Pervasive Listeria monocytogenes Is Common in the Norwegian Food System and Is Associated with Increased Prevalence of Stress Survival and Resistance Determinants

Abstract

To investigate the diversity, distribution, persistence, and prevalence of stress survival and resistance genes of Listeria monocytogenes clones dominating in food processing environments in Norway, genome sequences from 769 L. monocytogenes isolates from food industry environments, foods, and raw materials (512 of which were sequenced in the present study) were subjected to whole-genome multilocus sequence typing (wgMLST), single-nucleotide polymorphism (SNP), and comparative genomic analyses. The data set comprised isolates from nine meat and six salmon processing facilities in Norway collected over a period of three decades. The most prevalent clonal complex (CC) was CC121, found in 10 factories, followed by CC7, CC8, and CC9, found in 7 factories each. Overall, 72% of the isolates were classified as persistent, showing 20 or fewer wgMLST allelic differences toward an isolate found in the same factory in a different calendar year. Moreover, over half of the isolates (56%) showed this level of genetic similarity toward an isolate collected from a different food processing facility. These were designated as pervasive strains, defined as clusters with the same level of genetic similarity as persistent strains but isolated from different factories. The prevalence of genetic determinants associated with increased survival in food processing environments, including heavy metal and biocide resistance determinants, stress response genes, and inlA truncation mutations, showed a highly significant increase among pervasive isolates but not among persistent isolates. Furthermore, these genes were significantly more prevalent among the isolates from food processing environments compared to in isolates from natural and rural environments (n = 218) and clinical isolates (n = 111) from Norway. IMPORTANCE Listeria monocytogenes can persist in food processing environments for months to decades and spread through the food system by, e.g., contaminated raw materials. Knowledge of the distribution and diversity of L. monocytogenes is important in outbreak investigations and is essential to effectively track and control this pathogen in the food system. The present study presents a comprehensive overview of the prevalence of persistent clones and of the diversity of L. monocytogenes in Norwegian food processing facilities. The results demonstrate extensive spread of highly similar strains throughout the Norwegian food system, in that 56% of the 769 collected isolates from food processing factories belonged to clusters of L. monocytogenes identified in more than one facility. These strains were associated with an overall increase in the prevalence of plasmids and determinants of heavy metal and biocide resistance, as well as other genetic elements associated with stress survival mechanisms and persistence.

Keywords: Listeria monocytogenes; WGS; antibiotic resistance; food processing environment; food safety; inlA; meat industry; meat processing; persistence; pervasive; plasmids; salmon industry; salmon processing; stress resistance; stress survival; whole-genome sequencing.

FIG 1

Minimum-spanning tree based on wgMLST analysis for the 769 L. monocytogenes from food processing environments. The area of each node is proportional to the number of isolates represented, and the number of allelic differences between isolates is indicated on the edges connecting two nodes. The nodes are colored by factory of origin (meat production plants M1 to M9; salmon processing plants S1 to S6), and CCs and STs are indicated (the ST number is the same as the CC unless specified).

FIG 2

Presence of accessory genetic determinants associated with stress survival, resistance, or persistence in L. monocytogenes from food processing environments. (A) Unique combinations of ST and the variable stress response loci in isolates from food processing environments. All 769 isolates are shown individually in Fig. S1. The phylogeny is a midpoint-rooted NJ tree based on wgMLST analysis and shows one arbitrarily selected genome from each of the groups of genomes containing the same unique ST and gene combination. The number of genomes harboring the same unique combination is indicated in the right column. (B) Prevalence of genetic determinants in isolates from different sources. Cadmium resistance: cadA1C1, cadA2C2, cadA4A4, or cadA5C5. Arsenic resistance: arsA1D1R1D2R2A2B1B2 on LGI2 or arsCBADR on Tn554-like transposon. QAC resistance: bcrABC or qacH. Biofilm-associated genes: bapL and/or inlL. Asterisks represent significant differences (Pearson’s chi-squared test; *, P < 0.03; **, P ≤ 0.001; see also Table S5).

FIG 3

Different CCs detected in various proportions of factories. The data are presented as a stacked bar plot showing the percentage of examined meat and salmon processing plants in which each CC was detected. Fifteen factories were included.

FIG 4

Prevalence of subgroups of accessory genetic determinants associated with stress survival, resistance, and persistence in L. monocytogenes classified as persistent (A) and/or pervasive (B). Cadmium: cadmium resistance loci cadA1C1, cadA2C2, cadA4C4, or cadA5C5. Arsenic: arsenic resistance operons arsA1D1R1D2R2A2B1B2 on LGI2 or arsCBADR on the Tn554-like transposon. QAC: QAC resistance loci qacH or bcrABC. Biofilm: biofilm-associated genes bapL and/or inlL. SSI: stress survival islets SSI-1 or SSI-2. inlA m.: PMSC mutation in the inlA gene. Asterisks represent significant differences (Pearson’s chi-squared test; *, P < 0.05; **, P < 0.001; see also Table S6).

FIG 5

Pervasive strains, present in more than one processing plant. (A) NJ phylogenetic tree based on wgMLST analysis, showing CC (inner ring), factory of origin (middle ring), and isolates for which genetically similar isolates (≤20 wgMLST allelic differences) were found in at least one other factory (outer ring). Red branches indicate lineage I; black branches indicate lineage II. (B) Genetic associations between isolates from different factories illustrated by a chord diagram. The outer sectors represent the factories for which genetically similar isolates (≤20 wgMLST allelic differences) were found in at least one other factory, and the links between the factories represent the pairs of genetically similar isolates, colored by CC group. The thickness of each arc represents the sum of the similarity scores for each pair of isolates found in the two factories, weighted by their similarity. (C) Minimum-spanning tree showing the relationship between the 290 CC9 isolates. Nodes are colored by factory of origin (same colors as in panel A; the white-colored isolate is from a domestic kitchen). The area of each node is proportional to the number of isolates represented, and the number of allelic differences between isolates is indicated on the branches connecting two nodes. Branch lengths are square root scaled.

FIG 6

Presence of accessory genetic determinants associated with stress survival, resistance, or persistence in L. monocytogenes from different sources. (A and B) Norwegian clinical isolates from 2010 to 2015 (56) (A) and isolates from natural (rural/urban/farm/slug) environments in Norway (B) (2). For each ST, one arbitrarily selected genome from each of the groups of genomes containing the same unique combination of stress response loci is shown. (C) Prevalence of genetic determinants in isolates from different sources. See the legend to Fig. 2 for details on the categories. Statistical analysis results for differences in categories using Pearson’s chi-squared test are presented in Table S7 in the supplemental material.