Combinational Deletions of MGF360-9L and MGF505-7R Attenuated Highly Virulent African Swine Fever Virus and Conferred Protection against Homologous Challenge

Abstract

Multigene family (MGF) gene products are increasingly reported to be implicated in African swine fever virus (ASFV) virulence and attenuation of host defenses, among which the MGF360-9L and MGF505-7R gene products are characterized by convergent but distinct mechanisms of immune evasion. Herein, a recombinant ASFV mutant, ASFV-Δ9L/Δ7R, bearing combinational deletions of MGF360-9L and MGF505-7R, was constructed from the highly virulent ASFV strain CN/GS/2018 of genotype II that is currently circulating in China. Pigs inoculated intramuscularly with 10⁴ 50% hemadsorption doses (HAD₅₀) of the mutant remained clinically healthy without any serious side effects. Importantly, in a virulence challenge, all four within-pen contact pigs demonstrated clinical signs and pathological findings consistent with ASF. In contrast, vaccinated pigs (5/6) were protected and clinical indicators tended to be normal, accompanied by extensive tissue repairs. Similar to most viral infections, innate immunity and both humoral and cellular immune responses appeared to be vital for protection. Notably, transcriptome sequencing (RNA-seq) and quantitative PCR (qPCR) analysis revealed a regulatory function of the mutant in dramatic and sustained expression of type I/III interferons and inflammatory and innate immune genes in vitro. Furthermore, infection with the mutant elicited an early and robust p30-specific IgG response, which coincided and was strongly correlated with the protective efficacy. Analysis of the cellular response revealed a strong ASFV-specific interferon gamma (IFN-γ) response and immunostaining of CD4⁺ T cells coupled with a high level of CD163⁺ macrophage infiltration in spleens of vaccinated pigs. Our study identifies a new mechanism of immunological regulation by ASFV MGFs that rationalizes the design of live attenuated vaccine for implementation of improved control strategies to eradicate ASFV. IMPORTANCE Currently, the deficiency in commercially available vaccines or therapeutic options against African swine fever constitutes a matter of major concern in the swine industry globally. Here, we report the design and construction of a recombinant ASFV mutant harboring combinational deletions of interferon inhibitors MGF360-9L and MGF505-7R based on a genotype II ASFV CN/GS/2018 strain currently circulating in China. The mutant was completely attenuated when inoculated at a high dose of 10⁴ HAD₅₀. In the virulence challenge with homologous virus, sterile immunity was achieved, demonstrating the mutant's potential as a promising vaccine candidate. This sufficiency of effectiveness supports the claim that this live attenuated virus may be a viable vaccine option with which to fight ASF.

Keywords: ASFV; MGF360-9L; MGF505-7R; attenuated phenotype; immune response; protective efficacy.

FIG 1

Construction and characterization of a double-gene-deletion ASFV-Δ9L/Δ7R mutant. (A) Schematic diagrams of recombinant transfer vectors pASFV-Δ9L and pASFV-Δ7R. Open reading frames are indicated by arrows, with nucleotide positions labeled above. The deleted MGF360-9L and MGF505-7R were replaced with the p72eGFP and p72mCherry reporter gene cassettes, respectively. (B) Schematic diagrams of designs for centering primers and flanking primers located within or flanking the target gene, respectively. The expected DNA sizes of PCR products are indicated above. The X symbols in the left panel indicate that no bands were obtained by PCR. WT, wild type. (C) PCR analysis of ASFV-Δ9L/Δ7R using two sets of primer pairs, with the p72 primer pair used as an indicator of loading ASFV genomic DNA. ddH₂O, double-distilled water. (D) Replication kinetics of parental ASFV, ASFV-Δ9L mutant, and ASFV-Δ9L/Δ7R mutant. BMDM were infected with ASFV at an MOI of 0.5. Following the indicated durations (2 h, 24 h, 48 h, and 96 h), whole-cell cultures were subjected to repeated freeze-thawing processes. Virus titers in the supernatant were titrated by HAD₅₀.

FIG 2

Assessment of virulence of ASFV-Δ9L/Δ7R and its protective efficacy against parental ASFV challenge. (A) Schematic diagram of design and outcomes of animal experiment. Four naive pigs (C1, C2, C3, and C4) were placed in direct contact with six pigs (V1, V2, V3, V4, V5, and V6) intramuscularly inoculated with 10⁴ HAD₅₀ of ASFV-Δ9L/Δ7R mutant in room 1. At 23 days postvaccination, the four contact pigs were transferred to a separate room (room 2) for a further challenge experiment. All the pigs were challenged with 10² HAD₅₀ of highly virulent parental ASFV CN/GS/2018 and monitored for an extra 18 days. The four contact pigs either died or were euthanized in extremis between 7 and 8 days postchallenge (dpc). In contrast, only one vaccinated pig (V4) died, at 8 dpc. Rectal temperature was monitored daily. EDTA plasma samples, serum samples, and swab samples were collected at 2-day intervals. At necropsy, the indicated tissues were acquired. (B) Observational clinical signs of ASVF in contact animal C3. Arrows indicate representative cyanosis and necrotic lesions on the skin of the abdomen. (C) Descriptive survival outcomes of the pigs after vaccination and challenge.

