Overcoming microbial inhibition of S. sonnei through the exploitation of genomically-predicted antibiotic resistance profiles for the development of food enrichment media

Abstract

Linking outbreaks of Shigella spp. to specific foods is challenging due to poor selectivity of current enrichment media. We have previously shown that enrichment media, tailored to the genomically-predicted antimicrobial resistance (AMR) of Shiga toxigenic E. coli strains, enhances their isolation from foods. This study investigates the application of this approach for Shigella isolation. The AMR gene profiles of 21,908 published S. sonnei genomes indicated a high prevalence of genes conferring resistance to streptomycin (aadA, aph(3”)-Ib, aph(6)-Id , 92.8%), sulfonamides (sul1, sul2, 74.8%) and/or trimethoprim (dfrA, 96.2%). Genomic analysis and antibiotic susceptibility testing conducted with a panel of 17 outbreak-associated S. sonnei strains confirmed the correlation of AMR gene detection with resistance phenotypes. Supplementation of Shigella Broth (SB) with up to 400 µg/mL of trimethoprim or sulfadiazine did not supress growth of sensitive strains, whereas 100 µg/mL of streptomycin increased the selectivity of this broth. All three antibiotics increased the selectivity of modified Tryptone Soya Broth (mTSB). Based on these results, supplemented media formulations were developed and assessed by measuring the relative growth of S. sonnei in cultures co-inoculated with a strain of bacteriocin-producing E. coli that is inhibitory to Shigella growth. S. sonnei was not recovered from co-cultures grown in SB or mTSB without antibiotics. In contrast, media supplemented with streptomycin at 50 and 100 µg/mL, trimethoprim at 25 and 50 µg/mL, and sulfadiazine at 100 µg/mL increased the relative proportion of S. sonnei in post-enrichment cultures. The enhanced recovery of resistant S. sonnei strains achieved in this study indicates that, in cases where genomic data is available for clinical S. sonnei isolates, customization of selective enrichment media based on AMR gene detection could be a valuable tool for supporting the investigation of foodborne shigellosis outbreaks.