Skip to main content
ARS Home » Southeast Area » Athens, Georgia » U.S. National Poultry Research Center » Bacterial Epidemiology & Antimicrobial Resistance Research » Research » Research Project #430181

Research Project: Control Strategies and Evaluation of the Microbial Ecology Associated with Foodborne Pathogens and Poultry Processing

Location: Bacterial Epidemiology & Antimicrobial Resistance Research

2018 Annual Report

1a. Objectives (from AD-416):
Objective 1: Develop, evaluate and optimize processing treatments to reduce, control and potentially eliminate foodborne pathogens in poultry processing. Objective 2: Evaluate and define the potential role of protozoa in shaping the ecology of bacterial pathogens in controlling foodborne pathogens in poultry processing environments. Objective 3: Develop algorithms for interpreting and handling sequencing data to aid in epidemiological tracking, defining differences in isolates of foodborne pathogens, including antibiotic resistance patterns, and predicting and determining the source of the isolate.

1b. Approach (from AD-416):
The goals of this project fit into two major approaches: 1) development of alternative methods for processing poultry products, and 2) development of methods that accurately monitor the microbial quality of poultry products processed by alternative methods. The alternative methods include testing several novel chemical and physical decontamination procedures. The approach for most of this work is to apply the intervention strategy and compare the microbial quality of the treated poultry product with control product treated by standard methods. A long term objective is to develop systems of using protozoa as natural controllers of foodborne pathogens. This will involve studying the ecology of protozoa that feed on the pathogens and determining methods to enrich the processing environment with effective protozoa. Approaches for monitoring microbial quality will include enhancing the sensitivity and specificity of microbial detection. The project will also use genetic typing methods including whole genome sequencing and metagenomic sequencing to track specific clones of pathogens in and around poultry processing environments.

