Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Genetic adaptation to amoxicillin in Escherichia coli: The limited role of dinB and katE

  • Lisa Teichmann,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft

    Affiliation Swammerdam Institute of Life Sciences, Molecular Biology and Microbial Food Safety, University of Amsterdam, Amsterdam, The Netherlands

    ⨯
  • Marcus Wenne,

    Roles Data curation, Formal analysis, Software, Writing – review & editing

    Affiliations Department of Life Sciences, Division of Systems and Synthetic Biology, SciLifeLab, Chalmers University of Technology, Gothenburg, Sweden, Centre for Antibiotic Resistance Research (CARe) in Gothenburg, Gothenburg, Sweden

    ⨯
  • Sam Luitwieler,

    Roles Investigation, Methodology, Writing – review & editing

    Affiliation Swammerdam Institute of Life Sciences, Molecular Biology and Microbial Food Safety, University of Amsterdam, Amsterdam, The Netherlands

    ⨯
  • Gaurav Dugar,

    Roles Data curation, Formal analysis, Writing – review & editing

    Affiliation Swammerdam Institute of Life Sciences, Molecular Biology and Microbial Food Safety, University of Amsterdam, Amsterdam, The Netherlands

    ⨯
  • Johan Bengtsson-Palme,

    Roles Conceptualization, Formal analysis, Writing – review & editing

    Affiliations Department of Life Sciences, Division of Systems and Synthetic Biology, SciLifeLab, Chalmers University of Technology, Gothenburg, Sweden, Centre for Antibiotic Resistance Research (CARe) in Gothenburg, Gothenburg, Sweden, Department of Infectious Diseases, Institute of Biomedicine, University of Gothenburg, Gothenburg, Sweden

    ⨯
  • Benno ter Kuile

    Roles Conceptualization, Funding acquisition, Investigation, Methodology, Supervision, Writing – review & editing

    b.h.terkuile@uva.nl

    Affiliation Swammerdam Institute of Life Sciences, Molecular Biology and Microbial Food Safety, University of Amsterdam, Amsterdam, The Netherlands

    ⨯

Abstract

Bacteria can quickly adapt to sub-lethal concentrations of antibiotics. Several stress and DNA repair genes contribute to this adaptation process. However, the pathways leading to adaptation by acquisition of de novo mutations remain poorly understood. This study explored the roles of DNA polymerase IV (dinB) and catalase HP2 (katE) in E. coli’s adaptation to amoxicillin. These genes are thought to play essential roles in beta-lactam resistance—dinB in increasing mutation rates and katE in managing oxidative stress. By comparing the adaptation rates, transcriptomic profiles, and genetic changes of wild-type and knockout strains, we aimed to clarify the contributions of these genes to beta-lactam resistance. While all strains exhibited similar adaptation rates and mutations in the frdD gene and ampC operon, several unique mutations were acquired in the ΔkatE and ΔdinB strains. Overall, this study distinguishes the contributions of general stress-related genes on the one hand, and dinB, and katE on the other hand, in development of beta-lactam resistance.

Citation: Teichmann L, Wenne M, Luitwieler S, Dugar G, Bengtsson-Palme J, ter Kuile B (2025) Genetic adaptation to amoxicillin in Escherichia coli: The limited role of dinB and katE. PLoS ONE 20(2): e0312223. https://doi.org/10.1371/journal.pone.0312223

Editor: Bashir Sajo Mienda, Federal University Dutse, NIGERIA

Received: October 3, 2024; Accepted: December 27, 2024; Published: February 19, 2025

Copyright: © 2025 Teichmann et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data can be found in the following GitHub link: https://github.com/mnyt-aqw/Teichmann_2024_Amoxicillin.

Funding: This study was financed by The Netherlands Food and Consumer Product Safety Authority.

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

Since bacterial resistance mechanisms were first described, considerable efforts have been devoted to revealing the underlying molecular machinery involved in developing antimicrobial resistance. Understanding the cellular pathways involved is essential for addressing the growing threat looming over modern medicine–an era where previously effective antibiotics are increasingly unable to treat resistant infections (https://www.who.int/publications/i/item/no-time-to-wait-securing-the-future-from-drug-resistant-infections). Over billions of years, bacteria have evolved distinct and sophisticated defence mechanisms that not only enable survival but often allow them to thrive under adverse conditions. Many of these complex defence systems are versatile, protecting against multiple stressors and effectively pre-adapting bacteria to various forms of harm [1–3]. Environmental stressors, such as starvation, have been shown to activate mutagenic pathways, which, in turn, contribute to the acquisition of antibiotic resistance [4–6]. This connection underscores the complex interplay between bacterial survival strategies and the emergence of resistance, highlighting the importance of an integrative approach when investigating these adaptive mechanisms.

Exposure to beta-lactam antibiotics such as amoxicillin is known to trigger a range of stress responses in bacterial cells, including the SOS and oxidative stress response [7,8]. Amoxicillin, with its broad-spectrum activity and relative safety, holds a central position in human healthcare. It is crucial for treating a wide array of bacterial infections, from common respiratory and urinary tract infections to pneumonia and skin infections [9]. The extensive use of amoxicillin underscores the importance of understanding how bacterial adaptation to this antibiotic occurs.

