Research Project: Reduction of Foodborne Pathogens and Antimicrobial Resistance in Poultry Production Environments

Location: Egg and Poultry Production Safety Research Unit

2022 Annual Report

Objectives
1. Identify factors within the hatchery and brooder phase that induce serotype diversity and homologous recombination within Salmonella enterica subspecies I for the purpose of facilitating reduction of colonization of chicks through environmental remediation. 1.A. Investigate Salmonella ecology within commercial hatchery environments through the use of bio-mapping. 1.B. Identify conditions that facilitate either Homologous Recombination (HR) and/or Clonal Expansion (CX) by Salmonella enterica within the hatchery environment. 1.C. Develop intervention strategies for impeding emergence of new serotypes of Salmonella enterica in the hatchery and brooder environments. 2. Identification, characterization, and application of probiotic commensal microbes as an alternative to antibiotics to reduce Salmonella prevalence within commercial poultry houses. 2.A. Identification of non-pathogenic bacterial species that are modulated upon Salmonella Heidelberg infection. 2.B. Evaluate and characterize potential non-pathogenic strains to be included in the proLitterbiotic (pLb) culture collection. 2.C. Evaluate efficacy of proLitterbiotic (pLb) under live production scenarios. 3. Identify environmental and management drivers of foodborne pathogen ecology under pastured poultry rearing systems. 3.A. Environmental and microbiological characterization of pastured poultry farms to identify drivers of foodborne pathogen ecology. 3.B. Predict Salmonella prevalence during live production within pastured poultry flocks. 3.C. Evaluate the effectiveness of implementing probiotic/all-natural products within the diet of very young chicks (<1 week of age) on poultry gut health and product safety.

Approach
Our goal is to reduce pathogenic and antibiotic resistant Salmonella in eggs and poultry products entering the processing/post-harvest environment by generating research that identifies the drivers of Salmonella ecology in pre-harvest environments. This investigation will begin at the nexus of commercial poultry management (hatchery) and extend onto the farms where the birds are reared to processing weight, and we will investigate the variables that genotypically and phenotypically affect the presence of Salmonella pre-harvest. A better understanding of Salmonella ecology and diversity through the pre-harvest phase of poultry production will reduce Salmonella loads entering the processing environment and result in a safer product for the consumer. We will test alternative hypotheses about which pre-harvest environmental factors and management practices influence genomic (clonality, diversity) and phenomic (growth potential, antibiotic resistance) attributes of Salmonella. We will develop and test a “proLitterbiotic” culture to evaluate its efficiency to reduce the development of multidrug resistant Salmonella in live broiler chickens. Live production studies will use pastured poultry farms as a model for poultry management as we have access to working and experimental pastured poultry farms for more controlled research experiments. Expected outcomes for regulatory agencies, the conventional and pastured poultry industries and the consumer include: i) data-supported approaches for identifying risks associated with contamination of poultry entering the processing/post-harvest environments; ii) tools that facilitate characterization of Salmonella serovars and how mixtures correlate to epidemiological trends; iii) new approaches to interventions intended to disrupt the ability of Salmonella to maintain an optimized genome; iv) correlation of genomic markers to antimicrobial resistances present between and within Salmonella serovars within pre-harvest environments; and v) identification of best pre-harvest practices and alternatives to antibiotics that will help producers reduce foodborne pathogens in consumer products.

