Molecular epidemiology of coxsackievirus A16: intratype and prevalent intertype recombination identified

Abstract

Coxsackievirus A16 (CVA16) is responsible for nearly 50% of all the confirmed hand, foot, and mouth disease (HFMD) cases in mainland China, sometimes it could also cause severe complications, and even death. To clarify the genetic characteristics and the epidemic patterns of CVA16 in mainland China, comprehensive bioinfomatics analyses were performed by using 35 CVA16 whole genome sequences from 1998 to 2011, 593 complete CVA16 VP1 sequences from 1981 to 2011, and prototype strains of human enterovirus species A (EV-A). Analysis on complete VP1 sequences revealed that subgenotypes B1a and B1b were prevalent strains and have been co-circulating in many Asian countries since 2000, especially in mainland China for at least 13 years. While the prevalence of subgenotype B1c (totally 20 strains) was much limited, only found in Malaysia from 2005 to 2007 and in France in 2010. Genotype B2 only caused epidemic in Japan and Malaysia from 1981 to 2000. Both subgenotypes B1a and B1b were potential recombinant viruses containing sequences from other EV-A donors in the 5'-untranslated region and P2, P3 non-structural protein encoding regions.

Figure 1. The Neighbor-joining tree of all the 629 CVA16 viruses based on complete VP1 encoding sequences.

The branches showed in red color highlight the sequences from China.

Figure 2. Global distribution of subgenotypes of CVA16.

The time and spatial distribution of all genotypes, except genotype A, were shown in this chart. All the 628 genotype B strains from all other countries were shown in this chart except CVA16 prototype strain. Bar in blue, red, purple, and green indicate the subgenotype B2, B1a, B1b, and B1c, respectively. The length of the bar correlated with the numbers of the isolates in corresponding year.

Figure 3. Phylogenetic dendrogram showing the relationships between the cluster B1a, B1b and EV-A prototypes.

These were constructed by the different genomic regions of 35 complete CVA16 genomic sequences and EV-A prototype sequences. The neighbour-joining trees were reconstructed based on the P1 (a), P2 (b), P3 (c) genomic regions and complete genomes (d), respectively. Triangles indicated the subgenotype B1a. Rhombuses indicated the subgenotype B1b. Squares indicated the CVA16 prototype G-10 strain. The percentage of bootstrap (percentage of 1000 pseudoreplicate datasets) supporting the trees are indicated at the nodes; only values over 80% are shown.

Figure 4. Similarity plot and bootscanning analyses of all the two subgenotypes B1a, B1b of CVA16 and other EV-A prototype strains on the basis of full-length genomes.

Two strains were chosen as the representative strains of each subgenotype, respectively. Each of the four viruses was used as the query sequence. A sliding window of 500 nucleotides moving in 20 nucleotides steps was used in this analysis. (a and b) Tainan/5079/98/Taiwan/1998; (c and d) BJ/11/11 /BJ/CHN/2011; (e and f) shzh00-1/GD/CHN/2000; (g and h) BJ11/12/BJ/CHN/2011. The a, c, e, g were the results of similarity plot analyses; the b, d, f, h were the results of bootscanning analyses.