DNA polymerase IV, encoded by the dinB gene, plays a vital role in the SOS response [10]. The dinB gene facilitates error-prone DNA synthesis, introducing mutations that, while potentially harmful, can also allow bacteria to adapt and survive under challenging conditions [11–15]. The role of dinB in beta-lactam resistance is complex. Although beta-lactam antibiotics do not directly damage DNA, they have been shown to induce a two-component signal transduction system that activates the SOS response [7]. Moreover, dinB has been implicated in ampicillin-induced mutagenesis in Escherichia coli and other bacterial species and appears to be responsible for many of the mutations that arise following beta-lactam exposure [16]. Interestingly, these mutations occur independently of the typical SOS regulon and RpoS sigma factor activation, suggesting that other, yet-to-be-identified mechanisms might be involved [17]. One possible link could be the role of intracellular reactive oxygen species (ROS), which have been shown to increase following sub-inhibitory exposure to various bactericidal antibiotics, including aminoglycosides and beta-lactams like ampicillin [13,18,19].

This increase in ROS is hypothesized to result from increased metabolic rates, causing perturbations of the TCA cycle [20,21], potentially leading to an increase in mutations. On the one hand, these mutations may contribute to antibiotic resistance. On the other hand, they could eventually function as a secondary killing mechanism in addition to the antimicrobial properties of the antibiotics [8,22]. However, this ‘ROS theory’ remains a subject of debate and the ROS-induced lethal effect of bactericidal antibiotics seems to depend on antibiotic concentrations [23,24]. While it is well-established that bactericidal antibiotics elevate intracellular ROS levels, the exact role of ROS in antibiotic-induced mutagenesis and adaptive mutations is still not fully understood.

In this context, the enzyme KatE (catalase HP2) plays a crucial role in E. coli’s defence against oxidative stress. Catalase enzymes, such as KatE, are essential for detoxifying ROS by breaking down hydrogen peroxide (H2O2) into water and molecular oxygen [25]. This detoxification process significantly enhances bacterial fitness and resilience, particularly under conditions where antibiotic exposure induces ROS production [26]. KatE is induced by the stringent response in E. coli, which is activated under environmental stress conditions such as nutrient starvation [26]. The stringent response leads to the downregulation of cell growth-related transcription and the upregulation of survival mechanisms [27]. Interestingly, E. coli in a stringent state shows increased tolerance to many antibiotic classes [28,29]. Elevated levels of KatE have been proposed as a contributing factor to this increased tolerance, as a correlation between catalase activity and antibiotic tolerance has been observed during the stringent response [26]. However, the precise cause of this association remains unclear. A likely explanation could be that the catalase reduces the intracellular ROS levels induced by the antibiotic enough to enable survival [26].

In this study, we investigated the genetic and transcriptomic alterations in E.coli MG1655, a well-established model organism, and two corresponding knockout mutants (ΔdinB, ΔkatE) when exposed to sub-lethal concentrations of amoxicillin. We initially hypothesized that the absence of dinB would result in a reduced mutation rate, hindering E. coli’s ability to adapt to these conditions. Meanwhile, the deletion of katE was expected to lead to difficulties in adaptation due to increased oxidative stress, although it might also cause an elevated mutation rate which could elevate adaptation. This would impair E. coli’s ability to adapt to these conditions, given the previously reported lower mutation rates in dinB knockouts and the increased oxidative stress observed in katE-deficient strains [30,31]. Contrary to our expectations, the knockout strains adapted at a similar rate to the wild-type strain. This suggests that dinB and katE may not be as critical for beta-lactam adaptation as generally assumed.

Consistent mutations in the ampC operon were found across all three strains, highlighting its importance in amoxicillin resistance acquisition. Additionally, other genes were frequently mutated, with some identical mutations observed across different strains and replicates, including a notable mutation in rpoD, suggesting a significant role of this sigma factor under antibiotic stress.

Furthermore, the role of frdD mutations and regulation seemed to be more multifaceted than previously assumed, indicating potential involvement in resistance mechanisms beyond ampC regulation. The consistent changes observed in the transcriptomic regulation of the toxin/antitoxin system prlF/yhaV further suggests a role of this system in the acquisition of resistance.

2. Material and methods

Strains, growth conditions, and antimicrobial agents

The Escherichia coli strain MG1655 was utilized as the wild-type strain. Single gene knockout mutants JW0221 (ΔdinB749::kan) and JW1721 (ΔkatE731::kan) were obtained from the KEIO collection, supplied by Horizon Discovery Ltd. These knockout mutants contained kanamycin-resistant cassettes flanked by FLP recognition target (FRT) sites, which were removed using the pCP20 method prior to the experiments [32,33].

Bacterial cultures were grown in lysogeny broth (LB) containing 10 g/L NaCl, either in liquid or solid form. All strains were initially cultured to an OD600 of 0.1 and incubated at 37°C with shaking at 200 rpm overnight. For longer weekend incubations, the starting OD600 was reduced to 0.01, and the incubation temperature was adjusted to 30°C. Amoxicillin was supplied by Merck KGaA. A stock solution (10 mM) was prepared which was stored at 4°C and used within three days of preparation and filter sterilization.

