Research Project: Elucidating the Factors that Determine the Ecology of Human Pathogens in Foods

Location: Produce Safety and Microbiology Research

2023 Annual Report

Objectives
Objective 1: Identify and characterize factors associated with virulence and/or environmental adaptation of human bacterial pathogens using genomic and transcriptomic analyses. Sub-objective 1.A: Develop source attribution models for Campylobacter infections using frequency matching and population genetics-based approaches. Sub-objective 1.B: Identify ganglioside-like structures associated with Guillain-Barré syndrome in non-jejuni Campylobacter taxa. Sub-objective 1.C: Identify specific Campylobacter factors that contribute to the development of post infectious-Irritable Bowel syndrome (PI-IBS) and links to host response. Sub-objective 1.D: Identify the transcriptional network patterns of bacterial pathogens under stress and during adaptation to different environments. Sub-objective 1.E: Characterize mobile elements linked to the transfer of antimicrobial resistance (AMR) genes in Campylobacter. Objective 2: Evaluate microbiomes of produce production sites and their role in antimicrobial resistance gene reservoirs and bacterial pathogen fitness. Sub-objective 2.A: Investigate the utilization of fecal microbiomes to determine the role of indigenous fauna in the spread of Salmonella and AMR. Sub-objective 2.B: Evaluate the effects of irrigation water treatment on the microbial community and foodborne pathogens. Sub-objective 2.C: Evaluate the microbiomes of produce production environments to identify the role bacteriophages play in the development of AMR in bacteria. Objective 3: Assess virulence and antimicrobial resistance of foodborne pathogens using mass spectrometry-based proteomics. Sub-objective 3.A: Perform top-down proteomic identification of toxins, antibacterial and antimicrobial resistance proteins expressed by plasmids and bacteriophage carried by foodborne pathogens. Sub-objective 3.B: Investigate biofilms of pathogens using MALDI MSI, MALDI-TOF-TOF-MS/MS and top-down proteomic analysis. Objective 4: Characterize biomarkers for the development of automated detection platforms for onsite monitoring of foodborne pathogens. Sub-objective 4.A: Develop and evaluate immuno-biosensors for the detection of C. jejuni and C. coli using a liquid crystal-based biosensor. Sub-objective 4.B: Characterize outer membrane antigens in C. jejuni as a novel single ligand for detecting Shiga toxins. Objective 5: Elucidate the interplay between bacteriophages and their bacterial hosts in the environment to enhance the safety of food products and the prevention of emerging foodborne pathogens. Sub-objective 5.A: Determine the induction parameters and the mechanisms of transduction through lysogenic bacteriophages that contribute to the potential emergence of new pathogens. Sub-objective 5.B: Investigate the role of lytic bacteriophages against their host strains and other serogroups.

