Animal & Veterinary
Surveying Feed Commodities for Antibiotic Resistance
by David Wagner, Ph.D.
FDA Veterinarian Newsletter January/February 2002 Volume XVII, No 1
Widespread dissemination of resistance to antibiotics resulting from the selective effect of drug use in food animals may have important ramifications for both human and animal health. A critical question relevant to this ecological issue is: What is the source of the resistance determinants? Are they always present in the environment at low, virtually undetectable levels or are they introduced from outside? Are there management practices that could minimize or eliminate resistance development? Answers to these questions could lead to improvements in our ability to control development and dissemination of resistance to antibiotics and thereby optimize their efficacy, extend their useful lifespan, and minimize possible negative health consequences.
Animal feeds and feed commodities may serve as vectors for the dissemination and maintenance of resistance determinants in the animal production environment and thereby in the food supply. The commercial feed industry, during 2000, supplied an estimated 119 million tons of feed that were required to support intensive integrated animal production in the United States. This figure does not include the millions of tons of feed mixed on farms. With the exception of the cereal grains and grasses, virtually all other commodities that comprise animal feeds are by-products of other industries including the animal production industry. The ren
dering industry reported the production of 8.8 billion pounds of protein meals in 2000. A large portion of these products is incorporated into animal feeds. It is therefore not unreasonable to view the feed industry as a recycling business that utilizes by-products of high nutritional value from other industries, to provide complex nutrient sources for animal production.
Recently, the issue of antibiotic resistance resulting from the use of antibiotics in animal production has once again become a significant concern to the Center for Veterinary Medicine (CVM) and other national and international health agencies. The potential for resistance development in animal production settings to negatively impact human therapeutic efficacy is being revisited. The role that feed may play in the dynamics of antibiotic resistance development and dissemination in the animal production environment is essentially unknown. The possibility that feed may serve not only as a vector for resistance but also may function to maintain and concentrate resistance due to its recycling characteristic may be important. There are essentially no data available to assess the potential role of animal feed to serve as a vector for the transmission and/or maintenance of antibiotic resistant bacteria or resistance determinants in the animal environment.
During FY 2000, CVM's Division of Animal and Food Microbiology initiated a preliminary survey of animal feed commodities to assess their role in antibiotic resistance dissemination. Poultry by-product meal, meat and bone meal, blended animal proteins, and whole cereal grains (corn and oats) were the first products cultured. Enterococcus spp. were isolated, identified to species, and tested for susceptibilities to 17 antibiotics using a broth microdilution assay. The antibiotics are incorporated in a customized Gram positive panel that is used in the National Antimicrobial Resistance Monitoring System (NARMS) and produced by TREK Diagnostics, Inc. Eleven of the antibiotics are either used directly in the feed or water of animals or they are members of drug classes used in these ways.
A total of 175 samples were cultured. All were positive for enterococci except for corn (41%, 24/58) and poultry meal (95%, 19/20). Enterococcus faecium was by far the most prevalent Enterococcus species, accounting for 72% (167/232) of the isolates recovered. Enterococcus faecalis was rarely recovered from any of the samples tested. Nine of the twelve isolates recovered came from corn. This is interesting because E. faecalis is the most common cause of human infections accounting for more than 80% of reported cases. It is also the species most commonly isolated from animal production environments and from the intestinal contents of animals including humans. All of the samples of oats were positive for enterococci (36) but no E. faecalis were isolated. The results thus far show that enterococci are widely disseminated in the environment and can readily be isolated from feed commodities.
Resistance to antibiotics used in human therapy was primarily seen in isolates recovered from meat and bone meal and was observed for tetracycline, streptomycin, and erythromycin. Reduced susceptibilities to antibiotics used only in animal production were observed in significant proportions of the isolates from all the rendered animal by-products tested especially bacitracin susceptibility.
Resistance to ciprofloxacin was detected, but only in the isolates recovered from the cereal grains. None of the 167 isolates of Enterococcus faecium were determined to have minimum inhibitory concentrations (MIC) for Synercid above the NCCLS breakpoint of 4 ug/ml. This observation stands in stark contrast to E. faecium isolates recovered from litter taken from poultry production houses where about 65% of E. faecium are resistant to Synercid at concentrations up to 32ug/ml. This difference between feed and production environmental isolates probably demonstrates the selective effect elicited by virginiamycin use in feed. No resistance to Linezolid or penicillin, drugs of critical importance to human therapy was observed in any of the feed isolates tested.
What seems to be most remarkable about the data collected to date is the limited amount of antibiotic resistance associated with isolates from feed commodities relative to that seen in animal production environments. For example, Enterococcus isolates from poultry by-product meal are very susceptible to drugs such as penicillin and Synercid; whereas, isolates recovered from poultry production houses are very resistant with 75% and 65% of the isolates being resistant to the two drugs respectively. One question of interest resulting from these data is: Why are the isolates associated with poultry meal not resistant when those from live poultry are? The apparent decrease in susceptibility of isolates recovered from feed to the commonly used production drugs bacitracin, tylosin, and lincomycin may be an issue of concern to animal production.
This survey activity is continuing and will be expanded during 2002 with the involvement of the FDA's Office of Regulatory Affairs which will conduct a national sampling program of rendered feed commodities. These samples will be submitted to CVM's Office of Research for culturing and analysis. In addition to Enterococcus, E. coli, salmonellae, and Campylobacter will be selectively cultured and tested for antibiotic susceptibility profiles.
Dr. Wagner is a Research Animal Scientist in CVM's Division of Animal and Food Microbiology.