Whole genome sequencing provides possible explanations for the difference in phage susceptibility among two Salmonella Typhimurium phage types (DT8 and DT30) associated with a single foodborne…

Manal Mohammed, Martin Cormican

Phage typing has been used for decades as a rapid, low cost approach for the epidemiological surveillance of Salmonella enterica subsp. enterica serovar Typhimurium. Although molecular methods are replacing phage typing the system is still in use and provides a valuable model for study of phage-host interaction. Phage typing depends on the pattern of bacterial resistance or sensitivity to a panel of specific bacteriophages. In the phage typing scheme, S. Typhimurium definitive phage types (DT) 8 and 30 differ greatly in their susceptibility to the 30 typing phages of S. Typhimurium; DT8 is susceptible to 11 phages whereas DT30 is resistant to all typing phages except one phage although both DT8 and DT30 were reported to be associated with a single foodborne salmonellosis outbreak in Ireland between 2009 and 2011. We wished to study the genomic correlates of the DT8 and DT30 difference in phage susceptibility using the whole genome sequence (WGS) of S. Typhimurium DT8 and DT30 representatives. Comparative genome analysis revealed that both S. Typhimurium DT8 and DT30 are lysogenic for three prophages including two S. Typhimurium associated prophages (Gifsy-2 and ST64B) and one S. Enteritidis associated prophage (Enteritidis lysogenic phage S) which has not been detected previously in S. Typhimurium. Furthermore, DT8 and DT30 contain identical clustered regularly interspaced short palindromic repeats (CRISPRs). Interestingly, S. Typhimurium DT8 harbours an accessory genome represented by a virulence plasmid that is highly related to the pSLT plasmid of S. Typhimurium strain LT2 (phage typed as DT4) and codes a unique methyltransferase (MTase); M.EcoGIX related MTase. This plasmid is not detected in DT30. On the other hand, DT30 carries a unique genomic island similar to the integrative and conjugative element (ICE) of Enterotoxigenic Escherichia coli (ETEC) and encodes type IV secretion pathway system (T4SS) and several hypothetical proteins. This genomic island is not detected in DT8. We suggest that differences in phage susceptibility between DT8 and DT30 may be related to acquisition of ICE in DT30 and loss of pSLT like plasmid that might be associated with DT30 resistance to almost all phages used in the typing scheme. Additional studies are required to determine the significance of the differences among DT8 and DT30 in relation to the difference in phage susceptibility. This study represents an initial step toward understanding the molecular basis of this host-phage relationship.