential for triggering the signal that induces type I interferon (IFN), and a C-terminal repressor domain (RD) (Saito et al. 2007). Upon binding of pathogenic RNA to the helicase domain, the activated RIG-I undergoes a conformational change and is recruited to the mitochondrial antiviral signaling proteins (MAVS) adaptor, which forms a multilayered protein complex (Seth et al. 2005). The MAVS complex then catalyzes the interaction of the serine/threonine-protein kinase 1 (TBK1) and inhibitor of nuclear factor kapa-B kinase subunit epsilon (IKKε), which phosphorylate and activate the transcription factors interferon regulatory factor (IRF) 3 and NF-κB, and stimulate the expression of type I IFNs and proinflammatory cytokines for host defense (Fitzgerald et al. 2003;Loo and Gale 2011;Tenoever et al. 2007). The signaling activity of RIG-I is regulated by multiple mechanisms, among which ubiquitination modifications have been shown to play an important role. For example, the E3 ubiquitin ligase tripartite motif 25 (TRIM25) mediates K63-linked polyubiquitination of RIG-I, which promotes RIG-I-dependent antiviral signaling (Gack et al. 2007(Gack et al. , 2008. Whereas the E3 ubiquitin ligases c-Cbl and CHIP mediate K48-linked ubiquitination of RIG-I, which triggers RIG-I degradation and blunts RIG-I signaling Zhao et al. 2016). Therefore, distinct ubiquitin modifications of RIG-I modulate the antiviral immune response ). However, influenza viruses have evolved strategies that regulate RIG-I ubiquitination to evade host innate immune responses for survival Gack et al. 2009;Rajsbaum et al. 2012). Investigation of the mechanism by which RIG-I is regulated under different physio-pathological conditions will be valuable.
T cell immunoglobulin and mucin protein-3 (Tim-3) is an immune checkpoint inhibitor that induces T cell tolerance by binding to its ligands. After its discovery, Tim-3 was also found to be expressed in innate immune cells, including macrophages and dendritic cells Hou et al. 2014;Ju et al. 2010;Li et al. 2019;Sui et al. 2006). The association between the dysregulated upregulation of Tim-3 in immune cells and immune exhaustion made Tim-3 a therapeutic target following PD-1 and CTLA-4 for the therapy of immune disorders such as tumors and infectious diseases (Das et al. 2017). Structurally, Tim-3 is a type I membrane glycoprotein with a relatively short intracellular tail that lacks inhibitory motifs (Lee et al. 2011). So far, four molecules including Galectin-9, CEACAM1, phosphatidylserine, and HMGB1 have been reported to function as interaction partners of Tim-3. However, whether these interaction partners act as ligands of Tim-3 remain controversial (Chiba et al. 2012;De Sousa Linhares et al. 2020;Leitner et al. 2013). Compared to the relatively clear mechanisms by which Tim-3 induces tolerance in T cells, very little is known about Tim-3 signaling in innate immune cells. We previously demonstrated that Tim-3 inhibits NF-κB activation in macrophages in response to LPS stimulation (Yang et al. 2013) and that Tim-3 promotes M2 macrophage polarization by inhibiting the activity of STAT1 . These data suggest that Tim-3 also induces macrophage tolerance. Macrophages play critical roles in the first line of defense against virus infections. Whether Tim-3 promotes virus evasion by inducing macrophage tolerance needs to be investigated.
Here, for the first time, we identified a direct interaction between Tim-3 and RIG-I in macrophages and the molecular mechanisms by which Tim-3 promotes RIG-I degradation and bluntness. Thus, we provide a new mechanism by which Tim-3 induces virus tolerance in vivo, which may have therapeutic implications.

