Interferon Gamma Inhibits Equine Herpesvirus 1 Replication in a Cell Line-Dependent Manner

The sole equine herpesvirus 1 (EHV-1) immediate-early protein (IEP) is essential for viral replication by transactivating viral immediate-early (IE), early (E), and late (L) genes. Here, we report that treatment of mouse MH-S, equine NBL6, and human MRC-5 cells with 20 ng/mL of IFN-γ reduced EHV-1 yield by 1122-, 631-, and 10,000-fold, respectively. However, IFN-γ reduced virus yield by only 2–4-fold in mouse MLE12, mouse L-M, and human MeWo cells compared to those of untreated cells. In luciferase assays with the promoter of the EHV-1 early regulatory EICP0 gene, IFN-γ abrogated trans-activation activity of the IEP by 96% in MH-S cells, but only by 21% in L-M cells. Similar results were obtained in assays with the early regulatory UL5 and IR4 promoter reporter plasmids. IFN-γ treatment reduced IEP protein expression by greater than 99% in MH-S cells, but only by 43% in L-M cells. The expression of IEP and UL5P suppressed by IFN-γ was restored by JAK inhibitor treatment, indicating that the inhibition of EHV-1 replication is mediated by JAK/STAT1 signaling. These results suggest that IFN-γ blocks EHV-1 replication by inhibiting the production of the IEP in a cell line-dependent manner. Affymetrix microarray analyses of IFN-γ-treated MH-S and L-M cells revealed that five antiviral ISGs (MX1, SAMHD1, IFIT2, NAMPT, TREX1, and DDX60) were upregulated 3.2–18.1-fold only in MH-S cells.


Introduction
Equine herpesvirus 1 (EHV-1) is a member of the Alphaherpesvirinae subfamily of the Herpesviridae and is a major equine pathogen that causes respiratory disease, abortion, and, in some cases, the neurological disease [1][2][3]. The most devastating outcome of EHV-1 infection is the induction of abortion in pregnant mares, which has a major economic impact on the horse industry. EHV-1 infection generates a short-lived humoral immunity in the horse but does not confer long-term protection, as disease often occurs following infection [4,5].
Interferon gammas (Type II IFN) plays a major role in controlling the host immune response against viral and intracellular bacterial pathogens [6,7] and is produced at high levels by T helper Type 1 (Th1) cells, CD8 + CTL, and NK cells in response to virus infection [8]. IFN-γ induces the production of proinflammatory cytokines and chemokines in endothelial cells, epithelial cells, and fibroblasts. IFN-γ has its own receptor of ligand-binding IFNGR1 and IFNGR2 subunits [9,10] and the intracellular carboxy termini of IFNGR1 and IFNGR2 carry the tyrosine kinases JAK1 and JAK2, respectively, which phosphorylate the receptor upon ligand binding [11][12][13]. Phosphorylation leads to translocation of the signal transducer and activator of transcription 1 (STAT1) homodimers into the nucleus, where they bind to interferon gamma-activated sequence (GAS) sites on the promoters of downstream target genes. One of the major primary response genes is the transcription factor IRF1 The infected cells were also harvested at 30 hpi for Western blot analyses using anti-IEP OC33 [29] and anti-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Data are representative of three independent experiments. Numbers to the left represent molecular weight standards (kDa) (Bio-Rad Laboratories, Hercules, CA, USA). Quantification was done with Image Studio Lite software (LI-COR Biosciences, Lincoln, NE, USA). *, p < 0.01 for comparison with the untreated control. **, p < 0.05 for comparison with the untreated control. A two-tailed t-test was performed. (C) Detection of IFN-γ receptor in three mouse cells (MH-S, L-M, and MLE12). Cell extracts were used for Western blot analyses with the anti-mouse IFN-γRα monoclonal antibody (Santa Cruz Biotechnology) and anti-β-actin polyclonal antibody.

