UTILIZING A GENOMIC AND PHENOTYPIC PIPELINE FOR PREDICTING RISK AMONG EMERGENT SALMONELLA IN POULTRY PRODUCTION

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: AFRI COMPETITIVE GRANT
• Reporting Frequency: Annual
• Accession No.: 1024091
• Grant No.: 2020-67017-33235
• Cumulative Award Amt.: $479,676.00
• Proposal No.: 2020-03332
• Project Start Date: Jan 1, 2021
• Project End Date: Dec 31, 2024
• Grant Year: 2021
• Program Code: [A1332]- Food Safety and Defense

Recipient Organization
UNIV OF MINNESOTA
(N/A)
ST PAUL,MN 55108

Performing Department
(N/A)

Non Technical Summary
Salmonella continues to be a major cause of human foodborne illness, and consumption of poultry products is one of the major contributors to this problem. Despite decades of research on Salmonella, poultry producers still grapple with its control. Many Salmonella outbreaks attributed to poultry products involve new or different types of Salmonella not previously found by producers, or not previously causing major issues related to human illness. The goal of this project is to enable poultry producers to be predictive, rather than reactive, to the changing landscape of Salmonella in poultry operations. We will analyze the genomes of over 25,000 Salmonella from poultry and humans, and use this information to build a comprehensive genomic database. This database will allow anyone to rapidly analyze the DNA from Salmonella isolates coming directly from poultry farms, and predict if changes have occurred in these Salmonella that now make them a greater risk for persistence in the birds, or an enhanced risk for causing human disease. This project will empower poultry producers to respond proactively to emerging Salmonella in real time. As a result, poultry producers will be able to better mitigate Salmonella on farm before they cause large outbreaks in humans, thus reducing the burden of illness from the consumption of contaminated poultry products.

Animal Health Component

20%

Research Effort Categories

Basic

60%

Applied

20%

Developmental

20%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
712	3299	1100	100%

Knowledge Area
712 - Protect Food from Contamination by Pathogenic Microorganisms, Parasites, and Naturally Occurring Toxins;

Subject Of Investigation
3299 - Poultry, general/other;

Field Of Science
1100 - Bacteriology;

Keywords

salmonella

genomics

predictive

phenotype

genotype

poultry

Goals / Objectives
Human illness due to foodborne consumption of Salmonella continues to occur, despite efforts to regulate and reduce its occurrence in food. Poultry is one food animal that contributes significantly to this problem. Current molecular tools in place are quite effective at identifying and tracing Salmonella through production chains, and companies have developed extensive mitigation strategies to reduce or eliminate problematic clones once they occur. However, this reactive approach often results in human illness burden before mitigation is achieved. Reducing the gap between identifying and mitigating high-risk clones is a pressing need for producers. This project will address that need using high-resolution genomic analyses, combined with rapid phenotypic screens and predictive modeling. First, emergent or problematic Salmonella serovars in poultry will be investigated at the genomic level using phylogenetic and pangenome approaches. To aid in these analyses, a manually curated Salmonella virulence and fitness gene database will be developed and implemented. Second, genomic analyses will guide selection of strains within each serovar for phenotypic analyses for traits associated with adhesion, invasion, acid and disinfectant tolerance, biofilm formation, and competition within the avian gut niche. These data will be used to identify clades of higher risk, and to establish baseline risk across serovars of poultry relevance. Finally, predictive tools will be developed based upon these data to aid in surveillance by providing a risk score based on genomic and/or phenotypic attributes. This work will aid in understanding of how genomic shifts in poultry Salmonella can be used to proactively mitigate human risk.

