Co-infection of an animal with more than one genotype can occur in anthrax

Abstract

During the routine fingerprinting of outbreak strains of Bacillus anthracis of European and African origin by means of a 31-marker multi-locus variable number of tandem repeats analysis (MLVA), four cultures, two from the Etosha National Park (ENP), Namibia, and two from an outbreak in the Pyrenees in 1997, were found to harbour different genotypes (GTs). To investigate this further, isolates from 10 samples of blood-soaked soil from beneath anthrax carcasses and 18 clinical swabs taken from carcasses in the ENP were examined by a 31-marker MLVA. While only a single GT was found in any one of the 10 soil samples, four of the 18 swabs (22%) yielded different GTs. Two GTs were isolated from each of a zebra and a springbok and three GTs from each of a second zebra and an elephant. These animals had died in a region of the ENP where deaths caused by anthrax regularly occur every year. The results confirm the indications noted previously that co-infection with more than one GT is probably not especially uncommon. The results show that, for the purpose of analysing genotypes involved in an outbreak, it is important to examine more than a single colony from a clinical sample. Multi-locus variable number of tandem repeats analysis (MLVA)-based fingerprinting techniques have been used in many studies worldwide to characterize the occurrence of different genotypes of Bacillus anthracis in outbreaks of wildlife or livestock and to draw conclusions about the source, the possible routes of spread and the temporal and spatial distribution of outbreak strains. Simultaneous isolation of different genotypes from the same host revealed in our study by MLVA highlights the importance of examining more than a single colony from a clinical sample. This conclusion is not specific for MLVA but holds true for every high-resolution method, including full-genome sequencing.