The emergence of multi-drug resistant bacteria is a global problem that has arisen from widespread use of antimicrobials in both human and food animals. The subtherapeutic levels of antimicrobials used in poultry and livestock can cause the large populations of bacteria that live in the guts of these animals to be selected for antimicrobial resistance. The bacterial genome of these microorganisms provides a reservoir of potential resistance genes. Localization of multiple resistance genes on mobile genetic elements it thought to facilitate horizontal transfer of genetic material between bacteria providing a rapid means of dissemination at the molecular level. Horizontal transfer of antimicrobial resistance is one of the most significant issues with regard to our ability to curtail the emergence of multiple drug resistant pathogens.
Salmonella spp. are estimated to cause over 1.4 million cases of food-borne salmonellosis a year in the United States. Roughly half of these occur in children. Most cases of food-borne salmonellosis in humans are self-limiting; however, extended spectrum cephalosporins are often used for invasive infections, particularly for pediatric cases. Ceftiofur is an extended spectrum cephalosporin used to treat animals. In the last two decades multi-drug resistant Salmonella strains have emerged among humans and animals. More recently multi-drug resistant Salmonella enterica serotype Newport has emerged and has been associated with increased morbidity and mortality in both man and animals. The incidence of S. serotype Newport in animals increased from 106 reported cases in 1997 to 1,522 cases in 2003. Although the number of S. serotype Newport reported cases dropped in 2004 and 2005, it is the second most common Salmonella serotype isolated from animals (cattle, swine, poultry and horses) in the United States for the last two years. Because S. serotype Newport has emerged so rapidly and surpassed other Salmonella strains in incidence, it is of interest to know if this is due purely from selection by antimicrobials, or if S. serotype Newport is more fit than other Salmonella strains. This study intends to characterize and compare the metabolic fitness of Salmonella strains isolated from cattle to determine why S. serotype Newport has become a “superbug” so rapidly.
The objectives of this study were as follows:
Determination of Antimicrobial Susceptibility
The antimicrobial minimum inhibitory concentrations (MICs) were determined by broth microdilution according to methods described by the Clinical Laboratory Standards Institute (CLSI). Susceptibility testing was performed with the commercially available automated antimicrobial susceptibility system according to the manufacturer’s instructions. NARMS panels for Gram-negatives were used and the following antimicrobials were assayed: amikacin, amoxicillin/clavulanic acid, ampicillin, cefoxitin, ceftiofur, ceftriaxone, cefalotin, chloramphenicol, ciprofloxacin, gentamicin, kanamycin, nalidixic acid, streptomycin, sulfamethoxazole, tetracycline and trimethoprim/sulfamethoxazole. The following ATCC strains were used as controls for antimicrobial susceptibility testing: E. coli 35218, E. coli 25922, Staphylococcus aureus 29213, and Pseudomonas aeruginosa 27853.
Conjugation
All Salmonella Newport isolates were grown overnight in Brain Heart Infusion broth (BHI) medium plus 32 μg/ml tetracycline at 37oC without shaking. A plasmid-free nalidixic acid resistant E. coli JM109 was used as the recipient. All conjugation combinations were incubated overnight at 37oC. Tetracycline was used to select for horizontal transfer of the plasmid and nalidixic acid was used to counter select against donor cells.
Molecular Analysis
All Salmonella isolates were examined by pulsed-field gel electrophoresis (PFGE). Chromosomal DNA was digested with XbaI according to the standard CDC protocol. Electrophoresis was done and gels were stained with ethidium bromide and band patterns were analyzed.
Preparation of genomic DNA was done and plasmid DNA was extracted from TSB overnight cultures according to each manufacturer’s instructions.
Eighty-six of the 99 Salmonella enterica serotype Newport isolates obtained were resistant to multiple antibiotics. Thirteen were susceptible to all antibiotics tested (pansusceptible). Eleven (100% of the beef cattle isolates) of the pansusceptible isolates originated from beef cattle carcasses. Genetic analysis of the beef strains showed that there were only two clones. In order to establish that the multi-drug resistant Salmonella Newport strains are primarily coming from dairy cattle as opposed to beef cattle, a greater representation of strains from beef cattle will need to be characterized. Observation of this current trend probably means this is an early snapshot of the Newport serotype evolution and dissemination.
The plasmid DNA isolation procedure was done for the first 50 Salmonella Newport isolates. Plasmid DNA was isolated from 39 of 42 (93%) of multi-drug resistant dairy isolates. Plasmid DNA was also isolated from 6 of 10 of the pansusceptible Newport isolates (SN41-SN46); however, incompatibility group plasmids were not detected in these isolates by PCR. One hundred percent of the multi-drug resistant dairy isolates were positive by PCR for the incompatibility plasmid IncA/C. In addition to IncA/C, 4 dairy isolates contained a second plasmid; SN11 and SN14 contained the IncN incompatibility plasmid, and isolates, SN39 and SN40 contained the IncI incompatibility plasmid. This is the first known report of the broad-host range IncA/C incompatibility type plasmids in any Gram-negative field isolates. Broad host range plasmids have the capacity to replicate in many different Gram-negative bacterial species.
This study represents the first known identification of Salmonella field isolates carrying the IncA/C incompatibility type plasmid. Very little has been published in the scientific literature on IncA/C. An expanded characterization of Salmonella serogroups may aid in determining the origin and dissemination of the IncA/C plasmid acquired by the Newport isolates characterized in this study.
PFGE was done on all 99 isolates to show the relatedness of the bacterial genome. The isolates were clustered in three PFGE types. All of the antimicrobial susceptible isolates, 11 beef and 2 dairy, clustered in one PFGE type. PFGE analysis in combination with plasmid analysis is done to determine if the dissemination of resistance is due to the spread of a single Salmonella Newport strain (clone), or to the horizontal transfer of the bacterial plasmid among many unrelated strains. It is interesting to note that two pansusceptible dairy isolates clustered with the pansusceptible beef isolates rather than other dairy isolates. This may indicate that dissemination of multi-drug resistant Salmonella Newport has largely been due to clonal dissemination.
Overall, the possession of multi-drug resistance did not incur a significant fitness cost as compared to the pansusceptible Salmonella Newport isolates and the resistance genes were also transferable to E. coli. Other researchers have had difficulty transferring resistance from Salmonella Newport isolates; however, in this study an overnight broth technique was used that could detect very low levels of transfer. This indicates the plasmid was mobile; therefore, it is likely that plasmid transfer could occur between different species of bacteria in the bovine gastrointestinal tract.
The results of this study suggest that the dissemination of resistance is due to the spread of a single Salmonella Newport strain or clone. However, horizontal transfer of antimicrobial resistance is one of the most significant emerging issues with regard to the beef industry’s ability to curtail the emergence of multiple drug resistant pathogens.