Progress Report
Biomapping of a commercial broiler hatchery showed that Salmonella prevalence and serotype diversity were both facility area – and sample type-dependent. Hatcheries represent a nexus point in commercial broiler management between broiler egg production and live broiler production, therefore it represents an ideal area to target for in-depth Salmonella ecology experiments to improve commercial broiler food safety. ARS researchers in Athens, Georgia, biomapped a commercial broiler hatchery facility by targeting eggs from 2 flocks in peak production and sampling the major hatchery areas (egg inventory, pre in ovo incubation, post in ovo incubation, chick processing, and chick transport). For each area, 4 major sample types were recovered: eggs/chick; air filters/fans; floor drains; and rodent bait boxes. Salmonella was recovered from 27.1% (39/144) of the samples, with the highest prevalences in the processing (41.0%: 16/39) and transport (30.8%: 12/39) areas and from the floor drain samples (46.2%; 18/39). In terms of serotypes, the most dominant serotypes recovered were Kentucky (38.5%; 15/39). Only 12.8% (n = 5) of the isolates representing serotypes in the CDC top-32 serotypes of greatest interest to human health (Enteritidis, Mbandaka), with those isolates only found in the chick processing (Mbandaka) or transport (Enteritidis) areas of the facility. These results indicate potential critical control points within the hatchery facility (processing and transport areas) and where to focus within those areas (floor drains) to reduce Salmonella levels entering the live production farms and improve overall commercial broiler food safety. A rarely encountered plasmid of the foodborne pathogen Salmonella enterica carries cassettes of nucleotides with 99% similarity to a 23S rRNA gene on the chromosome. There is tremendous selection pressure within Salmonella enterica to maintain a core genome common to the 2% of serotypes that cause foodborne illness. ARS researchers in Athens, Georgia, used the dkgB intergenic sequence region from 96 Salmonella enterica strains encompassing 70 serovars to analyze the NCBI database of completed genomes of 1804 chromosomes and 1311 plasmids (last date of access July 6, 2022). Of the 1311 plasmids, 5 were found that had one or two different regions of approximately 2,500 nucleotides with greater than 99% similarity to sections of the chromosomal 23S ribosomal gene rrlH. The high degree of similarity of plasmid sequence with 23S ribosomal genes in the chromosome suggests that it is involved in the process of homologous recombination, which is important for maintaining the ability of Salmonella enterica to repair its genome and to maintain the ability to cause foodborne illness. Isolation of oxygen-sensitive non-pathogenic bacterial strains from chicken cecal contents. ARS researchers in Athens, Georgia, completed the testing of several selective media for the recovery of oxygen-sensitive Bifidobacterium and Lactobacillus species from the cecal contents of young broiler chickens. Polymerase chain reaction (PCR) primers targeting Bifidobacterium were used to confirm the identity of 66 presumptive isolates, while Biolog phenotype microarray MicroPlates were used for the identification of 11 Lactobacillus species. Two of the five Bifidobacterium isolates confirmed by PCR were sequenced and found to be Enterococcus species. Presumptive Lactobacillus species were identified by Biolog as acidophilus (n =7), lactis (n =2), hamster (n =1) and gasseri (n =1). One explanation for the incongruence between Bifidobacterium culture, PCR and whole genome sequencing results is that the media and/or PCR primers used were not specific for Bifidobacterium. Moreover, the fastidious nature of oxygen-sensitive bacteria makes them prone to loss and contamination during freeze-thaw cycles. This resulted in the loss of several presumptive Bifidobacterium and Lactobacillus isolates and could also explain the false Bifidobacterium positives reported. These results suggest that the oxygen-sensitive species are not ideal candidates to develop into proLitterbiotics; therefore, oxygen-sensitive non-pathogenic bacterial strains will no longer be considered for the development of proLitterbiotics. Selection of aerobic non-pathogenic litter bacterial strains for the development of a proLitterbiotic to reduce pathogenic Salmonella serotypes. ARS researchers in Athens, Georgia, began the characterization of twenty-nine Bacillus species isolated from reused litter previously inoculated with either Salmonella Heidelberg or Enteritidis. These isolates are being evaluated for their potentials as direct fed microbials and for proLitterbiotic development. Whole genome sequencing was used to confirm the species of Bacillus to be subtilis (n=17) and velezensis (n =12). All B. subtilis strains harbored antibiotic resistance genes on a mobile genomic island and are no longer under consideration for the development into proLitterbiotic. B. velezensis strains did not harbor a plasmid, and no antibiotic resistance or virulence gene was found on a mobile genomic island. B. velezensis strains showed high tolerance to acidic broiler litter extract (pH of 2) which suggests that they can survive the acidic pH of the stomach and acidified litter. Four unique B. velezensis strains named “LitMiChic” were selected and will be used for the development of proLitterbiotic. Preliminary results with LitMiChic were promising as demonstrated through conjugation experiments between E. coli and Salmonella Heidelberg. LitMiChic reduced the efficiency for the transfer of multidrug resistant plasmids from E. coli to Salmonella Heidelberg. These results suggested that LitMiChic has the characteristic of a “host-tailored” probiotic that can be used for pathogen reduction. Further studies will involve a complete genome assembly of the four B. velezensis strains with long read sequencing and investigation of their survival in fresh and reused litter microcosms. Determination of bacterial pathogen diversity in acidified re-used poultry litter during commercial live production. ARS researchers in Athens, Georgia, investigated the diversity and changes in abundances of several bacterial foodborne pathogens (Salmonella, Campylobacter), commensal bacteria (E coli, Enterococcus and Staphylococcus) and fungi from commercial poultry house acidified litter. Not only are these changes in abundance and diversity observed through a single flock within two houses but were followed over five flocks for an entire production year. Preliminary findings show that litter acidification does not allow for Campylobacter growth throughout grow-out, but appears to reduce Salmonella, E. coli, Staphylococcus, and Enterococcus populations during the first few weeks of grow-out, with those populations either returning to pre-acidification levels or exceeding them by the end of the 6-week grow-out period. Using machine learning algorithms to identify pastured poultry management variables that are predictive of pre-harvest Salmonella prevalence. Prevalence of Salmonella in pastured poultry production systems can lead to contamination of the final product. Therefore, the identification of farm practices that affect Salmonella prevalence is critical for implementing control measures to ensure the safety of these products. ARS researchers in Athens, Georgia, in collaboration with computer scientists from Mississippi State University developed predictive models based predominantly on deep learning approaches to identify key pre-harvest management variables in pastured poultry farms that contribute to Salmonella prevalence. This ensemble approach utilizing five different machine learning techniques predicted that physiochemical parameters of the soil and feces (metals such as sodium (Na), zinc (Zn), potassium (K), copper (Cu), electrical conductivity), the number of years that the farms have been in use, and flock size significantly influenced pre-harvest Salmonella prevalence. Egg source, feed type, breed, and manganese (Mn) levels in the soil/feces are other important variables identified to contribute to Salmonella prevalence on larger (=3 flocks reared per year) farms, while pasture feed and soil carbon-to-nitrogen ratio are predicted to be important for smaller/hobby (<3 flocks reared per year) farms. Predictive models such as the ones described here are important for developing science-based control measures for Salmonella to reduce the environmental, animal, and public health impacts from these types of poultry production systems.

