Development of methods to test drug sensitivity of fish pathogenic Flavobacterium columnare and drug sensitivity thresholds for F. columnare to the antimicrobial florfenicol.

Abstract

Antimicrobial drugs play a key role in managing the health of fish in aquaculture. However, public health concerns about antimicrobial resistance--the ability of microorganisms to resist standard antimicrobial treatments--include the potential for antimicrobial use in aquaculture to select for resistant bacteria in and around fish farms. Recent approval of two antimicrobial drugs to treat farmed freshwater fish with infections caused by the aquatic bacteria Flavobacterium columnare and F. psychrophilum created an important need for research to monitor these bacteria for changes in antimicrobial susceptibility and to decide when the antimicrobials should be used. Therefore three studies were conducted.

The initial study optimized methods for broth microdilution testing F. columnare and F. psychrophilum and conducted a multi-laboratory standardization trial that set quality control parameters for nine antimicrobials commonly used in aquaculture, thus creating the first standardized testing method. In the second study, we constructed frequency distributions using minimal inhibitory concentrations determined from testing 134 F. columnare with the standardized method. Analysis of the distributions determined the drug concentration--called an epidemiological cutoff value (ECV)--which separated the wild type isolates with no acquired or selected resistance from the non-wild type isolates. The ECV indicated 22 of 134 isolates had decreased their susceptibility to at least 1 antimicrobial. In addition, we developed a laboratory disease model with juvenile channel catfish, Ictalurus punctatus. We compared the virulence of three F. columnare isolates with wild type or three isolates with non-wild type florfenicol susceptibility using the model. We found that five isolates had similar high level virulence and were not affected by differences in florfenicol susceptibility.

Finally, we studied if decreased non-wild type florfenicol susceptibility affected the ability of the approved florfenicol treatment to control F. columnare infections. The approved treatment significantly reduced catfish mortality following exposure to an isolate with typical wild type florfenicol susceptibility but mortality was not reduced following exposure to an isolate with non-wild type susceptibility. Taken together, these studies provide methods and research needed to monitor F. columnare for changes in antimicrobial susceptibility and to rationally use florfenicol to control F. columnare infections.