Mice
All mice used in this study were bred under specific pathogen-free (SPF) conditions in our facilities. Experiments were performed in accordance with the national and institutional guidelines for animal care and approved by the Animal Ethics Committee of the Beijing Institute of Basic Medical Sciences, Beijing, China. Wild-type (WT) C57BL/6 mice were obtained from Beijing Vital River Laboratory Animal Technology Co. Ltd., China. C57BL/6 Tim-3 transgenic (Tim-3-TG) mice with high expression of Tim-3 ( Figure S1) were established in our laboratory as described previously (Zhao et al. 2014). For infection studies, 3-4-week-old mice were inoculated nasally with 1 × 10 6 PFU/kg H1N1 virus soluted in 20 uL PBS buffer (Gibco, USA). The control group was treated with the same volume of PBS. To investigate the role of Tim-3 in H1N1 infection, WT C57BL/6 mice were injected intraperitoneally with 10 mg/kg of soluble Tim-3 immune globulin (sTim-3-Ig) or its control Ig protein diluted in 200 µL PBS after infection with H1N1. Ninety-six hours post-infection, mice were sacrificed, and the lungs were collected and processed. The investigator was blinded when processing and assessing the outcome by giving a unique number to each animal, which was independent of genotype. Randomization to experimental groups was performed for these studies using a sample size of 5 mice per group.

Cell culture
The mouse RAW264.7 macrophage cell line, human U937 monocyte/macrophage cell line, and HEK-293T cells were obtained from the American Type Culture Collection (Manassas, VA, USA). RAW264.7 and HEK-293T cells were maintained in DMEM medium supplemented with 10% fetal bovine serum (FBS; Gibco) and penicillin-streptomycin (Invitrogen). U937 cells were maintained in RPMI-1640 medium supplemented with 10% FBS and penicillin-streptomycin. Mouse peritoneal macrophages were prepared as described previously .

Plasmids
Full-length human Tim-3, Tim-3-Y/F (Tim-3 intracellular domain 265/272 tyrosine was substituted by phenylalanine), Tim-3-ΔIC (Tim-3 intracellular domain was deleted) plasmids, human STAT1 plasmid, V5-tagged human RING finger protein 122 (RNF122) plasmids were constructed in our laboratory. The FLAG-tagged human RIG-I plasmid was kindly provided by Professor Danying Chen from Pecking University. The HA-tagged K48 Ub plasmid was kindly provided by Professor Yue Xiong from North Carolina University.

Dual-Luciferase reporter assay
The RIG-I luciferase promoter reporter was employed in HEK-293T cells (2 × 10 5 ) cultured in 0.5 mL antibiotic-free medium and transfected a day later with a mixture of 2 μg RIG-I luciferase promoter reporter (I-luc), STAT1 (control vector), Tim-3 (control vector), and 0.02 μg Renilla luciferase control (R-luc) using Lipofectamine 2000. Luciferase activity was quantified 72 h later using the dual-luciferase reporter assay system from Promega. Luminescence was measured using a microplate reader (Wellesley).

Quantitative reverse transcription-PCR (RT-qPCR)
Total RNA from mouse tissues (lungs) and macrophages (RAW264.7 cells) were extracted using TRIzol (Invitrogen), treated with DNase I (Qiagen), and purified using the RNeasy Mini kit (Qiagen). To measure IFN-α4, IFN-β1, Tim-3, and RIG-I expression, cDNA was prepared by following the instructions in the TransScript First-Strand cDNA Synthesis SuperMix kit (TransGen Biotech). RT-qPCR was carried out on a Bio-Rad iQ5 real-time PCR system with UltraSYBR Mixture (CWBIO). The primers used in this study are shown in Supplementary Table 1. Relative expression was determined using the comparative Ct (ΔΔCt) method.

Virus replication assay
All infection work was carried out according to the requirements for handling biological agents of the Advisory Committee on Dangerous Pathogens hazard groups. A/Beijing/501/ 2009 H1N1 (BJ501) virus was provided by Professor Deyan Luo (Beijing Institute of Basic Medical Sciences, China). 2 × 10 5 RAW264.7 cells were infected with H1N1 at an MOI of 0.1 PFU per cell for 24 h in the presence or absence of sTim-3-Ig. Then, the cells were analyzed to detect H1N1 viral replication using RT-qPCR.