Minimum inhibitory concentrations (MIC)

Minimum inhibitory concentrations (MICs) were measured twice a week in duplicate for each strain using the broth microdilution method [34]. Readings were taken every 10 minutes, with 5 minutes shaking intervals between measurements.

Evolution experiment

Evolution experiments were conducted as described previously [35]. At the start of the experiment, the MIC of each strain was determined. Amoxicillin exposure began at a concentration of 2 μg/mL, equivalent to ¼ of the MIC for all strains. Following overnight incubation, the OD600 was measured. If the OD600 was above 65% of the OD600 of the previous culture, the amoxicillin concentration was doubled in fresh medium. If the OD600 was below this threshold, the culture was transferred to fresh medium with the same amoxicillin concentration as the previous day. In parallel, each bacterial strain was cultured without antibiotic exposure as a biological control. Three technical replicates were performed. The experiment was terminated when the culture reached and tolerated an amoxicillin concentration of 1024 μg/mL for multiple days. If the culture could not adapt to this concentration, the experiment was stopped after 10 transfers without successful increase to a higher antibiotic concentration.

DNA isolation and whole genome sequencing

Genomic DNA was isolated from each culture at both the start and end of the evolution experiment using the PureLink Genomic DNA Mini Kit (Thermo Fisher Scientific), with some modifications to the manufacturer’s protocol. Cultures were pelleted by centrifugation at 12,000 × g for 1 minute, and the resulting pellet was resuspended in 300 μL of TE buffer. Then 40 μL of 10% SDS, 3 μL of 0.5 M EDTA, and 20 μL of proteinase K were added. Cell lysis was achieved by incubating the samples at 65°C for 5 minutes in a heating block. Following lysis, 20 μL of RNase A was added, and the samples were incubated at room temperature for 3 minutes. After RNase addition, the protocol was followed as recommended by the supplier.

The quality and quantity of the isolated genomic DNA were assessed using a NanoDrop spectrophotometer (Thermo Fisher Scientific) to measure absorbance at 260/280 nm. The integrity of the DNA was further confirmed by running the samples on a 1% agarose gel. The purified genomic DNA was then used for library preparation according to the manufacturer’s protocol for the NEBNext Ultra II FS DNA Library Prep Kit for Illumina (New England BioLabs). The libraries were subsequently sequenced using the Illumina sequencing platform.

RNA isolation and sequencing

Frozen cultures were thawed on ice, followed by overnight inoculation in LB with the appropriate antibiotic concentration (S2 Table). Total RNA was then extracted using PCI for RNA isolation. Ethanol precipitation was used to purify the RNA, and any residual genomic DNA was removed with DNase I (New England BioLabs) digestion according to the manufacturer’s instructions. RNA integrity was confirmed by visualizing the integrity of 16S and 23S ribosomal RNA bands on an 1% agarose gel. The NEBNext rRNA Depletion Kit (New England BioLabs) was employed to deplete ribosomal RNA, as per the manufacturer’s protocol. The rRNA-depleted RNA was then prepared for sequencing using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (New England BioLabs) and sequenced on an Illumina NextSeq 550 platform. Agencourt Ampure XP (Beckman Coulter) magnetic beads were used during library preparation.

Transcriptome analysis

The entire transcriptomics workflow was designed as a Nextflow pipeline v21.10.6 [36]. Single ended RNA-Seq data was quality controlled and trimmed using TrimGalore! (https://github.com/FelixKrueger/TrimGalore?tab=readme-ov-file) v0.6.7 with settings—phred33 -e 0.1—quality 28. Afterwards, MultiQC v1.13 [37] was used to create a combined quality report for all samples. Next, the reads that passed quality control for all samples were mapped against the MG1655 reference genome (RefSeq: GCF_000005845.2) using Bowtie2 v2.3.5.1 and the resulting bam files were sorted using Samtools v1.3.1. Finally, the sorted bam files were converted to transcript counts using featureCounts v2.0.1 with default settings in single-end mode [38]. The differential gene expression analysis, GO term and KEGG pathway analysis was performed using edgeR v4.0.16. A gene was considered differentially expressed if the FDR corrected p-value was equal or below 0.05 [39]. A different cutoff for the GO and KEGG pathway analysis was applied following recommendations by the authors of the goana and kegga packages, instead using a p-value of 0.001.

Genome analysis

Like the transcriptomic workflow, the genome analysis was also performed using Nextflow. Pair-end genomic reads were quality controlled, and adapters were removed the same way as for the RNA-Seq data. To identify mutations, as well as their frequency in the bacterial population, Breseq v0.37.1 [40] was used with the—polymorphism-prediction flag and supplied with either the MG1655 reference genome: NC_000913.3 or the corresponding reference genome for the knock out strains: NZ_CP009273.1. To analyze gene copy number the paired trimmed reads were mapped against the MG1655 reference genome using Bowtie2. The Samtools view function (v1.3.1) was used to convert the sam file to a binary bam file [41]. Potential PCR duplicates were removed using the Samtools rmdup function with the -S flag (pair-end mode). Finally, the positions of the mapped reads were extracted using the Samtools view function, in combination with a Perl script (v5.32.1). The CNOGpro R library (v1.1) was used to convert the mapping data to copy numbers using the CNOGpro function with windowlength = 100, followed by the functions normalizeGC and runBootstrap with settings ‘replicates = 1000, quantiles = c(0.025, 0.975)’ [42], generating bootstrapped copy numbers for each gene.

