Global Dynamics of Porcine Enteric Coronavirus PEDV Epidemiology, Evolution, and Transmission

Abstract

With a possible origin from bats, the alphacoronavirus Porcine epidemic diarrhea virus (PEDV) causes significant hazards and widespread epidemics in the swine population. However, the ecology, evolution, and spread of PEDV are still unclear. Here, from 149,869 fecal and intestinal tissue samples of pigs collected in an 11-year survey, we identified PEDV as the most dominant virus in diarrheal animals. Global whole genomic and evolutionary analyses of 672 PEDV strains revealed the fast-evolving PEDV genotype 2 (G2) strains as the main epidemic viruses worldwide, which seems to correlate with the use of G2-targeting vaccines. The evolving pattern of the G2 viruses presents geographic bias as they evolve tachytely in South Korea but undergo the highest recombination in China. Therefore, we clustered six PEDV haplotypes in China, whereas South Korea held five haplotypes, including a unique haplotype G. In addition, an assessment of the spatiotemporal spread route of PEDV indicates Germany and Japan as the primary hubs for PEDV dissemination in Europe and Asia, respectively. Overall, our findings provide novel insights into the epidemiology, evolution, and transmission of PEDV, and thus may lay a foundation for the prevention and control of PEDV and other coronaviruses.

Keywords: coronavirus; epidemiology; evolution; porcine epidemic diarrhea virus; transmission.

Fig. 1.

Characteristics of epidemics of PEDV from 2011 to 2021 in China. (A) A schematic representation of 149,869 fecal and intestinal tissue samples of pigs collected from the year 2011 to 2021. The upset plot arranged five common swine diarrhea-causing viruses: PEDV, TPEV, RV, PDCoV, SADS-CoV. The color shades and the size of circles represent the yearly sample size. (B) Phylogeographic distributions of 65 whole genomes sequenced PEDV strains from the year 2011 to 2021 in the study. The topological tree was built with the maximum likelihood method and 1,000 bootstrap replicates. The branch of the strain GDS09, belonging to the G1 genotype, is depicted in black in the phylogenetic tree. The size of circles in the provinces represents the sample size of sequenced PEDV full genomes. (C) Cytopathic effects of Vero-E6 cells infected with three representative PEDV strains (GDS29, JSS04, and GDS09) at 6, 12, 24, or 36 h postinfection (hpi), scale bar indicates 100 μm. (D) Immunofluorescence staining of the PEDV S protein in infected Vero-E6 cells. Cells were fixed at 6, 12, 24, or 36 hpi, and S protein was detected by indirect IFA. Nuclei were shown by 40,6-diamidino-2-phenylindole staining. The images of cells were acquired by a fluorescence microscope (Nikon Eclipse 80i), scale bar indicates 100 μm. (E) Viral titers in the culture supernatants were determined by TCID₅₀ assay. Data are shown as mean ± SEM and are representative of three independent experiments.

Fig. 2.

Evolution and recombination of PEDV strains of four geographical regions (China, South Korea, the US, and Europe). (A) The similarity and divergence of nucleotide sequences of six PEDV genes (ORF1a, ORF1b, spike, envelop, membrane, and nucleocapsid) among geographical region pairs. (B–E) Maximum likelihood time-scaled phylogenetic trees of full PEDV genomes of four geographical regions: (B) China, (C) South Korea, (D) the US, and (E) Europe. The branches of G1 PEDV strains are highlighted. Different marks of the dots represent different administrative regions or countries. The x-axis represents the time in years. The inner plots showed the PP densities of the most recent common ancestor (TMRCA) of four geographical regions. (F) Breakpoint distributions of recombination events in PEDV strains of four geographical regions. PP, posterior probability.

Fig. 3.

Haplotype dynamics of PEDV. Seven groups (A–E) of haplotypes were identified in supplementary figure S6, Supplementary Material online. (A–D) NSP8 haplotype distribution in (A) China, (B) the US, (C) Europe, and (D) South Korea. (E) Global distribution of various haplotype groups.

Fig. 4.

Spatiotemporal diffusion of PEDV strains around the world. (A) The discrete phylogeographic analysis was performed with the Bayesian stochastic search variable selection approach. We displayed the intensity of the estimated transition events associated with a BF support higher than three. The thickness of the dissemination links represented the PP. (B) Continuous phylogeographic analysis was performed with the lognormal RRW diffusion model. We mapped the MCC tree and 95% highest posterior density regions based on trees subsampled from the post-burn-in posterior distribution of trees. Nodes of the tree are scaled. A 95% highest posterior density regions were computed for successive time layers, superimposed using the same scale reflecting time cropped using worldwide international borders.

Fig. 5.

Phylogenetic analysis of complete PEDV genomes (n = 672) and estimated evolutionary rate of PEDV strains. (A) Phylogenetic tree constructed from 672 complete PEDV genomes using ML method. The outside layer denotes the genotypes. The highlighted dots indicate 65 PEDV strains sequenced in this study. The extra bars indicate the clusters of wild vaccine strains and vaccine-like strains. (B–E) Root-to-tip distance plots on sampling date and the estimated evolutionary rates of PEDV strains. (B) Strains (n = 33) of G1 genotype around the world, (C) strains (n = 639) of G2 genotype around the world, (D) Chinese strains of G2 genotype collected before the year 2018 (n = 173) and after the year 2018 (n = 34), and (E) Korean strains of G2 genotype collected before the year 2017 (n = 13) and after the year 2017 (n = 24). The highlighted strains in the outer plots were excluded in the linear fitting and evolutionary rate estimation according to TreeTime (Sagulenko et al. 2018). The inner plots show the posterior probability densities of the mean estimated evolutionary rate and the x-axis represents the evoltionary rate (×10⁻⁴ substitutions per site per year).

Fig. 6.

Comparison of the spike protein of genotype G1 and G2 viruses. Strains CV777 (Genbank accession No., AF353511) and JSS04 (Genbank accession No., MW143083) were used as representative strain of genotypes G1 and G2. (A) Schematic of PEDV spike protein. (B–D) Cartoon representation of amino acid sequences of the spike protein of CV777 with different perspectives. (D) Amino acid sequences at positions 55–60 and 152–159 of the spike protein of CV777. (E–G) Cartoon representation of amino acid sequences of the spike protein of JSS04 with different perspectives. (G) Amino acid sequences at positions 55–64 and 157–162 of the spike protein of JSS04. (H) Identification and validation of PEDV genotype-specific epitopes.