Genome analysis provides insights into the epidemiology of infection with Flavobacterium psychrophilum among farmed salmonid fish in Sweden

Abstract

The pathogen Flavobacterium psychrophilum is a major problem for the expanding salmonid fish farming industry in Sweden as well as worldwide. A better understanding of the phylogeography and infection routes of F. psychrophilum outbreaks could help to improve aquaculture profitability and the welfare of farmed fish while reducing the need for antibiotics. In the present study, high-throughput genome sequencing was applied to a collection of F. psychrophilum isolates (n=38) from outbreaks on fish farms in different regions of Sweden between 1988 and 2016. Antibiotic susceptibility tests were applied to a subset of the isolates and the results correlated to the presence of genetic resistance markers. We show that F. psychrophilum clones are not regionally biased and that new clones with a higher degree of antibiotic resistance have emerged nationwide during the study period. This supports previous theories of the importance of live fish and egg trade as a route of infection. Continuous monitoring of recovered isolates by high-throughput sequencing techniques in the future could facilitate tracing of clones within and between countries, as well as the detection of emergent virulent or antibiotic-resistant clones. This article contains data hosted by Microreact.

Keywords: Flavobacterium psychrophilum; antimicrobial resistance; aquaculture; genomics; molecular epidemiology; rainbow trout.

Fig. 1.

Population structure and spatio-temporal distribution of lineages of F. psychrophilum among Swedish farmed salmonid fish. Upper: NeighborNet representation of the genetic relationship between all included isolates and the reference genome JIP02/86 based on whole-genome SNP typing, annotated with SNP cluster designation and with the four major clusters highlighted. Lower: number of observations of the four major SNP clusters per year between 1988 and 2016. Each bar is annotated with county codes (in italics) showing the locations of observations of a given cluster in a given year. Right: map of Sweden showing county borders and county codes. Counties of origin of one or more isolates in the study are highlighted in darker grey.