Data analysis

The data analysis was performed using Python v 3.11.8 or R v4.3.2 in combination with Jupyter Lab v 4.0.11 (https://github.com/jupyterlab/jupyterlab) and Quarto v1.4.553 (https://github.com/quarto-dev/quarto-cli). Finally, ggplot2 (https://ggplot2.tidyverse.org) and Seaborn were used for data visualization [43].

Protein structure prediction

Protein structure changes resulting from the mutated genes were predicted using DDMut [44]. Initially, the protein sequence in PDB format was downloaded from UniProt (1SIG). Next, the identified mutation from the whole genome sequencing data was incorporated into this sequence. The DDMut online tool was then used to predict the new protein structure and stability (https://biosig.lab.uq.edu.au/ddmut/).

3. Results

Deletion of katE or dinB does not alter antibiotic adaptation rates

We investigated the adaptation rate of the wild-type compared to the dinB and katE knockout strains and found no significant differences (Fig 1). Despite their limitations in stress responses considered crucial for beta-lactam resistance, their adaptation dynamics closely reflected that of the wild-type E. coli. Indeed, the biological and technical variations between replicates were comparable to the differences observed between the strains, suggesting that knocking out dinB and katE did not confer any notable disadvantage in adaptation to amoxicillin. All strains exhibited a uniform adaptation rate until the clinical resistance threshold of 8 μg/mL (https://mic.eucast.org) and above.

thumbnail
Download:
Fig 1. Adaptation to amoxicillin.

Comparative analysis of the adaptation to amoxicillin across multiple transfers in three bacterial strains: MG1655 (black/grey), ΔkatE (purple), and ΔdinB (green) with three biological replicates for each strain. On the left: The y-axis represents the antibiotic concentration in micrograms per milliliter on a logarithmic scale, while the x-axis represents the number of transfers. The Venn diagram illustrates the overlapping mutations found in the three bacterial strains. Only mutations with >60% frequency and occurrence in at least 2 replicates are displayed. All strains shared a mutation in frdD. MG1655 and ΔdinB had a mutation between ampC/frdD in common. More detailed information can be found in Table 1.

https://doi.org/10.1371/journal.pone.0312223.g001

Mutations in the ampC Operon and rpoD are associated with amoxicillin resistance

Whole genome sequencing of the adapted cultures revealed that the strains only gained a small set of partially overlapping and even identical mutations (see Fig 1 and Table 1). We focused on mutations present in at least 60% of the population and only considered mutations that appeared in at least two out of the three replicates for each strain. This threshold was chosen based on the frequency distribution observed within the samples.

thumbnail
Download:
Table 1. Mutations acquired during amoxicillin exposure.

https://doi.org/10.1371/journal.pone.0312223.t001

All three strains acquired a mutation in frdD, a gene within the frd operon which contains the promoter of the ampC beta-lactamase gene [45]. Mutations in this gene have been found to influence transcription levels of ampC [46–48]. Furthermore, both the wild-type strain and dinB knockout gained a mutation in an intergenic region near ampC. We also identified a mutation in rpoD, the gene encoding sigma factor 70, in two of the katE replicates, as well as in two of the dinB and one of the MG1655 replicates. Despite this mutation not meeting our primary criteria of 60% frequency and occurrence in at least two replicates, we found it worth investigating due to its remarkable consistency and recurrence across different strains, all showing an identical mutation. This suggests an important role for rpoD in the adaptation process.

A transposon region containing ampC is amplified by amoxicillin exposure

Gene copy analysis revealed that all three strains amplified the same collection of genes after adaptation to amoxicillin (Fig 2), including the frd cluster, ampC, blc, gdx, ecnA, ecnB, efp, epmA, epmB, and a gene coding for the uncharacterized protein YjeJ, some of them by more than 10 additional copies (Fig 2). Parts of this region have been previously described as a potential transposon region and are transferable between E. coli cells [49]. We additionally observed unique gene amplifications in each strain. In the wild-type strain MG1655, unique amplifications were identified in mscM and yjeO. The katE knockout strain showed unique amplifications in frsA, and nhaR. The strain also amplified various transposable elements, namely insA6, insB1, insB5, insB6, and insB9. The dinB knockout strain exhibited unique amplifications in aspA, cutA, dcuA, dsbD, fxsA, groL, groS, insG, psd, yjdC, yjeH, and yjeI.

thumbnail
Download:
Fig 2. Genes with a bootstrapped abundance above 1.5.

