Genetic relationship between cocirculating Human enteroviruses species C

Abstract

Recombination events between human enteroviruses (HEV) are known to occur frequently and to participate in the evolution of these viruses. In a previous study, we reported the isolation of a panel of viruses belonging to the Human enterovirus species C (HEV-C) that had been cocirculating in a small geographic area of Madagascar in 2002. This panel included type 2 vaccine-derived polioviruses (PV) that had caused several cases of acute flaccid paralysis in humans. Previous partial sequencing of the genome of these HEV-C isolates revealed considerable genetic diversity, mostly due to recombination. In the work presented herein, we carried out a more detailed characterization of the genomes of viruses from this collection. First, we determined the full VP1 sequence of 41 of these isolates of different types. These sequences were compared with those of HEV-C isolates obtained from other countries or in other contexts. The sequences of the Madagascan isolates of a given type formed specific clusters clearly differentiated from those formed by other strains of the same type isolated elsewhere. Second, we sequenced the entire genome of 10 viruses representing most of the lineages present in this panel. All but one of the genomes appeared to be mosaic assemblies of different genomic fragments generated by intra- and intertypic recombination. The location of the breakpoints suggested potential preferred genomic regions for recombination. Our results also suggest that recombination between type HEV-99 and other HEV-C may be quite rare. This first exhaustive genomic analysis of a panel of non-PV HEV-C cocirculating in a small human population highlights the high frequency of inter and intra-typic genetic recombination, constituting a widespread mechanism of genetic plasticity and continually shifting the HEV-C biodiversity.

Figure 1. Phylogenetic relationships between Madagascan HEV-C field isolates collected in 2002 and other isolates for which sequences are available in GenBank, based on full-length VP1 sequences.

The length of the branches is proportional to the number of nucleotide changes (percent divergence). The percentage of bootstrap replicates is indicated for the main nodes. Each area of grey shading corresponds to a serotype. The field strains isolated in Madagascar in 2002 are indicated by circles; the isolates whose full-length genome was subsequently sequenced are indicated in bold. For the other isolates, the location and year of isolation are indicated in the tree. Triangles indicate the prototype strains.

Figure 2. Phylogenetic relationships between Madagascan CV-A13 field strains and other CV-A13 isolates for which sequences are available in GenBank, based on 3′ one-third of the VP1 region (∼300 nt).

The length of the branches is proportional to the number of nucleotide changes (percent divergence). The percentage of bootstrap replicates is indicated if higher than 70%. The field strains isolated in Madagascar are indicated by full circles if isolated in 2002, by open circles if isolated in other years. For the other isolates, the location and year of isolation are indicated in the tree. Triangles indicate the prototype strains. The CV-A11 G9 sequence was introduced for correct rooting of the tree.

Figure 3. Phylograms depicting the relationships between the studied HEV-C in different genomic regions.

The percentage of bootstrap replicates is indicated at nodes if higher than 70%. The length of branches is proportional to the number of nucleotide changes (percent divergence). The sequences of CV-A10 and EV-19 (members of HEV species A and B, respectively) were introduced for correct rooting of the tree. Triangles indicate the prototype strains, circles the field strains; the VDPV strain MAD004 is labelled with open circles. Each color corresponds to a given HEV-C serotype. Below each tree, the region taken in consideration for alignment is highlighted in red in the schematic diagram of the genome.

Figure 4. Pairwise comparison and bootscanning analysis of CV-A11 66122 (A), CV-A13 68095 (B) and CV-A17 68154 (C) full-length genomes.

The dotted rectangles indicate the putative recombination hotspots.

Figure 5. Pairwise comparison and bootscanning analysis of CV-A17 67610 (A), HEV-99 68229 (B) and MAD004 (C) full-length genomes.

The dotted rectangles indicate the putative recombination hotspots.