IFN-γ Blocks the IEP-Mediated Trans-Activation by Inhibiting the Expression of IEP in MH-S Cells
Interestingly, IFN-γ was not able to reduce the levels of IEP expression at 6 hpi [21], indicating that IFN-γ may inhibit viral gene expression after the immediate-early (IE) stage of infection. EHV-1 IEP is a major viral trans-activator and is essential for viral growth [27,30,31]. EHV-1 gene transcription occurs in a cascade that leads to the synthesis of viral proteins that are classified as IE, E, and L [32][33][34][35]. Based on these results, we hypothesized that IFN-γ blocks viral E and L gene expression by inhibiting IEP function. To address this possibility, we performed luciferase reporter assays with EHV-1 promoter reporter plasmids.
In luciferase assays with EHV-1 early EICP0 promoter reporter plasmids, IFN-γ abrogated the trans-activation activity of IEP by 96% and 21% in MH-S and L-M cells ( Figure 2A and Figure 2B, respectively). IFN-γ treatment reduced the level of the IEP by greater than 99% in MH-S cells, but only by 17% in L-M cells ( Figure 2C). Very similar results were obtained with the early UL5 and IR4 promoter reporter plasmids ( Figure 2D-G). These results suggest that IFN-γ blocks EHV-1 gene expression by inhibiting the production of the IEP. The infected cells were also harvested at 30 hpi for Western blot analyses using anti-IEP OC33 [29] and anti-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Data are representative of three independent experiments. Numbers to the left represent molecular weight standards (kDa) (Bio-Rad Laboratories, Hercules, CA, USA). Quantification was done with Image Studio Lite software (LI-COR Biosciences, Lincoln, NE, USA). *, p < 0.01 for comparison with the untreated control. **, p < 0.05 for comparison with the untreated control. A two-tailed t-test was performed. (C) Detection of IFN-γ receptor in three mouse cells (MH-S, L-M, and MLE12). Cell extracts were used for Western blot analyses with the anti-mouse IFN-γRα monoclonal antibody (Santa Cruz Biotechnology) and anti-β-actin polyclonal antibody.

IFN-γ Blocks the IEP-Mediated Trans-Activation by Inhibiting the Expression of IEP in MH-S Cells
Interestingly, IFN-γ was not able to reduce the levels of IEP expression at 6 hpi [21], indicating that IFN-γ may inhibit viral gene expression after the immediate-early (IE) stage of infection. EHV-1 IEP is a major viral trans-activator and is essential for viral growth [27,30,31]. EHV-1 gene transcription occurs in a cascade that leads to the synthesis of viral proteins that are classified as IE, E, and L [32][33][34][35]. Based on these results, we hypothesized that IFN-γ blocks viral E and L gene expression by inhibiting IEP function. To address this possibility, we performed luciferase reporter assays with EHV-1 promoter reporter plasmids.
In luciferase assays with EHV-1 early EICP0 promoter reporter plasmids, IFN-γ abrogated the trans-activation activity of IEP by 96% and 21% in MH-S and L-M cells (Figures 2A and 2B, respectively). IFN-γ treatment reduced the level of the IEP by greater than 99% in MH-S cells, but only by 17% in L-M cells ( Figure 2C). Very similar results were obtained with the early UL5 and IR4 promoter reporter plasmids ( Figure 2D-G). These results suggest that IFN-γ blocks EHV-1 gene expression by inhibiting the production of the IEP.  [11] and β-actin pAb. Control, pSVSPORT-transfected cells. *, p < 0.01 for comparison with the untreated control. **, p < 0.05 for comparison with the untreated control.

