Pseudorabies Virus Tegument Protein UL13 Suppresses RLR-Mediated Antiviral Innate Immunity through Regulating Receptor Transcription

Pseudorabies virus (PRV) has evolved various strategies to escape host antiviral immune responses. However, it remains unclear whether and how PRV-encoded proteins modulate the RIG-I-like receptor (RLR)-mediated signals for immune evasion. Here, we show that the PRV tegument protein UL13 functions as an antagonist of RLR-mediated antiviral responses via suppression of the transcription of RIG-I and MDA5, but not LGP2. UL13 overexpression significantly inhibits both the mRNA and protein levels of RIG-I and MDA5, along with RIG-I- or MDA5-mediated antiviral immune responses, whereas overexpression of RIG-I or MDA5 counteracts such UL13-induced suppression. Mechanistically, UL13 suppresses the expression of RIG-I and MDA5 by inhibiting activation of the transcription factor NF-κB. Consequently, overexpression of p65 promotes the activation of RIG-I and MDA5 promoters. Moreover, deletion of the p65-binding sites in the promoters of RIG-I or MDA5 abolishes the suppression role of UL13. As a result, mutant PRV lacking UL13 elicits stronger host antiviral immune responses than PRV-WT. Hence, our results provide a novel functional role of UL13-induced suppression of host antiviral immunity through modulating receptors’ transcription.


Introduction
Pseudorabies virus (PRV), a member of the Alphaherpesvirus subfamily, is the pathogen of Aujeszky's disease, which causes abortions and stillbirths in sows, central nervous system disorders in young piglets, and respiratory disease in older pigs [1][2][3], generating considerable economic losses worldwide in the swine industry worldwide [4]. Recent studies report that PRV can also infect humans, thus raising great concern about crossspecies PRV transmission [5,6]. Similar to other alphaherpesviruses, PRV has evolved multiple strategies to dismantle the host's innate antiviral response [7][8][9][10], such as blocking pattern-recognition receptor (PRR)-induced type-I interferon (IFN-I) and neutralizing the antiviral functions of IFN-stimulated genes (ISGs), resulting in the establishment of lifelong latent infection in the host. Nevertheless, viruses may also be released during latent infection, which causes persistent infection in the host, and poses potential risks to the breeding industry and human health. In addition, the constant mutation of PRV leads to classical attenuated vaccines failing to provide sufficient protection against PRV infection. Therefore, it is desirable to explore the strategies of innate immune escape of PRV for drug target selection.

Antibodies and Reagents
The antibodies used and the sources were as follows: Anti-Flag M2 mouse mAb (1:5000,

Lentiviral Infection and Stable Cell Line Generation
Stable cell lines were generated as described previously [24]. Briefly, lentiviral particles were produced in HEK293T cells transfected with two packaging plasmids (psPAX2 and pVSVG) and an empty vector or pCDH-Flag-UL13 plasmid using Lipofectamine 2000 (Invitrogen). After 24 h, the recombinant viruses were filtered, then infected HEK293T cells again and supplemented with Polybrene (6 µg/mL, Cat#H8641, Solarbio, Beijing, China). Cells were selected with puromycin (Cat# IP1160, Solarbio, China) at a final concentration of 6 µg/mL for 5 days. Monoclonal cells were obtained in 96-well plates via the limited dilution method, and UL13-HEK293T cells were identified by immunoblotting with anti-Flag monoclonal antibody (Sigma).

Reverse Transcription and Quantitative Real-Time PCR (qPCR)
Total RNA isolation was carried out using an RNA purification Kit (Promega, Madison, WI, USA) and reverse-transcribed to cDNA using M-MLV reverse transcriptase with RNase inhibitor (Takara Bio, Beijing, China). qPCR was performed in triplicate with RealStar Green Fast Mixture (A303, GenStar, Beijing, China) and on a StepOnePlus thermal cycler (ABI, Thermo Fisher, MA, USA). Threshold cycle numbers were normalized to triplicate samples amplified with primers specific to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). qPCR primers for the target genes are listed in Table 1.

