A fecal specimen was obtained from a piglet experimentally infected with strain SKA-1, which was first isolated from a pig with diarrhea in Tottori Prefecture, Japan. RNA was extracted from a 20% fecal suspension in phosphate-buffered saline by using the ISOGEN-LS Kit (Nippon Gene Ltd., Toyama, Japan). Cloning was performed according to the method described by Lambden et al. (13) with some modifications (14). During single-primer amplification, several DNA bands were detected that appeared to correspond to RNA segments of a rotavirus.

DNA was purified by using a Wizard SV Gel and a PCR Clean-Up System (Promega, Madison, WI, USA) and cloned into the PCR-TOPO vector by using a TOPO TA Cloning Kit (Invitrogen, Carslbad, CA, USA). Six clones for each of the segments were selected and used for sequencing. Sequence analysis and comparisons were performed by using GENETYX-WIN software (GENETYX, Tokyo, Japan) and MEGA software (www.megasoftware.net/).

We selected 4 genes encoding VP4, VP6, VP7, and NSP4 for sequence analysis. Sequence data were obtained for comparative sequence analysis among group A and nongroup A rotaviruses. There was little identity for any of the 4 genes between SKA-1 and group A or C rotaviruses. In contrast, nucleotide and amino acid sequences of the 4 genes of SKA-1 showed relatively high identities with those of a novel group of rotavirus strains (J19 and B219). The VP6 gene, which is associated with group specificity, of SKA-1 showed highest identities among the 4 genes with those of the novel group human rotaviruses: 72%–73% at the nucleotide level and 76%–77% at the amino acid level. VP6 also showed relatively high identities with those of group B rotaviruses: 52%–53% at the nucleotide level and 36%–40% at the amino acid level (Table). The other 3 genes (VP4, VP7, and NSP4), also showed high identities with those of the novel group of human rotaviruses, although identity values were lower than those for the VP6 gene (Table).

Lengths and nucleotide sequences of the 5′ and 3′ noncoding regions of the 4 genes of SKA-1 were similar to those of the novel group of rotaviruses. Using the VP7 gene as a reference, we identified sequences of the 5′ noncoding region of VP7 as 5′-GGAACTTTAAAGCC-3′ for strain SKA-1, 5′-GGCAATTTGAAGCC-3′ for the novel group of human rotaviruses, and 5′-GGCAATAAA-3′ for group B rotaviruses. Phylogenetic analysis of the 4 genes also showed that SKA-1 is closely related to the novel group of human rotavirus strains (J19 and B219) (Figure).

Although group A rotaviruses have been extensively studied, nongroup A rotaviruses have not been extensively studied. In particular, there is little information on group D, E, F, and G rotaviruses. A novel group of human rotaviruses (J19 and B219) are not related to any other groups of rotaviruses.

Rotavirus strain SKA-1 was detected in a fecal specimen from a piglet with diarrhea in Japan, and was isolated in MA-104 cells in 1999. Its group specificity was suggested to be group B on the basis of reverse transcription PCR results with a group B–specific primer. However, our sequence analysis showed that the porcine SKA-1 strain is different from group B human and porcine rotaviruses.

Nucleotide and amino acid sequences of SKA-1 did not show identities with those of group A or C rotaviruses, although they showed some relatedness to those of group B rotaviruses. However, high identities were observed between SKA-1 and a novel group of human rotaviruses. In addition, there are similarities in the 4 genes analyzed in this study between SKA-1 and the novel group of human rotaviruses: 1) nucleotide sequences and nucleotide numbers of noncoding regions at the 5′ and 3′ ends are similar to each other; 2) lengths of nucleotide and deduced amino acid sequences are similar to each other; and 3) phylogenetic analysis showed that these viruses are in the same cluster.

Except for porcine strain SKA-1, there has been no report of animal rotaviruses being classified into a novel group of human rotaviruses. A survey of the prevalence of antibodies against SKA-1 among humans and animals, including pigs, would be useful. A classification system for nongroup A rotaviruses has not been established because 1) information on nucleotide sequences of group D, E, F, and G rotaviruses is lacking; 2) no expressed reference VP6 proteins are available; 3) reference strains of group D–G rotaviruses have not been adapted to cell culture; and 4) it is not known whether fecal samples or RNA or extracted RNA of group E–G strains are available in any laboratories. On the basis of serologic characterization and sequence analysis, a classification system for nongroup A rotaviruses, which includes the novel group of human and porcine rotaviruses such as J19, B219, and SKA-1, should be established.