Approach
Objective 1: Campylobacter from poultry may be the source of infection in infants in low- and middle-income countries. Whole genome sequencing (WGS) of Campylobacter from various animals will be used in source attribution of infected infants. Non-jejuni Campylobacter may produce human ganglioside-like structures associated with Guillain-Barré syndrome. Using antisera, dot blot assays will use antibody binding to establish the presence of such structures. Campylobacter associated with post infectious-irritable bowel syndrome (PI-IBS) may have observable genomic signatures. WGS and gene-by-gene analysis will be compared between Campylobacter isolated from infections resulting in PI-IBS or no PI-IBS. Transcriptional patterns of C. lari may be altered under salt and oxidative stress. RNA sequencing will be used to determine the patterns that correlate with adaptation. C. coli mobile elements are potentially transferred into naïve strains via transmissible plasmids. Matings between C. coli strains containing mobile elements and naïve recipients will test lateral transfer of mobile elements. Objective 2: Microbiome WGS from animal feces might detect the presence of Salmonella and antimicrobial resistance (AMR) genes. Short- and long-read WGS of microbiomes from feces near produce will be used to determine presence and transmission of Salmonella and AMR genes. Irrigation treatments may affect the diversity of microbial communities and pathogens. WGS of irrigation samples will be used to learn the effects of disinfection on microbial communities and pathogens. Some bacteriophages may be associated with the transfer of AMR genes. WGS of environmental samples and metagenomic analysis will be used to understand transmission of AMR by bacteriophage. Objective 3: Induced toxins and AMR proteins may be identified by mass spectrometry (MS) and analysis. MS will be employed to determine conditions that cause the expression of toxins and AMR proteins. Also, mass spectrometry imaging and proteomic analysis will be used to spatially map Shiga toxin-producing Escherichia coli (STEC) biofilm-associated molecules. Objective 4: Campylobacters may potentially be detected in poultry products through use of liquid crystal system methodology. Monoclonal antibodies (mAb) that bind both C. jejuni and C. coli will be evaluated for sufficient selectivity and sensitivity. Using these mAb, a liquid crystal detection platform will be developed where the mAb-Campylobacter complex causes an observable deformation of lyotropic liquid crystals. The expression of certain LOS by C. jejuni may act as biosensors to detect Shiga toxins. In vitro binding assays will be used to identify C. jejuni strains that express LOS that mimic P-blood group antigens and quantify Shiga toxin (Stx)-binding ability. Objective 5: Stx-converting bacteriophage released by STEC may infect other bacteria to form new pathogens. Phages containing Stx genes will be used to lysogenize other E. coli. Bacteriophage cocktails may be developed into biocontrol alternatives to antibiotics. Lytic phages will be developed into multi-bacteriophage cocktail formulae for the reduction of target pathogens.

Progress Report
Under Sub-objective 1.A, progress continued in developing source attribution models for Campylobacter infections. The genomic sequences of Campylobacter jejuni and Campylobacter coli from diverse sources were obtained and used to test Machine Learning (ML) methods for probabilistic assignment of human cases of campylobacteriosis to possible source reservoirs. Genetic variation associated with adaptation to the most recent host was targeted using ML and probabilistic models to estimate the relative importance of different disease reservoirs. Probabilistic attribution identified poultry as the primary source of human clinical infections in the United States over the past 10 years. Under Sub-objective 1.C, research continued in the exploration for Campylobacter factors that contribute to the development of post infectious irritable bowel syndrome (PI-IBS) and links to host response. Whole genomic sequencing data from a collection of 120 clinical Campylobacter samples were annotated, and core genome, accessory genomes and pan-genomes were determined. Genomic assemblies were deposited into the pubMLST and NCBI databases. Unfortunately, Campylobacter from 2020 to 2021, and the required follow-ups to determine the incidence of PI-IBS cases were not performed during the COVID-19 pandemic by Colorado state laboratories. Under Sub-objective 1.D, progress was made on the adaptation of Campylobacter lari to oxidative stress. All Campylobacters encode the cytosolic superoxide dismutase SodB that is critical in the oxidative stress response. The well-characterized human pathogens C. jejuni and C. coli only contain SodB while C. lari encodes both the cytosolic SodB and the periplasmic SodC. The relatively impermeable Campylobacters membrane to hydrogen peroxide (H2O2), suggests that SodB detoxifies H2O2 generated by cytoplasmic metabolic reactions while SodC detoxifies exogenous H2O2, such as that produced by macrophages. C. lari strains grown in liquid culture are only minimally impacted by the addition of H2O2, whereas C. jejuni strains are rapidly killed