EHV-1 Inhibition by IFN-γ in MH-S, NBL6, and MRC-5 Cells Is Not Due to Nitric Oxide
Nitric oxide (NO)-induced by IFN-γ inhibited late vaccinia protein synthesis and DNA replication in infected RAW 264.7 cells [28]. Our previous results [36] showed that nitric oxide (NO) is not involved in the inhibition of varicella-zoster virus (VZV) replication by IFN-γ in A549, MRC-5, and ARPE-19 cells. To investigate whether the inhibition of EHV-1 replication in MH-S, NBL6, and MLE12 cells ( Figure 1) could be attributed to an NO-mediated mechanism, NO synthesis in cultures was determined by measurement of nitrite (NO2−), a stable product of NO. When mouse macrophage RAW264.7 cells were treated with 20 ng/mL of IFN-γ, NO production increased by 2-fold ( Figure 3). However, IFN-γ did not induce NO production in MH-S, L-M, MLE12, and NBL6 cells (Figure 3), suggesting that nitric oxide (NO) is not involved in the EHV-1 inhibition by IFN-γ in MH-S, NBL6, and MRC-5 cells.  Nitric oxide (NO)-induced by IFN-γ inhibited late vaccinia protein synthesis and DNA replication in infected RAW 264.7 cells [28]. Our previous results [36] showed that nitric oxide (NO) is not involved in the inhibition of varicella-zoster virus (VZV) replication by IFN-γ in A549, MRC-5, and ARPE-19 cells. To investigate whether the inhibition of EHV-1 replication in MH-S, NBL6, and MLE12 cells ( Figure 1) could be attributed to an NO-mediated mechanism, NO synthesis in cultures was determined by measurement of nitrite (NO 2 −), a stable product of NO. When mouse macrophage RAW264.7 cells were treated with 20 ng/mL of IFN-γ, NO production increased by 2-fold ( Figure 3). However, IFN-γ did not induce NO production in MH-S, L-M, MLE12, and NBL6 cells (Figure 3), suggesting that nitric oxide (NO) is not involved in the EHV-1 inhibition by IFN-γ in MH-S, NBL6, and MRC-5 cells.

JAK/STAT1 Signaling Pathway Mediates EHV-1 Inhibition
The inhibition of viral replication by IFN-γ depends on the activation of the JAK-STAT1 signaling pathway [37]. Our previous results [36] showed that all three human cell lines MRC-5, A549, and MeWo showed similar levels of IFN-γ receptor protein. Thus, we investigated the role of IFN-γ signaling through the interferon gamma receptor and JAK in EHV-1 inhibition. To ascertain if IFN-γ utilizes the JAK signaling to inhibit EHV-1 replication, human lung fibroblast MRC-5 cells were treated with JAK inhibitor (#420099, EMD Millipore) and IFN-γ (20 ng/mL) and infected with 0.05 MOI of EHV-1 KyA. Treatment with IFN-γ increased the levels of JAK2, STAT1, and phosphorylated STAT1, which resulted in a reduction in the expression of viral immediate-early proteins IEP and UL5P ( Figure 5, lanes 4 and 6). Treatment with the JAK inhibitor reduced by~4-fold (as determined by the density of the bands) the level of phosphorylated STAT1 (pSTAT1) without a change in the total STAT1 protein level ( Figure 5, lane 7). The expression of IEP and UL5P suppressed by IFN-γ was recovered by JAK inhibitor treatment ( Figure 5, compare lanes 5 and 7), while JAK inhibitor alone had no effect on the expression of IEP and UL5P (data not shown). In contrast, PI3K inhibitor (LY294002) did not reduce the level of pSTAT1 and had no effect on viral gene expression ( Figure 5, compare lanes 4 and 6). These results indicated that the loss of the IFN-γ antiviral effect during EHV-1 infection is due to a reduction in STAT1 phosphorylation. Taken together, these results suggest that IFN-γ-induced JAK/STAT1 signaling is regulated during EHV-1 infection.
2.6. Innate Immunity Related Genes and Antiviral Interferon-Stimulated Genes Are Upregulated in the IFN-γ-Treated MH-S Cells IFN-γ pretreatment inhibited EHV-1 replication in MH-S cells by 1200-fold ( [24]; Figure 1A), suggesting that IFN-γ-stimulated protein(s) (excluding its direct effect) mediate the inhibition of viral replication. In contrast, IFN-γ treatment reduced EHV-1 replication by only 2.5-fold in L-M cells ( Figure 1B). This significant difference of the effect of IFN-γ-on EHV-1 replication in these two cell types offers an approach to identify candidate anti-EHV-1 genes stimulated by IFN-γ. To identify IFN-γ-stimulated anti-EHV-1 genes, Affymetrix microarray analyses were performed with RNA from IFN-γ-treated MH-S and L-M cells. As shown in the Venn diagram, 152 of the genes whose expression was significantly altered were identified in both arrays ( Figure 6A). In IFN-γ-treated MH-S cells, 551 and 136 genes were upregulated and downregulated, respectively ( Figure 6B). In IFN-γ-treated L-M cells, 225 and 2 genes were upregulated and downregulated, respectively ( Figure 6B).

