P1-41 MALDI-TOF MS Biotyping for Characterization of Antimicrobial-resistant Escherichia coli from Concentrated Animal Feeding Operations and Associated Wildlife

Sunday, July 26, 2015
Exhibit Hall (Oregon Convention Center)
Jennifer Anders , University of Wyoming , Laramie , WY
Baolin Wang , University of Wyoming , Laramie , WY
Jeffrey Chandler , University of Wyoming , Laramie , WY
Jessica Prenni , Colorado State University , Fort Collins , CO
Alan Franklin , U.S. Department of Agriculture-NWRC-WS , Ft. Collins , CO
James Carlson , U.S. Department of Agriculture-NWRC-WS , Ft. Collins , CO
Jeffrey LeJeune , The Ohio State University , Wooster , OH
Bledar Bisha , University of Wyoming , Laramie , WY
Introduction: Antimicrobial resistant (AMR) bacteria represent one of the most significant challenges to food safety and security. Wildlife serve as one potential vector in the spread of AMR bacteria, thus interventions which limit wildlife contact with livestock or produce can control the dissemination of AMR bacteria. Accordingly, to pinpoint the problem wildlife vectors, methods are needed to rapidly characterize the AMR phenotypes of associated bacterial isolates.

Purpose: Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) Biotyping and bioinformatics were used to identify bacteria and correlate mass spectra to AMR phenotypes. AMR Escherichia coli collected from wildlife and their associated concentrated animal feeding operations (CAFOs) were used as the test bed.

Methods: MALDI Biotyping of 237 presumptive E. coli isolates grown on selective media was conducted using the Bruker Ultraflex II TOF/TOF configured in positive ion reflector mode with an accelerating voltage of 20 kV, following an ethanol/formic acid extraction of the bacteria. Antimicrobial susceptibility testing with 17 relevant antibiotics was performed on isolates confirmed as E. coli using the disk diffusion method. Correlations of mass spectra with the determined AMR phenotypes were accomplished using the principal component analysis (PCA) clustering feature of the MALDI Biotyper RTC software (Ver. 3.1.) and with novel algorithms developed by our laboratories.

Results: MALDI-TOF MS confidently identified (Biotyping score ≥ 1.8) 156 of the 237 isolates as E.coli, 35 isolates as beta- or gamma- proteobacteria sp., and 46 isolates could not be identified. Detailed integrations of mass spectral PCA clustering patterns proved useful for differentiation of isolates with specific AMR phenotypes. Further, within this dataset, multiple ions were identified that uniquely correlated with a specific antimicrobial resistance.

Significance: This study demonstrates that MALDI-TOF MS is a viable strategy for the identification of E. coli associated with wildlife/CAFOs and for discriminating between their AMR phenotypes.