Project Methods
Objective 1. Define the phylogenomic landscape of Salmonella in poultry production. Based upon existing data and industry feedback, we have identified serotypes of interest from which to build baseline genomic data, and to assess how genome changes correlate with increased prevalence in poultry or increased human illness. We will apply this analysis towards recent outbreak scenarios, dominant serovars of poultry irrespective of emergence, and serovars deemed emergent based upon stakeholder feedback and surveillance data. These data will represent the genomic framework for assessing potential risk.Objective 2. Use phenotypic and refined genomic analyses to establish predictive virulence and fitness scores. We will construct a Salmonella virulence and fitness gene database with emphasis on poultry fitness and human virulence. This manually curated database will help determine, at clade-specific and serovar-specific levels, which clades/serovars present greatest risk and what genotypes correlate with phenotypes assessed. Using multiple representative strains from each serovar of interest, selected using phylogenomic approaches, phenotypic assays relevant to human virulence and avian persistence will be performed. These data will enable better understanding of genomic changes that have resulted in successful poultry-associated outbreak lineages, and prediction of genomic traits conferring phenotypes that represent increased potential for persistence and human virulence.Objective 3. Develop and implement predictive tools for enhanced poultry Salmonella surveillance. We will develop and implement pipeline tools specific for Salmonella surveillance by poultry producers. This pipeline will be comprehensive and adaptable. The pipeline will be made available through a web-based tool enabling the user to input draft or completed Salmonella genome assemblies, returning key aspects of that genome's attributes. The tool will enable the use to easily track Salmonella strains within a production system in the context of currently circulating strains in poultry production and human disease across the US.

Progress 01/01/21 to 12/31/24

Outputs
Target Audience:This project involves heavy collaboration with the poultry industries, specifically the National Turkey Federation and Minnesota Turkey Growers Association. Presentations have been given to both of these groups, and we have actively involved them in this project. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has allowed for a graduate student to develop the necessary in vitro screens for this project, and apply them towards her thesis work. The student defended her thesis in March 2024. How have the results been disseminated to communities of interest?Results have been disseminated in multiple meetings to the National Turkey Federation and Minnesota Turkey Growers Association, and also at several national meetings including the North Central Avian Disease Conference and Poultry Science Association annual conference. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? We have developed a bioinformatics pipeline that has been be used to analyze 36 total Salmonella serovars implicated in food animal-associate human illness. This pipeline covers all steps laid out in Goal 1: from sample selection and quality filtering through SNP-based phylogenetic analyses, identification of accessory features, and pangenome analyses. Given the lack of a standardized classification system for isolate source information, we also developed a detailed protocol for classifying isolate source. We have developed an automated workflow for downloading of publicly available data from NCBI based upon lab source and host source. We have successfully integrated FDA's Salmonella virulence database into our own SPRAT database (sprat.onrender.com). Isolates have been run through the database prototype to determine if genomic changes have occurred that make the emergent strains more virulent or persistent. This database has been coupled with our existing pipeline for analyzing field strains of Salmonella for risk potential. Graduate student Abby Johnson in the Johnson laboratory has completed in vitro screens for adhesion/invasion in human epithelial cells; biofilm formation; acid tolerance; gastric survival; gut survival; bile tolerance; and poult colonization. Some of these experiments were performed on more than 100 isolates representing the 25 most dominant serovars found in poultry. We have developed a composite risk score model based upon the combination of in vitro traits at the serovar level, and genomic traits at the strain level. This score is provided as 0-10 and provides the ability to compare scores between serovars, but also between groups of strains within a serovar. At the serovar level, we determined the top five serovars with the highest 0-10 score were Infantis, Hadar, Johannesburg, Reading, and Breanderup. Overall, this project has provided a useful tool for incorporating gene-level Salmonella differences towards assessing the predicted virulence of an isolate towards humans. This tool can be used to proactively predict in a Salmonella found in animal production will present an enhanced risk to human health and potential foodborne outbreaks.

Publications

Progress 01/01/23 to 12/31/23

Outputs
Target Audience:This project involves heavy collaboration with the poultry industries, specifically the National Turkey Federation and Minnesota Turkey Growers Association. Presentations have been given to both of these groups, and we have actively involved them in this project. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has allowed for a graduate student to develop the necessary in vitro screens for this project, and apply them towards her thesis work. The student defended her thesis in March 2024. How have the results been disseminated to communities of interest?Results have been disseminated in multiple meetings to the National Turkey Federation and Minnesota Turkey Growers Association. What do you plan to do during the next reporting period to accomplish the goals?The web-based tool will be offically made public during the next project period (sprat.umn.edu).

