Molecular Pathogenesis and Immune Evasion of Vesicular Sto- matitis Virus Inferred from Genes Expression Changes in In- fected Porcine Macrophages

Molecular mechanisms associated with the pathogenesis of Vesicular stomatitis virus (VSV) in livestock remain poorly understood. Several studies have highlighted the relevant role of macrophages in controlling the systemic dissemination of VSV during infection in different animal models, including mice, cattle and pigs. To gain more insight on the molecular mechanisms used by VSV to impair the immune response in macrophages, we used microarrays to determine the transcriptomic changes produced by VSV infection in primary cultures of porcine macrophages. The results indicated that VSV infection induced the massive expression of multiple anorexic, pyrogenic, proinflammatory and immunosuppressive genes. Overall, the interferon (IFN) response appeared suppressed, leading to the absence of stimulation of interferon-stimulated genes (ISG). Interestingly, VSV infection promoted the expression of several genes known to downregulate the expression of IFN. This represents an alternate mechanism for VSV control of the IFN response, beyond the recognized mechanisms mediated by the matrix protein. Although there was no significant differential gene expression in macrophages infected with a highly virulent epidemic strain compared to a less virulent endemic strain, the endemic strain consistently induced higher expression of all upregulated cytokines and chemokines. Collectively, this study provides novel insights into VSV molecular pathogenesis and immune evasion that warrants further investigation. Citation: Zhu J.J; Velazquez-Salinas. L; Canter A.J.; Zhu J.J; Rodriguez L.L. Molecular Pathogenesis and Immune Evasion of Vesicular Stomatitis Virus Inferred from Genes Expression Changes in Infected porcine Macrophages. Pathogens 2021, 10, x. https://doi.org/10.3390/xxxxx Academic Editor: Firstname Lastname Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 30 July 2021 doi:10.20944/preprints202107.0687.v1 © 2021 by the author(s). Distributed under a Creative Commons CC BY license.


Introduction
Vesicular stomatitis virus (VSV) infection causes fever and vesicular stomatitis, one of four clinically indistinguishable viral vesicular diseases. VSV (family Rhabdoviridae, genus Vesiculovirus) is comprised of a non-segmented RNA viral genome encoding five structural proteins: nucleocapsid (N), phosphoprotein (P), matrix (M), glycoprotein (G), and the large RNAdependent RNA polymerase (L) (1, 2) along with two non-structural proteins (C and C0) of undetermined function encoded in overlapping reading frames of the P gene (3). VSV causes most of the cases of vesicular diseases reported in livestock resulting in economic losses associated to quarantines imposed by animal health authorities due to its similar clinical presentation with foot and mouth disease virus (FMDV) (4,5).
VSV has a broad host range and cell tropism due to its glycoprotein binding to host LDLR family members that are ubiquitously expressed on host cells and conserved among mammalian species (6,7). Besides typical vesicular lesions in specific tissues, infected animals also show systemic signs such as anorexia, lethargy and fever (https://en.wikivet.net/Vesicular_Stomatitis_Virus). Despite these clinical signs, VSV infection typically does not result in host mortality (4,5). After infection via insect bites, animals show limited virus replication, primarily in specific tissues where the vesicular lesions occur. Infected animals usually recover completely within 2-3 weeks of infection (4,5).
Unlike gross pathogenesis, the molecular pathogenesis of VSV is not very clear. Based on literature review, it appears that only TNF has been investigated in VSV pathogenesis, showing more rapid induction of TNF by an attenuated VSV mutant after infection, but more drastic TNF induction later in infection by wildtype VSV in mice (8). TNF knockout mice showed diminished weight loss following wildtype VSV infection, and the rapid weight loss seen in wildtype VSV infection was less pronounced in C57BL/6 mice infected by an attenuated mutant virus (8). In mice, interferons produced by VSV-infected macrophages play a key role in protection against neuropathogenesis of the virus (9). In the natural hosts such as cattle, VSV antigens were colocalized with an antibody against a marker molecule (MAC387, MRP14 or S100A9) of myeloid cells including macrophages using immunohistochemistry (10). Both wildtype VSV and matrix protein mutants productively replicate in porcine immune cells and non-immune cells (11,12).
Infection with wildtype VSV induced weaker proinflammatory cytokine responses and downregulated the expression of the costimulatory molecule complex CD80/86 and MHC class II 3 of 28 compared to the matrix protein mutant virus (11). A matrix protein (M51R) VSV mutant virus replicated ~1000 times less in cultured primary porcine macrophages than its wildtype counterpart and showed significantly diminished virulence in pigs (13). The molecular pathogenesis and immune evasion in natural hosts such as pigs and cattle have yet to be investigated.
It is well-known that VSV can inhibit the host interferon response primarily via its matrix proteins (14,15). VSV matrix protein mRNA can be translated into three proteins starting at three in-frame start codons (16). Transfection with plasmids containing the M protein gene alone can induce CPE in transfected cells (16). VSV M proteins can delay apoptosis induced by other viral components (17) and suppress transcription in infected cells by inhibiting the basal transcription factors TFIID and TFIIH and interacting with host Rae1 and Nup98 (18,19,20,21,22). VSV M proteins can also inhibit nuclear export of host mRNA and snRNAs (23) and NFκB activation (24).
The suppression of IFNβ expression by the matrix protein is correlated with the inhibition of host RNA and protein synthesis (25). A systems biology approach including transcriptomic analysis has been conducted to study VSV infection in a murine macrophage cell line (26,27 ); however, VSV pathogenesis and immune evasion were not explicitly explored based on transcriptional changes after virus infection. Although mice have been extensively used as an experimental model for VSV infection, they are not natural hosts for VSV infection. The transcriptomic analysis of VSV infection has not been investigated in the primary macrophages of its natural livestock hosts .
Macrophages play an important role in host defense against pathogens via positioning in all tissues where they can effectively sense danger signals with highly expressedPAMP receptors and producing a large quantity of both pro-and anti-inflammatory cytokines such as IL-1, IL-10, TGFβ and TNF via cell polarization and differentiation to regulate the immune response (28). Our previous study showed that primary porcine macrophages expressed higher levels of IFNβ and cytokines than primary fetal porcine kidney cell cultures after VSV infection (13). Given that VSV can infect and replicate in macrophages and the important role of macrophages in the immune response, ex vivo porcine macrophages were used as model cells to extrapolate the molecular mechanisms of VSV pathogenesis and immune evasion. The objective of this study was to formulate hypotheses for the molecular mechanisms of VSV pathogenesis and immune evasion based on gene expression changes in porcine macrophages after infection for further investigation.