FIG 3

Rectal temperatures, viremia titers, and virus shedding. (A) Rectal temperatures of pigs vaccinated with 10⁴ HAD₅₀ of ASFV-Δ9L/Δ7R (n = 6, black) or exposed through direct contact (n = 4, red) (left) and subsequently challenged with 10² HAD₅₀ of highly virulent parental ASFV (right). Rectal temperature of ≥40°C was defined as fever. (B) Virus titers in EDTA plasma samples. Values are expressed as log₁₀ copy numbers per milliliter. (C) Virus shedding. Swabs were soaked in phosphate-buffered saline (PBS) medium overnight. The following day, the fluids were vortexed, repeatedly freeze-thawed, clarified, and subjected to copy number detection.

FIG 4

Correlation between viral loads and postmortem lesions. (A) Virus titers in tissues from vaccinated pigs (black, n = 6) and contact pigs (red, n = 4). Totals of 30 mg of indicated types of tissue samples were homogenized, vortexed, clarified, and subjected to copy number detection. (B) Immunostaining of ASFV p72 antigens in the livers and spleens of one healthy animal, two vaccinated animals (V1 and V6), and one contact animal (C3). Nuclear staining of cells in the contact pig is almost completely absent due to advanced autolysis. (C) Comparative postmortem lesions. The images show representative organs from contact pig C3 (left) and vaccinated pig V1 (right) as follows: 1, submandibular lymph node; 2, gastrohepatic lymph node; 3, mesenteric lymph node; 4, kidney; 5, spleen; 6, lung; 7, heart; and 8, liver.

FIG 5

Characterization of histopathological lesions. Representative histopathological lesions in different tissue samples of a healthy pig (far left), vaccinated pigs V3 (survived) and V4 (died), and contact pig C3. Arrows indicate severe acute and diffuse hemorrhages and large numbers of karyorrhectic cells. Triangles indicate lymphoid depletion and loss of lymphocytes. Stars indicate infiltration of inflammatory cells.

FIG 6

Longitudinal p30 IgG responses in vaccinated (black) versus contact (red) pigs.

FIG 7

ASFV-Δ9L/Δ7R induced a more pronounced innate immune response in vitro. BMDM were mocked infected or infected with parental ASFV and ASFV-Δ9L/Δ7R at an MOI of 0.5. Following the indicated durations (6 h, 18 h, and 30 h), cell cultures were subjected to RNA-seq analysis. (A) Volcano plot of gene changes in ASFV-infected BMDM compared to the expression in mock-infected BMDM. Red dots and blue dots denote upregulated and downregulated DEGs (P < 0.01, log₂(fold change) > 1 or < −1), respectively. (B) Hierarchical clustering of the DEGs identified in ASFV-infected BMDM. A total of 221 genes implicated in the innate immune response were probed over time and are displayed in a heat map. Each panel represents a particular gene, and the color depicts the fold change (FC) at the indicated time points. (C) Histogram of significantly enriched GO classifications of upregulated DEGs. The top 10 upregulated GOTerms involved in the three main categories, namely, biological process, molecular function, and cellular component, are ranked based on the counts of upregulated DEGs in ASFV-Δ9L/Δ7R-infected BMDM compared to parental ASFV-infected samples at 18 h. The y axis indicates the number of DEGs in a specific category.

FIG 8

Validation of RNA-seq results. (A) BMDM were mock infected or infected with ASFV at an MOI of 0.5. A total of 48 genes encoding cytokines, NF-κB pathway, complement pathway, and IFN pathway were detected by qPCR for changes in expression with ASFV infection over time. Fold change values (relative to mock infection) were calculated using the ΔΔC_T method and are presented as the mean values in a heatmap (n = 3 independent experiments with replicates). Statistical analysis was performed between the results from parental ASFV-infected and ASFV-9L7R-infected BMDM. (B, C) Correlation analysis of fold changes in transcript abundance obtained from RNA-seq and qPCR analysis in parental ASFV-infected samples (B) and ASFV-Δ9L/Δ7R-infected samples (C). Gray shading denotes the 95% confidence intervals for linear regression analysis (red lines). (D) qPCR analysis of expression levels of IFNs in ASFV-infected samples over time. The results are displayed as fold change values relative to expression in mock samples (n = 3 independent experiments with replicates).

FIG 9

KEGG analysis.

FIG 10

ASFV-Δ9L/Δ7R induced T cell activation and macrophage infiltration in spleens. (A) Interferon gamma (IFN-γ) status in sections of spleens from one healthy animal (H), 5 vaccinated animals (V1, V2, V3, V5, and V6), and one contact animal (C3), as shown by IHC staining. Results for IFN-γ were negative in examined sections from healthy and contact pigs. (B) Integrated optical density (IOD) results from the experiment whose results are shown in panel A. (C, D) Immunohistochemical staining of CD4⁺ T cells (C) and CD163⁺ macrophages (D) in the spleens of two vaccinated animals (V2 and V3) and one contact animal (C3). Scale bars, 100 μm.