Swine testis (ST) cells were maintained in our laboratory in DMEM (Invitrogen Gibco, United States) supplemented with antibiotics (100 U/mL of penicillin and 100 μg/mL of streptomycin) and 10% fetal bovine serum (FBS) (Invitrogen Gibco, United States) at 37°C with 5% CO₂. The cell lines present in this study were obtained from Xiamen Immocell Biotechnology Co., Ltd, Guangdong province, China. The mouse anti-GETV-E2 polyclonal antibody was prepared in our laboratory (data not shown); the CF^®488A goat anti-mouse IgG (H+L) (green) antibody was purchased from Biotium, China; and the HRP-labeled goat anti-mouse IgG antibody was purchased from ZSGB-BIO, China. Porcine interferons were prepared and stocked in our laboratory (Liu et al., 2016). The virus, GETV strain BJ0304 (GenBank accession number OM363683), was stored at our laboratory.

The ST cells were treated with IFN-α, IFN-ω (referred specifically to as IFN-ω7 in this study), IFN-γ, IFN-λ3, or PBS for 24 hours (h) either 24 h before or 0 and 12 h post-infection (hpi) with GETV [multiplicity of infection (MOI) = 0.1], and mock-infected cells were used as controls. At the indicated time points after infection, after observing and recording the cytopathic effects (CPEs), the cell samples and the supernatant were collected. The former was used to analyze the expression levels of E2 by immunological methods (i.e., IFA and Western blot) and to detect the copies of GETV nsp3 (Supplementary Table S1) by RT-qPCR, while the supernatant samples were used for the TCID₅₀ assay according to the Reed and Muench method (Reed and Muench, 1938).

ST cell monolayers with or without IFN pro-treatment were inoculated with GETV at an MOI of 0.1 for 1.5 h and then cultured with DMEM with 2% FBS for the indicated time points after infection before analyzing the GETV infection using IFA as described by Li et al. (2017).

The IFN-γ-treated and/or GETV-infected and mock-infected ST cells were collected for RNA extraction using TRIzol reagent (Invitrogen, America), followed by cDNA synthesis. The GETV nsp3 and some genes selected from statistical analysis of RNA-seq were determined by RT-qPCR using the ChamQ Universal SYBR qPCR Master Mix (Vazyme, China), and GAPDH was used as an endogenous control. All primers used are listed in Supplementary Table S1.

Mock- and GETV-infected (MOI=0.1) ST cells with or without IFN-γ were collected at 20 hpi for RNA-sequencing. Then, 12 samples consisting of three repetitions in each group were used in this study. Illumina RNA-Seq library preparation, deep sequencing, and sequence analysis were performed as described by Liu et al. (2020). mRNA was selected using oligo(dT)-attached magnetic beads from purified total RNA. After treatment with fragmentation buffer, approximately 300 bp mRNA fragments were selected and converted to cDNA. Adaptors were ligated to the ends of the cDNA fragments before end-repair and dA-tailing and then amplified using an Illumina TruseqTM RNA sample Prep Kit. Libraries sequencing was carried out on an Illumina Novaseq 6000 Sequencing System, completed by Meggie Biotechnology Co., Ltd.

The raw data (reads) obtained from the Illumina Novaseq 6000 Sequencing System were processed and counted by fastp to obtain the clean data (reads), followed by being mapped to the Sus scrofa genome (RefSeq GCF_000003025.6, https://www.ncbi.nlm.nih.gov/genome/?term=Sus) to obtain the mapped data (reads) using the HISAT2 (v2.1.0) program with default parameters. To identify differentially expressed genes (DEGs) between two different groups, the expression level of each transcript was calculated according to the transcripts per million reads (TPM) method. DESeq2 (version 1.24.0) with parameters including P_adjust < 0.05 and |log2FC|≥1 was used to perform DEG analysis before gene expression quantitative analysis was carried out using the RSEM (v1.3.3) program. The DEGs were further analyzed by Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, which were performed using the GOATOOLS (v0.6.5) program and R script with adjusted P_adjust < 0.05, respectively. The P-value was calculated using Fisher's exact test. The heatmap analysis of DEGs was conducted with ClustVis (https://biit.cs.ut.ee/clustvis/). The Venn diagram of DEGs among pairwise comparisons was performed in imageGP (https://www.bic.ac.cn/ImageGP/index.php/Home/Index/VennDiagram.html). Moreover, protein–protein interaction (PPI) network analysis was performed using the STRING server (https://cn.string-db.org/cgi/input?sessionId=bYlaUtwqKXnr&input_page_show_search=on). The raw data were submitted to the Sequence Read Archive (SRA) database (accession: PRJNA799863).