JAK/STAT1 Signaling Pathway Mediates EHV-1 Inhibition
The inhibition of viral replication by IFN-γ depends on the activation of the JAK-STAT1 signaling pathway [37]. Our previous results [36] showed that all three human cell lines MRC-5, A549, and MeWo showed similar levels of IFN-γ receptor protein. Thus, we investigated the role of IFN-γ signaling through the interferon gamma receptor and JAK is due to a reduction in STAT1 phosphorylation. Taken together, these results suggest tha IFN-γ-induced JAK/STAT1 signaling is regulated during EHV-1 infection.

Innate Immunity Related Genes and Antiviral Interferon-Stimulated Genes Are Upregulate in the IFN-γ-Treated MH-S Cells
IFN-γ pretreatment inhibited EHV-1 replication in MH-S cells by 1200-fold ( [24]; Fig  ure 1A), suggesting that IFN-γ-stimulated protein(s) (excluding its direct effect) mediat the inhibition of viral replication. In contrast, IFN-γ treatment reduced EHV-1 replicatio by only 2.5-fold in L-M cells ( Figure 1B). This significant difference of the effect of IFN-γ on EHV-1 replication in these two cell types offers an approach to identify candidate ant EHV-1 genes stimulated by IFN-γ. To identify IFN-γ-stimulated anti-EHV-1 genes, Affy metrix microarray analyses were performed with RNA from IFN-γ-treated MH-S and L M cells. As shown in the Venn diagram, 152 of the genes whose expression was signif cantly altered were identified in both arrays ( Figure 6A). In IFN-γ-treated MH-S cells, 55 and 136 genes were upregulated and downregulated, respectively ( Figure 6B). In IFN-γ treated L-M cells, 225 and 2 genes were upregulated and downregulated, respectivel ( Figure 6B).  The numbers indicate the genes from the total microarray datasets that showed a greater than 2.0-fold increase or more than a 2.0-fold reduction (p-value ≤ 0.05). Up, upregulated genes; down, downregulated genes.
As shown in Table 1, genes associated with immune and inflammatory responses were significantly upregulated in IFN-γ-treated MH-S cells as compared to those of IFNγ-treated L-M cells. C1qa and C1qb, which are involved in the recognition component of complement activation, were upregulated 4.6 and 38.1-fold only in IFN-γ-treated MH-S The numbers indicate the genes from the total microarray datasets that showed a greater than 2.0-fold increase or more than a 2.0-fold reduction (p-value ≤ 0.05). Up, upregulated genes; down, downregulated genes.