under the same conditions. The negative effect on C. jejuni is eliminated when catalase, which detoxifies H2O2, is added to the cultures. Thus, the presence of SodC in C. lari has a protective effect against exogenous H2O2, and C. lari strains might be expected to have a higher survival rate within human macrophages. Under Sub-objective 1.E, progress was made in the transfer of mobile element-containing, drug-resistance conjugative plasmids in Campylobacter. Conjugations involved a streptomycin-sensitive C. coli donor strain that contains a large conjugative plasmid that confers resistance to kanamycin (kanr) and tetracycline (tetr) along with C. coli or C. jejuni recipient strains that were streptomycin-resistant (strepr) leading to tri-resistant (i.e., kanr, tetr, strepr) recipient transconjugants. However, analysis of the tri-resistant strains indicated that they were generated by natural transformation occurring in the donor from pieces of recipient chromosomal DNA that conferred strepr. To remove the natural transformants, an additional C. jejuni strain was added. This strain produces a secreted DNAse, which degrades the external recipient chromosomal DNA in the cell mixture, thus preventing it from being available for natural transformation. Bacterial conjugation is not affected by the presence of the external DNAse and the addition of the DNAse-producing strain greatly reduced the C. coli natural transformants. This permitted the identification of the true tri-resistant transconjugants among the recipient population. For Sub-objective 2.A, significant progress was obtained on sampling foodborne bacterial isolates from various sources in the agricultural fields in Northwestern Mexico, an important region for fresh produce commodities exported into the United States. Using statistics, genomics, and bioinformatics, over 200 pathogenic Escherichia coli (E. coli) isolated from clinical, animal and food sources and belonging to various phylogenetic groups were characterized. A correspondence analysis revealed that the detected E. coli phylogroups were not dispersed randomly and were associated with a particular isolation source. In particular, isolates belonging to phylogroup A were significantly associated with human sources, and those in phylogroup B1 showed a relationship with food sources. Phylogroup D isolates were also related to human sources and were found to contain the largest virulence gene repertoire conferring for persistence and survival in the host. These results provided fundamental information on the relationship of genomic profiles of pathogenic E. coli with their isolation source and disease outcome in humans, enabling the development of better tools for identifying potential transmission routes of contamination. For Sub-objective 2.B, we performed whole metagenomic sequencing on pathogen spiked irrigation samples that were treated or not treated with the disinfectant, calcium hypochlorite. Irrigation water samples were spiked with different levels of Shiga toxin-producing E. coli (STEC) and Salmonella and half of the samples were treated with the disinfectant. After seven days, microbial growth was concentrated using ultrafilters. Samples were split in half with one half treated with propidium monoazide (PMA), a chemical that prevents sequencing of DNA from dead bacteria (live). The other half was untreated allowing sequencing from total bacteria (live and dead). The DNA from the samples was then extracted and sequenced. Metagenomic assemblies were performed on these sequences to determine the viable (live) microbial communities and the total microbial communities (live and dead) for each sample. Future analyses are to be performed on these metagenomic data to determine the effect on pathogen and microbial community survival. For Sub-objective 2.C, researchers continued to investigate the role of bacteriophages in the development of antibiotic resistance bacteria. Diverse agricultural samples, including farm animal feces, soil, and agricultural water, were collected from an organic farm. The bacterial and viral DNAs were extracted separately from each sample type for metagenomic sequencing. Further, the metagenomic data were used to establish a bioinformatic pipeline for screening antibiotic-resistant genes (ARG). The analysis revealed a viral population, especially bacteriophages, that contained diverse ARG types and interacted with various bacterial hosts on ARG transfer. Future studies will investigate the correlation of specific ARG transfer with bacteriophage types and their bacterial hosts in agricultural environments. Under Sub-objective 3.A, progress was made on proteomic characterization of pathogens. Salmonella enterica (SE) strain carrying a megaplasmid (pESI) with several ARG. Salmonella enterica was exposed to subinhibitory levels of antibiotics to induce ARG and identify the proteins by top-down proteomics. Further experiments are planned on SE strains with and without pESI. We also examined a STEC strain that is a strong curli and biofilm producer. In addition to Shiga toxin (Stx1a and Stx2a), we identified the osmotically inducible protein (OsmY) in its full (18.2 kDa) and a truncated (11.4 kDa) form using top-down proteomic analysis. Alphafold2 predictions of