The data were analyzed using GraphPad Prism 8. Comparisons were determined using a two-way ANOVA. Differences were considered significant if the P < 0.05. P-values are indicated as follows: ns > 0.05; ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001.

To determine the replication efficiency of GETV treated with different doses of IFN-α, IFN-ω, IFN-γ, and IFN-λ3, ST cells were infected with GETV (MOI=0.1). Among the three different treatment methods, pre-treatment with IFNs showed the best inhibitory effect on viral replication (data not shown) and was selected for further research. An immunofluorescence assay (IFA) and cytopathic effect (CPE) assay showed that pre-treatment with a high dose (1,000 IU/mL) of recombinant porcine IFN-γ had a significant inhibitory effect on GETV replication in ST cells. In contrast, viral proliferation was not significantly affected by IFN-I (IFN-α, ω) and IFN-III (IFN-λ3), even with a high dose (1,000 IU/mL), as determined through IFA and CPE assays (Figures 1A, B). To verify the antiviral property of IFN-γ against GETV, the copy number of the virus, viral titer, and the expression level of structural protein E2 at different time points under the condition of a single (1,000 IU/mL) or gradient dose (from 1,000 IU/mL to 200 IU/mL) of IFN-γ pre-treatment were detected. As shown in Figure 1F, at 6–48 hpi, the transcriptional level of GETV in GETV-infected ST cells treated with the high dose (1,000 IU/mL) of IFN-γ was significantly lower than that in GETV-infected ST cells without IFN-γ treatment. The suppression effect on the virus titer of GETV was similar to the transcriptional level and appeared to be more remarkable at 12–48 hpi (Figure 1G). IFA and WB assays showed that pre-treatment with a high dose (1,000 IU/mL) of IFN-γ almost completely inhibited the translation level of GETV E2 at 0–48 hpi (Figures 1C, D). These results indicated that the high dose of IFN-γ could exert an obvious suppression on the transcriptional level of GETV, virus titer, and the expression level of the structural protein E2. By comparing the results between Figures 1D, F, almost no protein E2 was detected during 0–48hpi, indicating that the inhibitory effect of IFN-γ on the translation level appears to be stronger than on the transcriptional level if the sensitivity biases between qRT-PCR and WB were not taken into account. Moreover, the expression of E2 could be inhibited by IFN-γ in a dose-dependent manner in the range of 0–800 IU/mL, and complete inhibition could be obtained at all observed time points when the dose is up to 800 IU/mL (Figure 1E). These data may indicate that GETV has evolved mechanisms to antagonize the IFN-I/III responses by targeting the IFN-induced downstream instead of its induction signaling pathway. Furthermore, pre-treatment with IFN-γ could suppress GETV replication, and the inhibiting effect lasted for at least 48 h.

The mRNA-transcriptional levels of mock- and GETV-infected ST cells with or without IFN-γ pre-treatment were determined by high-throughput sequencing. DEGs in host cells from four groups were identified based on a defined criteria (fold change > 2.0 or < 0.5; p_adjust < 0.05). To validate mRNA-transcriptional levels obtained from RNA-Seq, seven genes were randomly selected and simultaneously verified using RT-qPCR. The results showed good correlations of all genes between assays (Figure 2), indicating the accuracy and validity of the RNA-Seq data, which could be used for further analysis, such as GO, KEGG, heatmap, and STRING analyses.