Accomplishments
1. A genomic motif was identified that differentiates pathogenic bacteria by Genus and species regardless of strain, lineage, or serovar differences. Some classes of antibiotics kill bacteria by inserting within the double strand of the organism’s DNA and interfering with replication; however, the physics and chemistry of DNA can interfere with the lethality of antibiotics and reduce efficiency. ARS researchers in Athens, Georgia, identified that an 8-nucleotide motif, namely “AAAAAAAA” and by inference its complement “TTTTTTTT”, is dispersed within the bacterial genome in a manner that is specific to the Genus and species of the organism, and it is not substantially changed by smaller scale genomic differences. Therefore, there is tremendous selection pressure to maintain the integrity of the motif despite these AT rich regions being susceptible to single point mutations. Defining a role for the AT 8-nucleotide motif as preserving the Genus and species of bacteria causing disease suggests that it is a rich target for development of new antibiotics that might avoid side effects in the host.

2. Raising chickens with a resilient microbiome. Reusing litter to raise consecutive flocks of broiler chickens is a widespread and well accepted practice in the United States and Brazil – two major world producers of broiler chickens. However, this practice of reusing litter is not acceptable in some parts of world including Canada and Europe. ARS researchers in Athens, Georgia, University of Georgia, Colorado State University, Centers for Disease Control and Prevention, Phase genomics. Inc., and Institute of Food Science, Vienna, Austria infected neonatal broiler chicks with Salmonella enterica serovar Heidelberg, a model pathogen. Afterwards, the chicks were raised on fresh pine shavings or reused litter for 14 days. The reuse of litter as bedding enriched for Bifidobacterium in the chick’s microbiome protected the animals from Salmonella infections, and reduced the spread of antimicrobial resistance from E. coli reservoirs. The study showed that litter management practices need to be considered as a strategy to naturally protect livestock and prevent the spread of antibiotic resistance.

3. A farm-to-fork perspective of Salmonella prevalence and diversity with pastured poultry management systems. A farm-to-fork perspective of Salmonella prevalence and diversity with pastured poultry management systems. Greater consumer demand for all natural, antibiotic-free poultry products has led to an increase in pastured poultry operations which has increased the level of environmental interaction, and potentially increased the exposure to foodborne pathogens. ARS researchers in Athens, Georgia, are researching the prevalence and diversity of Salmonella populations inherent within pastured poultry flocks. Salmonella was isolated and characterized from pre-harvest, post-harvest, and final product samples from flocks raised antibiotic-free. Salmonella was recovered from ~18% of the over 2300 samples, with Kentucky being the serotype most commonly isolated (~72% of all isolates). Even in the antibiotic-free pastured management system, approximately two-thirds of the Salmonella isolates exhibited resistances to tetracycline and streptomycin, as well as other clinically important antibiotics. These analyses showed that Salmonella prevalence and diversity were related more to the farm location than to the type of sample being tested, indicating the need for more tailored intervention strategies to continue to enhance the safety of these products.

