Oral Presentation Australian Society for Microbiology Annual Scientific Meeting 2021

Genomic epidemiology links Burkholderia pseudomallei from environmental sources to human melioidosis cases located in northern Australia over an 18-year period. (#121)

Jessica R Webb 1 , Mark Mayo 1 , Audrey Rachlin 1 , Celeste Woerle 1 , Ella Meumann 1 2 , Vanessa Rigas 1 , Glenda Harrington 1 , Mirjam Kaestli 1 3 , Bart J Currie 1 2
  1. Global and Tropical Health Division , Menzies School of Health Research, Charles Darwin University, Darwin, Northern Territory, Australia
  2. Department of Infectious Diseases, Royal Darwin Hospital, Darwin, Northern Territory, Australia
  3. Research Institute for the Environment and Livelihoods, Charles Darwin University, Darwin, Northern Territory, Australia

The environmental pathogen Burkholderia pseudomallei (Bp) is the causative agent of melioidosis, with mortality still over 40% in some regions. Melioidosis is a tropical and subtropical disease, with the highest global incidence reported in Darwin, Northern Territory (NT). Bp is ecologically established in the NT environment and is highly prevalent in soil and drain water in the Darwin city region. Despite this it is unclear where the main environmental reservoirs of infection and disease can be found, with consequences for public health. The Darwin Prospective Melioidosis Study commenced in 1989 and documents all culture confirmed melioidosis cases, each with corresponding epidemiological, clinical data and bacterial genomes. From the year 2000 to 2018, we performed environmental sampling at melioidosis patient residential addresses or linked exposure sites to pinpoint infection location and mode. Subsequent genomic epidemiology methods were used to determine if patient clinical isolates matched linked environmental isolates. Environmental sampling was performed at 98 locations linked to 98 melioidosis patients within 6 months (n=84 sites; median sampling period=19 days) or 6 months-5 years (n=14 sites; median sampling period=585 days) after a case was reported. We collected 975 environmental samples (742 soil [median=3]; 233 water [median=8]) from the 98 sites. Strain genotyping matched the genotype of Bp clinical isolate to epi-linked Bp environmental isolate for 18 patients, in which clinical and environmental isolates for 17 of the 18 patients clustered on a local SNP phylogeny. Subsequent pairwise comparative genomics of environmental Bp isolates from each patient’s residential address matched the clinical Bp isolate for 16 patients (range 0 to 15 SNPs; median SNP difference=4 [large deletions in the accessory and core genome were not identified]) and pairwise genomics reveled additional SNP differences compared to the SNP phylogeny. Results on inoculating scenarios, location and sources of infection will be presented for these 16 clinical-environmental Bp matches. These findings are directing further studies to identify disease hotspots for public health surveillance.