In pairwise comparisons, 280 DEGs (234 upregulated and 46 downregulated) in GETV vs. mock, 546 DEGs (358 upregulated and 188 downregulated) in IFN-γ vs. mock, 701 DEGs (514 upregulated and 187 downregulated) in IFN-γ+GETV vs. mock, 42 DEGs (41 upregulated and 1 downregulated) in IFN-γ+GETV vs. IFN-γ, and 342 DEGs (187 upregulated and 155 downregulated) in IFN-γ+GETV vs. GETV were identified, respectively (Figure 3A). Volcano plots were also drawn (Figures 3B–F). Using this approach, we visualized the transcriptional levels and statistical significance of all genes among groups. Meanwhile, by mapping the clean data to the GETV BJ0304 genome, few clean data in IFN-γ+GETV could be mapped to the GETV genome (Table 1), indicating the good inhibitory effect of IFN-γ on GETV at 20 hpi, which further attested the results observed in Figure 1.

To obtain systematic insights into the host responses to GETV infection and/or IFN-γ stimulation, DEGs in GETV vs. mock, IFN-γ vs. mock, and IFN-γ+GETV vs. GETV were selected and analyzed using the GO and KEGG databases. GO analysis of GETV vs. mock indicated that most of the identified DEGs were involved in cellular processes, biological regulation, and the response to stimuli; were located in the cell part, organelle, and organelle parts; and possessed binding, catalytic activity, and molecular function regulators (Figure 4A). GO analyses of both IFN-γ vs. mock and IFN-γ+GETV vs. GETV showed that the top three enriched cellular components, molecular function, and biological process were almost identical to the corresponding terms of GETV vs. mock (Figures 4A, C, E).

The 20 most significantly enriched pathways of the three pairwise comparisons revealed broad changes in the different biological states. At 20 hpi, DEGs were abundant in host pattern recognition receptor (PRR)-related pathways (i.e., NOD-like receptors (NLRs) (Supplementary Table S2), toll-like receptor (TLR) (Supplementary Table S3), and RIG-I-like receptor (RLR) (Supplementary Table S4) signaling pathway and its downstream pathways, including the NF-kappa B signaling pathway (Supplementary Table S5), the tumor necrosis factor (TNF) signaling pathway (Supplementary Table S6), the IL-17 signaling pathway (Supplementary Table S7), viral protein interaction with cytokine and cytokine receptor (Supplementary Table S8), cytokine-cytokine receptor interaction (Supplementary Table S9), the chemokine signaling pathway (Supplementary Table S10), and the JAK-STAT signaling pathway (Supplementary Table S11). Meanwhile, pathways involved in cell proliferation and differentiation, such as the mitogen-activated protein kinase (MAPK) signaling pathway (Supplementary Table S12) and cellular senescence (Supplementary Table S13), were also significantly affected by GETV infection (Figure 4B). The most significantly enriched signaling pathway of IFN-γ+GETV vs. GETV was antigen processing and presentation (Supplementary Table S14), which was followed by some intriguing pathways such as complement and coagulation cascades (Supplementary Table S15), cell adhesion molecules (Supplementary Table S16), cytokine-cytokine receptor interaction (Supplementary Table S17), and phagosome (Supplementary Table S18), all of which were overlapped in the IFN-γ vs. mock (Supplementary Tables S19–S23). This result indicated that the basic and vital pathways of the host induced by IFN-γ could not be changed by GETV (Figures 4D, F). In addition, IL-17 (Supplementary Table S24), TNF (Supplementary Table S25), and TLR signaling pathways (Supplementary Table S26) were identified in IFN-γ vs. mock (Figure 4D), while hematopoietic cell lineage (Supplementary Table S27) was activated in IFN-γ+GETV vs. GETV (Figure 4F).

To map the transcriptomic profiling of GETV-infected ST cells and preliminarily explore possible key factors or pathways involved in the inhibitory effect on the virus replication of IFN-γ, intriguing DEGs above in GETV vs. mock and IFN-γ+GETV vs. GETV were selected and analyzed.