Lung histology
Lungs from control and virus-infected mice were dissected, fixed in 4% phosphate-buffered formalin, embedded in paraffin, sectioned, stained with hematoxylin and eosin (H&E) staining solution, and analyzed under light microscopy for histological changes.

Microarray analysis
Differentially expressed mRNA in control RAW264.7 cells or RAW264.7 cells with stable knockdown of Tim-3 were profiled using Affymetrix mRNA microarray chips following the manufacturer's protocol. Briefly, collected samples was subjected to Gene Chip mRNA array analysis using a Gene Chip Scanner 3000 with Gene Chip Operating Software (GCOS) and analyzed. Candidate mRNA was further identified by qPCR.

Western blot analysis
Cells were lysed in lysis buffer (1% Triton X-100, 20 mM Tris-HCl pH 8.0, 250 mM NaCl, 3 mM EDTA pH 8.0), 3 mM EGTA (pH 8.0) with the pH adjusted to 7.6, and complete protease inhibitor cocktail (Roche, pH 7.5). Lysates were incubated on ice for 30 min and centrifuged at 12,000 × g for 15 min to remove cell debris, and then eluted by boiling for 10 min with 5× sample buffer (100 mM Tris-HCl pH 6.8, 2% SDS, 10% glycerol, 0.1% bromophenol blue, 1% β-mercaptoethanol). Protein samples were electrophoretically separated on 10% SDS-PAGE and transferred to a PVDF membrane (Millipore, IPVH00010). The membranes were blocked in TBST containing 5% fat-free dry milk for 1 h at room temperature and incubated with primary antibodies, including rabbit antibodies against Tim-3, GAPDH, β-actin, β-Tubulin, IFN-α4, IFN-β1 and HA-tag and a mouse antibody against human RIG-I, overnight at 4°C. After washing with TBST for 30 min, the membranes were incubated with anti-rabbit IgG or anti-mouse IgG for 1 h at room temperature. The bound antibodies were visualized using ECL kits (Amersham Biosciences, UK).

Statistical analysis
Statistical analyses were performed using unpaired, two-tailed, Student's t-tests, one-way ANOVAs, or two-way ANOVAs. P-values of less than 0.05 were considered statistically significant.

Tim-3 signaling inhibits immune response against H1N1 infection in vivo
To test whether Tim-3 is involved in the innate immunity against influenza virus, we challenged macrophages with H1N1, an RNA virus widely used for investigating antiviral immunity in both mouse and human models. Shortly after H1N1 challenge, the expression of Tim-3 was upregulated in RAW264.7 cells and BV2 cells (Figures 1a,b). To further investigate the role of Tim-3 in the host immune response against influenza virus, we challenged WT or Tim-3-TG mice with H1N1 virus via nasal infection, then examined viral replication using RT-qPCR, and detected lung damage using H&E staining. Figures 1c, d shows that 96 h after H1N1 infection, Tim-3-TG mice exhibited higher viral replication and more infiltration of inflammatory cells in the lungs than WT mice, suggesting that Tim-3 signaling attenuates the antiviral innate immune response in vivo. As Tim-3 signaling could be blocked by sTim-3-Ig , we further challenged the mice with H1N1 virus in the presence or absence of sTim-3-Ig. Figures 1e,f shows that lower virus replication and less infiltration of inflammatory cells were observed in the lungs of sTim-3-Ig-treated mice than in those of control Ig protein-treated mice. Innate immune cells play a critical role in the fight against viruses within a 96 h period. Thus, these data demonstrate that Tim-3 signaling is involved in the innate immune response against H1N1 infection and functions as a negative immune regulator.

