Global Distribution and Evolutionary History of Enterovirus D68, with Emphasis on the 2014 Outbreak in Ontario, Canada

Abstract

Despite its first appearance in 1962, human enterovirus D68 (EV-D68) has been recognized as an emerging respiratory pathogen in the last decade when it caused outbreaks and clusters in several countries including Japan, the Philippines, and the Netherlands. The most recent and largest outbreak of EV-D68 associated with severe respiratory illness took place in North America between August 2014 and January 2015. Between September 1 and October 31 2014, EV-D68 infection was laboratory confirmed among 153/907 (16.9%) persons tested for the virus in Ontario, Canada, using real time RT-PCR and subsequent genotyping by sequencing of partial VP1 gene. In order to understand the evolutionary history of the 2014 North American EV-D68 outbreak, we conducted phylogenetic and phylodynamic analyses using available partial VP1 genes (n = 469) and NCBI available whole genome sequences (WGS) (n = 38). The global EV-D68 phylogenetic tree (n = 469) reconfirms the divergence of three distinct clades A, B, and C from the prototype EV-D68 Fermon strain as previously documented. Two sub-clades (B1 and B2) were identified, with most 2014 EV-D68 outbreak strains belonging to sub-cluster B2b2 (one of the two emerging clusters within sub-clade B2), with two signature substitutions T650A and M700V in BC and DE loops of VP1 gene, respectively. The close homology between WGS of strains from Ontario (n = 2) and USA (n = 21) in the recent EV-D68 outbreak suggests genetic relatedness and also a common source for the outbreak. The time of most recent common ancestor of EV-D68 and the 2014 EV-D68 outbreak strain suggest that the viruses possibly emerged during 1960-1961 and 2012-2013, respectively. We observed lower mean evolutionary rates of global EV-D68 using WGS data than estimated with partial VP1 gene sequences. Based on WGS data, the estimated mean rate of evolution of the EV-D68 B2b cluster was 9.75 × 10^-3 substitutions/site/year (95% BCI 4.11 × 10^-3 to 16 × 10^-3).

Keywords: cluster lineage; enterovirus D68; next-generation sequencing; outbreak; phylogenetic analysis; whole genome sequencing.

FIGURE 1

Phylogenetic tree of EV-D68 detected globally and from Ontario, Canada. The phylogenetic tree based on the alignment of partial VP1 sequences (339 bp) of EV-D68 was inferred using the Neighbor-Joining method, evolutionary distances were calculated with the Maximum Composite Likelihood method (MCL), and 1,000 Bootstrap replicates were computed to estimate the accuracy of the phylogenetic inference using MEGA v6.06 (http://www.megasoftware.net). Ontario sequences from this study are identified with red filled circles. Sequences collected in 2014 from the USA, Italy, and Netherlands are marked with a blue square, green diamond, and light blue triangle, respectively. AFP sequences collected in 2013 obtained from GenBank are labeled with a green circle. Bootstrap values > 60% are shown next to the nodes. Branch lengths are drawn to the indicated scale, proportion of nucleotide substitutions per site. Some sub-clades are collapsed and labeled in bold to assist with visualization of other clades/sub-clades.

FIGURE 2

(A) Global distribution of EV-D68 as of December 2014. The location and corresponding year of specimen collection for EV-D68 VP1 sequences were obtained from the NCBI GenBank sequence database and color coded by country. EV-D68 clades and sub-clades documented in each country were assigned based on phylogenetic analysis of available partial (339 bp) VP1 nucleotide sequences as shown in Figure 1. Strains collected during 2014 are marked for Ontario (red), USA (blue), Netherlands (cyan), and Italy (green). The editable vector map of the world template was downloaded (http://www.presentationmagazine.com/world16maps-vector-editable-507.htm) and edited using PowerPoint and Adobe Photoshop. (B) The maximum clade credibility (MCC) tree of VP1 sequences of available strains including the ones from Ontario visualized by FigTree. Branches are color coded according to the world map in (A). Temporal x-axis scale represents the sampling dates. (C) Global distribution of clades, clusters and sub-clusters.

FIGURE 3

The phylogenetic relationships of the complete EV-D68 polyprotein gene (6567 nt) of Ontario strains (n = 2) to those available in NCBI’s GenBank sequence database (n = 36). Topology was constructed using the Neighbor-Joining analysis and the evolutionary distances were computed using the Maximum Composite Likelihood method available in MEGA v6.06. Bootstrap values (>70%) are shown as percentages derived from 1,000 samplings at the nodes of the tree. The scale bar represents the number of nucleotide substitutions per site. Strains from Ontario are marked with solid black circles.

FIGURE 4

(A) Schematic representation of EV-D68 genome. The black bars represent the location of nucleotide variation within strains of clade B2.b2. Height represents number of strains with specific nucleotide variation shown. (B) Nucleotide and amino acid identities between sub-clusters B2.b1 and B2.b2 (white area) and between the complete genomes of two strains from Ontario belonging to sub-clades B2.b1 and B2.b2 (gray area).