"Microbial Prevalence And Antibiotic Resistance Associated With Urban Food Production . . ." by Nirosha Ruwani Amarasekara

Access Type

Open Access Dissertation

Date of Award

January 2024

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Nutrition and Food Science

First Advisor

Yifan Zhang

Abstract

Despite the increasing popularity of urban agriculture and farmers’ markets, there is a lack of scientific data on microbial prevalence and antibiotic resistance associated with this unique food production system. The overall goal of this research is to understand antibiotic resistance and microbial prevalence in soil and fresh produce associated with urban agriculture. Specifically, we aim to investigate ARG in association with resistance genes to metals and other compounds in urban agricultural soil as well as antibiotic resistome and Listeria contamination in fresh produce distributed in urban communities.

In the initial phase of this research, soil samples were collected from an urban community garden in Detroit to explore the co-occurrence of antibiotic resistance genes (ARG) and resistance genes to other compounds, including heavy metals and biocides. Total soil DNA was extracted and subjected to metagenomic analysis on ARG, metal resistance genes (MRG), biocide resistance genes (BRG), multi-compound resistance genes (MCRG), and mobile genetic elements (MGE). The most abundant ARGs included multidrug, vancomycin, bacitracin, fosmidomycin, and macrolide-lincosamide-streptogramin (MLS) resistance genes. Common MRGs consisted of genes with resistance to copper (dnaK, copA, cueA), arsenic (arsM), multi-metal (recG_BAC0356, ruvB, cueA), iron (acn), and chromium (recG). Multidrug resistance genes (mexF, mexW, multidrug transporter gene), MLS resistance gene (macB), and MCRG (ruvB) showed the most correlations with ARGs and resistance genes to other compounds. The data demonstrate extensive co-occurrence of ARGs, MRGs, BRGs, and MCRGs in urban agricultural soil and suggest the great potential of soil contaminants as co-selection agents for antibiotic resistance.

In the second phase, 16S rRNA gene amplicon sequencing and shotgun metagenomic analysis was performed to investigate microbiome and the antibiotic resistome associated with fresh produce obtained from urban communities. A total of 432 fresh produce samples were collected from six farmers' markets and urban gardens in Michigan and West Virginia, resulting in 109 composite samples. The most prevalent microbial phyla identified were Proteobacteria, Actinobacteria, and Firmicutes. Resistance genes were against a broad spectrum of compounds, including antibiotics (aminocoumarins, aminoglycoside, beta-lactams, fluoroquinolones, fosfomycin, lipopeptides, multi-drug resistance, and oxazolidinone), biocides (biguanide, multi-biocide, peroxide, and phenolic compound), metals (arsenic, copper, zinc, cadmium, and multi-metal resistance), and multi-compound. Market variation was observed for microbiome and antibiotic resistome. ARG abundance is not necessarily correlated to bacterial abundance. The findings suggest possible multiple contamination sources for microbiome and ARG.

The third phase of the study focused on Listeria characterization in fresh produce. L. monocytogenes recovered from fresh produce was subjected to serotype analysis by PCR and genotyping by pulsed-field gel electrophoresis (PFGE). In-silico multi-locus sequence typing (MLST) and core-genome multi-locus sequence typing (cgMLST) were conducted for further gene profiling based on whole genome sequencing (WGS). The prevalence of Listeria spp. was 44.0% (48/109) that included 32 samples positive for L. welshimeri, 13 for L. innocua, and three for L. monocytogenes. Twelve L. monocytogenes isolates of serotype 1/2a and 4b were recovered from radish, potato, and romaine lettuce. Four clonal complexes (CC) were identified and included hypervirulent CC1 (ST1) and CC4 (ST219) of lineage I as well as CC7 (ST7) and CC11 (ST451) of lineage II. Clones CC4 and CC7 were present in the same romaine lettuce sample. CC1 carried Listeria pathogenicity island LIPI-1 and LIPI-3 whereas CC4 contained LIPI-1, LIPI-3, and LIPI-4. CC7 and CC11 had LIPI-1 only. Antibiotic resistance genes fosX, lin, mprF, norB and sul were detected in all L. monocytogenes genomes. This is the first report of hypervirulent CC4 in US food supply. Given its previous implication in foodborne outbreaks in Europe and its recovery in the environment and animals, future research should focus on understanding the contribution of food, animal, and environmental exposure to listeriosis.

To summarize, urban agricultural soil and fresh produce can carry extensive ARG and genes resistant to multiple compounds. The common occurrence of ARG classes in soil and fresh produce suggests that soil may be a likely origin of ARG in fresh produce, though farm and market practices may also play a role. Continued research is essential to explore the interplay of environmental contaminants in shaping antibiotic resistance in soil and fresh produce, and the role of food, animal, and environmental exposure in contributing to foodborne illnesses.

Download

Share

COinS