After GETV infection, a large number of genes involved in the RLR signaling pathway, the NLR signaling pathway, the cytokine-related pathways, and the IFN-response signaling pathway were identified as DEGs and upregulated, some of which were validated by RT-qPCR (Figures 5A, C, D, F, G). Besides, some DEGs (GADD45B, GADD45G, PIK3R1, PMAIP1, MCL1, TNF, and NFKBIA) were enriched in the apoptotic pathway (Figures 5A, E). Furthermore, the results of RT-qPCR showed that GETV infection could upregulate the transcriptional levels of IFN-λ3, as well as IFN-α, despite the existence of a temporal difference (Figure 5G). By combining the results of the antiviral assay (Figures 1A, B), it could be speculated that GETV has evolved the mechanisms to antagonize the IFN-I/III responses by targeting the IFNs induced downstream instead of its induction signaling pathways.

With the pre-treatment of IFN-γ, numerous genes are involved in the antigen processing and presentation pathway (e.g., SLA-1, SLA-2, SLA-3, SLA-DRA, SLA-DRB1, SLA-DQA1, SLA-DQB1, SLA-DOA, SLA-DMA, and SLA-DMB) and complement and coagulation cascades (e.g., C2, C3, C5, C1R, C1S, C4A, CFH, and F2). However, some important antiviral ISGs (ISG12(A), NLRC5, IFI6, OASL, RBP4, and DDIT3) were significantly upregulated in GETV-infected ST cells (Figure 5B). Based on the bioinformatics analysis and the fold changes in mRNA level, 16 genes were selected and verified by RT-qPCR (Figures 5H–J).

After the analysis of each pairwise comparison, the combined analyses were carried out to further understand how IFN-γ inhibits the replication of GETV.

PRRs are predominantly expressed in innate immune cells, such as dendritic cells (DCs) and macrophages, as well as non-immune cells, such as fibroblast cells and epithelial cells. Host PRRs include a special domain such as the ectodomain with leucine-rich repeats (LRRs), which could recognize viral proteins or nucleic acid components termed pathogen-related molecular patterns (PAMPs), thereby initiating intracellular signal transduction through the adaptor protein (STING, MAVS, MyD88) and finally activating the innate immune response to exert an antiviral effect (Thompson et al., 2011; Kawasaki and Kawai, 2014). Nevertheless, few studies about GETV have focused on this issue, and the mechanism of host-GETV interaction remains unknown. In this study, an in-depth insight into the transcriptional changes of GETV-infected ST cells was obtained using RNA-seq, which would help us to explore mechanisms underlying the host-pathogen interaction. Using RNA-seq, 280 DEGs in total, including 234 upregulated and 46 downregulated genes, were identified (Figures 3A, B), and the main PRRs and core genes involved in compelling pathways such as the cytokine-, apoptosis-, and antiviral-related pathways screened by comprehensive analyses (Figures 4B, 5A) triggered our interest.

Among the PRRs, RLRs, including three cytoplasmic RNA helicases (i.e., DDX58/RIG-I, IFIH1/MDA-5, and DHX58/LGP2), are essential for sensing viral RNA and initiating the innate immune response against virus infection. Both DDX58 and IFIH1 are composed of two N-terminal caspase recruitment domains (CARDs), a central DEAD-box helicase/ATPase domain, and a C-terminal regulatory domain (RD). Unlike them, DHX58 lacks the CARD, containing only the helicase domain and RD (Yoneyama et al., 2005; Takeuchi and Akira, 2009). DDX58 and IFIH1 recognize different RNA viruses by detecting short dsRNAs (up to 1kb) with 5′ triphosphate ends and long dsRNAs (>2 kb), respectively (Takeuchi and Akira, 2010). Studies have shown that DDX58 is able to recognize hepatitis C virus (HCV), vesicular stomatitis virus (VSV), Japanese encephalitis virus (JEV), Sendai virus (SV), and influenza virus. IFIH1 can recognize encephalomyocarditis virus (EMCV), mengo virus, Semliki Forest virus (SFV), and poliovirus (Kato et al., 2006; Saito et al., 2008; Pichlmair et al., 2009; Linder et al., 2021). DDX58 and IFIH1 can interact with IPS-1, which then activates signaling cascades leading to the expression of type I IFN genes via EYA4, TRAF3, NAP1/SINTBAD, TBK1/IKK-i, and IRF3/7. DHX58 functions as a positive regulator in RIG-I-mediated and MDA5-mediated virus recognition (Takeuchi and Akira, 2010). In this study, all members of RLRs were significantly upregulated in RNA-seq as expected and in further confirmation at different time points (Figures 5A, C); however, some important dsRNA/ssRNA sensors (e.g., TLR3 and TLR7) were not identified as DEGs. TLR3, a member of the TLR family proteins that is present in the endosome and recognizes dsRNA, can induce type I IFNs via the adaptor protein TIR-domain-containing adapter inducing IFN-β (TRIF) (Matsumoto and Seya, 2008). TLR7 is essential for viral ssRNA-induced type I IFN production (Kato et al., 2005; Takeuchi and Akira, 2010). Scientists are still split about the concept that ZBP1 (DAI) is a dsDNA sensor. Although no dsDNA occurred in the life cycle of GETV as an ssRNA virus, ZBP1 was also identified and significantly upregulated (Figures 5A, C), which may, through recognizing Z-RNA (Jiao et al., 2020), activate the downstream signaling pathway.