Primer sequences for regular PCR
AGCATGTGATTTAAAGGGGAAGTG CACTTCCCCTTTAAATCACATGCT

Cytoplasmic and Nuclear Protein Extraction
Cells were lysed on ice with a lysis buffer (10 mM HEPES at pH 7.9, 50 mM NaCl, 0.5 mM Sucrose, 0.1 mM EDTA, 0.5% Triton X-100, 1 mM DTT, 10 mM Sodium pyrophosphate decahydrate, 0.5 M NaF, 0.2 M Na3VO4, 1 mM PMSF, and protease inhibitor mixture) for 60 min. The supernatant was collected for the cytoplasmic extract after centrifuging for 5 min at 1500 rpm. The pellet was resuspended with Buffer A (10 mM HEPES at pH 7.9, 10 mM KCl, 0.1 mM EGTA, 0.1 mM EDTA, 1 mM DTT, 1 mM PMSF, and protease inhibitor mixtures), centrifuged for 5 min at 1500 rpm, and then the supernatant was removed. Then, four volumes of buffer C (10 mM HEPES at pH 7.9, 500 mM NaCl, 0.1 mM EGTA, 0.1 mM EDTA, 0.1% Nonidet P-40, 1 mM DTT, 1 mM PMSF, and protease inhibitor mixtures) were added, and vortexed for 30 min at 4 • C. The supernatant was collected as the nuclear extract after centrifugation for 10 min at 14,000 rpm. β-actin and Lamin B1 were used as loading controls for the cytoplasm and nucleus, respectively.

Immunofluorescence
UL13-HEK293T cells or control cells were transfected with poly(I:C) for 3 h and then fixed for 20 min with 4% cold paraformaldehyde. Cells were then permeabilized for 10 min with 0.1% Triton X-100, and then blocked with 5% bovine serum albumin (BSA, Sigma) for 30 min. Cells were incubated with the appropriate primary antibodies for 120 min, followed by staining with Alexa Fluor 594-conjugated secondary antibodies (Proteintech Group Inc. China) or 647-conjugated secondary antibodies (TIANGEN, Beijing, China) for 60 min. Nuclei were stained with DAPI. Images were visualized and acquired using a laser scanning confocal microscope with LAS X software (Leica, Wetzlar, Germany).

Statistical Analysis
All statistical analyses were carried out using GraphPad Prism v8.0 software; p-values were calculated with a two-tailed paired or unpaired Student's t-test, and p-values ≤ 0.05 were considered significant.

UL13 Inhibits RLR-mediated Antiviral Immune Responses
The role of the PRV tegument protein UL13 has been indicated in the suppression of host antiviral immune responses [24,29]. To confirm the suppressive effect of UL13 on RLR-mediated IFN-I signaling, we generated stable HEK293T cells ectopically expressing Flag-tagged UL13 (UL13-HEK293T cells). Immunoblotting analysis showed that Flag-UL13 was successfully expressed in UL13-HEK293T cells ( Figure 1A). Upon poly(I:C) transfection or Sendai virus (SeV) infection, which typically activate RLR-mediated immune responses, UL13-HEK293T cells showed impaired gene expression of IFNB1 and downstream ISGsincluding ISG56, MX1, and OASL-compared with control cells ( Figure 1B,C), indicating that UL13 inhibits RLR-mediated expression of type-I IFN and downstream ISGs. This observation was further confirmed in PK-15 cells stably expressing UL13 (UL13-PK-15 cells) ( Figure 1D,E). These results suggest that UL13 functions as an antagonist of RLR-mediated type-I IFN responses.

UL13 Inhibits RLR-Mediated Antiviral Response by Suppressing Transcription of RIG-I and MDA5
The transfected cytosolic RNA is mainly recognized by RLRs, and activates the adaptor protein MAVS to initiate innate antiviral immune response [11]. To know whether UL13 targets the RLR-mediated signaling pathway, we next examined the effect of UL13 on transfected poly(I:C)-mediated immune responses in HEK293T cells or PK-15 cells. The results showed that UL13 expression markedly inhibited the expression of RIG-I and MDA5, as well as phosphorylation of TBK1, in response to poly(I:C) transfection (Figure 2A), indicating that UL13 likely targets the expression of RIG-I-like receptors to suppress RNAmediated immune response. Similarly, upon SeV infection, UL13 expression also inhibited expression of RLR receptors-including RIG-I and MDA5-and phosphorylation of TBK1 and IRF3 in both HEK293T cells and PK-15 cells ( Figure 2B), further confirming the suppressive effect of UL13 on RNA-mediated immune responses. Notably, UL13 did not alter total protein levels of MAVS, TBK1, and IRF3 (Figure 2A,B), suggesting that UL13 possibly targets the upstream RLRs to suppress RLR-mediated signaling activation. Indeed, UL13 expression strikingly suppressed gene transcription of RIG-I and MDA5 under poly(I:C) transfection conditions ( Figure 2C). However, expression of UL13 did not affect mRNA expression of LGP2-another RIG-I like receptor. To confirm the effect of UL13 on RLR transcription, we first constructed luciferase reporter plasmids by introducing RIG-I, MDA5, or LGP2 promoter sequences into the pGL3-basic vector. Luciferase assays showed that UL13 expression remarkably restrained activation of RIG-I or MDA5 promoters in HEK293T cells, but not that of the LGP2 promoter ( Figure 2D), which is consistent with the suppressive effect of UL13 on transcription of RIG-I and MDA5 ( Figure 2C). Together,