3. Progress Report:
This is the final report for Project Number 6040-41420-005-00-D, it is being merged with project 6040-32000-009-00-D, "Monitoring and Molecular Characterization of Antimicrobial Resistance in Foodborne Bacteria." A study was conducted to measure the flock prevalence of Campylobacter by weekly sampling of gut contents collected in a commercial broiler slaughter plant. Using data from three years of this collection and readings from a nearby weather station, we tested for a relationship between month of the year, rain fall, and temperature and the prevalence of Campylobacter in broiler flocks. Ceca have been collected from more than 300 flocks and now have data from four continuous years. These data are being continually collated and used to further examine flock prevalence and seasonality of Campylobacter. Gut contents are also being archived and analysis is underway to compare the microbiome of Campylobacter positive and negative flocks to determine if there is a previously undescribed relationship between other bacteria and the presence or absence of Campylobacter. There have been several outbreaks of Campylobacteriosis linked with broiler chicken livers. We have undertaken a study to determine the prevalence of Campylobacter on the outside and inside surfaces of chicken liver both in a processing plant and at retail. We cultured exudate, the outer surface, inner tissue and whole blended liver chicken livers from packages purchased at retail. Campylobacter was detected inside and outside of livers in some samples from each of 15 packages. All isolates were collected and subjected to whole genome sequencing and multi-locus sequence typing. Multiple subtypes, including types previously reported associated with human disease, were found in all packages and even within some livers. There was no clear difference in subtype detected on the outer surface or inner tissue of chicken livers. Studies were undertaken to improve detection and culture of Campylobacter. Control of Campylobacter relies on sensitive and accurate detection of the organism. Three types of Campylobacter selective plating media were tested for selection of specific subtypes from broiler gut contents and carcass rinse. Such a selection could skew data towards or away from one subtype and have potential to confound our understanding of Campylobacter ecology. Forty nine of 100 samples had a Campylobacter positive result on all three media (campy-cefex, campy-line and Rf-Campylobacter agars). Isolates were subjected to whole genome sequencing and the results of sequencing were used to determine MLST sequence type of each isolate. It was far more likely that all three media recovered the same subtype of Campylobacter from each sample than for any one media to result in the recovery of a different type. We have shown that a 0.45 micron nitrocellulose filter placed on top of solid plating medium is useful as a means to eliminate non-Campylobacter contamination from samples with high numbers of both Campylobacter and background bacteria such as broiler feces or gut contents. A study was performed to determine the rate of Campylobacter passage through filters and onto the growth medium below. We also began to measure the proportion of an inoculum that makes passage through the filter. The data will allow us to use the filter method to quantitatively estimate the cellular density of Campylobacter in a complex, highly contaminated sample. Antimicrobial processing aids used to kill bacteria on chicken carcasses in the processing plant can have carryover of the killing effect on a carcass rinse sample. This can result in a false negative evaluation of the carcass when the rinsate is later cultured for Salmonella. Such a false negative can confound the processor or regulators in ongoing efforts to monitor Salmonella prevalence. We defined and tested neutralizers and developed a new culture medium that can be used to detect Salmonella in the presence of the carry-over killing effect of peroxyacetic acid, cetylpyridinium chloride and several other commonly used antimicrobials and we determined the efficacy of the mew medium relative to the traditional carcass sampling medium. This new medium, neutralizing buffered peptone water (n-BPW), has been accepted by regulators and is in use nationwide. However, the medium has not been validated with Campylobacter. Floor drains in poultry processing plants are commonly colonized by Listeria monocytogenes, which can contaminate finished product. Experiments were performed to test a novel chlorine dioxide generation and delivery system as a means to sanitize floor drains. The work documented the utility of self-contained pods to significantly lower the number of total aerobic bacteria in floor drains. Further testing was conducted in model floor drains in which a Listeria monocytogenes biofilm was previously established. The chlorine dioxide was very effective at killing Listeria in the drains. A pre-chill cooking procedure was conceived to eliminate all viable bacteria from broiler carcasses prior to shipment to a further processing plant. A process to cook whole broilers before rigor was developed and tested to determine the effect on meat quality. Overall, meat quality as measured by tenderness was not substantially impacted by cooking before chilling. This represents a potential method to eliminate bacterial contamination on broiler meat prior to shipment to a further processing facility for deboning, dicing and other further processing. By avoiding immersion chilling, a processor may also save energy, water, plant space and equipment costs. The numbers of Campylobacter on broiler carcass skin surface increases due to leakage of gut contents during automated defeathering. To control the contamination of poultry carcasses during processing, a novel method was designed and developed for plugging the cloaca of killed broiler carcasses prior to scalding. A tampon sewed into the cloaca of a carcass between killing and picking was an effective means of plugging cloacae for experimental purposes but is not logistically practical in commercial setting. We developed a shredded absorbent sponge material to test as a means to plug the cloaca. It was found that the shredded sponge material can be placed in the cloaca and much of it stays in place while plugged carcasses are in the feather picking machinery. This system may be a practical means to prevent Campylobacter escape from the cloaca during defeathering. Listeria monocytogenes can form biofilms and colonize floor drains in poultry processing plants causing potential to contaminate finished poultry meat products or product contact surfaces. Some protozoa, including Tetrahymena, can ingest and kill bacteria. ARS scientists in Athens, Georgia tested Tetrahymena as a means to control L. monocytogenes numbers in model drain systems both as free swimming planktonic cells and attached to a surface as a biofilm. Preliminary work showed that co-incubation with Tetrahymena can effectively lower the number of viable Listeria monocytogenes cells both in liquid and attached to a surface. Experiments have also revealed that adding Tetrahymena to a pre-formed biofilm can lower the numbers of Listeria monocytogenes. Killing of the bacteria by the protozoa begins almost immediately and is substantially completed before 24 hours. A panel of L. monocytogenes representing a diversity of the organism were tested and all strains were equally destroyed by the protozoa. Natural streams were tested and some were found to harbor Listeria monocytogenes. Wild protozoa were collected from these streams and used in experiments along with Tetrahymena and Euglena for the ability to kill L. monocytogenes as both free-swimming planktonic and biofilm cells. We found that Tetrahymena out performs the other kinds of protozoa tested for this purpose. A microfluidic device was developed for evaluating the motility and trapping of protozoa. Euglena was tested in the device and it was shown that the protozoa were attracted to an extract from Listeria monocytogenes, a food-borne pathogen. The device may be used to select protozoa that are preferentially attracted to pathogens that are then destroyed by the protozoa. Using this device it was demonstrated that specific volatile organic compounds were released by the Listeria that were responsible for attracting the protozoa. Whole-genome multi-locus sequence typing (wg-MLST) is a data-intensive method for evaluating populations of bacteria. Transforming raw DNA sequence data into a matrix that can be readily analyzed is a key step that needs more automation. We have developed a script that can use reference databases for any desired organism to create the analysis matrix. This script has been applied to two populations of Listeria monocytogenes. One population was captured from a North Georgia river and there was evident segregation of types according to the environment surrounding the capture site. In another study of the genomic content of L. monocytogenes isolated from processing plants, a total of 352 genes were significantly associated with repeated isolation of the specific strain versus sporadic strain isolation. It remains unknown how these genes may contribute to colonization of the processing plant. A gene that confers resistance to colistin (polymyxin E) has frequently been found on a type of plasmid known as IncI2. A multi-locus sequence typing (MLST) scheme was developed for IncI2 plasmids and has been applied to 110 known IncI2 plasmids. Preliminary analyses have shown that the plasmid is more stable than expected, meaning that the core of the plasmid does not readily change even though it can readily shift to other host bacteria. Control of this plasmid needs to start while it is still relatively rare.

