Abstract
Paenibacillus larvae is the causative agent of American foulbrood (AFB) in honeybees (Apis mellifera) and a devastating pathogen for honey and pollination industries worldwide. Despite this threat, a genomic survey of P. larvae has not been attempted within Australia. To examine the diversity of Australian populations, we sequenced 368 P. larvae genomes sourced primarily from south-eastern Australia. Multilocus sequencing typing analysis identified only 4 sequence types across all 368 samples, with 2 sequence types (ST18 and ST5) representing 96% of all isolates. In comparison to European-sourced P. larvae, sequences revealed much less genetic diversity in Australian isolates. However, Australian genotypes were very similar to those found in New Zealand populations. All Australian isolates were identified as enterobacterial repetitive intergenic consensus (ERIC) type I. To determine the feasibility of a genomic tracing system in a low-diversity genetic background, we investigated core-genome SNP (cgSNP) genotyping of isolates from a single beekeeper and from isolates across multiple apiaries and sample sites. We identified highly related cgSNP clusters, one with known epidemiological links, but another highly related cluster spanned several decades. Results strongly suggest that cgSNP analysis does have the discriminatory power to assist in the trace-forward and trace-back of AFB outbreaks, but importantly, the inclusion of background sequences and careful consideration of multiple analysis methods are required to avoid erroneous conclusions.
Keywords:
American foulbrood; Apis mellifera; Paenibacillus larvae; SNPs; core genome; honeybee; microbial pathogens; multilocus sequence type; phylogenetics; whole-genome sequencing.