NAPRRS: Evolutionary dynamics of PEDV
Two strains of the porcine epidemic diarrhea virus were identified in the early stages of the epidemic
20 August 2025
3 minute read
By: The porcine epidemic diarrhea virus (PEDV) (Coronaviridae family) is responsible for significant losses in the swine industry worldwide. PEDV was introduced to the United States in 2013 and quickly spread across the country. Evolutionary dynamics studies of this virus since its emergence have focused on the early stages of the epidemic, identifying two strains: indel and the more virulent non-indel strains.
However, the understanding of PEDVs evolution and molecular epidemiology in the current phase of endemic circulation in the US remains limited. Joao Paulo Herrera da Silva and colleagues at the University of Minnesota explored the temporal dynamics of PEDV in the US and provided an overview of its evolutionary status a decade since its introduction. The information was presented at the 2024 North American PRRS Symposium.
The incidence of PED was assessed based on the number of positive sow farms reported by the Morrison Swine Health Monitoring Project (MSHMP). A total of 556 spike protein subunit 1 sequences, 328 from MSHMP and 228 from GenBank (2013-2024) were analyzed with 313 of these being nearly complete genomes. For the spike protein, Bayesian methods were employed to reconstruct time-scaled phylogenetic trees and estimate the effective population sizes using BEAST v1.10.4. Recombination analyses were performed using RDP5 software. Analyses for both PEDV strains were conducted separately.
Sequence availability was proportionate to incidence data, with more sequences available during the epidemic period (May 2013 - December 2014), followed by a reduction in the number of both cases and sequences in subsequent years.
The indel and non-indel strains showed a low effective population size at the beginning of the PEDV epidemic, reflecting a population founder effect. Subsequently, both entered a period of expansion. Reliable effective population size estimates for the indel strain (99.7% nucleotide identity) could only be obtained until 2017 due to data availability. The non-indel strain (n=499) exhibited high nucleotide identity (mean: 99.7 ± 0.004). All recent sequences (2017 onwards) are grouped into two small and distinct clades (99.8% identity within clades, 98.4% between clades), each occurring in specific regions in the US.
This suggests that there is strong regional structuring or even compartmentalization of PEDV spread, which has implications of disease elimination efforts. Other clades existing during the early epidemic period appeared to have been displaced by 2019. Fluctuations in the effective population size for the non-indel strain reveal cycles of population contractions and expansions, perhaps associated with seasonality.
The indel ORF-S was under negative selection, while the non-indel strain evolved neutrally with several sites that were under positive selection; these substitutions appear randomly distributed across the phylogeny, suggesting they may be recurrent mutations. One site under positive selection was in an epitope of the spike protein.
Preliminary analysis detected thirteen recombination events in complete genome sequences, two of which left numerous descendants. The ORF-S of the non-indel strain showed high genetic similarity, patterns of temporal structuring, and several sites under positive selection. The possible displacement of older clades by more recent ones may reflect fitness differences, highlights the need for continued molecular surveillance. In addition, more representative sampling across regions is needed to confirm observed patterns.
