The emergence of H5N1 highly pathogenic avian influenza (HPAI) has raised worldwide concern because it has posed impacts affecting not only animal health, but also human health. Ducks have been suggested to contribute to H5N1 HPAI spread in Asia; however, there is a lack of evidence to support this hypothesis due to limited field and experimental data, particularly from countries where the disease remains endemic. The aim of this study is to determine whether ducks play a role in the maintenance, evolution and transmission of Indonesian H5N1 HPAI virus.
This study was initiated with a longitudinal field study on 96 smallholder duck farms in Central Java, Indonesia from March 2007 to March 2008. Thirty-four study farms experienced HPAI outbreaks where the majority of in-contact chickens, which intermingle with ducks, succumbed to the infection, while most of the ducks survived. Bird-level seroprevalence was 5.2 times higher in ducks (2.6%) than in chickens (0.5%), and duck flocks were 12.4 times more likely than chickens flocks to have antibodies against H5, suggesting that H5 subtype virus circulated more successfully among ducks than chickens. Incorporating further molecular and antigenic analyses on the virus isolates collected in this study was important to improve the understanding on the diversity of circulating viruses and the pattern of H5N1 spread in these farms.
Phylogenetic analysis of the haemagglutinin (HA) gene showed that 84 characterised H5N1 viruses, isolated from those farms, belonged to clade 2.1 with three virus sublineages being identified: clade 2.1.1 (N=1), clade 2.1.3 (N=80), and IDN/6/05-like viruses (N=3). Although all virus isolates possessed HA sequence characteristic of highly pathogenic viruses, there were no significant mutations found in either HA or neuraminidase (NA) proteins that are associated with the binding specificity to human cell receptors and the resistance to NA inhibitors, respectively, and no evidence of major antigenic variation among the isolates. Identical HA virus variants could be found at relatively distant geographic locations, whereas genetically distinct HA variants could be isolated from either chickens or ducks on the same farm at the same time, suggesting that a range of variants can circulate simultaneously within a short period during HPAI outbreaks.
Two viruses belonging to clades 2.1.1 (IDN34368) and 2.1.3 (IDN32226) were inoculated into 4-week old broiler chickens and 7- to 9-week old Pekin ducks through mucosal routes to determine the pathobiology. Both viruses produced an acute and severe disease in chickens, beginning around 24 hours post inoculation. Broad cell tropism with abundant virus in varied tissues was seen in chickens, but the disease appeared to be predominantly vascular. The disease in ducks was mostly sub-clinical, with most viral antigen located within the mucosa of the respiratory tract, but limited in other tissues. The clade 2.1.1 virus caused slightly greater tissue tropism in ducks than the clade 2.1.3, and one duck infected with clade 2.1.1 died due to brain infection. Chickens shed high virus titres around the time of clinical disease, while ducks shed low virus titres intermittently, for 1-8 day post inoculation (dpi). This suggests that chickens are more likely than ducks to be the main source of H5N1 infection, but ducks are probably more effective maintenance hosts since they can shed virus, with less disease appearance, for a longer period post infection.
The clade 2.1.1 virus (IDN34368) was inoculated into two groups of 4-week-old Pekin ducks (N total=15) to determine duration and level of viral shedding. A group of contact ducks (N=5) was mixed with one of the inoculated duck groups (N=5) at 1 dpi and another group of contact ducks (N=9) was mixed with the other inoculated group (N=10) at 10 dpi to investigate virus transmission during acute and post acute infection stages, respectively. All the acute stage-contact ducks became antibody- and virus/RNA-positive after the contact, whereas none of post acute stage-contact ducks seroconverted. A higher proportion of sera was classified antibody-positive by the haemagglutination inhibition (HI) test using horse rather than chicken erythrocytes, especially during the early response against H5N1 virus. Although H5 RNA was detected in oral or cloacal swabs up to 34 dpi, infectious virus was no longer detected once antibodies detected at 8 dpi, indicating that domestic ducks cannot maintain long-term shedding of Indonesian H5N1 HPAI virus.
In summary, this thesis produced the most comprehensive body of knowledge on H5N1 HPAI infection in domestic ducks to date. The integrated results of epidemiology, molecular biology, pathobiology and transmission studies indicated that ducks are not an exclusive species or factor responsible for the H5N1 HPAI endemic in Indonesia. Although ducks are well suited for harbouring and circulating the virus, other factors, particularly those related to management in smallholder farms, including mixed rearing and scavenging practices seem to contribute to the continuing disease outbreaks and to increase the risk for virus transmission to other species, including humans.