Tim-3 signaling inhibits type I IFN in H1N1-infected mice and in macrophages in vitro
Type I IFN plays critical roles in the immune response against virus infection. To test whether it is involved in Tim-3-induced infection tolerance, we examined the expression of the type I IFNs IFN-α4 and IFN-β1 in WT and Tim-3-TG mice following H1N1 infection. IFN-α4 and IFN-β1 expression in the infected alveolar macrophages of Tim-3-TG mice was significantly lower than that in WT mice (Figures 2a,b). We also detected the expression of these type I IFNs in H1N1-infected mice with or without Tim-3 blockade. The expression of IFN-α4 and IFN-β1 in alveolar macrophages was significantly upregulated in sTim-3-Igtreated mice compared to Ig-treated controls (Figures 2c,d). These results show that Tim-3 inhibits type I IFN production in H1N1-infected mice.
As macrophages play a critical role in the innate immune response against viruses, we infected the macrophage cell-line RAW264.7 with H1N1 in the presence of sTim-3-Ig or control Ig protein and then examined type I IFN expression and virus replication.
Figures 2e,f shows that blockade of Tim-3 led to decreased virus replication and increased IFN-α4 and IFN-β1 expression in macrophages compared to the Ig-treated control group. These data demonstrate that Tim-3 may inhibit the antiviral innate immune response by suppressing the type I IFN pathway.

Tim-3 signaling inhibits type I IFN expression through RIG-I
To investigate the mechanisms by which Tim-3 inhibits type I IFN, we knocked down Tim-3 in macrophages ( Figure 3a) and then analyzed gene expression profiles related to type I IFNs. According to the microarray data, Tim-3 inhibited RIG-I expression ( Figure 3b). As hours later, mice were sacrificed. (c) Lung tissues were collected, and H1N1 virus replication in the lungs was quantified using qPCR. (d) Tissue damage in the lungs was examined using hematoxylin and eosin (H&E) staining. (e and f) WT C57BL/6 mice were infected with 1 × 10 6 PFU/kg H1N1 virus and injected with 10 mg/kg soluble Tim-3 immune globulin (sTim-3-Ig) or its control Ig protein. Ninety-six hours later, mice were sacrificed. (e) Lung tissues were collected, and H1N1 virus replication in the lungs was quantified using qPCR. (f) Tissue damage in the lung was examined using H&E staining. The data are representative of three independent experiments using 5 mice in each group. The results are expressed as the mean ± SD; ** P < 0.01; *** P < 0.001. RIG-I is a cytoplasmic viral RNA sensor that triggers the signal to induce type I IFN production in response to viral infection  and an upstream molecule of the type I IFN pathway (Cao et al. 2015;Luo et al. 2008), we hypothesized that Tim-3 may inhibit type I IFN by regulating RIG-I. To test this hypothesis, we silenced RIG-I in macrophages and examined type I IFN expression with or without Tim-3 blockade. When RIG-I was silenced (Figure 3c), Tim-3 blockade did not increase IFN-α4 and IFN- (a-b) WT C57BL/6 mice and Tim-3-TG mice were challenged with 1 × 10 6 PFU/kg H1N1 virus or administered PBS as a control for 96 h. Then, mice were sacrificed (5 mice per group). (a) the expression of type I interferons (IFNs; IFN-α4 and IFN-β1) in alveolar macrophage cells was quantified using qPCR. (b) the expression of type I interferons in alveolar macrophage cells was examined by western blot. (c-d) WT C57BL/6 mice were infected with 1 × 10 6 PFU/kg H1N1 virus and injected with 10 mg/kg sTim-3-Ig or Ig protein. Ninety-six hours later, mice were sacrificed (5 mice per group). (c) the expression of type I interferons (IFNs; IFN-α4 and IFN-β1) in alveolar macrophage cells was quantified using qPCR. (d) the expression of type I interferons in alveolar macrophage cells was examined by western blot. (e) RAW264.7 cells were infected with H1N1 virus for 24 h, and then H1N1 viral replication and type I IFNs expression were examined using qPCR. (f) RAW264.7 cells were infected with H1N1 virus for 24 h, and the expression of type I IFNs was examined by western blot. In A, C and E, the results are expressed as the mean ± SD and are representative of three independent experiments. ** P < 0.01; *** P < 0.001.  (Figure 3d). These results indicate that Tim-3 may inhibit type I IFN expression through RIG-I.