4. Accomplishments
1. Pre-chill application of antimicrobials to broiler carcasses. Broiler carcasses are treated by cool water pre-chill to begin the chilling process in a commercial slaughter plant. Carcasses take up water during this procedure improving yield. However, prechill water does not generally include an antimicrobial chemical; therefore it may allow viable bacteria to be taken up by carcasses which could be released during sampling of carcasses or cut up parts. ARS researchers at Athens, Georgia, developed and used a pilot scale model system to test application of antimicrobial chemicals during immersion prechill. Peracetic acid (PAA), chlorine, and a combination of both with and without a chlorine stabilizing compound (T-128) were tested as prechill antimicrobial treatments. Carcasses were subjected to treated prechill, subsequent full chill and then cut up for sampling as parts. Results show that addition of chlorine or PAA to the prechill can be effective to lessen bacterial contamination of carcasses and related cut up parts.

2. Treatment of chicken parts before incorporation into ground meat products. Ground chicken meat can be heavily contaminated with human bacterial pathogens. Much of this contamination comes from the skin and meat surface which has been handled and subject to potential cross contamination. ARS researchers at Athens, Georgia, developed a pilot scale method to treat inoculated skin and meat with antimicrobial chemicals prior to grinding. Researchers tested this system to determine the efficacy of chlorine and peracetic acid to lower numbers of skin borne Campylobacter, Salmonella and Listeria monocytogenes in ground chicken meat product. A 1200 ppm peracetic acid dip was effective to lessen the numbers of human pathogens on chicken meat and skin and the ground chicken meat product made from it.

3. Treatment of floor drains to prevent contamination by Listeria monocytogenes. Floor drains tend to collect bacteria in poultry processing plants and can become colonized with Listeria monocytogenes. This deadly human pathogen can linger in a drain for years and has potential to spread around a processing facility and even contaminate fully cooked RTE meat. ARS researchers at Athens, Georgia, tested a self-contained chlorine dioxide generating pod for decontamination of floor drains. Researchers found that the pods work well to greatly lower natural bacterial numbers in floor drains. Researchers further tested the technology against L. monocytogenes inoculated in a model floor drain system. The easy to use and safe to deploy pods were extremely effective lowering the numbers of L. monocytogenes both in the drain water and as a biofilm attached to the inner wall of the drain pipe by more than 99.9999%.

Review Publications
Berrang, M.E., Gamble, G.R., Hinton Jr, A., Johnson, J. 2018. Neutralization of residual antimicrobial processing chemicals in broiler carcass rinse for improved detection of Campylobacter. Journal of Applied Poultry Research. doi:10.3382/japr/pfx071.
Park, S., Harrison, M., Berrang, M.E. 2017. Post-chill antimicrobial treatments to control Salmonella, Listeria and Campylobacter contamination on chicken skin used in ground chicken. Journal of Food Protection. 80:857-862.
Richardson, K.E., Cox Jr, N.A., Cosby, D.E., Berrang, M.E. 2018. Impact of desiccation and heat exposure stress on Salmonella tolerance to acidic conditions. Journal of Environmental Science and Health. Part B, 53:2:141-144. 10.1080/03601234.2017.1397467.
Steininger, C., Harrison, M., Berrang, M.E. 2018. Application of antimicrobial treatment to whole carcasses during pre-chill can improve microbial quality of broiler parts. Food Microbiology.
Berrang, M.E., Harrison, M., Meinersmann, R.J., Gamble, G.R. 2017. Self-contained chlorine dioxide generation and delivery pods for decontamination of floor drains. Journal of Applied Poultry Research. 26(3):410-415. doi: 10.3382/japr/pfx009.
Berrang, M.E., Meinersmann, R.J., Cox Jr, N.A. 2017. Passage of Campylobacter jejuni and Campylobacter coli subtypes through 0.45 and 0.65 µm pore size nitro-cellulose filters. Journal of Food Protection. 80(12):2029-2032.
Tang, Y., Meinersmann, R.J., Sahin, O., Wu, Z., Dai, L., Carlson, J., Plumblee Lawrence, J.R., Genzlinger, L.L., Lejeune, J., Zhang, Q. 2017. Wide but variable distribution of a hypervirulent Campylobacter jejuni clone in beef and dairy cattle in the united states. Applied and Environmental Microbiology. 83(24):e01425-17. doi: 10.1128/AEM.01425-17.