Differential gene expression
There were no genes differentially expressed between macrophages infected with epidemic VS New Jersey (NJ0612NME6) and endemic New Jersey (NJ0806VCB) strains (minimal FDR = 0.13).
There was a total of 4,346 significant differentially expressed genes at a false-discovery rate (FDR) of ≤ 0.05 with at least a 50% difference and a total of 3,345 with a difference of 2-fold or greater between epidemic VSV and mock infected macrophages. Between epidemic VSV and mock infected macrophages, there were a total of 3,345 significant differentially expressed genes (DEG) by at least 50% at a false-discovery rate (FDR) of ≤ 0.05. Among these DEG, the majority were detected as being downregulated (2,179 DEG) compared to 1,166 upregulated genes between VSV-infected and mockinfected cells, which was at approximately 2:1 ratio. Forty four percent of DEG were differentially expressed by 1.5-to 2.0-fold with 841 of these genes being downregulated, and 618 being upregulated ( Figure 1). There were 54% of DEG with a fold change between 2.0 and 5.0, and the largest proportion of genes, 1288, were downregulated compared to 521 genes upregulated. Finally, the remaining 2.3% of the DEG were differentially expressed with a fold change greater than 5 (27 genes downregulated and 50 genes upregulated) ( Figure 1). The most drastic differences were at a fold change of 10.2 for a downregulated gene, and 32.8 for an upregulated gene between these VSV-and mock-infected macrophages. Figure 1. The distribution of fold-changes of annotated genes with significantly differential expression equal to or greater than 2-fold between VSV-and mock-infected macrophages from most upregulated (32.8-fold) to most downregulated (-10.2-fold) genes in VSV-infected macrophage compared to mock-infection