Tim-3 signaling inhibits RIG-I expression through STAT1
The above microarray data showing that Tim-3 inhibits RIG-I expression were further confirmed using qPCR and western blot. The results showed that knockdown of Tim-3 (Figure 4a) or blockade of Tim-3 signaling (Figure 4b) in RAW264.7 cells led to increased RIG-I expression compared to that in control groups, whereas overexpression of Tim-3 in macrophages of Tim-3-TG mice (Figure 4c) led to decreased RIG-I expression compared to that in WT mice. These data further demonstrate that Tim-3 can inhibit RIG-I expression in macrophages.
We then examined the mechanism by which Tim-3 inhibits RIG-I expression. As we previously reported the mechanisms by which Tim-3 inhibits STAT1 activity Li et al. 2019), and STAT1 is a transcription factor of RIG-I (Elion et al. 2018;Wu et al. 2020), we hypothesized that Tim-3 may inhibit RIG-I expression through STAT1. We tested this assumption using a dual-luciferase reporter system. Transfection of STAT1 mediated the upregulation of RIG-I, which was reversed by co-transfection of Tim-3 ( Figure 4d). In addition, Tim-3 knockdown-induced upregulation of RIG-I was reversed in the presence of a STAT1 inhibitor (Figure 4e). These data suggest that Tim-3 inhibits RIG-I expression through STAT1, which leads to suppressed expression of type I IFNs.

Tim-3 interacts with RIG-I and induces proteasomal degradation of RIG-I through E3 ligase RNF-122
Given the role of RIG-I in the Tim-3-mediated antiviral immune response, we examined whether Tim-3 interacts with RIG-I. A co-immunoprecipitation assay showed that there was an interaction between exogenous Tim-3 and RIG-I in RAW264.7 cells both in the presence or absence of virus stimulation (Figure 5a). To provide more direct evidence that Tim-3 interacts with RIG-I, we labeled Tim-3 and RIG-I in RAW264.7 cells with indicated antibodies, and then detected the localization of Tim-3 and RIG-I using a confocal laser scanning microscope. Figure 5b shows that there was co-localization of Tim-3 and RIG-I. These data demonstrate co-localization and a direct interaction between Tim-3 and RIG-I.
RIG-I activity is tightly regulated at different levels besides the transcriptional level. Here, we found that Tim-3 is also involved in the post-translational regulation of RIG-I, as the data in Figure 5c show that transfection of Tim-3 decreased the half-life of endogenous RIG-I protein within 6 h in the presence of cycloheximide (CHX), a ribosome inhibitor that   a, b, and c) Expression of RIG-I was measured using qPCR and western blot in RAW264.7 cells or RAW264.7 cells with stable knockdown of blocks protein synthesis in ribosomes. Furthermore, we found that the decrease in RIG-I protein could be reversed by MG132, a proteasome inhibitor that blocks proteasomedependent protein degradation (Figure 5d). Collectively, these results suggest that Tim-3 may modulate RIG-I post-translational regulation and promote RIG-I degradation in a proteasome-dependent manner.
Post-translational regulation, particularly ubiquitination, plays a critical role in the regulation of RIG-I activity . Reports have shown that K48-linked ubiquitination of RIG-I leads to RIG-I degradation by the proteasome Oshiumi et al. 2012). To test whether Tim-3 regulates RIG-I ubiquitination, we transfected HEK-293T cells with Flag-tagged RIG-I and HA-tagged K48-Ub along with varying doses of plasmids expressing Tim-3 (Tim-3-WT) or plasmids expressing Y265F/Y272F of Tim-3 (Tim-3-Y/F), two residues that are indispensable for Tim-3 signaling (Tomkowicz et al. 2015), or Tim-3 lacking an intracellular tail (Tim-3-ΔIC). Our results show that Tim-3 signaling promoted RIG-I K48linked ubiquitination in a dose-dependent manner (Figure 5e), which is responsible for RIG-I degradation. However, when Tim-3 265/272 tyrosine residues of the intracellular domain were replaced with phenylalanine or the intracellular domain was deleted, Tim-3-enhanced ubiquitination of RIG-I was significantly reversed (Figure 5e). Moreover, the Y/F mutants of Tim-3 (Y265F/Y272F) weakened the binding of Tim-3 with RIG-I, while deletion of the intracellular domain of Tim-3 (ΔIC) resulted in the loss of binding activity of Tim-3 with RIG-I. These data suggest that Tim-3 catalyzes K48-linked ubiquitination to promote the degradation of RIG-I, and that Tim-3 interacts with RIG-I through its intracellular domain, in which Y265/Y272 play an important role.
The mechanisms by which Tim-3 promotes RIG-I ubiquitination were then investigated. The E3 ubiquitin ligase RNF122 has been reported to be involved in regulating RIG-I signaling . We speculated that RNF122 may be involved in Tim-3-promoted ubiquitination of RIG-I. To test this hypothesis, HEK-293T cells were transfected with plasmids encoding Flag-RIG-I, HA-K48-Ub, and V5-RNF122 and varying doses of plasmids encoding Tim-3, followed by a co-immunoprecipitation assay. The results show that RNF122 can mediate K48-linked ubiquitination of RIG-I (Figure 5f). In addition, Tim-3 promoted RNF122-mediated K48-linked ubiquitination of RIG-I in a dose-dependent manner (Figure 5f). Collectively, these data demonstrate that Tim-3 interacts with RIG-I and induces its proteasomal degradation through the E3 ligase RNF-122.

