Combinational Deletions of MGF110-9L and MGF505-7R Genes from the African Swine Fever Virus Inhibit TBK1 Degradation by an Autophagy Activator PIK3C2B To Promote Type I Interferon Production

ABSTRACT African swine fever (ASF), caused by the African swine fever virus (ASFV), is a transboundary infectious disease of domestic pigs and wild boars, resulting in significant swine production losses. Currently, no effective commercial ASF vaccines or therapeutic options are available. A previous study has shown that deletions of ASFV MGF110-9L and MGF505-7R genes (ASFV-Δ110-9L/505-7R) attenuated virulence in pigs and provided complete protection against parental lethal ASFV CN/GS/2018 (wild-type ASFV [ASFV-WT]) challenge, but the underlying mechanism is unclear. This study found that ASFV-Δ110-9L/505-7R weakened TBK1 degradation compared with ASFV-WT through RNA sequencing (RNA-seq) and Western blotting analyses. Furthermore, we confirmed that ASFV-Δ110-9L/505-7R blocked the degradation of TBK1 through the autophagy pathway. We also identified that the downregulation of an autophagy-related protein PIK3C2B was involved in the inhibition of TBK1 degradation induced by ASFV-Δ110-9L/505-7R. Additionally, we also confirmed that PIK3C2B promoted ASFV-Δ110-9L/505-7R replication in vitro. Together, this study elucidated a novel mechanism of virulence change of ASFV-Δ110-9L/505-7R, revealing a new mechanism of ASF live attenuated vaccines (LAVs) and providing theoretical guidance for the development of ASF vaccines. IMPORTANCE African swine fever (ASF) is a contagious and lethal hemorrhagic disease of pigs caused by the African swine fever virus (ASFV), leading to significant economic consequences for the global pig industry. The development of an effective and safe ASF vaccine has been unsuccessful. Previous studies have shown that live attenuated vaccines (LAVs) of ASFV are the most effective vaccine candidates to prevent ASF. Understanding the host responses caused by LAVs of ASFV is important in optimizing vaccine design and diversifying the resources available to control ASF. Recently, our laboratory found that the live attenuated ASFV-Δ110-9L/505-7R provided complete protection against parental ASFV-WT challenge. This study further demonstrated that ASFV-Δ110-9L/505-7R inhibits TBK1 degradation mediated by an autophagy activator PIK3C2B to increase type I interferon production. These results revealed an important mechanism for candidate vaccine ASFV-Δ110-9L/505-7R, providing strategies for exploring the virulence of multigene-deleted live attenuated ASFV strains and the development of vaccines.

DEGs are mainly enriched in cytokine response and interaction. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis was conducted to further explore the differential changes in biological signaling pathways during ASFV-WT and ASFV-D110-9L/505-7R infection, and the results showed that DEGs were mainly enriched in the cytokine-cytokine receptor interaction at different stages of infection ( Fig. 2A to C). At the later stages of infection (24 hpi), DEGs were also enriched in the NF-k B signaling pathway and other immune-related pathways (Fig. 2C). Notably, ASFV-D110-9L/505-7R infection induced more DEGs than ASFV-WT infection, among which 206, 145, and 376 DEGs were downregulated and 24, 24, and 227 DEGs were upregulated at 5, 12, and 24 hpi, respectively ( Fig. 2D to F). These results preliminarily indicated that ASFV-D110-9L/ 505-7R infection induces stronger host responses than ASFV-WT infection.
To determine whether the attenuated virulence of ASFV-D110-9L/505-7R is related to the induction of type I IFN and type II IFN, porcine alveolar macrophage (PAM) cells were first infected with ASFV-WT and ASFV-D110-9L/505-7R and then were treated or not treated with Herring testis DNA (HT-DNA, an IFN agonist, activates the cyclic GMP-AMP synthase-STING-dependent IFN signaling pathway) (31). We observed that ASFV- Further, the effect of ASFV-WT and ASFV-D110-9L/505-7R on IFN-band IFN-g -stimulated gene expression was also detected, and the results showed that the expression of Isg15 and PKR (as representative IFN-b-induced genes), and Irf1 as well as Gbp1 (as representative IFN-g -induced genes), has no significant difference in ASFV-WT-and ASFV-D110-9L/505-7R-infected cells ( Fig. 4E to H). These results revealed that ASFV-D110-9L/505-7R infection specifically promotes the expression of genes associated with the type I interferon pathway compared with ASFV-WT infection.
