A balanced game: chicken macrophage response to ALV-J infection

Avian leukosis virus subgroup J (ALV-J) infection can cause tumors and immunosuppression in infected chickens. Macrophages play a central role in host defense against invading pathogens. In this study, we discovered an interesting phenomenon: ALV-J replication is weakened from 3 hours post-infection (hpi) to 36 hpi, which was verified using Western blotting and RT-PCR. To further investigate the interaction between ALV-J and macrophages, transcriptome analysis was performed to analyze the host genes’ function in chicken primary monocyte-derived macrophages (MDM). Compared to the uninfected control, 624 up-regulated differentially expressed genes (DEG) and 341 down-regulated DEG at 3 hpi, and 174 up-regulated DEG and 87 down-regulated DEG at 36 hpi were identified in chicken MDM, respectively. ALV-J infection induced strong innate immune responses in chicken MDM at 3 hpi, instead of 36 hpi, according to the analysis results of Gene Ontology and KEGG pathway. Importantly, the host factors, such as up-regulated MIP-3α, IL-1β, iNOS, K60, IRG1, CH25H, NFKBIZ, lysozyme and OASL were involved in the host defense response during the course of ALV-J infection. On the contrary, up-regulated EX-FABP, IL4I1, COX-2, NFKBIA, TNFAIP3 and the Jak STAT pathway inhibitors including CISH, SOCS1 and SOCS3 are beneficial to ALV-J survival in chicken macrophages. We speculated that ALV-J tropism for macrophages helps to establish a latent infection in chicken MDM from 6 to 36 hpi. The present study provides a comprehensive view of the interactions between macrophages and ALV-J. It suggests the mechanisms of defense of chicken macrophages against ALV-J invasion and how ALV-J escape the host innate immune responses. Electronic supplementary material The online version of this article (10.1186/s13567-019-0638-y) contains supplementary material, which is available to authorized users.


Introduction
Avian leukosis virus subgroup J (ALV-J) is an oncogenic retrovirus, primarily inducing neoplastic diseases and reproduction problems in infected chickens. It is well known that ALV-J causes enormous economic loss in the global poultry industries. To date, there are no vaccines or treatments to protect against ALV-J infection. Since little is known about the interaction between ALV-J and the host, current strategies are focused on ALV-J eradication. RNA viruses are prone to mutations. In contrast to the virus, the host does not change quickly. It is therefore an enticing strategy to try to overpower ALV-J by finding ways to make chickens less permissive to viral replication.
Studies concerning host innate and adaptive immune responses to ALV-J are in their infancy [1]. Macrophages are found in all tissues and have a well-defined role in host responses against viral infection [2]. However, macrophages are susceptible to infection for human immunodeficiency virus (HIV) [3], dengue virus [4], and porcine reproductive and respiratory syndrome viruses [5]. In particular, macrophages serve as a reservoir throughout HIV infection [3]. Importantly, macrophages also play a key role in avian viral infections including infectious bursal disease virus (IBDV) [6], avian influenza virus (AIV) [7], Newcastle disease virus (NDV) [8] and infectious bronchitis virus (IBV) [9]. However, the role of macrophages in ALV-J infection remains unclear.
In our previous study, we found that primary chicken monocyte-derived macrophages (MDM) were susceptible to ALV-J and infection resulted in expression of immune-related genes [10]. However, the number of genes we examined was too low to comprehensively map the involvement of immune host factors in an ALV-J infection. The goal of the current study was to examine host gene expression profile to improve our understanding of the relationship between macrophages and ALV-J during infection.
In this study, RNA-seq analysis platform and gene overexpression verification were employed to analyze chicken MDM gene expression after ALV-J infection. Our findings provide a comprehensive view of ALV-J immune escape and the host defense response to ALV-J infection in chicken macrophages.

Animals and viruses
A total of 12 six-week-old specific-pathogen-free (SPF) White Leghorn chickens, half females and half males, were purchased from Guangdong DaHuaNong Animal Health Products Co., Ltd (Guangzhou, China) and housed under pathogen free conditions. Laboratory ALV-J strain SCAU-HN06 was kindly provided by Prof. Weisheng Cao, South China Agricultural University. All animal experiments were performed with approval and guidance from South China Agricultural University Institutional Animal Care and Use Committee.