The graph illustrates the number of genes that are amplified ≥ 1.5 times. The rhomboid shape represents the bootstrapped values. The transparent circles represent the corresponding lower and upper bounds.

https://doi.org/10.1371/journal.pone.0312223.g002

Transcriptomic Changes in Knockout Mutants highlight potential compensatory Mechanisms

The deletion of the dinB and katE genes resulted in distinct transcriptomic changes compared to the wild type (Fig 3). In the ΔdinB mutant, genes associated with error-free translesion synthesis were significantly downregulated, as well as genes involved in error-prone translesion synthesis. The ΔkatE mutant exhibited a downregulation of hydrogen peroxide catabolic and metabolic processes, but this effect was less pronounced than the impact observed on pathways related to the ΔdinB knockout. This suggests that while katE is involved in managing oxidative stress, other pathways can partially compensate for its absence, whereas the loss of dinB appears to be less easily compensated.

thumbnail
Download:
Fig 3. Transcriptomic changes in knockout mutants—Downregulation.

The figure illustrates the proportion of significant transcriptomic changes across the knockout mutants. The size of the data points relates to the proportion of significant changes, whereas the colour code indicates p-values. NT stands for non-treated, while T means treated with ¼ MIC during overnight incubation and before RNA isolation.

https://doi.org/10.1371/journal.pone.0312223.g003

Influence of adaptation and mutations on the transcriptome

All three strains—wild type (MG1655), ΔdinB, and ΔkatE—demonstrated a significant upregulation of ampC at the end of the evolution experiment (Fig 4). In the wild-type MG1655, the most significant downregulation in the transcriptome could be found in waaB (LPS synthesis, involved in host invasion [50]), asnA (asparagine synthetase), and gstB (oxidative stress), alongside the significant upregulation of mdtK (multidrug efflux pump) and yhaV (toxin of the YhaV-PrlF toxin-antitoxin system).

thumbnail
Download:
Fig 4. Top three up-and down-regulated genes after AMO adaptation.

Blue dots illustrate significant downregulation, red dots significant upregulation.

https://doi.org/10.1371/journal.pone.0312223.g004

The ΔdinB strain showed significant downregulation of alaE (alanine exporter), tatE (transport of folded proteins, virulence factor [51]), and hdeA (acid stress response), with upregulation of panB (pantothenate biosynthesis pathway, involved in resistance in Edwardsiella tarda [52]) and efp (elongation factor, peptide bond synthesis). The ΔkatE strain displayed downregulation of asnA, lsrK (quorum sensing), and rppH (mRNA degradation), and upregulation of efp and ecnA (antidote, programmed cell death).

Across all strains, there was a general trend of more upregulated pathways than downregulated during amoxicillin adaptation. The ΔdinB strain exhibited the most extensive changes in pathway regulation, indicating a significant transcriptomic shift in response to both, the gene knockout and the amoxicillin exposure.

Evolved Knockout strains both downregulated prlF and yhaV, but upregulated distinct sets of genes

In the ΔdinB strain, the transcriptomic data revealed a significant downregulation of prlF and yhaV compared to the evolved wild-type. PrlF and YhaV are known to be a toxin-antitoxin system in E. coli [53]. The ΔdinB strain furthermore exhibited the most significant upregulation of fxsA and yjeI. The fxsA gene is implicated in stress response pathways [54], while yjeI is a Ser/Thr kinase [55].

The ΔkatE strain demonstrated a different set of genes that were the most significantly upregulated, including ecnA and efp. The ecnA gene is part of the entericidin operon, involved in bacterial programmed cell death and stress responses [56]. The efp gene encodes an elongation factor and mutations in the gene have been associated with suppressed lethality of a rep/uvrD double mutant [57].

Overall, comparing the evolved knockout strains to the evolved wild-type disclosed that both ΔdinB and ΔkatE mutants adapted by downregulating the common stress response regulators prlF and yhaV. Despite this shared downregulation, each strain utilized distinct upregulated pathways. The ΔdinB strain strongly focused on upregulating genes related to cell envelope stability and stress tolerance, while the ΔkatE strain mostly adapted expression of genes associated with stress response and iron metabolism.

RpoD mutation impacts protein structure

To better understand the potential impact of the mutation in rpoD on the protein structure, we compared the predicted structures of the mutated protein with the wild-type protein. The expected stability change of the protein after the mutation (D445V) was 0.06 kcal/mol (stabilizing). Based on the model, the mutation D445V causes the protein to lose numerous bonds to the neighbouring amino acids (Fig 5). Notably, the D445V mutation occurs within region 2 of the sigma 70 factor, a highly conserved region critical for both DNA binding (specifically, -10 promoter recognition) and RNA polymerase (RNAP) binding. The mutation could potentially affect both RNAP binding and promoter recognition, which are essential for the transcription of numerous genes.

thumbnail
Download:
Fig 5. RpoD protein prediction.

