Comparative micro-epidemiology of pathogenic avian influenza virus outbreaks in a wild bird population

Abstract

Understanding the epidemiological dynamics of highly pathogenic avian influenza virus (HPAIV) in wild birds is crucial for guiding effective surveillance and control measures. The spread of H5 HPAIV has been well characterized over large geographical and temporal scales. However, information about the detailed dynamics and demographics of individual outbreaks in wild birds is rare and important epidemiological parameters remain unknown. We present data from a wild population of long-lived birds (mute swans; Cygnus olor) that has experienced three outbreaks of related H5 HPAIVs in the past decade, specifically, H5N1 (2007), H5N8 (2016) and H5N6 (2017). Detailed demographic data were available and intense sampling was conducted before and after the outbreaks; hence the population is unusually suitable for exploring the natural epidemiology, evolution and ecology of HPAIV in wild birds. We show that key epidemiological features remain remarkably consistent across multiple outbreaks, including the timing of virus incursion and outbreak duration, and the presence of a strong age-structure in morbidity that likely arises from an equivalent age-structure in immunological responses. The predictability of these features across a series of outbreaks in a complex natural population is striking and contributes to our understanding of HPAIV in wild birds. This article is part of the theme issue 'Modelling infectious disease outbreaks in humans, animals and plants: approaches and important themes'. This issue is linked with the subsequent theme issue 'Modelling infectious disease outbreaks in humans, animals and plants: epidemic forecasting and control'.

Keywords: H5NX; avian influenza virus; epidemiology; genetics; serology; wild birds.

Figure 1.

Mortality among swans on the Fleet Lagoon. (a) Mortality among swans greater than approximately four months old on the Fleet Lagoon during H5N1 (2007/08), H5N8 (2016/17) and H5N6 (2017/18). Brightly coloured bars indicate the number of dead swans observed during HPAIV ‘outbreak periods’ on the Fleet Lagoon (period between the first and last confirmed positive cases in swans at the site for each HPAIV subtype). Brightly coloured bars include all mortality observed during the outbreak period, regardless of whether the carcasses were tested for AIV or AIV positivity. Pale-coloured bars indicate the mortality observed among birds on the Fleet Lagoon in periods when HPAIV was not detected, and therefore indicate the typical level of mortality observed among swans on the Fleet Lagoon. While the last HPAIV H5N1-positive swan was found in the vicinity of the Fleet Lagoon during week 4 of 2008, a Canada goose (Branta canadensis) was found positive for HPAIV less than 1 km from the Fleet Lagoon during week 7 of 2008 and is marked with an asterisk. Dots and horizontal lines indicate the median and 95% HPD interval dates for the MRCA of the Fleet Lagoon outbreak clade, as estimated using phylodynamic methods. (b) Estimated proportion of birds of each age group that died of HPAIV infection during each outbreak, based on birds believed to be alive in the population at the time each outbreak started. Colours indicate the respective outbreaks. Adjusted Wald 95% confidence intervals are given.

Figure 2.

Bayesian phylogenetic trees of HA sequences. (a) Bayesian time-scaled phylogenetic tree of the HA segment of 421 Eurasian H5 HPAIVs. Colours at tips indicate the location of sampling (Asia: red, Europe (including Russia): dark blue, Fleet Lagoon: bright blue). (b) Expansion of the well-supported clade that contains the H5N8 viruses sampled on the Fleet Lagoon (bright blue). The location of this clade within the larger phylogeny is indicated by the linked vertical grey line in (a). Nodes with posterior support values greater than 0.5 or greater than 0.75 are marked with white and black circles, respectively. (c) As for (b), but showing the phylogenetic position of the H5N6 viruses from the Fleet Lagoon.

Figure 3.

Seropositivity of swans. (a) Percentage of swans with antibodies targeting AIV NP by year of hatching and date of sampling. Colours represent birds hatched in different years. Adjusted Wald 95% confidence intervals are given. The date of the HPAI H5N6 outbreak is marked. (b) Percentage of swans with titres greater than 8 for HPAIV H5N8 (A/Turkey/England/16) (blue) and LPAIV H5N3 (A/Teal/England/06) (orange). Birds that were sampled on more than one occasion have been removed. If a bird was found to be seronegative for previous AIV exposure by NP-ELISA, it is included here as having a titre less than or equal to 8, despite HI assays not being conducted for that sample. Samples were collected during June, July and November 2017, but birds are grouped by hatch year only as no seasonal trend in change in titre was observed in these data.