In this study, over 15 cytokines/chemokines were upregulated in response to GETV infection. TNF is a key mediator and regulator of immune responses, mainly via the induction of NF-κB and MAP pro-survival kinases and activation of cell death in certain pathological situations to exert its biochemical functions (Varfolomeev and Vucic, 2018). The transcriptional level of TNF was upregulated in a time-dependent manner after GETV infection (Figures 5A, D). Besides, some other important pro-inflammatory cytokines, IL-6 and IL-1α, as well as NLRP3 (NOD-like receptor protein 3), were upregulated to a top degree at 36 hpi (Figures 5A, D), indicating a higher inflammation response for viral clearance at the late phase after infection, which was supported by the report that the expression level of NLRP3 was positively correlated with the peak inflammatory symptoms (Chen et al., 2017). Alphavirus is well known for its capability to trigger the host cell to activate its genetically programmed cell death pathway, leading to the morphological features of apoptosis (Griffin and Hardwick, 1997; Dhanwani et al., 2012; Wang et al., 2012). However, as a member of the genus Alphavirus, this issue about GETV remains poorly reported. In this study, GADD45B, a promoter of apoptosis, was gradually upregulated over time after GETV infection, and a largely consistent situation was also observed in NOXA (PMAIP1) and MCL1 (Figures 5A, E). However, the MCL1/NOXA axis is involved in the intrinsic pathway of apoptosis, where NOXA selectively binds to MCL1 and prevents it from inhibiting apoptosis. Thus, it could be speculated that GETV may regulate host apoptosis in multiple ways or in some feedback manners. The cellular mechanisms involved in GETV-induced apoptosis are complex, and much information needs to be further explored.

Furthermore, induction of IFN-α and IFN-λ3 (Figures 5A, G) and activation of downstream antiviral responses with a series of upregulated ISGs that limit viral proliferation were observed after infection (Figures 5A, F). Among the ISGs, the OAS family could activate the latent ribonuclease (RNase L) to degrade basal mRNAs (Floyd-Smith et al., 1981) and escape the antiviral mRNAs from Rnase L-mediated mRNA decay (Burke et al., 2019), while Mx1 could block the entry of viral nucleic acids into the cell (Nigg and Pavlovic, 2015; Verhelst et al., 2017). Interestingly, Mx1 can also inhibit the production of type I IFN by regulating the phosphorylation of IRF3 in a negative-feedback manner (Schattgen et al., 2016). Thus, it could be speculated that GETV may evolve multiple mechanisms to enable the evasion of IFN-related immune responses.

