Salmonella Detectives

By Annel K. Greene, PhD
Center Director, Clemson University
Animal Co-Products Research and Education Center

With the recent outbreak of Salmonella in eggs, questions quickly arose about the source of the contamination and the culpability for the outbreak. In such an episode, it is important for the rendering industry to have data to validate safety. Dr. Xiuping Jiang, associate professor of Food Science and Microbiology and a member of the Clemson University Animal Co-Products Research and Education Center (ACREC) team, is working on ACREC-funded Salmonella studies as well as on several different Salmonella-related research projects on rendered animal co-products with the Animal Protein Producers Industry (APPI) and NP Analytical Laboratories.

In 1984, APPI initiated its Salmonella Reduction Education Program and a microbiological survey of animal protein meals. NP Analytical Laboratories analyzed more than 3,400 protein meal samples from 31 rendering facilities in the first five months of 2010. In reviewing the data from NP Analytical Laboratories, 257 of the samples (7.5%) were positive for Salmonella. Of these samples, 32 (0.9 percent) were enumerated and the isolates were identified to serotype. None of the tested samples contained the foodborne pathogens Salmonella enteriditis or Salmonella typhimurium.

Most Salmonella serovars are pathogenic; however, none of the serovars isolated from animal protein meals were listed as the most dangerous serovars in the recently released Food and Drug Administration (FDA) draft Compliance Policy Guide, Section 690.800, Salmonella in Animal Feed. The 32 Salmonella positive samples were shipped to Jiang’s laboratory for further analysis for antimicrobial resistance. Jiang and her team determined that approximately 94 percent of the Salmonella isolates from animal protein meals were susceptible to the most commonly used antibiotics.

Jiang, along with graduate student Brandon Kinley, fellow ACREC researcher Dr. Paul Dawson, and Clemson statistician Dr. James Rieck, recently published a study in the Canadian Journal of Microbiology(1) that involved a survey of bacterial contamination rates in rendered animal products. Total bacterial counts as well as a number of specific bacterial species were enumerated. The researchers further studied the prevalence of Salmonella and enterococci in the samples. The team analyzed 150 samples (including meat and bone meal, poultry meal, feather meal, and blood meal) provided by U.S. rendering companies. Results of this study indicated moisture contents ranging from 1.9 to 11.5 percent. Total bacterial counts were reported and the researchers determined that 81.3 percent of the samples contained Enterococcus species. However, only 8.7 percent of the samples contained Salmonella.

These findings are similar to the results from NP Analytical Laboratories in which there was a 7.5 percent incidence rate of Salmonella in animal protein meals. Both studies indicate a considerably lower incidence of Salmonella in rendered products than the 17 percent rate reported in a study in 1968(2), the 56 percent rate reported by FDA in 1995(3), the 25 percent rate of Salmonella enterica present in samples reported by FDA in 2002(4), and the 25 percent rate previously reported in an earlier APPI survey(5). The study also indicated that poultry and feather meal samples were more likely to contain Salmonella than other protein meals. Further study by Jiang and Kinley indicated that Salmonella was not persistent in the plant’s environment over time and the presence of Salmonella in finished protein meal products may be due to post-process contamination. Jiang’s study was conducted on meals collected directly from the end of the processing line in the rendering facility as compared to the samples in the FDA studies that were transported and subjected to secondary contamination prior to sample collection.

Jiang and Kinley continued their study by testing six Salmonella isolates for thermal resistance. The researchers determined that all Salmonella isolates from animal protein meals had D values within nine minutes at 55 degrees Celsius ©, two minutes at 60 degrees C, and under 45 seconds at 65 degrees C. The D value is defined as the time required to destroy 90 percent of a bacterial population at a specific temperature. Since rendering cookers typically reach 121 to 138 degrees C and products are thermally processed for times of one hour or more, the rigor of the thermal treatment in the rendering cooker far exceeds the thermal conditions used in the study. This indicated that the rendering process should destroy all Salmonella, provided there are no cold spots within the cooking system. Also, since the inherent Salmonella in the meals had a low heat resistance, Salmonella contamination was most likely due to post-process contamination.

Low incidence of Salmonella and absence of E. coli in all of the animal protein meal samples indicate the rendering industry is doing a good job of destroying microorganisms in its processing. The study was of importance in providing baseline microbiological data for fresh rendered protein meals and because the samples were collected from production lines prior to transport to and mixing with other feed components in a feed mill, the results are more accurate in representing microbial load than previous studies.

Jiang reported that use of heat resistant enterococci could be useful as a marker to determine adequate heating and/or post process contamination of products. Enterococcus species is a common bacterium that is found in the environment and in animals and has been previously reported in animal protein meals. The organism is heat tolerant and, therefore, presence of this organism in 81.3 percent of the samples is not surprising. Previous studies by FDA have indicated similar results. Enterococci are used to indicate the presence of antimicrobial resistance genes. The two resistance genes (vanA and vanB) associated with plasmids have been reported as conferring high vancomycin resistance. In the study, Jiang and her team determined that none of the enterococci isolates from animal protein meals contained either vanA or vanB genes, indicating a low potential for vancomycin-resistant genes transferring from the enterococci in the meal to other species in the product.

In other studies, Jiang is continuing her work on producing bacteriophages against Salmonella for large scale application to animal protein meals and to treat processing environments. Bacteriophages are viruses that are specific to bacterial species and are capable of quickly killing host bacteria. Bacteriophages are self-replicating antimicrobials that are safe to humans, animals, and plants since they are only capable of affecting specific host bacteria. Jiang’s team has isolated bacteriophages capable of destroying Salmonella.

The project is now moving into studies on how to efficiently produce large volumes of the bacteriophages and subsequent purification. Results of Jiang’s project indicate that this biological control method has potential for reducing Salmonella populations within the rendering processing plant environment and in animal proteins. The study also indicates that the bacteriophages may protect products against future Salmonella contamination and may be effective in reducing Salmonella infection in animals who consume the bacteriophages.

Jiang and her research team are microbiological detectives on the trail of Salmonella knowledge related to rendered products. Their ultimate goal is to eliminate Salmonella from these materials. The data gathered by Jiang and her team is rapidly increasing the database of knowledge concerning the microbial world of rendered products. This knowledge will help improve the biosafety of these materials and help protect the industry from biosecurity issues. R

1. Kinley, B., J. Rieck, P. Dawson, and X. Jiang. 2010. Analysis of Salmonella and enterococci isolated from rendered animal products. Canadian Journal of Microbiology. 56:65–73.
2. Loken, K.I., K.H. Culbert, R.E. Solee, and B.S. Pomeroy. 1968. Microbiological quality of protein feed supplements produced by rendering plants. Applied Microbiology. 16(7):1002–1005.
3. McChesney, D.G., G. Kaplan, and P. Gardner. 1995. FDA survey determines Salmonella contamination. Feedstuffs. 67:20–23.
4. Crump, J.A., P.M. Griffin, and F.J. Angulo. 2002. Bacterial contamination of animal feed and its relationship to human foodborne illness. Clinical Infectious Diseases. 35(7):859–865.
5. Franco, D.A. 2005. A survey of Salmonella serovars and most probable numbers in rendered animal protein meals: inferences for animal and human health. Journal of Environmental Health. 67(6):18–22.

ACREC Solutions – October 2010 RENDER | back