UL13 Inhibits RLR-Mediated Antiviral Response by Suppressing Transcription of RIG-I and MDA5
The transfected cytosolic RNA is mainly recognized by RLRs, and activates the adaptor protein MAVS to initiate innate antiviral immune response [11]. To know whether UL13 targets the RLR-mediated signaling pathway, we next examined the effect of UL13 on transfected poly(I:C)-mediated immune responses in HEK293T cells or PK-15 cells. The results showed that UL13 expression markedly inhibited the expression of RIG-I and MDA5, as well as phosphorylation of TBK1, in response to poly(I:C) transfection ( Figure  2A), indicating that UL13 likely targets the expression of RIG-I-like receptors to suppress RNA-mediated immune response. Similarly, upon SeV infection, UL13 expression also inhibited expression of RLR receptors-including RIG-I and MDA5-and phosphorylation of TBK1 and IRF3 in both HEK293T cells and PK-15 cells ( Figure 2B), further confirming the suppressive effect of UL13 on RNA-mediated immune responses. Notably, UL13 did not alter total protein levels of MAVS, TBK1, and IRF3 (Figure 2A,B), suggesting that

Enforced Expression of RIG-I and MDA5 Counteracts UL13-Suppressed Antiviral Immune Responses
Having known that UL13 expression inhibits transcription of RIG-I and MDA5, leading to repression of RNA-triggered induction of IFNs and downstream antiviral responses, we next examined whether UL13 directly targets the mRNA of RLRs. qPCR analysis showed that UL13 overexpression has no effect on exogenous RIG-I/MDA5 transcription ( Figure 3A), suggesting that UL13 may repress endogenous transcription of RLRs. Then, we investigated whether overexpression of RIG-I or MDA5 could rescue the phenotype. As expected, enforced expression of RIG-I or MDA5 counteracted UL13-induced suppression of antiviral immune responses, including expression of IFNB1 and downstream ISGs (ISG56 and OASL), both at the basal level and in response to stimulation of transfected poly(I:C) ( Figure 3B), showing that RIG-I and MDA5 are the key targets for UL13 during RLRmediated immune responses. In line with this observation, overexpression of RIG-I or MDA5 also enhanced phosphorylation of TBK1 induced by transfected poly(I:C) in UL13-HEK293T cells, without altering the expression of total TBK1 ( Figure 3C). Collectively, these data demonstrate that UL13 inhibits RLR-mediated antiviral immune responses via downregulation of RLR expression. ing RIG-I, MDA5, or LGP2 promoter sequences into the pGL3-basic vector. Luciferase assays showed that UL13 expression remarkably restrained activation of RIG-I or MDA5 promoters in HEK293T cells, but not that of the LGP2 promoter ( Figure 2D), which is consistent with the suppressive effect of UL13 on transcription of RIG-I and MDA5 ( Figure  2C). Together, these data demonstrate that UL13 negatively regulates the RLR signaling pathway via suppression of the transcription of RIG-I and MDA5. sion of RIG-I or MDA5 also enhanced phosphorylation of TBK1 induced by tran poly(I:C) in UL13-HEK293T cells, without altering the expression of total TBK1 3C). Collectively, these data demonstrate that UL13 inhibits RLR-mediated antiv mune responses via downregulation of RLR expression.

