The enteric pathogens, Shiga toxin-producing E. coli (STEC) and non-typhoidal Salmonella (NTS), are leading causes of foodborne infections in the US, resulting in 265,000 and 1.2 million illnesses every year, respectively. The emergence of antibiotic resistance in these pathogens has been documented and is of great concern due to negative patient health outcomes and the possibility of transfer of resistance genes to other clinically relevant pathogens. However, there is a scarcity in information about frequencies of antibiotic resistant and factors associated with resistant STEC and NTS infections in Michigan. It is necessary to have a complete understanding about the of emerging antibiotic resistance and factors driving the rise of resistance in STEC and NTS to help develop effective control strategies. In this dissertation, 980 STEC isolates collected from patients in Michigan between 2001 and 2014 were examined for resistance to clinically relevant antibiotics. The examination of STEC strains for resistance, revealed high frequencies of resistance to ampicillin and trimethoprim-sulfamethoxazole, with significant increases in antibiotic resistance rates observed over this 14-year period. Multivariate logistic regression analysis identified non-O157 serotypes to be independently associated with antibiotic resistance. The recent increase in incidence of non-O157 serotypes observed in the US, coupled with the high frequencies of antibiotic resistance observed in this study, suggest the emergence of antibiotic resistant non-O157s as important human pathogens. Additionally, antibiotic resistant STEC isolates from patients in recent years (2010-2014) were more likely to cause hospitalizations than pansusceptible STEC isolates, suggesting that resistant STEC infections may result in adverse patient outcomes. Using whole genome sequencing, we also identified chromosomal mutations and 33 horizontally acquired genes present in the genomes of non-O157 STEC, likely conferring resistance. Importantly, by creating a co-occurrence network of these genes, we identified the co-occurrence of certain resistance genes, which are possibly present on the same mobile genetic element, thus resulting in multi-drug resistance. In addition to examining resistance in STEC, a total of 198 clinical NTS isolates collected between 2011 and 2014 were also examined for antibiotic resistance in this dissertation. Resistance to tetracycline, trimethoprim-sulfamethoxazole and ampicillin were commonly observed. Concerningly, high frequencies of multidrug resistant NTS were also observed with significant increases in their prevalence observed between 2011 and 2014. These high multidrug resistant rates have important implications on patient care as the efficacy of multiple antibiotics is reduced. Antibiotic resistant NTS isolates were also found to result in significantly longer mean hospital stays compared to pansusceptible NTS. Serovar specific differences in frequencies of antibiotic resistance were observed; S. Enteritidis were observed to have lower resistance frequencies than other serovars. Lastly, to better understand the role that cattle reservoirs play in harbouring antibiotic resistant STEC strains, we examined 121 STEC isolates collected in 2012 from six cattle farms in Michigan for antibiotic resistance. While high resistance frequencies to tetracycline and trimethoprim-sulfamethoxazole were observed in certain herds, no resistance to ampicillin was observed, unlike what was observed in STEC isolates collected from patients. While different populations of resistant STEC may be circulating in the clinical and agricultural environments, continuous monitoring of resistance in the cattle reservoir is warranted to determine if animal reservoirs can serve as potential sources of resistant infections in humans.
Read