Although GETV infection boosted inflammatory factors and ISG expression to achieve an antiviral niche or milieu in ST cells, it appears that this antiviral state could affect GETV proliferation only with low magnitude (Figures 1A, B). Meanwhile, viruses have evolved multiple mechanisms to enable evasion of host immune response, such as the IFN system in the process of virus-host interaction (Garcia-Sastre, 2017; Nelemans and Kikkert, 2019), which also occurred in alphaviruses mainly via nsP2, including the examples that CHIKV disables the unfolded protein response of the host (Fros et al., 2015) and that SINV inhibits IFN production (Frolova et al., 2002) by targeting STAT signaling directly (Fros et al., 2016). Following almost non-effective endogenous IFN-α/λ3 (Figures 5A, G), exogenous IFN-α/ω and IFN-λ3 pre-treatment with high biological activity failed to inhibit the viral infection and early replication in vitro (Figures 1A, B). It could be speculated that GETV may have developed strategies that avoid the action of IFN by preventing the binding of viral products to cellular sensors and by inactivating downstream cellular factors involved in IFN signal transduction or the establishment of the antiviral state because the depths of anti-virus ISG and other immune systems, such as complement and coagulation cascades, are limited in the host infected by GETV, indicating that GETV enables the evasion of host immunity through some uncertain mechanisms.

IFN-γ is a pleiotropic cytokine that has immunomodulatory effects, including enhancement of NK- and T-cell-mediated cytotoxicity, B-cell differentiation, surface antigen expression, and macrophage activation, as well as antiviral effects through partially overlapping but distinct signaling cascades with type I/III IFNs (Ramana et al., 2002). Although IFN-γ is an immune cell-confined type II IFN, its receptors (IFNGR1/2) are widely distributed in almost all cell types (Fan et al., 2020), including ST cells (Shan et al., 2019). As a broad-spectrum antiviral agent, besides human viruses such as the Ebola virus (Rhein et al., 2015), Zika virus (Chaudhary et al., 2017), and antiviral-resistant hepatitis C virus (Meissner et al., 2010), IFN-γ has been shown to enable the direct inhibition of numerous porcine viruses, such as the porcine reproductive and respiratory syndrome virus (PRRSV) (Hamana et al., 2017), foot-and-mouth disease virus (FDMV) (Bautista and Molitor, 1999), and African swine fever virus (ASFV) (Fan et al., 2020), indicating significant application potential prospects in the therapeutic or preventive field, which also extends to GETV for the first time in our study. Contrary to IFN-α/λ3/IFN-ω, IFN-γ presented effective inhibitory activity against GETV in a dose-dependent manner in ST cells in this study, and the inhibiting effect did not weaken significantly over a long period (Figure 1), which may lead to the development of novel strategies for the treatment of GETV.

A total of 342 DEGs (187 upregulated and 155 downregulated) between GETV and GETV treated with IFN-γ were identified (Figure 3A). Among them, KEGG and heatmap analysis combined with RT-qPCR showed that, besides antigen processing and presentation (Figures 4F, 5B, H), complement and coagulation cascades (Figures 4F, 5B, I) coupled with some ISGs such as NLRC5, OASL, ISG12A, DDIT3, RBP4, and IFI6 (Figures 5B, J) were activated in response to IFN-γ treatment. Moreover, the combined analysis among pairwise comparisons of GETV vs. mock, IFN-γ vs. mock, and IFN-γ+GETV vs. GETV was performed to further investigate how IFN-γ inhibits the replication of GETV. A total of 20 DEGs were shared among groups (Figure 6A) and clustered in cytokine-related factors, ISGs, and complements (Figure 6B). These results revealed that the activation of complement and coagulation cascades, coupled with some important ISGs, may play synergetic roles in the anti-GETV activity of IFN-γ. Intriguingly, NLRC5, a specific IFN-γ-induced atypical NLR family member that is essential for the IFN-γ-induced activation of MHC class I (Kanneganti, 2010; Meissner et al., 2010), mounted among pairwise comparisons (Table 2), speculating that this ISG may be a cross-talker in the synergetic antiviral effect induced by IFN-γ. Even more noteworthy, OASL possesses an antiviral effect against the single-stranded RNA virus (Marques et al., 2008) and can potentially be used to overcome viral evasion and enhance innate immunity (Zhu et al., 2015). It was upregulated in all pairwise comparisons (Table 2), suggesting that it may be a key molecule for IFN-γ to function in its anti-GETV activity. Previous studies have indicated that the mechanisms underlying the antiviral activity of IFN-γ may vary in different target cell types, and in ST cells, IRF1 plays a critical role in IFN-γ-induced inhibition against transmissible gastroenteritis virus (TGEV) (Shan et al., 2019). In contrast, IRF1 was not significantly upregulated in our study. This inconsistent result may be caused by the different viruses used and requires further exploration and clarification.

