As part of a project to identify viral etiology of diarrhea, 393 fecal samples were collected from children <5 years of age with watery diarrhea attending 2 hospitals in Bhutan. During February 2010–January 2012, 381 fecal samples (109 in 2010, 185 in 2011, and 87 in 2012) were collected from hospitalized patients or outpatients of Jigme Dorji Wangchuk National Referral Hospital. This hospital mainly serves the population of Thimphu and is the only national reference hospital in the country. In April 2010, 12 more samples were obtained from children hospitalized in Mongar Regional Referral Hospital. This hospital mainly serves the population of Mongar and is the reference hospital for the eastern part of the country. The Research Ethics Board of Health in Bhutan approved this study.

Viral genomic DNA was extracted from fecal samples by using the QIAamp Viral RNA Mini Kit (QIAGEN Sciences, Valencia, CA, USA) according to the manufacturer’s instructions. The presence of BuV was determined by nested PCR targeting the NS1 region (1). For whole-genome sequencing of BuV3, primers were constructed from consensus regions after aligning the whole-genome sequences of BuV1. BuV-positive samples were tested for norovirus, bocavirus, adenovirus, astrovirus, salivirus, cosavirus, and aichivirus by PCR, and for rotavirus by enzyme immunoassay (5–8). No tests for diarrheic bacteria were done on these samples. PCR amplicons were directly sequenced to confirm the findings and to compare the sequences. Nucleotide sequencing was performed by using the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA USA) according to the manufacturer’s instructions. Multiple sequence alignment was done by using MUSCLE (9) and the phylogenetic tree was constructed by using the neighbor-joining method by MEGA 5 (10). A bootstrap analysis of 1,000 replicates was done to find the significance of branching.

Three of 393 samples (0.8%) were positive for BuV DNA. The overall proportion of positive samples of BuV in Thimphu was 0.5% (2 samples); 1 (0.9%) was detected in 2010 and the other (1.1%) in 2011. No samples from Thimphu were BuV-positive in 2012. In addition, 1 sample (8.3%) from Mongar was BuV-positive. The demographic data of these patients are shown in Table 1. In 1 of the 3 BuV-positive fecal samples, norovirus GII.6 was also detected; in the other 2 samples, the enteric viruses tested for were not detected.

The near-complete genome sequences of the 3 BuVs collected in Bhutan were obtained. The lengths of the open reading frames of NS1, VP1, and VP2 of the Bhutanese BuV genes were different from those of BuV1 and BuV2 (Table 2). Phylogenetic analyses, with a bootstrap value of 100, of the deduced amino acid sequence of NS1, VP1 (Figure), and VP2 genes showed that the Bhutanese BuVs consistently formed a cluster different from the genotypes BuV1 and BuV2. The NS1, VP1, and VP2 of the Bhutanese isolates showed ≈99% nucleotide and amino acid identities among themselves. Compared with BuV1, the respective NS1, VP1, and VP2 nt (amino acid) identities of the Bhutanese BuVs were 96% (95%–96%), 83% (77%–78%), and 79%–80% (73%), and compared with BuV2, 96% (95%), 80% (71%), and 76% (65%). Phylogenetic analysis and identity data support the finding of a new genotype, BuV3, in Bhutan. Similar to BuV1 and BuV2, all isolates collected in Bhutan contained an ATP- or GTP-binding Walker loop, phospholipase A₂, 2 conserved replication initiator motifs, and a glycine-rich sequence; they also had the same splice sites. In the Bhutanese BuVs, 3 tandem repeats, TAGTTGATAAGT, TAGTTTATAAGT, and TAGTTTATAAAT, in the 3′ untranslated region (UTR) occurred at a frequency of 0, 6, or 3 times, and 0 or 1 times, respectively (Table 2). These results show that BuV3 3′UTR differs from 3′UTR of the other 2 genotypes.

Genetic analyses of BuVs from Bhutan revealed a new genotype that may cause severe diarrhea in children. The absence of other diarrhea-causing viruses tested for in 2 of the 3 samples containing BuV3 supports a possible role for this parvovirus in diarrhea in infants; however, direct proof of causation is necessary (11,12). A more extensive testing for other known pathogens by using virochip microarrays, metagenomics, and direct testing for pathogenic bacteria will help determine whether BuVs alone can cause diarrhea.

Phylogenetic analyses and identity data indicate that BuV3 is a new genotype. The identity data between BuV3 and other genotypes are comparable to the data found by Phan et al., wherein genotypes BuV1 and BuV2 showed 95% identity in the NS1 region, but only 72% identity in the capsid region (1). In BuV1, the 3 tandem repeats each occurred with a frequency of 3 or 2 times, once, and once, respectively. This differs substantially from that of BuV3 found in Bhutan, further revealing the molecular characteristics of the new genotype.

In conclusion, we have detected BuV DNA in the feces of children with severe diarrhea in Bhutan. Additional work is needed to determine this is indeed a human causal pathogen of diarrhea or of any illness. Identification of BuV3 provides evidence of a more diverse population of BuVs than previously documented. Whether this novel genotype is circulating globally or only in Bhutan requires further investigation.