Culture of primary chicken MDM
Chicken primary MDM were cultured and identified according to previous studies [10,11]. Briefly, peripheral blood mononuclear cells (PBMC) were isolated from blood obtained from SPF chickens using chicken lymphocyte separation medium (Solarbio, Beijing, China) according to the manufacturer's instructions. The supernatant was removed and adherent cells were washed twice with PBS to remove thrombocytes, non-adherent lymphocytes and other semi-adherent cells after 6 h of incubation. These adherent cells were primarily chicken monocytes. Subsequently, fresh RPMI-1640 medium with 15% chicken serum, 100 U/mL penicillin and 100 mg/mL streptomycin were added to the remaining monocytes. Chicken monocytes were then cultured for 6 days to generate mature macrophage differentiation. The culture medium was changed every 2 days in order to ensure stable and consistent conditions.

Detection of ALV-J replication in MDM
Chicken MDM were infected with a 10 5 TCID 50 /mL of ALV-J strain SCAU-HN06. DNA, RNA and total proteins were extracted from the ALV-J infected MDM at 3, 6, 12, 24 and 36 h post-infection (hpi). RT-PCR was employed to detect the ALV-J replication using specific PCR primers H5/H7 [12]. Western blotting was performed with ALV-J envelope protein specific mouse antibody JE9 (kindly provided by Dr Aijian Qin, Yangzhou University, Yangzhou, China) and rabbit anti-β-actin antibody (Bioworld, Louis Park, USA) according to the method described previously [13]. IRDye 700DX-conjugated antirabbit IgG and IRDye 800-conjugated anti-mouse IgG (Rockland Immunochemicals, Limerick, PA, USA) was used as the secondary antibody. Membranes were visualized and analyzed with an Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE, USA). ALV-J provirus was detected by PCR with primers H5/H7 using DNA template.

Total RNA isolation
Total RNA for RNA sequencing (RNA-Seq) was isolated from pooled MDM (isolated and cultured from 12 SPF chickens) infected with ALV-J (10 5 TCID 50 /mL) at 3 and 36 hpi using TRIzol reagent (Invitrogen, CA, USA). Samples were collected from two independent experiments. Non-infected MDM were used as a control group. Purity and quantity of total RNA were assessed using the NanoPhotometer ® spectrophotometer (Implen, CA, USA) and the Bioanalyzer 2100 system (Agilent Technologies, CA, USA). RNA degradation and contamination were further monitored using agarose gel electrophoresis.

Library preparation for mRNA sequencing
After quality inspection, approximately 3 μg RNA per sample was used as input material for the RNA sample preparations. Briefly, ribosomal RNA was first removed using the Ribo-zero ™ rRNA Removal Kit (Epicentre, WI, USA), and rRNA free residue was cleaned up by ethanol precipitation. Subsequently, sequencing libraries were generated using the rRNA-depleted RNA by NEBNext ® Ultra ™ Directional RNA Library Prep Kit for Illumina ® (NEB, MA, USA) according to the manufacturer's recommendations. First strand cDNA was synthesized with random hexamers and M-MuLV Reverse Transcriptase. Second strand cDNA synthesis was subsequently performed using DNA Polymerase I and RNase H. In the reaction buffer, dNTP containing dTTP were replaced with dUTP. Remaining overhangs were converted into blunt ends via the exonuclease and polymerase activities. After adenylation of 3′ ends of DNA fragments, NEB-Next Adaptor with hairpin loop structure were ligated to prepare for hybridization activities. In order to select cDNA fragments of preferentially 150-200 bp in length, the library fragments were purified with AMPure XP system (Beckman Coulter, Beverly, USA). Then 3 μL USER Enzyme (NEB, Ipswich, MA, USA) was used with size-selected, adaptor-ligated cDNA at 37 °C for 15 min followed by 5 min at 95 °C before PCR. Then, PCR was performed with Phusion High-Fidelity DNA polymerase, universal PCR primers, and Index (X) Primers. At last, products were purified (AMPure XP system) and library quality was assessed on the Agilent Bioanalyzer 2100 system. The mRNA libraries were sequenced at the Novogene (Beijing, China) on an Illumina Hiseq 2000 platform.

Data analysis of mRNA
Raw data (raw reads) of fastq format were first processed through in-house perl scripts. In this step, clean data (clean reads) were obtained by removing adapter sequences as well as reads containing poly-N and low quality reads. Therefore, only high quality data were analyzed and quality scores (Q20 and Q30) and GC content were subsequently calculated. All the following analyses were based on the clean data with high quality. Reads were mapped to the chicken genome assembly [14] using Tophat (v2.0.9). The mapped reads of each sample were assembled by both Scripture (beta2) and Cufflinks (v2.1.1) in a reference-based approach.