Interestingly, a total of 124 DEGs shared in pairwise comparisons of GETV vs. mock and IFN-γ vs. mock presented the identical upregulated tendency (Figure 6A, Supplementary Figure S1), indicating that IFN-γ could strengthen the biological processes induced by GETV infection to play a role in controlling innate antiviral immunity, such as the PRR-related pathways (i.e., NLR and TLR signaling pathways) and its downstream pathways, including the TNF signaling pathway, viral protein interaction with cytokine and cytokine receptor, the IL-17 signaling pathway, and cytokine-cytokine receptor interaction (Figure 6C). Kajita reported that IFN-γ might contribute to the innate immune response to cutaneous viral infections by accelerating TLR3 expression and function in keratinocytes (Kajita et al., 2015). Although numerous DEGs involved in the TLR signaling pathway were identified, none of the TLRs themselves were identified as DEGs. This needs further investigation to be elucidated.

Compared to GETV infection, treatment with IFN-γ could activate additional immune-related genes, and the most significantly enriched pathways were antigen procession and presentation, cytokine-cytokine receptor interaction, complement and coagulation cascades, and the JAK-STAT signaling pathway (Figures 6D–F). It was, therefore, possible that the JAK-STAT signaling pathway, the classical pathway activated by interferons to initiate the transcription of ISGs, may also be indispensable to the antiviral effect of IFN-γ on alphavirus, which was supported by the mechanism for IFN-γ-mediated clearance of SINV infection from mature neurons (Platanias, 2005). Interestingly, several DEGs (i.e., IFNE, IL22RA1, and OSM) enriched in the cytokine-cytokine receptor interaction pathway were involved in the JAK-STAT signaling pathway in this study. Furthermore, complement and coagulation cascades play vital roles in initiating and regulating the innate immune system responses against virus infection, and the individual proteins of this pathway are strictly regulated (Maloney et al., 2020). In this study, DEGs involved in complement and coagulation cascades (i.e., C2, SERPING1, C4A, C1R, C1S, C3, C5, CFH, and F2) were upregulated in GETV-infected ST cells through treatment with IFN-γ (Figure 5B), indicating that the activation of the complement and coagulation cascades may be the pivotal mechanism for the IFN-γ-mediated inhibitory effect of GETV. To our knowledge, antigen processing and presentation, the cornerstones of adaptive immunity (Pishesha et al., 2022), are the traditional immunoregulatory functions of IFN-γ. However, they may solely play an inferior role in the process of defense against GETV replication by IFN-γ in non-immune ST cells. Therefore, it could be speculated that the activation of JAK-STAT signaling, coupled with complement and coagulation cascades, may play synergetic roles in the anti-GETV activity of IFN-γ in ST cells.

In summary, comparative transcriptome profiling between mock- and GETV-infected ST cells was obtained for the first time, and it was inferred that GETV could escape from some innate immune responses, which would help us reveal the precise mechanisms underlying virus-host interaction. Importantly, exogenous porcine IFN-γ showed a dose-dependent antiviral effect against GETV in vitro, which would contribute to the prevention and control of the virus. Furthermore, IFN-γ could exert biological processes analogous to those induced by GETV, and the primary mechanism for IFN-γ-mediated inhibition of GETV proliferation may be related to the synergetic effect of the JAK-STAT signaling pathway and complement and coagulation cascades; NLRC5 and OASL may be the cross-talkers during the synergetic antiviral effect induced by IFN-γ. The abovementioned mechanisms need to be addressed in the future, and the application prospects of IFN-γ on the prevention of GETV should be further investigated in vivo.