UL13 Overexpression Restrains NF-κB Activation to Modulate Transcription of RIG-I and MDA5
Next, we sought to investigate the mechanisms by which UL13 suppresses the transcription of RLRs and RLR-mediated antiviral gene expression. It has been established that activation of NF-κB and mitogen-activated protein kinases (MAPKs) modulates gene transcription. We therefore examined whether UL13 modulates the activation of NF-κB and/or MAPKs to regulate IFN responses induced by poly(I:C) transfection. qPCR analysis showed that inhibition of NF-κB by a chemical inhibitor compromised the expression of IFNB1 and downstream ISGs such as ISG56 and MX1, whereas inhibition of MAPKs did not alter the UL13-mediated suppression of IFNB1 and ISGs induced by transfected Poly(I:C) in HEK293T cells ( Figure 4A), indicating that UL13 regulates NF-κB activation to repress RLRmediated IFN responses. Consistently, immunoblotting analysis also showed that UL13 overexpression significantly suppressed phosphorylation of p65 and retarded degradation of IκBα in HEK23T cells transfected with poly(I:C) ( Figure 4B). Importantly, overexpression of UL13 strikingly decreased the protein expression of RIG-I under both basal and stimulated conditions ( Figure 4B), implying that suppression of NF-κB by UL13 is related to RIG-I expression. Moreover, UL13 overexpression impaired poly(I:C)-induced nuclear translocation of p65 ( Figure 4C,E) and mRNA expression of IL6 and TNF ( Figure 4F)-two key inflammatory cytokines that are dependent on NF-κB. Altogether, these results suggest that UL13 modulates NF-κB activation to suppress RLR expression and RLR-mediated antiviral immune responses.
To further clarify whether UL13 regulates transcription of RIG-I or MDA5 by inhibiting activation of NF-κB, we next examined the effects of p65 on the activation of RIG-I, MDA5, and LGP2 promoters. Luciferase assays showed that p65 significantly promoted the activity of RIG-I and MDA5 promoters, but not the LGP2 promoter ( Figure 4G), demonstrating that NF-κB activation is critical for transcription of RIG-I and MDA5. Interestingly, RIG-I and MDA5 promoters, but not the LGP2 promoter, contain several putative p65-binding sites ( Figure 4H). To confirm the direct role of NF-κB in the transcription regulation of RIG-I or MDA5, we constructed two mutant RIG-I and MDA5 luciferase reporter plasmids in which two putative p65-binding sequences were deleted ( Figure 4H). The results showed that while UL13 expression dramatically suppressed wild-type RIG-I-promoter-driven luciferase activity, UL13 failed to suppress mutant RIG-I-promoter-driven luciferase activity when the p65-binding motif between −421 and −412 was deleted ( Figure 4I). Similarly, UL13 inhibited wild-type MDA5 promoter activity but failed to suppress mutant MDA5 promoter activation when the p65-binding sites were deleted ( Figure 4I). Together, the above data suggest that the transcription of RIG-I and MDA5 is dependent on NF-κB, and that UL13 suppresses transcription of RIG-I and MDA5 by inhibiting NF-κB activation.

UL13 Deficiency Potentiates RLR-Mediated Antiviral Responses during PRV Infection
Next, we investigated the effects of UL13 on RIG-I and MDA5 expression during PRV infection in HEK293T cells. Compared with wild-type PRV (PRV-WT), HEK293T cells showed higher levels of RIG-I or MDA5 when infected with the mutant PRV lacking UL13 (PRV-∆UL13) ( Figure 5A). Consequently, PRV-∆UL13 infection enhanced the induction of IFNB1 and downstream ISGs, as well as the key inflammatory mediators, including IL6 and TNF ( Figure 5B,C). Consistent with these results, the expression of RIG-I and MDA5, along with the phosphorylation levels of TBK1 and p65-but not the key adaptor protein MAVS-were higher in HEK293T cells infected with PRV-∆UL13 than in PRV-WTinfected cells ( Figure 5D). Collectively, these data suggest that deficiency of UL13 promotes RLR-mediated antiviral immune responses during PRV infection.   of IFNB1 and downstream ISGs, as well as the key inflammatory mediators, including IL6 and TNF ( Figure 5B,C). Consistent with these results, the expression of RIG-I and MDA5, along with the phosphorylation levels of TBK1 and p65-but not the key adaptor protein MAVS-were higher in HEK293T cells infected with PRV-∆UL13 than in PRV-WT-infected cells ( Figure 5D). Collectively, these data suggest that deficiency of UL13 promotes RLR-mediated antiviral immune responses during PRV infection.