Previous studies have shown that gene-deleted LAVs of ASFV were still the most effective vaccine candidates to prevent ASF (44,45). Therefore, analyzing and clarifying the differences between LAVs and the parental virus strain are significant for vaccine development (46). This study used RNA-seq analysis to evaluate the transcription difference of genes in PAM cells during ASFV-D110-9L/505-7R or ASFV-WT infection. The RNA-seq results showed that DEGs were mainly enriched in the innate immune pathway, and the TBK1 protein, an important adaptor molecule of the type I interferon pathway, was further identified as an obvious DEG involved in innate immune signaling pathways. Finally, PIK3C2B, a positive regulator of autophagy (32), is identified as one of the key molecules mediating differential expression of TBK1 during ASFV-D110-9L/505-7R and ASFV-WT infection. This study provides new ideas for the study of LAV pathogenesis and has important significance for developing ASF vaccines. However, several issues need to be raised and discussed in this study as follows. (i) Previous studies have shown that the deletion of genes from different strains or isolates does not always have the same outcomes (19,47), and several different ASFV strains (including genotype I ASFV) are circulating in China (48)(49)(50)(51)(52). This study only focused on the type II ASFV strain (ASFV CN/GS/2018 isolate). Whether this study's results are consistent with the type I strain or other type II ASFV strains is necessary for further studies. (ii) Thus far, only LAVs have shown solid protection (some LAVs provided protection rates as high as 100%) against the lethal ASFV-WT challenge (28). An effective ASF vaccine must be accompanied by an effective protective immune mechanism and require  cross-involvement of both innate and adaptive immune cross talk (53). A previous study has found that MGF505-7R is not only involved in regulating the innate immune response (19) but also can be recognized by specific T cells (54). In future studies, we should pay more attention to illustrating the relationship and mechanism between innate and adaptive immune response, which plays an important role in understanding the immune protection mechanisms of ASFV-D110-9L/505-7R. (iii) Previous studies have shown cell apoptosis is necessary for ASFV pathogenesis (55) and inflammatory response is regulated during ASFV infection (56,57). This study mainly illustrated the interferon pathway (Fig. 4) in the immune response. Although the innate response is the most significant signaling pathway for GO enrichment, there is also enrichment in other pathways. Therefore, the influence of other factors on virus virulence is our future research direction.
In summary, this study reveals that the attenuated virulence of ASFV-D110-9L/505-7R is probably associated with restoring TBK1 autophagic degradation. TBK1 autophagic degradation is further identified to be regulated by a positive autophagy regulation molecule, PIK3C2B. Our findings show the involvement of the PIK3C2B-TBK1 axis in innate immune responses during ASFV infection, providing important clues for exploring and understanding the mechanisms of mutant ASFV strain infection. Notably, the identification of MGF110-9L and MGF505-7R as important inhibitors of innate immune responses and critical determinants of ASFV virulence points provides a new host-virus interaction mechanism of ASFV-D110-9L/505-7R and theoretical guidance for the development of ASF vaccines.

MATERIALS AND METHODS
Ethics statements. All experiments involved with ASFV-WT and ASFV-D110-9L/505-7R were conducted in the biosafety level 3 (BSL-3) facilities in the Lanzhou Veterinary Research Institute (LVRI) of the Chinese Academy of Agricultural Sciences, approved by the Ministry of Agriculture and Rural Affairs.
