Genetic characterization of norovirus GII.4 variants circulating in Canada using a metagenomic technique

Abstract

Background: Human norovirus is the leading cause of viral gastroenteritis globally, and the GII.4 has been the most predominant genotype for decades. This genotype has numerous variants that have caused repeated epidemics worldwide. However, the molecular evolutionary signatures among the GII.4 variants have not been elucidated throughout the viral genome.

Method: A metagenomic, next-generation sequencing method, based on Illumina RNA-Seq, was applied to determine norovirus sequences from clinical samples.

Results: Herein, the obtained deep-sequencing data was employed to analyze full-genomic sequences from GII.4 variants prevailing in Canada from 2012 to 2016. Phylogenetic analysis demonstrated that the majority of these sequences belong to New Orleans 2009 and Sydney 2012 strains, and a recombinant sequence was also identified. Genome-wide similarity analyses implied that while the capsid gene is highly diverse among the isolates, the viral protease and polymerase genes remain relatively conserved. Numerous amino acid substitutions were observed at each putative antigenic epitope of the VP1 protein, whereas few amino acid changes were identified in the polymerase protein. Co-infection with other enteric RNA viruses was investigated and the astrovirus genome was identified in one of the samples.

Conclusions: Overall this study demonstrated the application of whole genome sequencing as an important tool in molecular characterization of noroviruses.

Keywords: Antigenic drift; Co-infection; Metagenomics; Next-generation sequencing; Norovirus; Recombination.

Fig. 1

Phylogenetic analysis of individual ORFs. Consensus sequences obtained in this study along with certain full-length Canadian sequences and reference sequences were aligned and phylogenetic trees were constructed for ORF1 (a), ORF2 (b) and ORF3 (c) using the Maximum Likelihood method. The robustness of the phylogeny was assessed through bootstrap analysis of 1000 pseudo-replicates. Sequences in brown are isolated from Ontario, orange from Alberta, blue for Nova Scotia, green from Newfoundland and Labrador. The recombinant sequence is shown in bold. The sequences generated in this study are italicized

Fig. 2

Phylogenetic analysis of the full-length genome sequences. A phylogenetic tree was constructed using certain full-length genome sequences obtained in this study and several full-length sequences from different geographical regions using the Maximum Likelihood method. Sequences in red are isolated from Ontario, blue for Nova Scotia, green from Newfoundland. The recombinant sequence is shown in bold

Fig. 3

SimPlot analysis of the complete genomic sequence of BMH16–078 recombination. SimPlot was constructed using the RDP4 Software version 4.72 with a slide window width of 200 bp and a step size of 20 bp. At each position of the window, the query sequence was compared to each of the reference strains. The X-axis indicates the nucleotide positions in the multiple alignments of the NoV sequences; and the Y-axis indicates nucleotide identities between the query sequence and the NoV reference strains

Fig. 4

SimPlot analysis of the complete genomic sequences of GII.4 Sydney 2012. SimPlot was constructed using a Simplot software version 3.5 with a slide window width of 200 bp and a step size of 20 bp. At each position of the window, the query sequence was compared to other Sydney 2012 variants sequenced in this study. The X-axis indicates the nucleotide positions in the multiple alignments of the NoV sequences; and the Y-axis indicates nucleotide difference (%) between the query sequence (BMH15–58) and other sequenced Sydney 2012 strains

Fig. 5

Non-synonymous differences within the structural domains of the capsid protein (VP1, ORF2), which are the N-terminal (N), shell (S), P1, and P2 domains. Individual epitope sites are highlighted in different colors and putative conformational epitopes are shown as regions 2–4. Unique variants are shown by *. Accession numbers for Farmington Hills 2002, Hunter 2004, Den Haag 2006 1, Den Haag 2006 2, Apeldoorn 2007, New Orleans 2009 1, New Orleans 2009 2, Sydney 2012 1, Sydney 2012 2, and Sydney 2012 3 are JX445152, JX445153, JX445158, JX445155, JX445161, JX445164, JX445145, KF509946, KF509947, KJ96280, respectively

Fig. 6

Amino acid variations within the RNA-dependent RNA polymerase (RdRP) protein of the Sydney 2012 sequences. The Jalview histogram below the alignment indicates the conservation of the physico-chemical properties for each column (lower bars with lower numbers, lower conservation; completely conserved columns are in yellow)