Discussion
Innate immune responses have attracted much attention recently in the treatment of immune disorders, including infectious diseases. The RLR RIG-I is a cytoplasmic RNA Tim-3 (a), in RAW264.7 cells treated with 10 μg/mL sTim-3-Ig or Ig for 24 h (b), and in peritoneal macrophages isolated from WT mice or Tim-3-TG mice (5 mice per group) (c). (d) HEK-293T cells were transfected with pGL3-RIG-I-promoter and Renilla luciferase plasmids, with or without plasmids encoding STAT1 or Tim-3 and their corresponding control vectors. Forty-eight hours later, luminescence was detected. (e) RAW264.7 cells with or without Tim-3 knockdown were treated with a STAT1 inhibitor (STAT1-inh) or DMSO. Then, RIG-I mRNA levels were measured using qPCR. The data are expressed as the mean ± SD and are representative of three independent experiments. * P < 0.05; *** P < 0.001. sensor that detects viral RNA and triggers the signal to induce type I IFN and proinflammatory cytokine production during viral infection. As an important trigger of antiviral innate immunity, RIG-I should be tightly regulated to avoid autoimmune damage mediated by an excessive immune response. The mechanisms by which RIG-I is regulated under different physio-pathological conditions is of great interest to be determined. In this study, we found a new mechanism by which Tim-3 induces viral immune evasion: by suppressing RIG-I transcription and promoting its ubiquitination and degradation, Tim-3 inhibits type I IFN production and antiviral activity. Our findings confirm that Tim-3 may also contribute to infection tolerance by regulating the activity of innate immunity.
Tim-3 is an immune checkpoint inhibitor that was initially identified on activated T cells, including Th1, Tc1, and Th17 cells (Kanai et al. 2012;Ocana-Guzman et al. 2016). By binding its ligand Gal-9, Tim-3 induces T cell tolerance or exhaustion, thus negatively regulating the immune responses of these T cells (Sanchez-Fueyo et al. 2003). Recent research reports, including ours, have shown that Tim-3 is also a negative regulator of innate immune cells Yang et al. 2013), such as dendritic cells and macrophages. Our research group has been working on the mechanisms by which Tim-3 negatively regulates the immune response. So far, we have reported several pathways for Tim-3 inhibition of innate immunity. For example, by suppressing NF-κB signaling, Tim-3 inhibits Toll-like receptor (TLR) 4-mediated activation of macrophages (Yang et al. 2013). By binding to and inhibiting the nuclear translocation of STAT1, Tim-3 promotes tumorpromoting M2 macrophage polarization . Furthermore, our studies have demonstrated that Tim-3 inhibits macrophage killing and phagocytosis of Listeria monocytogenes (Wang et al. 2017) and promotes the immune evasion of Listeria monocytogenes at the stage of innate immunity (Wang et al. 2020). Collectively, these results confirm that Tim-3 is an important innate immune regulator. However, the precise regulatory mechanisms of Tim-3 signaling in antiviral innate immunity remain unclear. Here, we found a new mechanism by which Tim-3 inhibits the innate immune response against viral infection. Tim-3 promotes virus invasion, increases virus replication, and exacerbates tissue damage, and immunoprecipitated with Tim-3 antibody, followed by western blot analysis