Quantitative real-time PCR and RNA interference experiments. Total cellular RNA was isolated with TRIzol (Invitrogen Life Technologies) reagent from the treated cells, and then RNA was transcribed to cDNA using a reverse transcription kit (TaKaRa; code no. RR047A). Transcription levels of genes were assessed by qRT-PCR analysis using the SYBR green premix (Bio-Rad) following the manufacturer's instructions. Gene expression levels were normalized to GAPDH (glyceraldehyde-3-phosphate dehydrogenase), and relative mRNA expression levels were evaluated and calculated utilizing the threshold cycle (2 2DDCT ) method. All primers used in the qRT-PCR assay are listed in Table 3. Small interfering RNAs (siRNAs) and siRNA-control were designed and synthesized by Sangon Biotech (Shanghai, China). PAM cells were transfected with ATG5 siRNA for 48 h and then infected with ASFV-WT and ASFV-D110-9L/ 505-7R for the indicated time points at a multiplicity of infection (MOI) of 1. Expression of proteins and knockdown efficiency were detected by Western blotting and qRT-PCR analyses, respectively. Western blotting. The transfected or virus-infected cells were washed twice with cold phosphate-buffered saline (PBS), lysed in lysis buffer (pH 7.5, 50 mM Tris-HCl, 150 mM NaCl, 0.5% Triton X-100, 10% glycerol, 2% SDS, b-mercaptoethanol, bromophenol blue, and 1 mM EDTA), and heated at 90°C for 10 min. Then proteins were separated with SDS-PAGE and transferred to the nitrocellulose (NC) membrane (Pall). Transferred membranes were incubated overnight at 4°C with the primary antibodies and for 1 h at room temperature (RT) with the corresponding horseradish peroxidase-conjugated secondary antibodies. Proteins were detected with Pierce ECL Western blotting substrate (Thermo Fisher Scientific).
RNA sequencing analysis. PAM cells were mock infected or infected with ASFV-WT and ASFV-D110-9L/505-7R strains at indicated time points (0, 5, 12, and 24 hpi). Cells were harvested for RNA extraction using TRIzol reagent (Invitrogen, CA, USA). RNA integrity was assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Then, the libraries were constructed using VAHTS universal V6 RNA-seq library prep kit according to the manufacturer's instructions. The transcriptome sequencing and analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China). Finally, differential expression genes (DEGs) in the PAM cells infected with ASFV-WT and ASFV-D110-9L/505-7R were identified utilizing the edgeR R package (3.18.1). The P values were adjusted using the Benjamini-Hochberg method. A corrected P value of ,0.05 and absolute fold change (jlog 2 FCj)of .2 were set as the thresholds for significantly differential expression. HAD 50 assay. HAD 50 assay was used to quantify the median tissue culture infectivity of ASFV-WTand ASFV-D110-9L/505-7R-infected samples as previously described, with a minor modification (59). Briefly, PAM cells were seeded in 96-well plates for 2 h, and then 10-fold serially diluted samples were added into each well in triplicate. Subsequently, cells were incubated with 2% porcine red blood cells diluted in PBS, and the hemadsorption (HAD) phenomenon was observed for 5 to 7 days. Finally, HAD 50 was calculated utilizing the Reed and Muench method.
Transmission electron microscopy. For transmission electron microscopy (TEM) analysis of autophagosome, ASFV-WT-and ASFV-D110-9L/505-7R-infected PAM cells were fixed with 2.5% glutaraldehyde (Merck, Germany) in 0.1 M phosphate buffer for 1 h at RT. Cells were then postfixed with 2% osmium tetroxide and embedded in epoxy according to standard procedures. After polymerization, about 80-nm-thick sections were obtained and stained with uranyl acetate and lead citrate as previously described (60). Samples were observed under a transmission electron microscope (HT7700; Hitachi, Japan) operated at 80 kV.
Statistical analysis. All experiments were done at least in duplicate for a total of $3 biological replicates. Data are displayed as means 6 standard deviation (SD). Statistical significance was determined using the unpaired two-tailed Student's t test. Comparisons showing a P value of ,0.05 were considered statistically significant. Asterisks indicate the statistical significance between connected two bars; *, P , 0.05; **, P , 0.01; ***, P , 0.005; and ****, P , 0.001.
Data availability. The RNA sequencing results were deposited in the Sequence Read Archive database (BioProject accession no. PRJNA952836).

ACKNOWLEDGMENTS
This work was supported by grants from the National Key R&D Program of China