Discussion
IFN-mediated innate immune responses are crucial in protecting the host from viral infections and activating adaptive immunity [30,31]. In response to viral infection, host cells activate multiple signaling cascades to disrupt viral replication when specific molecular components of viruses are recognized [32,33]. Generally, cytosolic DNA sensors, such as cGAS, play important roles in recognition of DNA viruses [34][35][36]. Interestingly, recent studies suggest that RLRs sense RNA intermediates and host-derived RNAs during DNA virus infection [37][38][39][40]. As a DNA virus, PRV can establish long-term infection by evading cGAS-STING-and TLR3-induced antiviral innate immune responses [7,24,29]. However, how PRV evades RLR-mediated immune responses remains obscure. Here, we show that the PRV tegument protein UL13 suppresses type-I IFN production and downstream ISGs induced by RNA stimulation, indicating that UL13 functions as a novel antagonist of RLR-mediated antiviral responses. Hence, our study provides evidence that UL13 plays important roles in PRV's evasion of RLR-mediated antiviral innate immune responses.
In recognition of RNAs, RLRs activate the MAVS-TBK1-IRF3 axis to induce type-I IFN responses [21]. Viruses, such as alphaherpesviruses, have developed multitudinous strategies for the evasion of RLR-mediated signaling [21]. Among these strategies, targeting RIG-I and MDA5 receptors is an effective way to inhibit RLR-mediated signaling transduction. For example, the HSV-1 UL11 C-terminal RNA-binding domain binds to RIG-I and MDA5, resulting in inhibition of downstream signaling activation [22]. HSV-1 UL37 deamidates RIG-I, rendering it unable to sense viral RNA and, thus, blocking its ability to induce antiviral immune responses [23]. HSV-2 utilizes the virion host shutoff (vhs) protein to suppress RIG-I and MDA5 expression [41]. Consistently, we found that PRV infection could also significantly inhibit the gene transcription of RIG-I and MDA5, but not LGP2. Mechanistically, the PRV tegument protein UL13 inhibits transcription of RIG-I or MDA5 by suppressing activation of NF-κB. Notably, both RIG-I and MDA5 are ISGs, and their expression can be quickly induced upon viral infection and dsRNA stimulation [42,43]. As IFN-inducible genes, increases in RIG-I and MDA5 expression may amplify the effects of IFNs [44]. Therefore, targeting of RIG-I and MDA5 by PRV can be a novel and simple strategy to escape RLR-mediated antiviral innate immune response.
Gene expression is under tight control of transcription factors that bind to unique DNA enhancer/repressor elements [26]. For RIG-I, several transcription factors have been reported to be directly involved in its transcriptional regulation. For example, interferon regulatory factor-1 (IRF1) positively regulates interferon-or dsRNA-induced RIG-I transcription [45]. CCAAT/enhancer-binding protein beta (C/EBPβ) acts as a repressor element to bind the RIG-I promoter for RIG-I transcriptional inhibition [46]. In this study, we found a key p65-binding site in the RIG-I proximal promoter. Deletion of the p65-binding site resulted in dramatic loss of RIG-I promoter activation induced by p65, which is consistent with a previous report [46]. Unlike RIG-I, the molecular mechanism that induces MDA5 expression has not been presented. Here, we identified that p65 expression noticeably induced MDA5 promoter activity. Deletion of two key p65-binding sites abolished activity of the MDA5 promoter induced by p65, indicating that p65 is also the key transcriptional factor for MDA5 transcription. In contrast, no putative p65-binding site was found in the LGP2 promoter, which is consistent with the finding that UL13 did not affect LGP2 activation. UL13 is a serine/threonine protein kinase that targets several viral and cellular substrates [47]. Studies have shown that UL13 can participate in various processes depending on its kinase activity. For example, UL13 directly modulates the phosphorylation of viral proteins VP11/12, ICP22, and UL49 [48,49], promotes the assembly and release of mature infectious virions [50], and stabilizes the viral ICP0 protein against degradation. Furthermore, UL13 targets STING or IRF3 to escape IFN-mediated antiviral innate immune response [24,25,29]. Here, we found that UL13 inhibited the expression of RIG-I and MDA5 by blocking the activation of NF-κB. However, the detailed mechanisms of how UL13 prevents NF-κB activation require further investigation.
In summary, this study identifies novel evasion strategies for PRV via suppression of the transcription of RIG-I and MDA5 by UL13 through targeting NF-κB activation, further highlighting the importance of UL13 in the regulation of host innate antiviral immune responses.