(188da) Data-Driven Analysis of Antimicrobial Resistance of Foodborne Pathogens in Six States of USA

Foodborne pathogens are becoming increasingly problematic to public health, as they cause a significant increase in disease and mortality rates. It is reported that these pathogens cause thousands of infections and hundreds of deaths in USA every year [1]. These pathogens, which were able to be eliminated by traditional antimicrobials before, obtain drug-resistant genes and then become resistant to existing antimicrobials. Specifically, these drug-resistant genes encode molecules that are responsible for: 1) the degradation or modification of antimicrobials, 2) antimicrobial receptor modification, and 3) efflux pumps of antimicrobials [2]. One single gene may not be able to trigger antimicrobial resistance. Therefore, it is necessary to investigate: (1) the set of genes that are most related to drug resistance; (2) how drug resistant genes are carried and transferred in the food system of USA. Fortunately, antimicrobial resistance data in USA are actively monitored and collected in the following database: National Database of Antibiotic Resistance Organisms (NDARO), NCBI Pathogen Detection Isolates Browser (NPDIB), National Antibiotic Resistance Monitoring System (NARMS) [3]. In particular, the NPDIB database shows antimicrobial-resistant genes sampled from foodborne pathogens (e.g., Campylobacter, Escherichia, Klebsiella, Legionella, Listeria, Providencia, Salmonella, Pseudomonas, Acinetobacter, Cronobacter, Elizabethkingia, Citrobacter, Enterobacter) detected in meats (e.g., chicken, beef, pork, and turkey) in various states of US. The antimicrobials resisted by foodborne pathogens include, but not limited to, streptomycin, tetracycline, ampicillin, cefoxitin, chloramphenicol, gentamicin, sulfisoxazole, kanamycin, and ciprofloxacin.

While antimicrobial-resistant genes have been published in the aforementioned database, little work has been done to discover the meaningful information hidden in the data. For example, little work has been done to identify the antimicrobial gene set in different states responsible for antimicrobial resistance and to study the avenues that transfer these genes in the food system. In this work, we proposed the first systematic approach to analyze the antimicrobial-gene data from NPDIP database for six states that are geographically either close (e.g., Pennsylvania, Maryland, and New-York states) or far (e.g., New-Mexico, Minnesota, and California). Principal component analysis (PCA) [4] was used to visualize the high-dimensional data in a two-dimensional space. On the basis of data projected on the PCA space, the hierarchical clustering approach [5] was implemented to study the similarities of the following objectives in the six states: antimicrobials that are resisted by foodborne pathogens, the genes related to antimicrobial-resistance, the foodborne pathogens carrying antimicrobial-resistant genes, and the meats that carry antimicrobial-resistant genes. Preliminary results show that: 1) all six states have similar resistant cases for antimicrobials ampicillin, streptomycin, and gentamicin; 2) genes aadA, aph(3”), aph(3”)-Ib, aph(6)-I, aph(6)-Id, bla, blaCMY, tet, tet(A), sul2 are involved in most antimicrobial resistant cases in all six states; 3) Salmonella species are the major pathogens carrying those antimicrobial-resistant genes; 4) chicken and turkey are the two major meats carrying foodborne pathogens with antimicrobial resistant genes. In addition, the PCA and hierarchical clustering analysis indicate that adjacent states (e.g., PA, NY, and MD) are not necessary to have similar antimicrobial-resistant genes when compared to those faraway states (e.g., CA and MN). Meats may be the carriers of antimicrobial-resistant genes through the whole nation.

Reference

1. Crim, S.M., et al., Preliminary Incidence and Trends of Infection with Pathogens Transmitted Commonly Through Food - Foodborne Diseases Active Surveillance Network, 10 US Sites, 2006-2014. Mmwr-Morbidity and Mortality Weekly Report, 2015. 64(18): p. 495-499.
2. Walsh, C., Molecular mechanisms that confer antibacterial drug resistance. Nature, 2000. 406(6797): p. 775-781.
3. Agarwala, R., et al., Database resources of the National Center for Biotechnology Information. Nucleic Acids Research, 2018. 46(D1): p. D8-D13.
4. Wold, S., K. Esbensen, and P. Geladi, Principal Component Analysis. Chemometrics and Intelligent Laboratory Systems, 1987. 2(1-3): p. 37-52.
5. Chumwatana, T., Using Clustering Techniques for Non-segmented Language Document Management: A Comparison of K-mean and Self Organizing Map Techniques. Proceeding of Knowledge Management International Conference (Kmice) 2014, Vols 1 and 2, 2014: p. 600-605.