using the indicated antibodies. (b) RAW264.7 cells were labeled with Tim-3 and RIG-I antibodies, and then the cells were immunostained with the indicated secondary antibodies (Alexa Fluor® 488 and Alexa Fluor® 594) and analyzed under fluorescence microscopy. (c) HEK-293T cells transfected with Tim-3 or control vectors were treated with or without cycloheximide (CHX) for 6 h. Expression of RIG-I was analyzed using immunoblot with the indicated antibodies (c, upper). The value of the RIG-I/GAPDH protein ratio was measured by using ImageJ (c, lower). (d) HEK-293T cells transfected with Tim-3 or control vectors were treated with or without MG132 for 6 h in the presence of CHX. Expression of RIG-I was analyzed using immunoblotting with the indicated antibodies (d, upper). The value of the RIG-I/GAPDH protein ratio was measured by using ImageJ (d, lower). (e) HEK-293T cells were transfected with Flag-RIG-I, HA-K48-UB plasmids, and varying doses of plasmids encoding Tim-3 (WT: 0, 0.5 and 1 μg) or plasmids encoding Tim-3-Y/F mutation (Tim-3 265/272 tyrosine residues of intracellular domain were replaced with phenylalanine) or Tim-3-ΔIC (Tim-3 intracellular domain was deleted) for 24 h. Ubiquitination of RIG-I was analyzed using immunoprecipitation with Flag antibody, followed by western blot analysis using the indicated antibodies. (f) HEK-293T cells were transfected with Flag-RIG-I, HA-K48-Ub, V5-RNF122, and varying doses of plasmids encoding Tim-3 (0, 0.5 and 1 μg) for 24 h. Ubiquitination of RIG-I was analyzed using immunoprecipitation with Flag antibody, followed by western blot analysis using the indicated antibodies. The data are representative of three independent experiments.
suggesting that it regulates the antiviral immune response and functions as a negative immune regulator. Furthermore, we found that Tim-3 inhibits type I IFN expression both in H1N1-infected mice and in macrophages in vitro, indicating that Tim-3 may inhibit the antiviral innate immune response by suppressing the type I IFN pathway. Interestingly, from the RNA microarray data, we found that Tim-3 can inhibit the expression of RIG-I, which is a cytoplasmic sensor that triggers the induction of type I IFN production. We then silenced RIG-I and tested the type I IFN expression in macrophages with or without Tim-3 blockade. The results show that RIG-I silencing reduces the upregulation of IFNs induced by Tim-3 blockade, suggesting that Tim-3 inhibits type I IFN expression through RIG-I. Thus, we demonstrated a new mechanism by which Tim-3 inhibits antiviral innate immunity and provided direct evidence of how Tim-3 regulates the immune response indirectly through suppressing the activity of RIG-I.
Innate immunity provides the first line of host defense against viral infection. Pathogen recognition receptors (PRRs), which recognize a variety of viral components such as proteins, lipids, and viral nucleic acids, are the main sensors of invading viruses. PRRs include TLRs, RLRs, Nod-like receptors, and DNA sensors (Song et al. 2017). Since RIG-I is not only a key sensor in the antiviral immune response but also an important interference target for virus immune evasion, the mechanism of RIG-I inhibition and its potential application in antiviral response need to be widely investigated. Here, we found a novel negative regulation