Quantification of gene expression level
The FPKM (fragments per kilo-base of exon per million fragments mapped) was calculated based on the length of the fragments and reads count mapped to this fragment. Cuffdiff (v2.1.1) was used to calculate FPKM of coding genes in each sample. Gene FPKM were computed by summing the FPKM of transcripts in each gene group.

Differential expression analysis
Cuffdiff software was used to provide statistical routines for determining differential expression in digital transcripts or gene expression data using a model based on the negative binomial distribution. In the present study, for differentially expressed genes (DEG), the threshold was q value < 0.05, log2 |(fold change)| ≥ 1 with an FPKM value no less than 10 in infected or uninfected samples.

Gene ontology, and pathway analysis
DEG were subjected to Gene Ontology (GO) categorization and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) version 6.8 [15].

Validation of gene expression in RNA-seq by quantitative real-time PCR (qPCR)
Total RNA was extracted from ALV-J-infected (10 5 TCID 50 /mL) and uninfected MDM at 3 hpi and 36 hpi using RNAiso reagent (TaKaRa, Japan). For gene expression analysis, cDNA synthesis of mRNA was performed using a PrimeScript RT Reagent Kit (Perfect Real Time) (TaKaRa, Japan) according to the manufacturer's protocol. The qPCR primers were designed using the NCBI Primer BLAST program [16] and were based on published target sequences (Additional file 1A) [17][18][19][20]. The GAPDH gene was used as an internal control. qPCR was performed on a Bio-Rad CFX96 Real-Time Detection System using iTaqTM Universal SYBR ® Green Supermix Kit reagents (Bio-Rad, CA, USA) according to the manufacturer's specifications. Data analyses were performed using the 2 −ΔΔCt method.

Transfection of up-regulated DEG and detection of their function on ALV-J replication
MDM were cultured in 12-well plates and transfected with 1 μg plasmids including K60, IRG1, OASL, CH25H, CISH, EX-FABP, IL4I and SOCS3 using Lipofectamine 3000 reagent, respectively. EGFP was used as a control. After incubation for 4 h, Lipofectamine 3000 transfection reagent was removed, and the cells were replenished with RPMI-1640 medium with 15% chicken serum. 24 h later, the transfected MDM were infected with 10 5 TCID 50 /mL of ALV-J strain SCAU-HN06. At 3 hpi, ALV-J replication level was analyzed by Western blot and qPCR. The primers used in the construction of these plasmids are summarized in Additional file 1B.

Statistical analyses
Statistical comparisons were performed using GraphPad Prism 5 (GraphPad Software Inc., San Diego, CA, USA). The results are presented as the mean ± SEM. Two-way ANOVA analysis was used to analyze the statistical significance among multiple groups and unpaired Student's t-test between two groups. Statistical significance is indicated by p values of > 0.05 (non-significant, ns), < 0.05 (*), 0.01 (**) or 0.001(***).

Raw data information
The sequencing data obtained from RNA-Seq were released to the GEO database under the accession numbers GSE103207.

Detection of ALV-J in chicken MDM
ALV-J infections of MDM resulted in either genome integration or reverse transcription into cDNA during 3-36 hpi ( Figure 1A). The rate of ALV-J replication was very high at 3 hpi and gradually decreased from 6 to 36 hpi ( Figure 1B). Furthermore, the viral envelope protein was detectable by Western blotting at 3 and 6 hpi but not at later times. Similarly, env gene expression at 3 hpi was greater than that at 6 hpi ( Figure 1C). These results demonstrate that the replication rate of ALV-J was extremely high at 3 hpi but low after 6 hpi, ALV-J replication might be inhibited in chicken MDM.

Differentially expressed genes after ALV-J infection in chicken MDM
The Illumina HiSeq 2000 platform produced 814 412 986 raw reads. After discarding adaptor and low-quality sequences, we obtained 794 083 068 clean reads (119.19 Gb). The clean reads were mapped onto the chicken reference genome (Gallus_gallus-4.0), and the mapping rate of each library ranged from 79.29 to 83.27% (Additional file 2).
We found that at 3 and 36 hpi, 558 and 108 DEG were uniquely up-regulated, and 324 and 70 were uniquely down-regulated in ALV-J infected MDM, respectively. There were 66 up-regulated and 17 downregulated DEG in common at the two time points (Figures 2A and B). In addition, three DEG, OASL, FKBP51 and MCF2, were up-regulated at 3 hpi and down-regulated at 36 hpi ( Figure 2C). Nine DEG were down-regulated at 3 hpi but up-regulated at 36 hpi ( Figure 2D). More details of the DEG are shown in Additional file 3.