Discussion
The innate immune response is the first line of defense against viral pathogens, and in the horse, protective immunity to EHV-1 infection was characterized by a polarized IFN-γ dependent immunoregulatory cytokine response [22]. Equine IFN-γ increases major histocompatibility complex (MHC) Class II expression by monocytes and peripheral blood mononuclear cells (PBMCs) [42], and contributes significantly to antiviral activity [43]. Our previous results showed that nonpathogenic EHV-1 KyA immunization induced humoral and T cell immune responses that protected CBA mice from pathogenic RacL11 challenge infection at 4 weeks post-immunization [44][45][46]. Affymetrix microarray analysis revealed that the IFN-γ gene and 16 antiviral interferon-stimulated genes (ISGs) were upregulated 3.1-48.2-fold at 8 h post-challenge in the lungs of RacL11-challenged mice that had been immunized with KyA [21]. CpG oligodeoxynucleotides (CpG ODNs) can enhance innate immune responses [47] and induce type I IFN, IFN-γ, and TNF-α production in equine peripheral blood mononuclear cells (PBMC) [48]. In this regard, our recent results [24] showed that intranasal immunization with CpG-B ODN 1826 significantly increased expression of IFN-γ and 7 antiviral ISGs upon pathogenic RacL11 challenge, accelerated clearance of virus from the lungs of infected CBA mice, and protected mice at 1-5 days post-immunization. IFN-γ treatment reduced EHV-1 yields in murine alveolar macrophage MH-S cells and protected mice against lethal EHV-1 challenge [21]. The findings of this study showed that the inhibition of EHV-1 replication is mediated by JAK/STAT1 signaling, suggesting that that IFN-γ-stimulated genes (ISGs) are involved in the inhibition of EHV-1 replication.
The genes of EHV-1 are coordinately expressed and temporally regulated in an immediate-early (IE), early (E), and late (L) fashion [33][34][35][36], and the sole immediate-early protein (IEP) activates the expression of the E and L genes [25,[49][50][51]. Our findings showed that IFN-γ blocks EHV-1 replication by inhibiting IEP expression in a cell line-dependent manner. Varicella-zoster virus (VZV) and EHV-1 are members of the subfamily Alphaherpesvirinae and genus Varicellovirus, and have very similar genomic structure (group D) [52]. Our previously published results showed that IFN-γ inhibited VZV replication by inhibiting immediate-early IE62 expression.
High output nitric oxide (NO) induced by IFN-γ has potent antimicrobial activity against several classes of pathogens [53,54]. IFN-γ-induced nitric oxide synthase (iNOS) and NO were shown to inhibit replication of ectromelia virus, vaccinia virus (VACV), and herpes simplex-1 (HSV-1) in murine macrophage-like RAW 264.7 cells [55]. HSV-1, EHV-1, and VZV are members of the subfamily Alphaherpesvirinae [52]. Our findings in this study showed that IFN-γ treatment effectively inhibited EHV-1 replication in murine MH-S and equine NBL6 cells but failed to induce NO in either MH-S or NBL6 cells. Our previous results [37] showed that NO is not involved in the VZV inhibition by IFN-γ in human MRC-5 cells. These results suggest that NO is not involved in the EHV-1 inhibition by IFN-γ in MH-S, NBL6, and MRC-5 cells.
IFN-γ is a cytokine and a central regulator of the immune response and signals via the JAK-STAT pathway. Phosphorylated STAT1 homodimers translocate to the nucleus, where they bind to GAS (gamma activating sequence) sites and recruit additional factors to modulate gene expression [14]. STAT1 is essential for the IFN-γ-dependent inhibition of vaccinia virus replication [56]. Our result showed that JAK inhibitor abrogated IFNγ-induced STAT1 phosphorylation ( Figure 5). The expression of EHV-1 IEP and UL5P suppressed by IFN-γ was recovered by JAK inhibitor treatment ( Figure 5). These results indicated that JAK/STAT1 pathway involved in the controlling EHV-1 replication.
Cells of the innate immune system detect viral infection largely through pattern recognition receptors that can activate downstream signaling pathways that culminate in the activation of interferon regulatory factors and subsequent induction of interferons (IFNs). IFNs bind their cognate cell-surface receptors and induce hundreds of interferonstimulated genes (ISGs), many of which function by various mechanisms to inhibit virus replication [38,57,58]. Our data showed that IFN-γ treatment reduced EHV-1 yield by approximately 1200-fold in murine alveolar macrophage MH-S cells but only by 2.5-fold in mouse fibroblast L-M cells (Figure 1). Affymetrix microarray analysis revealed that five antiviral ISGs, MX1, SAMHD1, IFIT2, NAMPT, TREX1, and DDX60, were significantly upregulated only in IFN-γ-treated MH-S cells (Table 2). MX1, which is known to be antiviral against hepatitis C virus, influenza A virus, measles virus, and vesicular stomatitis virus was significantly upregulated by 18.1-fold. SAMHD1, which is known to be antiviral against human immunodeficiency virus, hepatitis B virus, and human T cell leukemia virus type 1 was upregulated by 5.9-fold. The six IFN-γ-stimulated genes will be assessed in controlling EHV-1 replication.
We are presently developing recombinant adenoviruses that express each of the ISGs induced by IFN-γ and are associated with the inhibition of EHV-1 replication. Our initial experiments with a recombinant adenovirus that expresses MX1 have validated this approach and revealed that MX1 has strong antiviral activity to EHV-1 in human ARPE-19 cells. Experiments with our panel of recombinant adenoviruses that express 15 candidate ISGs will identify individual ISGs with anti-EHV-1 activity in cell culture and allow experiments to reveal the mechanism of anti-EHV-1 activity. In addition, ISGs that effectively target specific steps in the virus replication cycle would be assayed for antiviral activity in the CBA mouse model of EHV-1 pathogenesis. Findings from these experiments would serve as the basis to assess promising candidate ISGs in the natural host in an effort to develop effective treatment for infections due to this major equine pathogen.