the full and truncated OsmY protein structures indicated that the truncation site is unexpectedly located in an alpha-helix secondary structure, suggesting that the sequence truncation may occur during pre-translation at the messenger RNA transcription level. Progress was made on Sub-objective 3.B. Biofilms of a STEC were analyzed by MALDI-TOF imaging mass spectrometry (IMS). A second MALDI-TOF-TOF instrument was used to perform top-down proteomic identification of specific protein biomarkers. MALDI-IMS determined that the primary protein component of curli, i.e. CsgA, was located primarily at the air/liquid interface, suggesting that curli production may be induced by oxidative stress or nutrient deprivation. Other proteins were detected at the interface, including cold-shock proteins and DNA-binding proteins HU-alpha and HU-beta. To address Objective 4, progress was achieved in developing a liquid crystal-based biosensor for detecting Campylobacter spp. in food samples with microspheres conjugated to an anti-Campylobacter monoclonal antibody (C731). ARS scientists optimized the biosensor detection threshold for Campylobacter in spiked chicken meat samples. The procedures were transferred to the industry stakeholders to develop high-throughput onsite sampling in food processing facilities. Ongoing research is currently aimed at optimizing the stringency in the buffer compositions by incorporating surfactants to improve background levels using the selected combination of microspheres conjugated to the C731 monoclonal antibody with the lyotropic liquid crystal and associated instrumentation for data analysis. Under Objective 5, research continued on the characterization of lysogenic phages and their association with bacterial hosts. The complete genomes of diverse STEC were obtained, and the presence and diversity of lysogenic phages within each STEC genome were determined. The completeness of lysogenic phage genomes was determined to investigate the potential transfer capability from one STEC host to other bacterial strains. Several genes, including virulence genes and ARG, were detected in the phage genomes. For lytic phage research, an encapsulation method was established using an alginate compound to increase the phage stability upon application in adverse environments, such as high temperature and low pH. New lytic bacteriophages were isolated to control Salmonella Infantis and were characterized genomically and tested for antimicrobial activity. Phage cocktail formulations were optimized to screen for synergistic antimicrobial effects.

Accomplishments
1. Improved categorization of pathogenic Escherichia coli (E. coli). Pathogenic Escherichia coli are causative agents of a broad range of enteric human diseases such as colitis, dysentery, hemolytic uremic syndrome as well as other extraintestinal diseases, including sepsis and urinary tract infections. With the combinatorial use of statistics, genomics and bioinformatics, ARS researchers in Albany, California, differentiated a collection of hundreds of pathogenic E. coli isolates recovered from food, animal, environmental and clinical samples based on genetic groups and disease outcome categorization. In particular, some genetic groups were significantly linked to isolation sources, and pathogenicity markers were statistically associated with each detected genetic group in the examined E. coli isolates. These findings provide fundamental information to regulatory agencies as well as public health, academic and food processing laboratories by identifying genetic markers specific to various pathogenic E. coli recovered from distinct isolation sources, including food, animal and human. and Consequently, this research will assist in developing better typing tools for identifying potential transmission routes of contamination.

2. A bacteriophage cocktail lytic against E. coli O157:H7 has been approved for patent application. Antimicrobial intervention, such as chlorination, is commonly used in the food industry because of its easy access, installation, and low cost. However, the major drawbacks are that frequent application results in the development of bacterial resistance, thus compromising the efficacies of these chemical antimicrobials. ARS researchers in Albany, California, have developed a lytic bacteriophage (phage) cocktail containing different well-characterized lytic phages that can effectively control and mitigate Shiga toxin-producing E. coli (STEC) O157:H7. The phage cocktail demonstrated a much stronger antimicrobial activity than a commercial phage product in reducing a four-strain E. coli O157:H7 cocktail by more than 3 logs. This phage cocktail (Invention Disclosure USDA Docket No. 0020.23) is in a patent application process. It will improve the antimicrobial activity of traditional intervention strategies against STEC O157 contamination in pre-harvest or post-harvest environments with better effectiveness, cost-efficiency, and reliability. It is expected that the phage cocktail can be utilized by farmers, food processing plants, and other regulatory agencies to mitigate the contamination of these pathogens in the area where the traditional interventions are not applicable or bacterial resistance to sanitizers have developed.