GO annotation of DEG after ALV-J infection in chicken MDM
The GO biological process analysis shows that the up-regulated DEG were mainly enriched for immunerelated terms but down-regulated DEG were not (Figure 3). For the up-regulated DEG at 3 hpi, the top three significant GO terms were inflammatory response, response to lipopolysaccharide and regulation of apoptotic process ( Figure 3A). The top three significant GO terms for up-regulated DEG at 36 hpi were inflammatory response, innate immune response and Toll-like receptor signaling pathway ( Figure 3B). Down-regulated DEG at 3 hpi were significantly enriched for transmembrane transport, endocytic recycling and positive regulation of interleukin-6 production ( Figure 3C). The five down-regulated DEG at 36 hpi included HMOX1, SLC11A1, SLC40A1, GAB1 and SLC25A4 were significantly enriched on four GO terms ( Figure 3D). More details of the DEG involved in GO enrichment analysis can be found in Additional file 4.

Pathway analysis of DEG after ALV-J infection in chicken MDM
KEGG analysis illustrated that up-regulated DEG induced by ALV-J in MDM at 3 hpi were involved in immune-related pathways including MAPK signaling, Toll-like, NOD-like, RIG-I-like and Jak-STAT signaling pathway, and etc. (Figure 4A). Up-regulated DEG identified at 36 hpi were significantly enriched in cell adhesion molecules, influenza A, Toll-like receptor and adipocytokine signaling pathway ( Figure 4B). However, just two pathways were significantly enriched by the down-regulated DEG at 3 hpi ( Figure 4C). Moreover, down-regulated DEG at 36 hpi did not enrich any pathway. More details of the DEG involved in KEGG enrichment analysis can be found in Additional file 5.

More immune-related DEG were induced by ALV-J at 3 hpi than at 36 hpi
The immune-related genes were selected according to gene function annotation. A greater number of immune-related DEG were found at 3 hpi than at 36 hpi, and most of these immune-related DEG were up-regulated by ALV-J infection at 3 hpi ( Figure 5A). According to published studies [21][22][23], 94 and 23 differentially expressed interferon-stimulated genes (ISG) were identified in ALV-J-infected MDM at 3 hpi and 36 hpi, respectively (Additional file 6). Similarly, the expression of most ISG (79) was significantly increased at 3 hpi, especially IRG1, RIPK2, CH25H, IRF7 and etc. ( Figure 5B).

RNA-seq data matched the qPCR data
To validate the RNA-Seq results, we chose eight immunerelated DEG for qPCR analysis. These included CH25H, PKR, SOCS5, NOD1, TLR7, IL-18, ISG12-2 and OASL. The qPCR data matched the RNA-Seq results and both methods indicate similar trends for these eight genes ( Figure 6).

Overexpression of K60, IRG1, OASL and CH25H inhibits ALV-J replication
Overexpression of K60, OASL, CH25H and IRG1 significantly decreased ALV-J replication at the protein (Figures 7A and B) and mRNA ( Figure 7C) levels in chicken MDM cells at 3 hpi when compared to the control group (EGFP).

Overexpression of CISH, EX-FABP, IL4I1 and SOCS3 promotes ALV-J replication
Overexpression of CISH, EX-FABP and SOCS3 significantly increased the expression of ALV-J env gene at protein levels ( Figures 8A and B) and mRNA levels ( Figure 8C) in chicken MDM cells at 3 hpi. However, overexpression of IL4I1 only significantly increased the expression of the ALV-J env gene at the mRNA level, but there was no difference at the protein level ( Figure 8).

