Molecular characterization of bovine enteric caliciviruses: a distinct third genogroup of noroviruses (Norwalk-like viruses) unlikely to be of risk to humans

Abstract

Bovine enteric caliciviruses (BoCVs) have been classified in the Norovirus (Norwalk-like virus) genus of the Caliciviridae, raising questions about zoonotic transmission and an animal reservoir for the human Norwalk-like viruses (NLVs), an important cause of nonbacterial gastroenteritis in humans. We examined the genetic relationship of human NLVs to BoCVs that were identified by using reverse transcription-PCR with primer pairs originally designed to detect human NLVs. Polymerase, capsid, and open reading frame 3 (ORF3) gene sequence analyses of BoCVs that were identified from 1976 to 2000 from throughout the United Kingdom showed that BoCVs formed a distinct third genogroup of closely related viruses distinct from the human genogroup I and II NLVs. Evidence was not obtained to support the concept that BoCVs are circulating in humans and pose a threat to human health.

FIG. 1.

Multiple amino acid sequence alignment of the four UK BoCV capsid proteins, the German Bo/Jena/80/DE, and a UK representative of human NLV genogroups I (Hu/NLV/Southampton/91/UK) and II (Hu/NLV/Melksham/89/UK). The dotted line beneath the sequences represents amino acids 50 to 225 of the S domain, the solid lines represent the P1 domain, and the dashed line represents the P2 domain of the Hu/Norwalk/68/US (36). The conserved 5-amino-acid LAGNA motifs located at the 3- and 5-fold axes of symmetry of Hu/Norwalk/68/US are highlighted in gray.

FIG. 2.

Schematic representation and nucleotide sequence of the reference UK BoCV (Bo/Newbury2/76/UK), the German BoCV (Bo/Jena/80/DE), the human NLV genogroup I Hu/NLV/Norwalk/68/US, and the human NLV genogroup II Hu/NLV/Melksham/89/UK viruses in the region of the ORF1-2 and ORF2-3 overlaps, showing start (in bold) and stop (underlined) codons as predicted by the NCBI ORF finder. The boxes on the left depict the 3′ ends of the ORF1s, the pairs of black boxes in the center depicts the 5′ and 3′ ends of the ORF2s, and the boxes on the right depict the complete ORF3s. The second initiation codon of the Bo/Newbury2/76/UK sequence is shown (bold italicized). The 173 bases not shown in the Bo/Newbury2/76/UK ORF2-3 overlap are labeled 173. The scale indicates nucleotide positions in relation to Bo/Jena/80/DE.

FIG. 3.

Amino acid identity plot generated by using SimPlot (29) with a window of 80 amino acids at 20-amino-acid intervals to compare the complete capsid protein of Bo/Newbury2/76/UK with the capsid protein of Bo/Jena/80/DE and the genogroup I human NLV Hu/Southampton/91/UK and the genogroup II NLV Hu/Melksham/89/UK. The dotted line beneath the plot represents the S domain, the solid line represents the P1 domain, and the dashed line represents the P2 domain of the Hu/NLV/Norwalk/68/US, as determined by X-ray crystallography (36).

FIG. 4.

Phylogenetic analysis generated by using the TreePuzzle maximum likelihood method for the translated amino acid sequences of (A) the long polymerase sequence (138 amino acids) and (B) 113 amino acids from the 5′ end of the ORF3 gene (starting from the Bo/Jena/80/DE ORF3 initiation codon) of the UK BoCVs, the German Bo/NLV/Jena/80/DE, and representatives of the human NLV genogroups I and II, with the Sapporo-like virus Hu/Manchester/93/UK used as the outgroup. The numbers at the nodes (quartet puzzling values) indicate the frequencies of occurrence for 1,000 replicate trees. The asterisk indicates the sample identified with the GLPSG1/YGDD1 primer pair.

FIG. 5.

Phylogenetic analysis generated by using the TreePuzzle maximum likelihood method for the translated amino acid sequences of the complete capsid protein gene, the UK BoCVs, the German Bo/Jena/80/DE, the two unpublished Dutch BoCVs Bo/CH126/98/NET and Bo/CH131/98/NET, and representatives of the seven clusters in the human NLV genogroups I and II, with the Sapporo-like virus Hu/Manchester/93/UK used as the outgroup. The numbers at the nodes (quartet puzzling values) indicate the frequencies of occurrence for 1,000 replicate trees. The superscripts represent the genetic clusters of the genogroup I and II viruses, as determined by Ando et al. (1).