Absence of Detectable Replication of Human Bocavirus Species 2 in Respiratory Tract

Human bocavirus (HBoV) commonly infects young children and is associated with respiratory disease; disease associations of the divergent HBoV-2 species are unknown. Frequent HBoV-2 detection in fecal samples indicated widespread circulation in the United Kingdom and Thailand, but its lack of detection among 6,524 respiratory samples indicates likely differences from HBoV-1 in tropism/pathogenesis.

S ince its discovery in 2005 (1), human bocavirus (HBoV) has been the subject of intense investigation as a potential cause of human respiratory disease (2). In addition to respiratory tract and systemic infections, HBoV DNA sequences are frequently detected in fecal samples during primary infections (3,4), although a causative role in viral gastroenteritis has not been established (5)(6)(7). Other parvoviruses, including canine and bovine members of the genus Bocavirus, can replicate in the gastrointestinal tract and are often linked to enteric disease (8,9).
Until recently, published genetic analyses reported minimal sequence variability of HBoV strains; 2 genetic lineages differed in nucleotide sequence by only 2% in the virus protein 2 (VP2) gene (10). However, more divergent HBoV-like variants, provisionally designated as HBoV species 2 (HBoV-2), have been identified in fecal samples from children in Pakistan and the United Kingdom. These viruses show >20% nt sequence divergence (11). Published primer sequences for HBoV contain several mismatches with HBoV-2 sequences that may prevent their amplification (11). Thus, published surveys of HBoV prevalence likely report only HBoV-1. Therefore, HBoV-2 may rep-resent an additional, currently undetected, agent in respiratory or enteric disease.

The Study
To investigate HBoV-2, we developed new PCR-based detection methods for HBoV by using primer sets highly conserved between HBoV-1 and HBoV-2 and speciesspecific primers for HBoV-2. Large-scale screening of persons in the United Kingdom and Thailand was performed to compare virus detection frequencies in respiratory and fecal samples.
A total of 6,138 respiratory samples from 3,754 persons (2,018 male, 1,722 female, 14 sex unknown) during January 1, 2007-June 30, 2008, were obtained from the Specialist Virology Centre (Edinburgh, UK). Samples were not identified but epidemiologic and demographic information was retained (12,13 DNA was extracted from 200-µL samples of pooled or individual specimens (respiratory samples, clarified fecal supernatant) into 40 µL Tris-EDTA buffer as described (13). Respiratory and fecal samples from Edinburgh were screened in pools of 10; both sample types from Bangkok were screened individually. Screening was performed by using nested primers conserved between HBoV-1 and HBoV-2 in the nucleoprotein (NP)-1 gene (universal primers: outer sense [position 2589 in DQ000496 st2 isolate (1) Amplication conditions were 94°C for 2 min and 35 cycles at 94°C for 18 s, 50°C for 21 s, and 72°C for 1.5 min. Amplicons were differentiated by digestion with RsaI. Fragments were sized by agarose gel electrophoresis. All known HBoV-1 sequences contain an RsaI site between nt 2772 and nt 2773, resulting in fragments of 46 bp and 91 bp; this site is absent in HBoV-2 (undigested amplicon length of 237 bp). Respiratory and fecal samples from both centers were screened by using universal primers, and positive samples were digested with RsaI. Undigested amplicons and some predicted HBoV-1 fragments (46 bp and 91 bp) were sequenced to confirm virus identity. All samples were additionally screened with HBoV-2-specific primers; 16 undigested samples were positive with HBoV-2-specific primers, and all samples identified as HBoV-1 were negative. Thus, species-specific primers enabled effective screening of HBoV-2 among samples with high frequencies of HBoV-1.
HBoV-positive fecal samples were generally restricted to children <5 years of age (25 from 30 infected children whose ages were known) (Figure, panel A; Table). Median age of children infected with HBoV-2 (7-12 months) was lower than that for those infected with HBoV-1 (1-2 years). Infections with HBoV-1 and HBoV-2 were observed at low frequencies in older persons (2 and 5 of 1,791 persons >35 years of age, respectively). For respiratory samples, HBoV-1 infections showed a similar peak incidence among children 1-2 years of age ( Figure, panel B), similar to that observed for fecal samples. This age group was most frequently infected in our previous analyses of respiratory samples from Edinburgh (12). There were no differences in frequencies of HBoV-1 or HBoV-2 infection between male and female participants. Samples from Bangkok were divided into those from persons with diarrhea (327) and asymptomatic controls (213); detection of HBoV-1 and HBoV-2 was restricted to persons with diarrhea (n = 12 and 2, respectively).
In contrast to its frequent detection in fecal samples, HBoV-2 was not detected in >6,500 respiratory samples (Table). However, high frequencies of HBoV-1 were recorded (14% among children in Bangkok and 3.4% among children in Edinburgh); the group from Edinburgh contained a substantial number of older children (37% >5 years of age).

Conclusions
Four conclusions can be drawn from this study. First, HBoV-2 circulates in 3 widely separated areas (United Kingdom, Thailand, and Pakistan [11]) and is likely distributed globally. Second, infections with HBoV-2 show a pattern of infecting young children, most <1 year of age. Third, absence of HBoV-2 in respiratory samples suggests a different tissue tropism that may influence its transmission route and ability to infect systemically and establish persistence. Determining the biologic basis for such differences will be useful in understanding the pathogenesis of HBoV-1-related respiratory disease. Fourth, at a practical level, absence of HBoV-2 in respiratory samples indicates no likely role for this virus in respiratory disease. Thus, screening methods may be adequate for detecting HBoVassociated respiratory disease. Nevertheless, the unexpectedly diverse human bocavirus group may contain additional variants with potential etiologic roles in respiratory or other diseases.
Since this study was completed, evidence for an interspecies HBoV-1/-2 recombinant associated with acute gastroenteritis has been obtained; the structural gene region was most closely related to HBoV-2, and NS1/NP-1 grouping with HBoV-1 (14). Although this recombinant would have been identified as HBoV-1 by using typing assays described in the current study, sequence analysis of HBoV-1-positive samples in this study and our previous study of respiratory samples from Edinburgh and Bangkok (12,15) identified only HBoV-1 in the study population, consistent with all other analyses of this sample type worldwide. Nevertheless, future typing assays should analyze both VP1/2 and NS/NP-1 to ensure that this and potentially other interspecies recombinants are identified. Investigation of genetic diversity of this group and development of effective screening methods for variants of HBoV is required for studies of human disease. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the Centers for Disease Control and Prevention or the institutions with which the authors are affiliated.
All material published in Emerging Infectious Diseases is in the public domain and may be used and reprinted without special permission; proper citation, however, is required.