Wild-Type D445 left, Mutant V445 right. Interactions are illustrated with the following colour code. Ionic (yellow), polar (orange), hydrogen bond (red), VDW (light blue).

https://doi.org/10.1371/journal.pone.0312223.g005

4. Discussion

The generally assumed roles of dinB and katE in beta-lactam resistance mechanisms in E. coli need to be reconsidered. Contrary to previous assumptions, the adaptation rates to amoxicillin in the absence of dinB and katE, were similar to those observed in wild-type E. coli. The uniformity observed in the adaptation rates of both wild-type and knockout strains indicates shared adaptation mechanisms or the activation of compensatory pathways within the bacterial genome. This collective adaptation hypothesis gains support from the limited set of mutations shared across strains and the consistent pattern of gene amplifications observed in a region previously identified as a ‘pre-plasmid’ [49]. Notably, this region encompasses the ampC gene and its corresponding operon. Amplification of the ampC gene serves as a survival mechanism in cells exposed to beta-lactams [58]. However, amplification alone does not confer high resistance levels unless accompanied by specific mutations [47,58,59]. Most described in clinical isolates are mutations changing the -35 and -10 box, and mutations in the attenuator region (+17 to +37) [60–63], which corresponds to our findings.

Some mutations in frdD were detected in at least two of the three biological replicates, suggesting a potential role in adaptation. The frd operon, which encodes the fumarate reductase enzyme, overlaps with the ampC promoter, and its terminator acts as an attenuator for ampC expression [45]. Mutations in the frd operon have been described to boost ampC amplification and contribute to beta-lactam resistance [47,59]. However, fumarate itself seems to be involved in persister formation, as cells carrying a plasmid containing both ampC and the frd operon show a higher resistance to beta-lactams than cells containing only ampC [64]. Additionally, cells lacking ampC still adapt to high concentrations of beta-lactam antibiotics like ampicillin, indicating the presence of alternative adaptation mechanisms that do not rely on ampC amplification [65].

Furthermore, our data show that the frdD gene was constantly upregulated in response to amoxicillin exposure, even in strains that had already adapted to the antibiotic. In contrast, ampC transcription remained unchanged upon amoxicillin exposure in these adapted cells. This suggests a more complex role of fumarate in amoxicillin adaptation beyond ampC regulation. The frdD gene is integral to anaerobic respiration in bacteria, and the efficacy of beta-lactam antibiotics is linked to bacterial respiration [66,67]. Therefore, changes in anaerobic respiration patterns mediated by frdD expression may affect bacterial fitness and susceptibility to antibiotics. Moreover, frdD also plays a role in energy metabolism. Beta-lactam antibiotics primarily target cell wall synthesis by inhibiting the activity of penicillin-binding proteins (PBPs). This inhibition has been associated with changes in TCA cycle activity and oxidative phosphorylation [19,68]. Consequently, changes in energy availability and cellular redox status, influenced by frdD expression, may affect the efficiency of cell wall synthesis, and by extension, bacterial susceptibility to beta-lactam antibiotics.

Additionally, frdD could be directly involved in beta-lactam adaptation as mediator of metabolic flexibility under stress conditions. Fumarate serves as an alternative electron acceptor under anaerobic conditions and is reduced to succinate by fumarate reductase [69]. In certain conditions, fumarate reductase can substitute for succinate dehydrogenase, a component of complex II in aerobic respiration that also participates in the Krebs cycle [70,71]. This substitution may reflect a metabolic adaptation to different growth conditions, as fumarate reductase is associated with higher superoxide anion production than succinate dehydrogenase [70,72].

Although producing more superoxide anions might seem counterintuitive during oxidative stress, it can be beneficial for two reasons. First, superoxide production can aid as a signal to activate stress response pathways, helping cells adapt to environmental changes. Superoxide is metabolized into oxygen and hydrogen peroxide by superoxide dismutase, and hydrogen peroxide is subsequently transformed by catalases and peroxidases [73]. Superoxide dismutase activity has been linked to antibiotic resistance by activating the stringent response, upregulating efflux pumps, downregulating of the outer membrane porin OmpF, and co-regulation of the multidrug-resistant locus mar [73–76]. Second, elevated superoxide levels can increase mutagenesis, which, while potentially harmful, also generates genetic diversity [76,77]. This diversity can be advantageous under selective pressures, providing a collection of mutations from which beneficial traits, like antibiotic resistance, can emerge.

The recurrent rpoD mutation identified in both knockout and wild-type strains suggests a key role for rpoD in the adaptive response to amoxicillin. Although the precise functioning of this mutated sigma factor under antibiotic stress is presently unclear, the consistent appearance of identical mutations across various strains points to a significant regulatory function. Being a highly conserved housekeeping gene across bacteria [78], rpoD modulates 60–95% of sigma factors during exponential growth and binds to over half of all sigma factor binding sites across the genome [79–81]. Mutants of rpoD generated through site-directed mutagenesis, exhibit genome-wide transcriptomic changes that lead to phenotypes with improved stress tolerance [82]. While mutating rpoD seems disadvantageous due to a likely fitness loss and a high chance of lethality, MAGE-seq measurements reveal that the specific D445V mutation has a limited impact on viability [83,84]. Surprisingly, this mutation occurs in one of the most conserved areas of the gene, region 2, which binds to the -10 motif and unwinds the DNA duplex [85]. However, such mutations in highly conserved regions of sigma factors have been noted in other studies on antibiotic resistance and long-term adaptation [86].

