Summary
Outbreaks of anthrax occur sporadically in Australia most commonly in the anthrax belt, a region which extends from Southern Queensland through the centre of New South Wales and into Northern Victoria. Little is known about the epidemiological links between Bacillus anthracis strains isolated from different outbreaks and the diversity of strains within Australia. This project aims to provide knowledge and tools to help optimise response strategies and assist in the prevention of anthrax within Australia. To enable these aims the project has focussed on improving our understanding of the epidemiology and ecology of Anthrax in Australia and development of enhanced diagnostics.
Data on outbreaks of disease (see Appendix 1) has been collated and temporally and spatially diverse strains of B. anthracis associated with recent outbreaks of disease collected and accessioned into the National Anthrax Reference Collection for use in molecular epidemiological studies. Genotyping of isolates for molecular epidemiological studies was undertaken using Multiple-Locus Variable-number tandem repeat Analysis (MLVA) employing 25 markers and Whole Genome Single Nucleotide Polymorphism (SNP) Typing (WGST). MLVA genotyping of 150 isolates of B. anthracis revealed 8 unique genotypes (MLVA groups, MG 1-8). The majority of isolates fell within two closely related genotypes (MG 1 and MG 2). Two genotypes (MG 7 and MG 8) were distinct from the major cluster of genotypes which was also reflected in their geographic origin outside of the anthrax belt. Genotyping of B. anthracis strains from outbreaks of disease in Victoria identified the presence of multiple genotypes associated with each outbreak. However, overall the genetic diversity of B. anthracis isolates from Australia is relatively low. Whole genome sequences were determined for 52 B. anthracis isolates which are representative of the isolates held within the National Anthrax Reference Collection. SNPs were identified between the whole genome sequences which were then used for WGST of these isolates. WGST was determined to be a more discriminatory method than Multi Locus VNTR analysis (MLVA) for genotyping B. anthracis isolates.
The WGST method discriminated 4 genotypes, WGST genotypes B, C, D and E, within a MLVA genotype MG2 and two genotypes, WGST genotypes H and I, within MLVA genotype MG6. However, there was one case where WGST was unable to discriminate between two MLVA genotypes, MG 1 and MG3. It may be possible to discriminate these two MLVA genotypes using less stringent criteria for the definition of SNPs than that used in the present analysis. Overall the WGST analysis confirmed the MLVA typing of B. anthracis isolates but revealed previously undetected diversity within B. anthracis isolates from NSW. The diversity of B. anthracis in NSW is consistent with the historical record of B. anthracis within Australia with the greatest diversity of isolates identified where the organism has been resident for the longest period of time.
The geographical distribution of genotypes within Australia suggests that a single genotype was introduced into the Eastern states of Australia and that this introduction was followed by spread of the pathogen throughout the anthrax belt. Following the spread of the pathogen, localized differentiation into other closely related genotypes has occurred. In contrast, unexplained occurrences of disease in areas outside of the anthrax belt, which are associated with different genotypes, indicate separate introductions of B. anthracis into Australia. The rapid lateral flow Immunochromatographic diagnostic test (ICT) kits are a highly valuable tool for active surveillance and disease diagnosis in areas with a history of anthrax and a routine tool in the investigation of sudden death in cattle. The kits are produced by the National Anthrax Reference Laboratory (NARL) have been distributed throughout Australia (Victoria, NSW, Western Australia, South Australia, Tasmania and Queensland). The number of kits distributed is increasing each year. The ability to produce these kits in-house at the NARL has provided Australia with the means to increase anthrax surveillance from the network of veterinarians and animal health officers. The simplicity of the kit and the ability to perform carcass-side has empowered veterinarians to implement greater caution when faced with sudden deaths on farm and provided a tool that can be used to justify to farmers the cascade of events that follow a diagnosis of anthrax. The high sensitivity, high specificity and simplicity of the ICT have made this test an invaluable field tool for veterinarians and animal health officers throughout Australia. In order to better predict the potential ecological and geographic distribution of B. anthracis in Australia an Ecological niche model (ENM) was developed .
The ENM approach predicts the potential ecological and geographic distribution of the pathogen based on outbreak locations and environmental conditions including rainfall, temperature and soil pH. The ENM developed in this study predicted the region where B. anthracis is presently found in the Anthrax Belt stretching from southern Queensland through the centre of NSW and into Northern Victoria. However, early records of disease in Australia, particularly along the southern Victorian coastline were not predicted by the model. It is hypothesised that these early disease incidents were caused by consistent introduction and re-introduction of B. anthracis into these areas, most likely through movement of infected animals or importation of contaminated material (for example bone meal) through ports in these areas without long term establishment of disease. The ENM analysis and WGST data is also consistent with the hypothesis that the unexplained occurrence of anthrax in Walpole in Western Australia and Rockhampton in Queensland were the result of recent introduction of the disease into these regions and not the exposure of animals to spores lying dormant in the soil. Overall, the consensus of the ENM with occurrences of anthrax in Australia gives confidence that a robust model has been developed to describe the distribution of the disease in Australia.