Impacts
What was accomplished under these goals? We have developed a bioinformatics pipeline that has been be used to analyze 36 total Salmonella serovars. This pipeline covers all steps laid out in Goal 1: from sample selection and quality filtering through SNP-based phylogenetic analyses, identification of accessory features, and pangenome analyses. Given the lack of a standardized classification system for isolate source information, we also developed a detailed protocol for classifying isolate source. We have developed an automated workflow for downloading of publicly available data from NCBI based upon lab source and host source. We have successfully integrated FDA'sSalmonella virulence database into our own SPRAT database. Isolates have been run through the database prototype to determine if genomic changes have occurred that make the emergent strains more virulent or persistent.This database has been coupled with our existing pipeline for analyzing field strains of Salmonella for risk potential. Graduate student Abby Johnson in the Johnson laboratory has completed in vitro screens for adhesion/invasion in human epithelial cells; biofilm formation; acid tolerance; gastric survival; gut survival; bile tolerance; and poult colonization. Some of these experiments were performed on more than 100 isolates representing the 25 most dominant serovars found in poultry. We have developed a composite risk score model based upon the combination of in vitro traits at the serovar level, and genomic traits at the strain level. This score is provided as 0-10 and provides the ability to compare scores between serovars, but also between groups of strains within a serovar. At the serovar level, we determined the top five serovars with the highest 0-10 score were Infantis, Hadar, Johannesburg, Reading, and Breanderup.

Publications

Progress 01/01/22 to 12/31/22

Outputs
Target Audience:This project involves heavy collaboration with the poultry industries, specifically the National Turkey Federation and Minnesota Turkey Growers Association. Presentations have been given to both of these groups, and we have actively involved them in this project. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has allowed for a graduate student to develop the necessary in vitro screens for this project, and apply them towards her thesis work How have the results been disseminated to communities of interest?Results have been disseminated in multiple meetings to the National Turkey Federation and Minnesota Turkey Growers Association. What do you plan to do during the next reporting period to accomplish the goals?We plan to complete the poult colonization trials for representative strains belonging to each Salmonella serovar of interest. We are also completing a database which houses the computational data developed in this project; this database is being used to populate an interactive web tool intended for visualization of current isolates using our tools.

Impacts
What was accomplished under these goals? • We have developed a bioinformatics pipeline that has been be used to analyze 36 total Salmonella serovars. This pipeline covers all steps laid out in Goal 1: from sample selection and quality filtering through SNP-based phylogenetic analyses, identification of accessory features, and pangenome analyses. • Given the lack of a standardized classification system for isolate source information, we also developed a detailed protocol for classifying isolate source. We have developed an automated workflow for downloading of publicly available data from NCBI based upon lab source and host source. • We have established a collaboration with FDA to develop a Salmonella virulence database. Isolates are being run through the database prototype to determine if genomic changes have occurred that make the emergent strains more virulent or persistent. Eventually, this database will be coupled with our existing pipeline for analyzing field strains of Salmonella for risk potential. • Graduate student Abby Johnson in the Johnson laboratory has completed in vitro screens for adhesion/invasion inhuman epithelial cells; biofilm formation; acid tolerance; gastric survival; gut survival; bile tolerance; and poult colonization (pending). These experiments were performed on more than 100 isolates representing the 25 most dominant serovars found in poultry. We have developed a composite risk score model based upon the combination of in vitro traits at the serovar level, and genomic traits at the strain level. This score is provided as 0-10 and provides the ability to compare scores between serovars, but also between groups of strains within a serovar. At the serovar level, we determined the top five serovars with the highest 0-10 score were Infantis, Hadar, Johannesburg, Reading, and Breanderup.

Publications

• Type: Journal Articles Status: Published Year Published: 2022 Citation: Peichel C, Manjankattil S, Dewi G, Akhtar L, Bina P, Creek M, Deniz K, Hobbs J, Wichmann E, Feist E, Schlosser M, Johnson TJ, Noll S, Johny A. PSV-A-9 reduction of Salmonella Reading using a dairy-originated probiotic bacterium, Propionibacterium freudenreichii in combination with a live attenuated Salmonella Typhimurium vaccine in growing turkeys. Journal of Animal Science 100:255-256.

Progress 01/01/21 to 12/31/21

Outputs
Target Audience:This project involves heavy collaboration with the poultry industries, specifically the National Turkey Federation and Minnesota Turkey Growers Association. Presentations have been given to both of these groups, and we have actively involved them in this project. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has allowed for a graduate student to develop the necessary in vitro screens for this project, and apply them towards her thesis work. How have the results been disseminated to communities of interest?Results have been disseminated in multiple meetings to the National Turkey Federation and Minnesota Turkey Growers Association. What do you plan to do during the next reporting period to accomplish the goals? Abby Johnson has developed all of the key assays for our screening pipeline, which will be used to screen multiple strains from each serovar of interest: Adhesion to avian cells (BAT/C) Adhesion/invasion in human cells (Caco-2) Tolerance to disinfectants (Biophene, bleach, Virocid, Decon-7, and quaternary ammonium compounds) Acid tolerance Heat tolerance Biofilm formation A rapid poult screening assay A prototype database has been developed. Currently, it is in beta testing with our industry stakeholders. The inclusion of phenotypic data, with genomic data, will be used to develop a risk tool to identify high-risk Salmonella strains in poultry production.