3. Bacteriophages involved in a new mechanism of ARG transfer within agricultural environments. The continuous emergence of antibiotic-resistance bacteria has caused a serious food safety issue in the United States. ARS researchers in Albany, California, have determined the antibiotic resistance genes (ARG) profile within various agricultural samples and discovered a new mechanism of ARG dissemination in agricultural-associated environments. The results indicate that bacteriophages carry various ARG types and can disperse these genes among the bacterial population, contributing to the development of antibiotic-resistant strains. This pivotal information unveils the controversial phage-associated ARG transmission and provides valuable information in preventing the spread of foodborne pathogens harboring ARG in produce-associated environments. The information could be used by researchers, farmers, and other regulatory agencies for tracking phage-mediated ARG transfer and preventing the emergence of ARG strains.

Review Publications
Lacombe, A.C., Quintela, I.A., Liao, Y., Wu, V.C. 2022. Shiga toxin-producing Escherichia coli outbreaks in California’s leafy greens production continuum. Frontiers In Food Science And Technology. 2. Article 1068690. https://doi.org/10.3389/frfst.2022.1068690.
Fagerquist, C.K. 2023. Top-down identification of Shiga toxin (and other virulence factors and biomarkers) from pathogenic E. coli using MALDI-TOF/TOF tandem mass spectrometry. In: Shah, H.N., Gharbia, S.E., Shah, A.J., Tranfield, E.Y., Thompson, K.C., editors. Microbiological Identification Using MALDI-TOF and Tandem Mass Spectrometry: Industrial and environmental applications. 1st edition. West Sussex, UK: John Wiley & Sons Ltd. p. 71-96. https://doi.org/10.1002/9781119814085.ch3.
Garcia Bardales, P., Schiaffino, F., Huynh, S., Paredes Olortegui, M., Penataro Yori, P., Pinedo Vasquez, T., Manzanares Villaneuva, K., Curico Huansi, G., Shapiama Lopez, W., Cooper, K.K., Parker, C.T., Kosek, M.N. 2022. “Candidatus Campylobacter infans” detection is not associated with diarrhea in children under the age of 2 in Peru. PLOS Neglected Tropical Diseases. 16(10). Article e0010869. https://doi.org/10.1371/journal.pntd.0010869.
Ndraha, N., Huang, L., Wu, V.C., Hsiao, H. 2022. Vibrio parahaemolyticus in seafood: Recent progress in understanding influential factors at harvest and food-safety intervention approaches. Current Opinion in Food Science. 48. Article 100927. https://doi.org/10.1016/j.cofs.2022.100927.
Quintela, I.A., Vasse, T., Lin, C., Wu, V.C. 2022. Advances, applications, and limitations of portable and rapid detection technologies for routinely encountered foodborne pathogens. Frontiers in Microbiology. 13. Article 1054782. https://doi.org/10.3389/fmicb.2022.1054782.
Gummalla, V., Zhang, Y., Liao, Y., Wu, V.C. 2023. The role of temperate phages in bacterial pathogenicity. Microorganisms. 11(3). Article 541. https://doi.org/10.3390/microorganisms11030541.
Parker, C.T., Schiaffino, F., Huynh, S., Paredes Olortegui, M., Penataro Yori, P., Garcia Bardales, P.F., Pinedo Vasquez, T., Curico Huansi, G.E., Manzanares Villaneuva, K., Shapiama Lopez, W.V., Cooper, K.K., Kosek, M.N. 2022. Shotgun metagenomics of fecal samples from children in Peru reveals frequent complex co-infections with multiple Campylobacter species. PLOS Neglected Tropical Diseases. 16(10). Article e0010815. https://doi.org/10.1371/journal.pntd.0010815.
