Highly lethal genotype I and II recombinant African swine fever viruses detected in pigs

Abstract

African swine fever virus (ASFV) poses a great threat to the global pig industry and food security. Currently, 24 ASFV genotypes have been reported but it is unclear whether recombination of different genotype viruses occurs in nature. In this study, we detect three recombinants of genotype I and II ASFVs in pigs in China. These recombinants are genetically similar and classified as genotype I according to their B646L gene, yet 10 discrete fragments accounting for over 56% of their genomes are derived from genotype II virus. Animal studies with one of the recombinant viruses indicate high lethality and transmissibility in pigs, and deletion of the virulence-related genes MGF_505/360 and EP402R derived from virulent genotype II virus highly attenuates its virulence. The live attenuated vaccine derived from genotype II ASFV is not protective against challenge of the recombinant virus. These naturally occurring recombinants of genotype I and II ASFVs have the potential to pose a challenge to the global pig industry.

Fig. 1. Genomic analysis of newly emerging recombinant African swine fever viruses.

a The phylogenetic tree was built using the maximum likelihood (ML) and IQ-Tree based on the full genome sequences of the three recombinant ASFVs and 56 reference ASFVs of eight different genotypes from the GenBank database. b Specific single nucleotide polymorphisms (SNPs) at the junction regions of two recombinant fragments. The whole genomes of three recombinant ASFVs were compared with those of the representative NH/P68-like genotype I ASFV SD/DY-I/21 and Georgia07-like genotype II ASFV HLJ/18 by using SnapGene software. The recombinant fragments derived from genotype I ASFV are labeled in blue, and the fragments derived from genotype II ASFV in pink. c The diagram of the recombinant ASFV was created using BLAST Ring Image Generator with the whole genome sequence of JS/LG/21.

Fig. 2. Genetic changes in the three recombinant ASFVs compared with genotype I and II ASFVs.

Nucleotide mutations, deletions, and insertions in the three recombinant ASFVs compared with the corresponding regions of genotype I virus SD/DY-I/21 (a, b) and genotype II virus HLJ/18 (c, d) in the OFRs (a, c) and the noncoding regions (b, d).

Fig. 3. Pathogenicity and transmissibility of the recombinant African swine fever virus JS/LG/21 in pigs.

Groups of six SPF pigs were inoculated with 10⁶ HAD₅₀ (a–f) or 10³ HAD₅₀ (g–l) of JS/LG/21, and two naive SPF pigs were cohoused with each group from the first day of inoculation. The rectal temperature (a, g) and survival of the pigs (b, h) were monitored daily. Oral swabs (c, i), rectal swabs (d, j), and blood (e, k) were collected at the indicated timepoints, and tissue samples (f, l) were collected from dead pigs for viral DNA detection using qPCR. The dashed black lines indicate the normal rectal temperature (40 °C) of pigs. LN1, inguinal lymph node; LN2, submaxillary lymph node; LN3, mediastinal lymph node. Source data are provided as a Source Data file.

Fig. 4. Virulence and transmissibility of JS/LG/21-7GD in Pigs.

a Schematic representation of the gene-deleted JS/LG/21-7GD. The deleted gene segments were replaced with the p72mCherry and p72Venus reporter gene cassettes as indicated, respectively. Nucleotide positions indicating the boundaries of the deletion relative to the ASFV JS/LG/21 genome are indicated. b Growth curve of JS/LG/21 and JS/LG/21-7GD in PAMs. PAMs were infected with JS/LG/21 and JS/LG/21-7GD at an MOI of 0.1, respectively. The supernatants were collected from three wells for viral DNA detection using qPCR at the indicated timepoints. Data were presented as mean values ± SD. c Fluorescence of PAMs infected with JS/LG/21-7GD at 72 h post-inoculation. This experiment was performed three times and the data from one independent experiment were shown. Ten SPF pigs were inoculated with 10⁶ TCID₅₀ of JS/LG/21-7GD. Rectal temperature (d) and survival (e) were monitored daily for 28 days after infection. f Sera were collected from ten inoculated pigs at the indicated timepoints for ASFV-specific antibody detection using a blocking ELISA kit (in-house). Data were presented as mean values ± SD. The dashed black lines indicate the normal rectal temperature (40 °C) of pigs. Source data are provided as a Source Data file.

Fig. 5. Protective efficacy of the HLJ/18-7GD vaccine against challenge with different ASFVs in pigs.

Groups of five SPF pigs were vaccinated with 10⁶ TCID₅₀ of HLJ/18-7GD, and then intramuscularly challenged with 10³ HAD₅₀ of HLJ/18 (a–c) or the recombinant virus JS/LG/21 (d–f) on day 28 post-vaccination. Groups of four unvaccinated SPF pigs were challenged as controls. The rectal temperature (a, d) and survival (b, e) were monitored daily, and tissue samples (c, f) were collected from dead pigs or euthanized pigs at the end of the observation period for viral DNA detection using qPCR. The dashed black lines indicate the normal rectal temperature (40 °C) of pigs. LN1, inguinal lymph node; LN2, submaxillary lymph node; LN3, mediastinal lymph node. Source data are provided as a Source Data file.