Plaque Assays
EHV-1 virus titers were determined on NBL6 cells as described previously [62]. Serial dilutions of samples from each passage were used to inoculate NBL6 monolayers that were incubated for 4 days in medium containing 1.5% methylcellulose. Plaques were quantitated by fixing with 10% formalin solution (Thermo Fisher Scientific, Waltham, MA, USA) and staining with 0.5% crystal violet.

Western Blot Analysis
Preparation of cytoplasmic and nuclear extracts of transfected cells and Western blot analysis were performed as previously described [51]. Briefly, blots were incubated with the antibodies indicated in the figure legends for 2 h, were washed three times for 5 min each in TBST, and incubated with secondary antibody (anti-rabbit IgG (Fc)-alkaline phosphatase (AP) conjugate (Promega, Madison, WI, USA)) for 30 min. Proteins were visualized by incubating the membranes containing blotted protein in AP conjugate substrate (Bio-Rad, Hercules, CA, USA) according to the manufacturer's directions. The density of the brands on the membrane was determined by scanning with a HP Scanjet 8300 (Hewlett-Packard, Palo Alto, CA, USA), and the scans were analyzed with Image Studio Lite software (LI-COR Biosciences, Lincoln, NE, USA).

Luciferase Reporter Assays
The luciferase reporter assay was performed by using Lipofectamine 3000 reagent (Invitrogen, San Diego, CA, USA) according to the manufacturer's protocol. The mouse, equine, and human cells were seeded at 50% confluency in 24-well plates, treated with 0 or 20 ng/mL (0.5 mL/well) of murine (Cell Sciences, Canton, MD, USA), equine (R&D Systems, Minneapolis, MN, USA), or human IFN-γ (Cell Sciences), respectively, and cotransfected with 0.07 pmol of reporter vector and 0.07 pmol of effectors in each well at 24 h post-treatment. Four microliters of Lipofectamine 3000 were diluted with 114 µL of Opti-MEM medium (Invitrogen, San Diego, CA, USA). DNA and 1.5 µL of P3000 reagent were mixed with 114 µL of Opti-MEM medium. The total amount of DNA was adjusted to the same amount with pSVSPORT1 DNA. The solutions were combined and incubated at room temperature for 15 min, and one-third volume was transferred into each of three wells of the four human cells. At 40 h post-transfection, luciferase activity was measured with a luciferase assay kit (Promega, Madison, WI, USA) and a Polarstar Optima plate reader (BMG LABTECH Inc., Cary, NC, USA).

Nitric Oxide (NO) Assay
Mouse and equine cell lines were seeded in 24-well plate (0.5 mL/well) and cultured in the presence and absence of murine or equine IFN-γ (20 ng/mL). After 24 h treatment, NO generation was evaluated by measuring the accumulation of nitrite in the culture medium by using the NO kit (EMD Millipore, Burlington, MA, USA).

Microarray Analysis
MH-S and L-M cells were treated with 20 ng/mL of murine IFN-γ (Cell Sciences, Canton, MA, USA). The untreated and treated cells were harvested at 8 h post-treatment and microarray analyses were performed as described previously [21]. The quality and quantity of RNA were determined, and RNA was processed with the Mouse Affymetrix Genome 430 2.0 arrays (Affymetrix, Santa Clara, CA, USA) at the LSUHSC-S CMTV Genomics/DNA Array Core Facility. Biotinylated cRNA was generated using the Affymetrix 3 IVT Kit (Affymetrix, CA, USA), per the manufacturer's instructions. Arrays were scanned using a GeneChip Scanner 3000 7G with autoloader. Pixel intensities were measured, expression signals were analyzed, and features were extracted using the commercial software package Transcriptome Analysis Console 3.0 (Affymetrix, Santa Clara, CA, USA). Gene expression changes were considered significant if the p value was less than 0.05, the fold change was at least 2.0 between treated and mock-treated lungs, and changes in gene expression were reproducible in all replicate comparisons. Genes expressed at different levels in untreated controls were excluded from analysis.

Microarray Data Accession Number
The microarray data were deposited in the NCBI Gene Expression Omnibus (GEO) database under accession number GSE126813 (http://www.ncbi.nlm.nih.gov/geo; 21 February 2019).