Pathway analyses
To identify biological pathways/processes most impacted by the differential expression, the lists of DEG with differential expression of a fold change of at least 2-fold to remove DEG with minor effects were used in the DAVID analysis. GO term analysis showed that NFκB signaling pathway was the most over-represented by the DEG with three other significant biological processes in protein ubiquitination, Toll-like receptor signaling, and mRNA transcription regulation (Table 1). KEGG pathway analysis identified eleven over-represented biological pathways with top five pathways (TNF, TLR, NFkB, RIG-I-like receptor, and NOD-like receptor signaling) that are known to play key roles in the immune response (Table 1). Only one biological pathway (TNFinduced apoptosis) was detected with REACTOME analysis.
When this list of DEG was further narrowed to those with a fold change of at least 4 (more biologically impactful DEG), GO term analysis identified thirteen over-represented biological processes; six involved in the immune response, four in apoptosis, two in signaling pathways and one in RNA transcription. The two most over-represented GO terms were in inflammatory response (GO_0006954) and apoptotic process (GO_0006915) ( Reactome-HSA-380108: Chemokine receptors bind chemokines 7 0.003 Table 1. Gene ontology terms, Kyoto Encyclopedia of Genes and Genomes (KEGG) and REACTOME biological pathways over-represented by genes differentially expressed by at least 2-and 4-fold between VSV-infected and mockinfected porcine macrophages using the NCBI DAVID program with Benjamini p-value correction

Interferon expression and signaling
VSV infection significantly induced the expression of IFNB by 6.2-fold, but not other interferons, in infected macrophages compared to mock-infected cells ( Table 2). The endemic strain induced higher IFNB expression (2-fold) than the epidemic strain but not at the significant level. VSV infections significantly suppressed the expression of an IFNA homologous to human IFNA17 by approximately 2-fold and did not significantly alter expression of other interferons including types II and III. The expression of type I (IFNAR1 and IFNAR2) and II (IFNGR1) receptors were suppressed in VSV-infected cells compared to mock-infection ( Table 2). The type III IFN receptor (IFNLR1) was expressed at a very low level in the macrophages (signal intensity = 69, SNR < 2). The expression of typical interferon stimulated genes (ISGs) was not significantly changed by VSV-infection (only 10 genes listed in Table 2). These results indicate that VSV infection suppressed type I IFN and II IFN signaling.

Immune signaling pathways
The expression of a transcription factor (ATF2) and five MAPK kinases (MAP2K5, MAPK14/p38, MAPK4, and MAP3K18) in the MAPK signaling pathways was significantly downregulated in VSV infected cells compared to mock-infected cells (Table 3). Likewise, the expression of six activator genes (CARD6, IKBKB, IRAK1, NLK, TAB1, TAB2, and TAK1) in the NFκB pathway was significantly downregulated in VSV-infected cells, whereas the expression of three inhibitors of NFκB [NFKBIA, NFKBID, and TNFIP3/A20 (38) was significantly upregulated (Table 3). Three genes (IRF5, MAVS and TBK1) in the RIG-I signaling pathway were expressed at significantly higher levels in VSV-infected cells than in mock-infected cells ( Table 3). The expression of four TLR receptors (TLR1, TLR2, TLR4 and TLR6) and two signal transducers (BTK and TRIF) was downregulated in VSV-infected cells compare to mock-infected cells, whereas TLR7 was upregulated (  Table 3. Expression levels (EXP), false discovery rates (FDR) and fold differences (epidemic vs mock infection: EP/M, epidemic vs endemic infection: EP/EN) of interferon expression regulating genes differentially expressed between infected-and/or mock-infected macrophages

Cytokines, chemokines, and receptors
VSV infection significantly induced the expression of five immune cytokines (CSF3, IL1A, IL10, IL27, TNF and TNFSF9) and suppressed TNFSF11 expression (Table 4). Four non-typical immune cytokines (AREG, HBEGF, LIF and VEGFA) were expressed at significantly higher levels in VSVinfected than mock-infected cells (Table 4). Among those cytokines, AREG, IL1A, IL10, LIF and TNF were upregulated by > 11-fold. Overall, the endemic strain induced consistently higher expression (averaging 1.7-fold) of the upregulated cytokines than the epidemic strain though at not significant levels, whereas the receptor expression was very similar (Table 4). There were three significantly downregulated (IL17RA, LTBR and TNFRSF1A) and three upregulated (TNFRSF10, IL1R2 and IL20RB) cytokine receptors in VSV-infected cells compared to mock-infected cells ( Table 4). All these DEG are proinflammatory genes with the exception of IL10, IL1R2, IL20RB, and the four non-typical immune cytokines. These results show that VSV infection induced both pro-and anti-inflammatory cytokine expression and suppressed the expression of IL-17 and TNF receptors.
The expression of eight autophagy-associated genes including seven ATGs and FLCN and positive autophagy regulator, ULK1 (47) and RB1CC1 (48) Rab33b (49) was significantly lower in VSV-infected cells than in mock-infected cells ( Table 6). Two autophagy inhibitors, BCL2L11/BIM (50) and Gadd45b (51) were expressed at significantly higher levels in VSV-infected cells than in Mock-infected cells (Table 6). GADD45B was one of the top 10 most induced genes after VSV infection (Table 10).
The expression of XBP1, a key regulator in stress-induced unfolded protein response (UPR) (52) and ERN1, the ER stress sensor of UPR (53), were significantly downregulated in VSV-infected cells compared to mock-infected cells (