mechanism of RIG-I, which could be employed by the H1N1 virus to evade the immune response. We previously demonstrated that Tim-3 inhibits the activity of STAT1, which is a transcription factor of RIG-I. In this study, we demonstrate that STAT1mediated RIG-I transcription can be inhibited by Tim-3. In addition, Tim-3 silencing induced upregulation of RIG-I, which could be reversed by STAT1 inhibition. There are four potential ligands of Tim-3, among which Galectin-9 is widely expressed on immune cells including on the RAW264.7 cells ). So we argue that Galectin-9 is the potential ligand of Tim-3 which stimulates the Tim-3 signaling to inhibit RIG-I expression.
In summary, we first demonstrate that Tim-3 inhibits RIG-I expression at the transcriptional level in a STAT1-dependent manner, which may to some extent reveal the mechanism of Tim-3 intervention in the type I IFN pathway.
Compared to regulation at the transcription level, the post-translational modification of RIG-I plays a more important role in immunity against virus as it provides faster control. Ubiquitination is an important post-translational modification involved in various cellular functions (Jiang and Chen 2011). E3 ubiquitin ligases have been reported to play critical roles in the regulation of RIG-I activity (Kawai and Akira 2011;Medvedev et al. 2015). The E3 ubiquitin ligase tripartite motif containing protein 25 (TRIM25), Riplet (also known as RNF135), and MEX3C deliver the K63-linked polyubiquitin moiety to RIG-I CARDs and the C-terminal domain, thus positively regulating RIG-I-mediated signaling (Gack et al. 2007;Kuniyoshi et al. 2014;Oshiumi et al. 2010). However, RING E3 ligase RNF122 and RNF125 mediate K48-linked ubiquitination of RIG-I, leading to RIG-I degradation by proteasomes, thus negatively regulating RIG-I-mediated signaling (Arimoto et al. 2007;Wang et al. 2016). Here, we found a direct interaction and co-localization between Tim-3 and RIG-I, which inspired us to investigate the mechanisms by which Tim-3 regulates RIG-I. Furthermore, we also found that Tim-3 promotes RIG-I degradation through the ubiquitin-proteasome system. RNF122, an E3 ubiquitin ligase involved in proteasome-mediated degradation of proteins, plays a critical role in the K48-linked ubiquitination of RIG-I enhanced by Tim-3. These findings demonstrate that Tim-3 interacts with RIG-I and induces proteasomal degradation of RIG-I through RNF-122.
In summary, we identified a novel mechanism by which Tim-3 induces immune tolerance in macrophages and promotes virus evasion. Tim-3 interacts with RIG-I, inhibits RIG-I expression through STAT1, promotes RIG-I ubiquitination and degradation through the E3 ligase RNF-122, and subsequently inhibits type I IFN production and antiviral activity. To the best of our knowledge, this is the first report demonstrating that Tim-3 inhibits antiviral innate immunity by suppressing the activity of RIG-I. Our data shed new light on how Tim-3 inhibits the immune response to induce virus tolerance, which may have therapeutic implications.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This study was supported by the National Natural Science Foundation of China (81771684 and 81971473); Beijing Natural Science Foundation (7192145); and Science and technology innovation 2030 Major projects s _ ṣ)2021ZD0201600).