Sequencing and analysis of globally obtained human respiratory syncytial virus A and B genomes

Abstract

Background: Human respiratory syncytial virus (RSV) is the leading cause of respiratory tract infections in children globally, with nearly all children experiencing at least one infection by the age of two. Partial sequencing of the attachment glycoprotein gene is conducted routinely for genotyping, but relatively few whole genome sequences are available for RSV. The goal of our study was to sequence the genomes of RSV strains collected from multiple countries to further understand the global diversity of RSV at a whole-genome level.

Methods: We collected RSV samples and isolates from Mexico, Argentina, Belgium, Italy, Germany, Australia, South Africa, and the USA from the years 1998-2010. Both Sanger and next-generation sequencing with the Illumina and 454 platforms were used to sequence the whole genomes of RSV A and B. Phylogenetic analyses were performed using the Bayesian and maximum likelihood methods of phylogenetic inference.

Results: We sequenced the genomes of 34 RSVA and 23 RSVB viruses. Phylogenetic analysis showed that the RSVA genome evolves at an estimated rate of 6.72 × 10(-4) substitutions/site/year (95% HPD 5.61 × 10(-4) to 7.6 × 10(-4)) and for RSVB the evolutionary rate was 7.69 × 10(-4) substitutions/site/year (95% HPD 6.81 × 10(-4) to 8.62 × 10(-4)). We found multiple clades co-circulating globally for both RSV A and B. The predominant clades were GA2 and GA5 for RSVA and BA for RSVB.

Conclusions: Our analyses showed that RSV circulates on a global scale with the same predominant clades of viruses being found in countries around the world. However, the distribution of clades can change rapidly as new strains emerge. We did not observe a strong spatial structure in our trees, with the same three main clades of RSV co-circulating globally, suggesting that the evolution of RSV is not strongly regionalized.

Fig 1. Protein Entropy Plots.

This is an entropy plot of the concatenated predicted protein sequences of all of the viruses sequenced in this study. Entropy values were calculated using BioEdit 7.0 and the plot was generated using Microsoft Excel. Black bars are for RSVA sequences and red bars are for RSVB sequences. The higher the bar is the greater the variation at that position in the protein sequence. Across the top of the plot are listed the abbreviated protein sequence names in the order in which the CDS sequences for the proteins appear in the genome.

Fig 2. G Protein Entropy Plot and Positive Selected Sites.

This is an entropy plot of the G protein sequences with positively selected sites also shown. Entropy values were calculated using BioEdit 7.0 from the alignments used for the positive selection analysis and the plot was generated using Microsoft Excel. Black bars are for RSVA sequences and red bars are for RSVB sequences. Sites predicted to be under positive selection in this study are shown with black diamonds for RSVA and red diamonds for RSVB. Near the top are shown sites predicted to be under positive selection or diversifying selection from previously published studies with black pluses for RSVA and red pluses for RSVB.

Fig 3. CDS Evolutionary Rates.

This plot shows the estimated evolutionary rates for each CDS in RSVA and RSVB. Error bars represent the 95% HPD values. Rates were determined using BEAST v1.8.0 with the GTR model of nucleotide substitution, a gamma-distributed rate variation among sites, an uncorrelated lognormal relaxed molecular clock, and a flexible Bayesian skyline tree prior. Each CDS was ran with a chain length of 50 million and sampled 10,000 times.

Fig 4. Tree of RSVA Genome Sequences.

This is a maximum clade credibility tree of RSVA genome sequences generated in this study and retrieved from GenBank. Tip times correspond to date of collection with the scale axis across the bottom showing the years. Tip labels show the accession number, country of isolation, and collection date. The labels are color coded with black for sequences from this study (FTS), grey for sequences with an undetermined genotype (UND), and the remaining colors corresponding to previous published genotypes as show in the key in the upper left corner. Brackets highlight the major clades. Bayesian posterior probabilities are shown for key nodes.

Fig 5. Tree of RSVB Genome Sequences.

This is a maximum clade credibility tree of RSVB genome sequences generated in this study and retrieved from GenBank. Tip times correspond to date of collection with the scale axis across the bottom showing the years. Tip labels show the accession number, country of isolation, and collection date. The labels are color coded with black for sequences from this study (FTS), grey for sequences with an undetermined genotype (UND), and the remaining colors corresponding to previous published genotypes as show in the key in the upper left corner. Brackets highlight the major clades. Bayesian posterior probabilities are shown for key nodes.