A field experiment was conducted using weaned piglets, which were randomized to pens by weight and allowed to acclimatize to their surroundings and diet. All piglets received normal dietary rations as per NRC recommendations. The pens were assigned randomly to six treatment groups (incomplete factorial design with low CTC, high CTC, high Zn being the factors; low and high CTC could not be interacted per FDA regulations). Swab samples (nasal, skin, and tonsils) were collected on Days 0, 21 and 42. Methicillin-resistant Staphylococcus aureus (MRSA) isolates were recovered after streaking swabs onto a selective medium, CHROMTM MRSA agar(Hardy Diagnostics) at 37ºC for 24 h. Bacterial isolation was performed by picking three distinct colonies from each plate and re-streaking onto blood agar plates. We isolated 720 bacterial colonies for each sampling site. All isolates were subjected to species confirmation by PCR followed by detection of mecA (methicillin resistance) and czrC (zinc resistance) genes. All MRSA isolates were also subjected for spa typing, sequencing the spa gene repeat region, to further study the molecular epidemiology (this added step resulted in a request for a one-year no-cost extension to the original grant).
Overall, the prevalence of mecA-positive MRSA was 42.8% (308/720), 37.2% (268/720), and 42.6% (307/720) among nasal, skin, and tonsillar samples, respectively. The prevalence of mecA- and czrC-positive MRSA was affected by Zn (P < 0.05) but not CTC supplementation (P > 0.05). The prevalence of czrC-positive MRSA was 20% (144/720), 21.1% (152/720), and 14.3% (103/720) in nasal, skin, and tonsillar swabs, respectively. The occurrence of the czrC gene was strongly associated with mecA-positive MRSA isolates (P < 0.0001). The median minimum inhibitory concentrations (MICs) of Zn for zinc resistant and susceptible isolates were 8 and 4 mM, respectively (P < 0.001). We conducted spa gene repeat sequence analysis on all MRSA isolates originating from the three anatomic pig collection sites. The preliminary results from several isolates are presented here as the sequence analysis is still in progress (July 2015). Among the nasal isolates (n=121), the majority of them belong to t034 (50; 41.3%), unknown spa type (42; 34.7%), and t007 (29; 23.9%) spa types. In case of tonsillar MRSA isolates (n=109), the major spa types were t034 (41; 37.6%), t007 (37; 33.9%), and 28.4% (31) of unknown spa types. In the case of skin isolates (n=105), the major spa types were t034 (89; 84.7%), t127 (8; 7.6%), t007 (7; 6.6%), with only a single isolate not belonging to any known spa types.
There was a clear epidemic spread in the first 21 days of the feeding trial which provided most of the increase in the prevalence of MRSA in nasal, tonsillar, and skin regions of piglets and across all treatment groups. This suggests that either: 1) the MRSA strains were endemic in the barn environments before pigs were introduced, or 2) the MRSA strains were present in a very few piglets upon arrival and the stress of transport, mixing, and new environs triggered epidemic spread of the organism. Since our study was longitudinal in nature with repeated sampling it helped us to identify infectious disease dynamics and a clearly discernable epidemic spread of MRSA. In contrast, most previous studies of MRSA in swine in North America have been cross-sectional in design; or, with spacing of visits and animal sampling such that infectious disease dynamics could not be observed.
The data generated from our study will help to clarify the link between zinc and chlortetracycline use in swine production to other antibiotic resistance in a major pathogen of public health significance. It is imperative that the speculated associations between the use of these antibiotic and non-antibiotic antimicrobials and MRSA be examined in realistic field settings so as to more quantitatively evaluate the potential risks to public health. These data are essential for conducting meaningful quantitative risks assessments (QRA) and to draw effective intervention strategies with the help of more epidemiological surveillance. More importantly, our research findings will help swine producers to make informed decisions in optimizing existing management practices (such as mineral supplementation and the use of feed grade antimicrobials) in order to best manage public health risks in their settings.
For more information please contact:
H. Morgan Scott DVM, PhD
Department of Veterinary Pathobiology
Mail Stop 4467
Texas A&M University
College Station, TX, USA 77843-4467