Identification of a novel picornavirus related to cosaviruses in a child with acute diarrhea

Abstract

Diarrhea, the third leading infectious cause of death worldwide, causes approximately 2 million deaths a year. Approximately 40% of these cases are of unknown etiology. We previously developed a metagenomic strategy for identification of novel viruses from diarrhea samples. By applying mass sequencing to a stool sample collected in Melbourne, Australia from a child with acute diarrhea, one 395 bp sequence read was identified that possessed only limited identity to known picornaviruses. This initial fragment shared only 55% amino acid identity to its top BLAST hit, the VP3 protein of Theiler's-like virus, suggesting that a novel picornavirus might be present in this sample. By using a combination of mass sequencing, RT-PCR, 5' RACE and 3' RACE, 6562 bp of the viral genome was sequenced, which includes the entire putative polyprotein. The overall genomic organization of this virus was similar to known picornaviruses. Phylogenetic analysis of the polyprotein demonstrated that the virus was divergent from previously described picornaviruses and appears to belong to the newly proposed picornavirus genus, Cosavirus. Based on the analysis discussed here, we propose that this virus represents a new species in the Cosavirus genus, and it has tentatively been named Human Cosavirus E1 (HCoSV-E1).

Figure 1

Genomic organization of Cosavirus. Schematic of initial protein products P1, P2, and P3 (A). Schematic of processed polyprotein (B). Representation of sequence obtained from Human Cosavirus-E1 (C).

Figure 2

Phylogenetic Analysis of HCoSV-E1. Multiple sequence alignments were generated with HCoSV-E1 P1 (A), P2 (B), and P3 (C) sequences and the corresponding regions of known picornaviruses using ClustalX. PAUP was used to generate phylogenetic trees and bootstrap values (> 700) from 1,000 replicates are shown.