Host mRNA transcription, modification, and stability
Thirteen genes involved in transcription of host RNA based on KEGG pathways were significantly downregulated in VSV-infected cells compared to mock-infected macrophages ( Table 7). The   Table 7. Expression levels (EXP), false discovery rates (FDR) and fold differences (epidemic vs mock infection: EP/M, epidemic vs endemic infection: EP/EN) of transcription-and translation-related genes differentially expressed between infected-and/or mock-infected macrophages

Inflammation-related genes
Two proinflammatory mediator genes, ADM, (57), and a prostaglandin E (PGE) synthase (PTGS2) (58) were expressed at significantly higher levels (5.6 and 15.6 times, respectively) after VSV infection (Table 8). MALAT1 and DUSP2 play a role in prostaglandin E2 production in macrophages (59, 60), and their expression was also upregulated by VSV infection ( A multitude of genes associated with pro-inflammatory responses were also differentially affected by VSV infection compared to mutant vs. mock infection (Table 8). Two genes MAP3K8   Table 8. Expression levels (EXP), false discovery rates (FDR) and fold differences (epidemic vs mock infection: EP/M, epidemic vs endemic infection: EP/EN) of genes that are associated with macrophage immunity and were differentially expressed between infected-and/or mock-infected macrophages

Discussion
The and CXCL2 (>20 times) and a PGE synthetase gene (PGTS2). CCL3, CCL4 and CCL5 share the same receptor (CCR5). Likewise, CXCL1 and CXCL3 also share the same receptor (CXCR1) with CXCL2 (39). These chemokines were also massively induced after VSV infection. High production of these cytokines and chemokines and PGE is known to induce fever (72,73,74). Interestingly, LIF expression was highly upregulated after VSV infection, and LIF injection can induce fever in animals (75,76). Based on these results, it is hypothesized that high fever is mainly caused by VSV-induced high production of CCR5 and CXCR1 ligands, IL1A, LIF, PGE and TNF. These cytokines, chemokines, and PGE are potent mediators of inflammation. IL1A and TNF are well-known potent proinflammatory cytokines. PGE induces vasodilatation and local recruitment of neutrophils, macrophages, and mast cells at early stages of inflammation (77,78). Chemokines induced by VSV infection including CCL3, CCL4, CCL5, CCL20, CXCL1, CXCL2 and CXCL3 recruit macrophages, NK cells, neutrophils and/or Th17 cells (39). CSF3 stimulates neutrophil generation in the bone marrow (79). The expression level of CSF3 was significantly increased after VSV infection. Neutrophils are known to play a key role in clearance of viruses via phagocytosis and neutrophil extracellular traps (80). Therefore, the high production of PGE, cytokines and chemokines Other cytokines highly induced by >10-fold include AREG, IL-10 and LIF, which are known to have immune-suppressive effects. High levels of IL-10 suppress the innate and adaptive immune responses (83). TNF leads to IL-10 production by monocytes and together with IL-10, inhibit of CD4 T-cell expansion and function (84). LIF can suppress IFNγ and LPS signaling (85,86). LIF appears to be an immune-tolerogenic cytokine based on promoting Treg differentiation and inhibiting pro-inflammatory Th17 cell differentiation (87). Several growth factor such as VEGF and EGFs can inhibit IFNB expression (35) or suppress the anti-VSV activity of IFNα and IFNβ (88).
The expression of VEGF and two EGFs (AREG and HBEGF) were induced after VSV infection in this study. PGE2 selectively suppresses effector functions of macrophages and neutrophils and the Th1-, CTL-, and NK cell-mediated type 1 immunity, but it promotes Th2, Th17, and regulatory T cell responses (77,78). Therefore, we hypothesize that high levels of EGFs, IL-10, LIF, PGE and VEGF could play a key role in suppressing the immune response of infected and non-infected cells to facilitate VSV infection and cause disease.
Our results also indicate that VSV can evade the immune response of infected cells by various mechanisms. It is known that VSV can activate IFN response via RIG-I-MAVS and TLR4/CD14 signaling pathways to induce an antiviral response (14,89). The expression of two key signaling transducers, TBK1 and MAVS (90,91) in the RIG-I-MAVS signaling pathway was downregulated in VSV-infected cells compared to mock-infected cells. It has been previously reported that VSV glycoprotein binds to the TLR4/CD14 dimer leading to the induction of interferon expression, mainly mediated by IFNB via a TICAM1/TICAM2-dependent but MyD88-and NFκB-independent signaling pathway (92). In our study, the expression of TLR4 and TICAM2 was significantly sup- Previous studies of IFNB promoters showed that ATF2-JUN, IRF, and NFκB transcription factors regulate IFNB expression (95). The expression of ATF2 and IRF5 transcription factors was downregulated by VSV-infection. There were several downregulated signaling transducer or upregulated signaling inhibitor genes that could inhibit MAPK, NFkB, signaling pathways (Table 2).