Review Publications
Guard, J.Y., Rivers, A.R., Vaughn, J.N., Rothrock Jr, M.J., Oladeinde, A.A., Shah, D. 2021. AT homopolymer strings in salmonella enterica subspecies I contribute to speciation and serovar diversity. Microorganisms. 9(10):2075. https://doi.org/10.3390/microorganisms9102075.
Rothrock Jr, M.J., Oladeinde, A.A., Guard, J.Y. 2021. Salmonella diversity along the farm-to-fork continuum of pastured poultry flocks in the southeastern United States. Frontiers in Animal Science. 2:761930. https://doi.org/10.3389/fanim.2021.761930.
Welch, C.W., Lourenco, J.M., Seidel, D.S., Krause, T.R., Rothrock Jr, M.J., Pringle, T., Callaway, T.R. 2021. The impact of pre-slaughter fasting on the ruminal microbial population of commercial angus steers. Microorganisms. 9(12):2625. https://doi.org/10.3390/microorganisms9122625.
Oladeinde, A.A., Abdo, Z., Zwirzitz, B., Woyda, R., Lakin, S.M., Press, M.O., Cook, K.L., Cox Jr, N.A., Thomas, J.C., Looft, T.P., Rothrock Jr, M.J., Zock, G.S., Plumblee Lawrence, J.R., Cudnik, D., Ritz, C. 2022. Litter commensal bacteria can limit the horizontal gene transfer of antimicrobial resistance to Salmonella in chickens. Applied and Environmental Microbiology.88(9). https://doi.org/10.1128/aem.02517-21
Oladeinde, A.A., Abdo, Z., Press, M.O., Cook, K.L., Cox Jr, N.A., Zwirtzitz, B., Woyda, R., Lakin, S.M., Thomas Iv, J.C., Looft, T.P., Cosby, D.E., Hinton Jr, A., Guard, J.Y., Line, J.E., Rothrock Jr, M.J., Berrang, M.E., Herrington, K., Zock, G.S., Plumblee Lawrence, J.R., Cudnik, D., House, S.L., Ingram, K.D., Lariscy, L., Wagner, R., Aggrey, S.E., Chai, L., Ritz, C. 2021. Horizontal gene transfer is the main driver of antimicrobial resistance in broiler chicks infected with Salmonella enterica Serovar Heidelberg. mSystems. 6(4):e00729-21. https://doi.org/10.1128/mSystems.00729-21.
Beaudry, M.S., Thomas, J.C., Baptista, R.P., Sullivan, A.H., Norfolk, W., Devault, A., Enk, J., Kieran, T.J., Rhodes, O.E., Perry-Dow, K., Rose, L., Bayona-Vasquez, N.J., Oladeinde, A.A., Lipp, E.K., Sanchez, S., Glenn, T.C. 2021. Escaping the fate of Sisyphus: assessing resistome hybridization baits for antimicrobial resistance gene capture. Environmental Microbiology. https://doi.org/10.1111/1462-2920.15767.
Plumblee Lawrence, J.R., Cudnik, D., Oladeinde, A.A. 2022. Bacterial detection and recovery from Poultry Litter. Frontiers in Microbiology. 12:803150. https://doi.org/10.3389/fmicb.2021.803150.
Rothrock Jr, M.J., Min, B., Castleberry, B., Waldrip, H., Parker, D.B., Brauer, D.K., Pitta, D., Indugu, N. 2021. Antibiotic resistance, antimicrobial residues and bacterial community diversity in pasture-raised poultry, swine and beef cattle manures. Journal of Animal Science. https://doi.org/10.1093/jas/skab144
Ricke, S.C., Dittoe, D.K., Tarcin, A.A., Rothrock Jr, M.J. 2022. Communicating the utility of the microbiome and bioinformatics to small flock poultry producers. Poultry Science. 101(5) https://doi.org/10.1016/j.psj.2022.101788.
Xu, X., Rothrock Jr, M.J., Reeves, J., Mishra, A. 2022. Using E. coli populations to predict foodborne pathogens in pastured poultry farms. Food Microbiology. https://doi.org/10.1016/j.fm.2022.104092.
Guard, J.Y. 2022. Through the looking glass: Genome, phenome, and interactome of salmonella enterica. Pathogens. 11(5):581. https://doi.org/10.3390/pathogens11050581.
Dorr, M., Silver, A., Smurlick, D., Arukha, A., Kariyawasam, S., Oladeinde, A.A., Cook, K., Denagamage, T. 2022. Transferability of ESBL-encoding IncN and IncI1 plasmids among field strains of different Salmonella serovars and Escherichia coli. Journal of Global Antimicrobial Resistance. 30:88-95. https://doi.org/10.1016/j.jgar.2022.04.015.