The Trade-Off Between Sanitizer Resistance and Virulence Genes: Genomic Insights into E. coli Adaptation

Abstract

Background: Escherichia coli is one of the most studied bacteria worldwide due to its genetic plasticity. Recently, in addition to characterizing its pathogenic potential, research has focused on understanding its resistance profile to inhibitory agents, whether these be antibiotics or sanitizers.

Objectives: The present study aimed to investigate six of the main serogroups of foodborne infection (O26, O45, O103, O111, O121, and O157) and to understand the dynamics of heterogeneity in resistance to sanitizers derived from quaternary ammonium compounds (QACs) and peracetic acid (PAA) using whole-genome sequencing (WGS).

Methods: Twenty-four E. coli strains with varied resistance profiles to QACs and PAA were analyzed by WGS using NovaSeq6000 (150 bp Paired End reads). Bioinformatic analyses included genome assembly (Shovill), annotation via Prokka, antimicrobial resistance gene identification using Abricate, and core-genome analysis using Roary. A multifactorial multiple correspondence analysis (MCA) was conducted to explore gene-sanitizer relationships. In addition, a large-scale analysis utilizing the NCBI Pathogen Detection database involved a 2 × 2 chi-square test to examine associations between the presence of qac and stx genes.

Results: The isolates exhibited varying antimicrobial resistance profiles, with O45 and O157 being the most resistant serogroups. In addition, the qac gene was identified in only one strain (S22), while four other strains carried the stx gene. Through multifactorial multiple correspondence analysis, the results obtained indicated that strains harboring genes encoding Shiga toxin (stx) presented profiles that were more likely to be sensitive to QACs. To further confirm these results, we analyzed 393,216 E. coli genomes from the NCBI Pathogen Detection database. Our results revealed a significant association (p < 0.001) between the presence of qac genes and the absence of stx1, stx2, or both toxin genes.

Conclusion: Our findings highlight the complexity of bacterial resistance mechanisms and suggest that non-pathogenic strains may exhibit greater tolerance to QAC sanitizer than those carrying pathogenicity genes, particularly Shiga toxin genes.

Keywords: bacterial communities; bacteriophage; evolutionary mechanisms; foodborne bacteria; sanitizer resistance.

Figure 1

Antimicrobial resistance genes in E. coli for six serogroups. Legend: The Venn diagram illustrates the distribution of antimicrobial resistance genes among E. coli serogroups O45, O26, O157, O103, O111, and O121. Each circle represents a specific serogroup, and the overlapping areas indicate genes shared between serogroups. The central region (O103, O111, O121) highlights the presence of the mdf(a) gene, while the outer regions show unique resistance genes, such as tet(b) in O157.

Figure 2

Phylogenetic tree and antimicrobial resistance gene profiles of E. coli strains with sanitizer resistance indicators. Legend: The phylogenetic tree illustrates the genetic relationships among E. coli isolates from cattle production environments, with branches representing different sequencing strains. The presence or absence of antimicrobial resistance genes is represented by black (present) and white (absent) blocks. Sanitizer resistance levels for QACs and PAA are depicted using a heatmap, where increasing red intensity corresponds to higher resistance. Also, the number inside each block indicates the MIC range for each compound for each strain, with values of 1 (6.25 ppm) 2 (12.50 ppm), and 3 (25 ppm). Finally, D60 represents the bacterial load inactivated after exposure to 60 °C for 1 min and was previously tested in a related study [23]. A lower intensity of color for D60 indicates greater resistance to heat. The numbers inside the blocks represent the logarithmic inactivation load after exposure to heat. The phylogenetic tree was edited using the ITOL website: https://itol.embl.de/ (accessed on 5 February 2025). The Newick tree from core-genome alignment was uploaded, and the metadata were included in binary format (AMR presence/absence) and as a heatmap (sanitizer or D60 value).

Figure 3

Multiple correspondence analysis (MCA) of E. coli sanitizer sensitivity in relation to gene presence. Legend: MCA plot visualizing the relationship between categorical variables (genes and phenotypes) associated with E. coli isolates and the presence or absence of the stx1 gene (indicated by red circles for no_stx and blue circles for stx). The two dimensions (Dim1 and Dim2) explain 21% and 17.8% of the variance, respectively. The ellipses show clustering based on the presence (blue) or absence (red) of the stx gene, highlighting associations between various antimicrobial resistance genes (e.g., blaTEM, sul, tet), sanitizer resistance (PAA and QAC), and other phenotypic traits such as biofilm formation and heat resistance. Key variables such as the presence of cbtA and biofilm formation are present within the stx-negative group, while the stx-positive group shows stronger associations with QAC sensitivity, colrnaI, and cold adaptation.

Figure 4

Multiple correspondence analysis (MCA) of E. coli resistance and gene presence in relation to sanitizer sensitivity. Legend: The Variable Factor Map demonstrates the grouping (associations) of the analyzed variables associated with E. coli isolates and the presence or absence of the stx1 gene. The two dimensions (Dim1 and Dim2) explain 21% and 17.8% of the variance, respectively. The closer together the points are grouped, the stronger the relationship between the variables. For example, stx1 and QAC sensitivity, and emrE and colRNAl.

Figure 5

Phage and gene insertion sites in E. coli strains: stx1 and qac gene localization. Legend: These genomic maps illustrate the insertion sites of phages and key genes (stx1 and qacE-delta1) in five E. coli strains (S9, S11, S22, S23, and S24). Each panel represents the genome of a strain, highlighting phage insertion regions (labeled in green, red and purple) and the presence of the stx1 gene (indicated in yellow) and qacE-delta1 gene (indicated in black). Regions 1 through 15 mark specific phage insertion sites. The presence of the stx1 gene correlates with multiple insertion regions, particularly in strains S9, S11, S23, and S24, while the qacE-delta1 gene is uniquely identified in strain S22.

Figure 6

Linear DNA sequences found to possess stx or qac genes in the present study and the associated neighbor genes. Legend: The sequences analyzed in the present study were S9, S11, S22, S23, S24, and S24.2. The red rectangles represent Shiga toxin genes (stx1 or stx2), while the yellow rectangles represent genes resistant to quaternary ammonium compounds (qacEdelta1) or the integron-integrase gene (intI1). This image was created using genoPlotR (version 0.8.11) in RStudio.