Host anti-ALV-J candidates
As an avian retrovirus, ALV-J has been studied for many years although many interesting scientific problems such as tumorigenesis, immunosuppression and immune responses induced by ALV-J infection are still not understood [1]. In our previous studies, we found that chicken MDM are susceptible to ALV-J infection [10]. In the present study, we observed that ALV-J replication in MDM was active at 3 hpi, but inhibited from 6 to 36 hpi. It is reported that recombinant chicken IFN-α as well as the ISG, CCCH type zinc finger antiviral protein (ZAP), could inhibit ALV-J replication in DF1 cells [24,25]. So, we speculated that ISG may also resist ALV-J replication in chicken MDM.
ALV-J infection induced most immune-related DEG in MDM at 3 hpi (Tables 1 and 2). Strikingly, the expression of 79 ISG including CH25H, PKR, OASL, Mx, and etc. were significantly increased in ALV-J-infected MDM at 3 hpi. ISG exert numerous antiviral effector functions by targeting almost any step in the virus life cycle [26]. For example, CH25H broadly inhibited growth of enveloped viruses including VSV, HSV, HIV and the acutely pathogenic viruses EBOV, RVFV and RSSEV by converting cholesterol to 25-hydroxycholesterol (25HC) [27]. Additionally, 18 up-regulated ISG were identified in ALV-J MDM at 36 hpi. All of these up-regulated ISG in  ALV-J-infected MDM at 3 and 36 hpi might serve as candidates resisting ALV-J infection.
In addition to the above ISG, some DEG have significant changes in expression at 3 hpi and 36 hpi (Table 2). MIP-3α, macrophage inflammatory protein-3 alpha, is responsible for the chemo-attraction of dendritic cells, and effector and memory B and T cells [28]. Moreover, MIP-3α exhibited anti-microbial and anti-HIV activities [28][29][30]. IL-1β is produced primarily by activated macrophages and possess multiple and diverse properties in their response to infection [31,32]. Thus, host damage following infection induces macrophage secretion of a variety of inflammatory mediators including IL-1 and NO that activate anti-pathogenic microorganism defense mechanisms [2]. NO production is primarily catalyzed by iNOS and is a part of innate host defenses [33]. ALV-J infection in MDM at 3 hpi increased expression of the two orthologues of chicken IL-8, K60 (IL8L1) and IL8 (IL8L2) [34,35]. IL-8 is a potent chemo-attractant and activator of macrophages [36]. Furthermore, IL-8 has been shown to attract and activate T lymphocytes [37,38], which would aid in raising an immunologically specific response against ALV-J. Immune response gene 1 (IRG1) was originally identified as a highly inducible gene in murine macrophages following LPS stimulation [39]. The role of IRG1 in the course of virus infection has not been extensively reported. IRG1 was identified as an ISG with antiviral effects against different neurotropic viruses [40]. OASL has been found to broadly inhibit the replication of viruses such as swine fever virus, RSV and HCV through a variety of mechanisms [41][42][43]. Lysozyme is a differentiation marker for macrophage, and is activated during macrophage differentiation [44]. We found that lysozyme expression increased incrementally from 3 hpi to 36 hpi in ALV-J-infected MDM. This result reminded us that ALV-J infection could stimulate chicken macrophage maturation. Lysozyme is a cornerstone of innate immunity due to its direct antimicrobial  activity through peptidoglycan hydrolysis and immune regulatory functions [45]. Interestingly, lysozyme also possesses antiviral properties [46,47]. NFKBIZ encodes the protein IκBz and is known as a partner of NFκB that regulates innate host defense factors [48].
Accordingly, the host genes MIP-3α, IL-1β, iNOS, K60, IRG1, CH25H, OASL, lysozyme and NFKBIZ served as restriction factor candidates against ALV-J infection in chicken macrophages. We further selected several of the anti-ALV-J candidates for verification. Exactly, the experiments in vitro show that overexpression of K60, IRG1, CH25H, and OASL could significantly decrease ALV-J replication in MDM at 3 hpi (Figure 7).

