Using whole genome sequencing to study American foulbrood epidemiology in honeybees

Abstract

American foulbrood (AFB), caused by Paenibacillus larvae, is a devastating disease in honeybees. In most countries, the disease is controlled through compulsory burning of symptomatic colonies causing major economic losses in apiculture. The pathogen is endemic to honeybees world-wide and is readily transmitted via the movement of hive equipment or bees. Molecular epidemiology of AFB currently largely relies on placing isolates in one of four ERIC-genotypes. However, a more powerful alternative is multi-locus sequence typing (MLST) using whole-genome sequencing (WGS), which allows for high-resolution studies of disease outbreaks. To evaluate WGS as a tool for AFB-epidemiology, we applied core genome MLST (cgMLST) on isolates from a recent outbreak of AFB in Sweden. The high resolution of the cgMLST allowed different bacterial clones involved in the disease outbreak to be identified and to trace the source of infection. The source was found to be a beekeeper who had sold bees to two other beekeepers, proving the epidemiological link between them. No such conclusion could have been made using conventional MLST or ERIC-typing. This is the first time that WGS has been used to study the epidemiology of AFB. The results show that the technique is very powerful for high-resolution tracing of AFB-outbreaks.

Fig 1. Core genome MLST results of Paenibacillus larvae isolates from the outbreak on Gotland in 2014.

Two minimum spanning trees generated in SeqSphere+ based on two cgMLST-analysis with ERIC II-isolates (A) and ERIC I-isolates (B), respectively. Results in A are based on 3,309 target genes and B is based on 3,155 target genes. The numbers represent the number of allele differences between isolates. The branch lengths are not proportional to the number of differences. The colors indicate the beekeeper origin of the isolates.

Fig 2. Core genome MLST results of ERIC I-type isolates from AFB outbreaks on Gotland (G) and from Uppland (U).

Minimum spanning tree generated in SeqSphere+ based on comparing the 3,052 genes present in the assemblies of the isolates. The numbers represent the numbers of allele differences between groups of isolates. The lengths of the lines are not proportional to the number of differences. The colors indicate the origin of the isolates (the beekeeper/owner of the apiaries). The circles show which MLST sequence type the isolates belonged to. The minimum spanning tree should not be over-interpreted as showing transmission routes, only the level of relatedness.