The D445V mutation has been highlighted in recent literature, in a study involving a Chron’s disease-associated E. coli strain [87]. It was shown not to affect the catalytic activity of RNA polymerase (RNAP), yet it influences genes related to gut colonization and beta-lactam resistance. The study found transcriptomic changes in key genes, including ampC, the ROS responder yggE, and the rpoS regulator rprA. Notably, D445V was sufficient to render the previously susceptible strain resistant to five beta-lactams and increase resistance to other tested antibiotics such as ciprofloxacin [87].

Given the D445V mutation’s position in region 2 and the predicted changes to the protein structure, it seems likely that this mutation affects the sigma factor’s ability to bind to RNA polymerase. RpoD and RpoS (Sigma 30) share several binding sites [88] and compete for RNAP [89,90]. The D445V mutation might enhance RpoD’s affinity to bind to RNAP, giving it an advantage over RpoS. Indeed, the previously described D445V mutant showed lower levels of sigma S [87]. While it may seem counterintuitive to inhibit RpoS, the sigma factor responsible for the general stress response in E. coli [91,92], especially due to its involvement in antibiotic-induced mutagenesis [16,93], this strategy might favour bacterial survival under specific conditions where maintaining fitness outweighs the need for broad stress response, making the attenuation or loss of RpoS advantageous.

One replicate of MG1655 and one replicate of ΔdinB additionally carried an identical prlF mutation (S2 Table). Nevertheless, MG1655 upregulated the toxin-antitoxin system (TA), and the dinB knockout downregulated it upon adaptation and exposure to amoxicillin. The mutation we observed in prlF causes downregulation of ompF, thus leading to resistance to multiple antibiotics, i.e. carbenicillin, and aztreonam [94]. OmpF is involved in beta-lactam uptake, and mutations and porin loss are known to be involved in beta-lactam resistance [95,96]. How the mutation in prlF changes its interaction with YhaV is unknown. YhaV does not only cleave cellular mRNAs like ompF [97], but it also causes reversible bacteriostasis that is neutralized by PrlF [53,98]. Bacteriostasis could indirectly impact amoxicillin efficacy. Beta-lactams are most effective against actively growing and dividing bacteria [99]. Thus, through growth arrest, the cell might counteract the antibiotic’s primary target–penicillin-binding proteins involved in the final stages of peptidoglycan cross-linking [100].

Our findings give rise to several hypotheses (Fig 6). First, the consistent mutations in frdD and the ampC operon suggest that the regulatory role of frdD on ampC is crucial for beta-lactam resistance, but fumarate’s role in this process is likely several pathways. Furthermore, the recurrent rpoD mutations indicate a significant regulatory role for this sigma factor under antibiotic stress. Third, the involvement of the TA system prlF/yhaV suggests that this system might have a role in the acquisition of antimicrobial resistance which is not very well understood in the framework of the presently available knowledge. Lastly, our data set shows that despite being limited in stress responses, the dinB and katE knockout strains had high adaptability, enabling a seemingly unhindered acquisition of amoxicillin resistance.

thumbnail
Download:
Fig 6. Conclusion.

Key genetic factors in amoxicillin resistance acquisition in E. coli. Created in BioRender.com.

https://doi.org/10.1371/journal.pone.0312223.g006

Supporting information

S1 Fig. Frequency distributions of mutations.

The graph illustrates the frequency distributions of mutations across the samples.

https://doi.org/10.1371/journal.pone.0312223.s001

(TIF)

S1 Table. PrlF mutation.

https://doi.org/10.1371/journal.pone.0312223.s002

(DOCX)

S2 Table. Antibiotic concentrations RNA extraction.

https://doi.org/10.1371/journal.pone.0312223.s003

(DOCX)

Acknowledgments

The authors thank Stanley Brul and Meike Wortel for stimulating discussions.