Impacts
What was accomplished under these goals? We have developed a bioinformatics pipeline that will be used to analyze all serovars. This pipeline covers all steps laid out in Goal 1: from sample selection and quality filtering through SNP-based phylogenetic analyses, identification of accessory features, and pangenome analyses. Given the lack of a standardized classification system for isolate source information, we also developed a detailed protocol for classifying isolate source. The sample selection and quality filtering process has been completed for serovars Reading, Infantis, Uganda, Hadar, and Senftenberg and is underway for serovars Agona, Typhimurium, Kentucky, and Heidelberg. Analyses on serovar Infantis are nearing completion. The primary dataset used for analyses was all Infantis genomes collected in the USA from humans and poultry (n = 4289). Core genome multilocus sequence typing (based on 2913 loci) identified two distinct clades. Clade A contains a larger number of human-source samples from earlier collection years, while clade B is dominated by poultry-source (primarily chicken-source) isolates from later collection years. Accessory feature identification was carried out on all 8211 genomes. Specific features looked for included antimicrobial resistance genes (ARGs) and antimicrobial resistance-conferring SNPs (ResFinder and PointFinder databases), plasmid replicons (PlasmidFinder), Salmonella Pathogenicity Islands (SPIFinder), and mobile genetic elements (MobileElementFinder). A large conjugative megaplasmid, pESI (plasmid for emerging S. Infantis) with multiple ARGs, was found to be associated with clade B, although there were some lineages within clade A that also carried pESI. For USA human- and poultry source genomes, there does not appear to be a consistent AMR profile associated with the plasmid, although some ARGs appear to be closely associated (e.g. ant(3")-Ia, sul1, and tet(A)) than others. We have established a collaboration with FDA to develop a Salmonella virulence database. Isolates are being run through the database prototype to determine if genomic changes have occurred that make the emergent strains more virulent or persistent. Eventually, this database will be coupled with our existing pipeline for analyzing field strains of Salmonella for risk potential. Graduate student Abby Johnson in the Johnson laboratory has collaborated with Dr. Anup Johny's lab to develop in vitro screens for adhesion/invasion to avian and human epithelial cells. We used this model to demonstrate that the acquisition of a key virulence factor by S. Reading, sopE, corresponded with an increase in invasive potential for Salmonella Reading belonging to circulating and emergent clades (including the outbreak strain of 2018-2019). Experiments are currently being conducted with strains from serovars Infantis and Hadar.

Publications

• Type: Journal Articles Status: Published Year Published: 2021 Citation: Manjankatti S, Nair DV, Peichel C, Noll S, Johnson TJ, Cox RB, Donoghue AM, Johny AK. Effect of caprylic acid alone or in combination with peracetic acid against multidrug-resistant Salmonella Heidelberg on chicken drumsticks in a soft scalding temperature-time setup. Poultry Science, 100:101421.
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Dewi G, Manjankattil S, Peichel C, Jia S, Nair D, Vickers Z, Johnson TJ, Cardona C, Noll S, Johny AK. Effect of plant-derived antimicrobials against multidrug-resistant Salmonella Heidelberg in ground turkey. Poultry Science, In Press.
• Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Peichel C, Manjankattil S, Dewi G, Johnson TJ, Noll S, Johny AK. Effect of a live attenuated Salmonella Typhimurium-vaccine against emerging Salmonella Reading in poults. Poultry Science Association Annual Meeting, July 2021, Virtual.
• Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Emergence of a novel Salmonella enterica serotype Reading clonal group is linked to its expansion in commercial turkey production. Invited presentation National Turkey Federation annual meeting, Nashville, TN, February 2020.
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Nair DVT, Johnson TJ, Noll SL, Johny AK. Effect of supplementation of a dairy-originated probiotic bacterium, Propionibacterium freudenreichii subsp. freudenreichii, on the cecal microbiome of turkeys challenged with multidrug-resistant Salmonella Heidelberg. Poultry Science 100:283-295.