The survival strategy of the ALV-J in MDM
Initially, we found that ALV-J infection activated many pattern recognition receptors (PRR) pathways including Toll-like receptors (TLR), RIG-I-like receptors (RLR), NOD-like receptors (NLR) and cytosolic DNA-sensing pathway at 3 hpi (Figure 4). Up to now, the specific innate sensors responding to ALV were unknown [1]. We speculated that ALV-J should theoretically be recognized by PRR such as TLR, RLR, IFI16, and cGAS, similar to HIV [1]. However, no functional PRR such as TLR3, TLR4, TLR7 or MDA5 were induced in ALV-J-infected MDM at 3 hpi. Only TLR15 was up-regulated by ALV-J at 3 hpi (Table 1). Therefore, we speculate that chicken macrophages lack functional PRR for ALV-J to escape host immune attack at the early stage of infection. Indeed, it has been reported that other retroviruses use this strategy to achieve immune escape in macrophages [49]. Interestingly, the expression of TLR1, TLR7 and TLR15 was significantly increased in ALV-J-infected MDM at 36 hpi (Table 1). As a result, ALV-J replication was inhibited and weak innate immune responses were induced at 36 hpi. At this infection stage, ALV-J may be recognized by TLR1, TLR7 and TLR15.
The Jak-STAT pathway is a major signaling pathway in the function of immune cells and is activated by cytokines and growth factors [50]. In this study, many negative feedback regulators of cytokine signaling mediated by this pathway were identified. CISH, SOCS1 and SOCS3 were significantly induced in ALV-J-infected MDM at 3 hpi and enriched on the Jak-STAT signaling  (Table 1). CISH and suppressor of cytokine signaling (SOCS) family proteins are Jak-STAT inhibitors, including 8 members, CISH, SOCS1, SOCS2, SOCS3, SOCS4, SOCS5, SOCS6 and SOCS7 [51]. CISH, SOCS1, SOCS2 and SOCS3 are the best characterized SOCS family members [52]. CISH is induced by cytokines that activate STAT5 and block the STAT binding to cytokine receptors. SOCS1 binds to the Jaks and inhibits catalytic activity, while SOCS3 binds to Jak-proximal sites on cytokine receptors and inhibits Jak activity [52]. It has been reported that SOCS3 enhances HIV-1 replication in macrophages by inhibiting antiviral IFN-β signaling [53]. The SOCS3 expression was also significantly increased in ALV-J-infected MDM (Table 1). Further verification found that overexpression of CISH and SOCS3 promoted ALV-J replication in MDM at 3 hpi (Figure 8). We speculated that ALV-J infection inhibited the Jak-STAT pathway via inducing expression of CISH, SOCS1 and SOCS3. Indeed, the key factors such as Jak and STAT in Jak-STAT pathway were not remarkably induced by ALV-J infection in MDM. The role of SOCS5 has not been well identified during viral infection. However, a novel role for SOCS5 has been found to restrain the early phase of influenza A infection by inhibiting EGFR activity [54]. Coincidently, SOCS5 expression was inhibited by ALV-J in MDM at 3 hpi (Table 1).

Table 1 The DEG listed involved in the pattern recognition receptor pathway and Jak-STAT signaling pathway in ALV-Jinfected MDM
In addition, the expression of NFKBIA and TNFAIP3 (A20) were significantly increased in ALV-J-infected MDM at 3 hpi (Table 1). The IκBα protein is encoded by NFKBIA and is an important negative regulatory factor in the NF-κB pathway [55]. The protein TNFAIP3 is known as a powerful suppressor of cytokine signaling and innate antiviral pathways, and it can inhibit the activity of NF-κB and NF-κB-mediated inflammatory responses [56,57]. TNFAIP3 deficiency in myeloid cells and lung epithelial cells could protect against influenza A virus infection [58,59]. We speculated that NFKBIA and TNFAIP3 were significantly induced by ALV-J infection in MDM to down-regulate cytokines in macrophages, resulting in viral persistence in the host. Altogether, NFKBIA, TNFAIP3, CISH, SOCS1 and SOCS3 were considered as a counterbalance to the antiviral immune responses induced by ALV-J infection in chicken MDM at 3 hpi.

IL4I1, PTGS2 (COX-2)
and EX-FABP significantly changed expression at 3 hpi and 36 hpi (Table 2). We also found that overexpression of IL4I1 and EX-FABP could enhance ALV-J replication in chicken MDM at 3 hpi (Figure 8). IL4I1 is an immunosuppressive enzyme and primarily expressed in professional antigen-presenting cells and it inhibits T-cell proliferation and activation [60,61]. COX-2 is one of the important mediators of inflammation in response to viral infection, which contributes to viral replication and this has been shown for HCV [62], HBV [63], dengue virus [64] and cytomegalovirus [65]. The EX-FABP gene encodes an extracellular fatty acid binding protein and it is significantly induced by Salmonella enteritidis infections in chickens [66]. This protein may provide fatty acids for mitochondrial respiration during infection. EX-FABP expression is enhanced after treatments with inflammatory stimuli and is repressed by anti-inflammatory agents, behaving as an acute phase and constitutively expressed survival protein [67]. EX-FABP was robustly induced in ALV-J-infected MDM at 36 hpi and therefore may provide protection for MDM after the chemokine and cytokine production induced by ALV-J at 3 hpi. There is a rule that the virus is lost when the cells die. Consequently, induction of EX-FABP at 36 hpi may be a strategy of ALV-J to live in harmony with chicken macrophages at the late stages of infection. Consequently, EX-FABP, IL4I1 and COX-2 together with NFKBIA, TNFAIP3, CISH, SOCS1 and SOCS3 enable ALV-J survival in chicken macrophages.