Interestingly, there were six suppressor genes of IFNB expression (AHR, ATF3, DUSP1, FOS, HES1 and PRDM1) upregulated by up to 32.8-fold in VSV-infected cells. This result supports published results that VSV suppresses the interferon response.
Our results show that VSV infection did not induce expression of other type I interferons except IFNB in pig macrophages. IFNB induction is known to induce IRF7 expression, which is needed for induction of IFNA (96). Although IFNB expression was induced in infected macrophages, the expression of IRF7 and IFNA was not increased by VSV infection. This could be explained by the suppression of interferon signaling mediated by downregulated expression of type I and II interferon receptors as shown in Table 6. This seems to a novel immune evasion mechanism of VSV in addition to the inhibition of mRNA nuclear export mediated by VSV matrix protein (14,15). It has been previously reported that VSV infection inhibits the expression of interferon-stimulated genes via miR-132 to facilitate viral replication (97,98). We found that miR-132 is highly upregulated by VSV infection (Table 11) which could also explain the lack of induction of interferon stimulated genes by IFNB.
Viral infections trigger three inter-connected biological processes including apoptosis, autophagy, and stress-induced unfolded protein response (UPR), which can inhibit virus replication (52). However, viruses can subvert or even manipulate these responses to promote infection, for example, VSV can delay the onset of apoptosis (17,99). Our results show that multiple genes associated with these three processes were differentially expressed during VSV infection as listed in Table 10. Among these genes, GADD45B which is one of top 10 most induced genes (22.3-fold) after VSV infection, suppresses apoptosis and autophagy (51). The expression of two key regulatory genes, XBP and PPP1R15A, in the UPR pathway was reduced after VSV infection. Additionally, genes in death receptor signaling including TNFRSF1A and several signaling transducers were downregulated in VSV-infected cells, which could delay the necroptosis/apoptosis induced by high expression of TNF. These results suggest that VSV can suppress these three important innate immune mechanisms during infection.
The chemokines associated with the CCR5 receptor activate macrophages to induce proinflammatory cytokine expression (100,101). Expression of CCR5 was significantly downregulated in VSV-infected macrophages, potentially reducing the immunostimulatory effect of CCR5. Chemokines. Additionally, high IL10 and LIF expression and downregulation of IL17RA in the infected cells could mitigate the effects of Th17 cells recruited by increased CCL20 expression.
Therefore, VSV appears to be able to evade the immune response associated with chemokines in- It has been reported that VSV interferes with host gene transcription (19). Our results showed that VSV downregulated several genes of the host transcription machinery as shown in Table 11. Additionally, VSV may control host protein translation by altering expression of genes dephosphorylation-mediated inhibition of viral protein translation. This study provides novel insights (summarized in Table 9) that warrant further investigation of VSV virulence factors and pathogenesis.  Table 9. Differentially expressed genes used to infer candidate mechanisms of VSV systemic and tissue pathogenesis and immune evasion in infected and non-infected cells

Cell culture of macrophages and viruses
Primary swine macrophage cell cultures were derived from pig peripheral blood as previously described (107) (108). Differences between these two viruses have been reported in previous studies suggesting that NJ0612NME6 has higher virulence than NJ0806VCB in inoculated pigs (12). VSV infection experiments were conducted with three biological replicates using ex vivo cultured primary macrophages isolated from three different commercial domestic pigs. Macrophages were infected with a MOI of 10 TCID50 of each virus, respectively. Mock infection was also performed in the cultured macrophages from these same pigs as non-infected controls.

