Phylogeny-based evolutionary, demographical, and geographical dissection of North American type 2 porcine reproductive and respiratory syndrome viruses

Abstract

Type 2 (or North American-like) porcine reproductive and respiratory syndrome virus (PRRSV) was first recorded in 1987 in the United States and now occurs in most commercial swine industries throughout the world. In this study, we investigated the epidemiological and evolutionary behaviors of type 2 PRRSV. Based on phylogenetic analyses of 8,624 ORF5 sequences, we described a comprehensive picture of the diversity of type 2 PRRSVs and systematically classified all available sequences into lineages and sublineages, including a number of previously undescribed lineages. With the rapid growth of sequence deposition into the databases, it would be technically difficult for veterinary researchers to genotype their sequences by reanalyzing all sequences in the databases. To this end, a set of reference sequences was established based on our classification system, which represents the principal diversity of all available sequences and can readily be used for further genotyping studies. In addition, we further investigated the demographic histories of these lineages and sublineages by using Bayesian coalescence analyses, providing evolutionary insights into several important epidemiological events of type 2 PRRSV. Moreover, by using a phylogeographic approach, we were able to estimate the transmission frequencies between the pig-producing states in the United States and identified several states as the major sources of viral spread, i.e., "transmission centers." In summary, this study represents the most extensive phylogenetic analyses of type 2 PRRSV to date, providing a basis for future genotyping studies and dissecting the epidemiology of type 2 PRRSV from phylogenetic perspectives.

FIG. 1.

Classification and evolutionary history of all type 2 PRRSV lineages. (Right) Overall phylogeny constructed using MrBayes, which was midpoint rooted and partitioned into lineages by dotted lines. Information for each lineage is indicated to the side of the phylogeny, including diversity of the lineage, its distribution within the United States, its international distribution, and some well-known isolates. For clarity, only the bootstrap values for selected and well-supported nodes are shown. (Left) Summary of coalescence analyses of the overall and lineage data sets, performed in BEAST. The summary is based on the backbone structure of maximum clade credibility (MCC) trees reconstructed in BEAST, with unsupported nodes collapsed together to prevent uncertainty. Each box represents a well-supported lineage. The left boundary of the box is aligned with the emergence time of the lineage, whereas the right boundary of the box is aligned with the sampling time of the most recent sample. The shaded area inside the box corresponds to the periods when sequences were sampled. The black curve inside the box is a BSP, and the gray curves represent upper and lower 95% HPD values. The y axis of each BSP is relative genetic diversity, represented on a log₁₀ scale. For some of the lineages, BSP was not performed because of limited sample size or a serious deviation from clock behavior (vaccine-associated lineages). In the cases of vaccine lineages, we used histograms instead, which revealed the sampling size during the course of their history. The y axis of each histogram shows the number of samples.

FIG. 2.

Classification and evolutionary history of lineage 1 PRRSVs. The Bayesian phylogeny is divided into sublineages. Figure descriptions are as described in the legend to Fig. 1.

FIG. 3.

Classification and evolutionary history of lineage 8 PRRSVs. Figure descriptions are as described in the legend to Fig. 2.

FIG. 4.

Classification and evolutionary history of lineage 9 PRRSVs. Figure descriptions are as described in the legend to Fig. 2.

FIG. 5.

Interstate PRRSV directional transmission frequencies for 10 states in the United States. The viral outflows from six states, namely, Minnesota (A), Iowa (B), North Carolina (C), Nebraska (D), Oklahoma (E), and Texas (F), are demonstrated on the maps. Frequencies of transmission from one state to another are divided into 5 categories, corresponding to the thickness of the connecting line. (G) Pairwise transmission frequencies between states, with the number of asterisks indicating the relative extent of transmission.