Aguirre-Sanchez, J.R., Valdez-Torres, J.B., Castro del Campo, N., Martinez-Urtaza, J., Castro del Campo, N., Lee, B.G., Quinones, B., Chaidez-Quiroz, C. 2022. Phylogenetic group and virulence profile classification in Escherichia coli from distinct isolation sources in Mexico. Infection, Genetics and Evolution. 106. Article 105380. https://doi.org/10.1016/j.meegid.2022.105380.
Fagerquist, C.K., Wallis, C.M., Chen, J. 2023. Top-down proteomic identification of protein biomarkers of Xylella fastidiosa subsp. fastidiosa using MALDI-TOF-TOF-MS and MS/MS. International Journal of Mass Spectrometry. 489. Article 117051. https://doi.org/10.1016/j.ijms.2023.117051.
Fagerquist, C.K., Shi, Y., Dodd, C.E. 2023. Toxin and phage production from pathogenic E. coli by antibiotic induction analyzed by chemical reduction, MALDI-TOF-TOF mass spectrometry and top-down proteomic analysis. Rapid Communications in Mass Spectrometry. 37(10). Article e9505. https://doi.org/10.1002/rcm.9505.
Talukdar, P., Crockett, T.M., Gloss, L.M., Huynh, S., Roberts, S.A., Turner, K.L., Lewis, S.T., Herup-Wheeler, T.L., Parker, C.T., Konkel, M. 2022. The bile salt deoxycholate induces Campylobacter jejuni genetic point mutations that promote increased antibiotic resistance and fitness. Frontiers in Microbiology. 13. Article 1062464. https://doi.org/10.3389/fmicb.2022.1062464.
Liao, Y., Zhang, Y., Salvador, A., Ho, K., Cooley, M.B., Wu, V.C. 2022. Characterization of polyvalent Escherichia phage Sa157lw for the biocontrol potential of Salmonella Typhimurium and Escherichia coli O157:H7 on contaminated mung bean seeds. Frontiers in Microbiology. 13. Article 1053583. https://doi.org/10.3389/fmicb.2022.1053583.
Sun, X., Liao, Y., Zhang, Y., Salvador, A., Ho, K., Wu, V.C. 2022. A new Kayfunavirus-like Escherichia phage vB_EcoP-Ro45lw with antimicrobial potential of Shiga toxin-producing Escherichia coli O45 strain. Microorganisms. 11(1). Article 77. https://doi.org/10.3390/microorganisms11010077.
Carter, M.Q., Laniohan, N.S., Pham, A., Quinones, B. 2022. Comparative genomic and phenotypic analyses of virulence potential in Shiga toxin-producing Escherichia coli O121:H7 and O121:H10. Frontiers in Cellular and Infection Microbiology. 12. Article 1043726. https://doi.org/10.3389/fcimb.2022.1043726.
Kirchner, M., Miller, W.G., Osborne, J., Badgley, B., Niedermeyer, J.A., Kathariou, S. 2023. Campylobacter colonization and diversity in young turkeys in the context of gastrointestinal distress and antimicrobial treatment. Microorganisms. 11(2). Article 252. https://doi.org/10.3390/microorganisms11020252.
Bolinger, H., Miller, W.G., Osborne, J., Niedermeyer, J., Kathariou, S. 2023. Campylobacter jejuni and Campylobacter coli from houseflies in commercial turkey farms are frequently resistant to multiple antimicrobials and exhibit pronounced genotypic diversity. Pathogens. 12(2). Article 230. https://doi.org/10.3390/pathogens12020230.
Meinersmann, R.J., Berrang, M.E., Shariat, N.W., Richards, A.K., Miller, W.G. 2023. Despite shared geography, Campylobacter isolated from surface water are genetically distinct from Campylobacter isolated from chickens. Microbiology Spectrum. 11(2). Article e04147-22. https://doi.org/10.1128/spectrum.04147-22.