RNA isolation
Total RNA was extracted from primary swine macrophage cell cultures infected with the indicated viruses, or mock infected at 5 hours post-infection. Cells were harvested and lysed with a cell lysis buffer (Qiagen, Valencia, CA) and RNA was isolated using a RNeasy mini kit (Qiagen) according to the manufacturer's instructions. The RNA quality was then determined using an Agilent 2100 bioanalyzer (Santa Clara, CA) using an RNA nanochip according to the procedures outlined by Agilent Technologies (Santa Clara, CA). RNA was quantified using a Nanodrop 1000 (Thermo Scientific, Waltham, MA). A 44,000 (44K) porcine whole genome expression microarray was designed based on pig expressed sequences (cDNA and EST) and porcine genome sequence homologous to non-porcine sequences as reported by Zhu et al. (109). All porcine EST and RNA sequences were downloaded from the NCBI database and assembled into unique sequences using the CAP3 software program (Huang and Madan, 1999). The assembled sequences were aligned to pig genome sequences using the UCSC genome browser to select 3' end RNA sequences or the genome sequences aligned with other expressed sequences of other species if no porcine expressed sequences were available. These selected sequences were used to design 60-mer oligonucleotide microarray probes with a low probability of cross-reacting with other genes and a bias to the 3'-end of RNA sequences using Array The custom designed porcine microarrays were manufactured by Agilent Technologies and used for this study. Both ASFV-infected and mock-infected RNA samples were labeled with Cy3 and Cy5 individually using an Agilent low-input RNA labeling kit (Agilent Technologies). A Cy5-labeled ASFV-infected or mock-infected sample was co-hybridized with a Cy3-labeled mockinfected or ASFV-infected in one array, respectively, for each time point using a dye-swap design.

DNA microarray analysis
The entire procedure of microarray analysis was conducted according to protocols, reagents and equipment provided or recommended by Agilent Technologies. Array slides were scanned using a GenePix 4000B scanner (Molecular Devices) with the GenePix Pro 6.0 software at 5 μM resolution.

Statistical and bioinformatic analyses of microarray data
Background signal correction and data normalization of the microarray signals and statistical analysis were performed using the LIMMA package (110). Log2 fold changes in signal intensity were used in the statistical analysis to identify deferentially expressed genes. To account for multiple testing, the p-values were adjusted using the Benjamini and Hochberg method and expressed as a false-discovery rate (FDR). The probe sequences were aligned to the porcine genome sequence displayed in the UCSC genome browser to validate the annotation by computational methods, such as BLAST. Gene expression differences with an FDR value of 0.05 or smaller and an expression difference ≥ 50% were considered statistically significant and were considered differentially expressed genes (DEG). Genes down-or upregulated in the infected macrophages compared to the non-infected macrophages were expressed as negative and positive values (fold), respectively.

Pathway analyses
The identified DEG were mapped to human reference genes. Two lists of upregulated and downregulated gene associated with human Entrez gene ID were analyzed with a NCBI online bioinformatics program (DAVID Bioinformatics Resources 6.8) to identify the biological pathways (GOTERM_BP_DIRECT, KEGG_PATHWAY and REACTOME_PATHWAY) significantly over-represented by DEG (P ≤ 0.05 with Benjamini correction). The DEG with differential expression of 2fold or greater and 4-fold greater were used in these analyses to take the magnitudes of differentiation expression into consideration.

Biological inference
The biological functions of DEG in the identified over-represented pathways associated with the immune response were based on scientific publications obtained from PubMed. Biological inferences were based on (i) the immunological functions of the DEG, (ii) gene expression levels based on microarray averaged signal intensity and (iii) magnitudes (fold) of the differential expression, assuming higher mean signal intensity and larger differentially expressed genes play a bigger biological role in the gene groups. Genes with no significantly differential expression but are known to play important roles in the biological pathways associated with the significant DEG were also used as supporting evidence. Genes down-or upregulated in the VSV-infected samples compared to the mock-infected samples were expressed as negative and positive values (fold), respectively. In this study, genes differentially expressed between infected and mock infected macrophages were used to infer the molecular mechanisms of VSV pathogenesis and immune evasion.

Data Availability Statement:
The microarray raw data are in the process of submitting to NCBI database. The data sets will be available to the public if this manuscript is accepted for publication.