References

  1. 1. Smith WPJ, Wucher BR, Nadell CD, Foster KR. Bacterial defences: mechanisms, evolution and antimicrobial resistance. Nat Rev Microbiol. 2023 Aug;21(8):519–534. Epub 2023 Apr 24. pmid:37095190.
  2. 2. Abranches J, Martinez AR, Kajfasz JK, Chávez V, Garsin DA, Lemos JA. The molecular alarmone (p)ppGpp mediates stress responses, vancomycin tolerance, and virulence in Enterococcus faecalis. J Bacteriol. 2009 Apr;191(7):2248–56. Epub 2009 Jan 23. pmid:19168608.
  3. 3. Gottesman S. Trouble is coming: Signaling pathways that regulate general stress responses in bacteria. J Biol Chem. 2019 Aug 2;294(31):11685–11700. Epub 2019 Jun 13. pmid:31197038.
  4. 4. Galhardo RS, Hastings PJ, Rosenberg SM. Mutation as a stress response and the regulation of evolvability. Crit Rev Biochem Mol Biol. 2007 Sep-Oct;42(5):399–435. pmid:17917874.
  5. 5. Martinez JL, Baquero F. Mutation frequencies and antibiotic resistance. Antimicrob Agents Chemother. 2000 Jul;44(7):1771–7. pmid:10858329.
  6. 6. Petrosino JF, Galhardo RS, Morales LD, Rosenberg SM. Stress-induced beta-lactam antibiotic resistance mutation and sequences of stationary-phase mutations in the Escherichia coli chromosome. J Bacteriol. 2009 Oct;191(19):5881–9. Epub 2009 Jul 31. pmid:19648247.
  7. 7. Miller C, Thomsen LE, Gaggero C, Mosseri R, Ingmer H, Cohen SN. SOS response induction by beta-lactams and bacterial defense against antibiotic lethality. Science. 2004 Sep 10;305(5690):1629–31. Epub 2004 Aug 12. pmid:15308764.
  8. 8. Kohanski MA, Dwyer DJ, Hayete B, Lawrence CA, Collins JJ. A common mechanism of cellular death induced by bactericidal antibiotics. Cell. 2007 Sep 7;130(5):797–810. pmid:17803904.
  9. 9. Huttner A, Bielicki J, Clements MN, Frimodt-Møller N, Muller AE, Paccaud JP, et al. Oral amoxicillin and amoxicillin-clavulanic acid: properties, indications and usage. Clin Microbiol Infect. 2020 Jul;26(7):871–879. Epub 2019 Dec 4. pmid:31811919.
  10. 10. Galhardo RS, Do R, Yamada M, Friedberg EC, Hastings PJ, Nohmi T, et al. DinB upregulation is the sole role of the SOS response in stress-induced mutagenesis in Escherichia coli. Genetics. 2009 May;182(1):55–68. Epub 2009 Mar 6. pmid:19270270.
  11. 11. Baharoglu Z, Mazel D. SOS, the formidable strategy of bacteria against aggressions. FEMS Microbiol Rev. 2014 Nov;38(6):1126–45. Epub 2014 Jun 30. pmid:24923554.
  12. 12. Andersson DI, Hughes D. Microbiological effects of sublethal levels of antibiotics. Nat Rev Microbiol. 2014 Jul;12(7):465–78. Epub 2014 May 27. pmid:24861036.
  13. 13. Kohanski MA, DePristo MA, Collins JJ. Sublethal antibiotic treatment leads to multidrug resistance via radical-induced mutagenesis. Mol Cell. 2010 Feb 12;37(3):311–20. pmid:20159551.
  14. 14. Yeiser B, Pepper ED, Goodman MF, Finkel SE. SOS-induced DNA polymerases enhance long-term survival and evolutionary fitness. Proc Natl Acad Sci U S A. 2002 Jun 25;99(13):8737–41. Epub 2002 Jun 11. pmid:12060704.
  15. 15. Corzett CH, Goodman MF, Finkel SE. Competitive fitness during feast and famine: how SOS DNA polymerases influence physiology and evolution in Escherichia coli. Genetics. 2013 Jun;194(2):409–20. Epub 2013 Apr 15. pmid:23589461.
  16. 16. Gutierrez A, Laureti L, Crussard S, Abida H, Rodríguez-Rojas A, Blázquez J, et al. β-Lactam antibiotics promote bacterial mutagenesis via an RpoS-mediated reduction in replication fidelity. Nat Commun. 2013;4:1610. pmid:23511474.
  17. 17. Pérez-Capilla T, Baquero MR, Gómez-Gómez JM, Ionel A, Martín S, Blázquez J. SOS-independent induction of dinB transcription by beta-lactam-mediated inhibition of cell wall synthesis in Escherichia coli. J Bacteriol. 2005 Feb;187(4):1515–8. pmid:15687217.
  18. 18. Händel N, Hoeksema M, Freijo Mata M, Brul S, ter Kuile BH. Effects of Stress, Reactive Oxygen Species, and the SOS Response on De Novo Acquisition of Antibiotic Resistance in Escherichia coli. Antimicrob Agents Chemother. 2015 Dec 14;60(3):1319–27. pmid:26666928.
  19. 19. Kohanski MA, Dwyer DJ, Collins JJ. How antibiotics kill bacteria: from targets to networks. Nat Rev Microbiol. 2010 Jun;8(6):423–35. Epub 2010 May 4. pmid:20440275.
  20. 20. Qi W, Jonker MJ, Katsavelis D, de Leeuw W, Wortel M, Ter Kuile BH. The Effect of the Stringent Response and Oxidative Stress Response on Fitness Costs of De Novo Acquisition of Antibiotic Resistance. Int J Mol Sci. 2024 Feb 23;25(5):2582. pmid:38473832.
  21. 21. Lopatkin AJ, Stokes JM, Zheng EJ, Yang JH, Takahashi MK, You L, et al. Bacterial metabolic state more accurately predicts antibiotic lethality than growth rate. Nat Microbiol. 2019 Dec;4(12):2109–2117. Epub 2019 Aug 26. pmid:31451773.
  22. 22. Dwyer DJ, Kohanski MA, Collins JJ. Role of reactive oxygen species in antibiotic action and resistance. Curr Opin Microbiol. 2009 Oct;12(5):482–9. Epub 2009 Jul 31. pmid:19647477.
  23. 23. Liu Y, Imlay JA. Cell death from antibiotics without the involvement of reactive oxygen species. Science. 2013 Mar 8;339(6124):1210–3. pmid:23471409.
  24. 24. Van Acker H, Coenye T. The Role of Reactive Oxygen Species in Antibiotic-Mediated Killing of Bacteria. Trends Microbiol. 2017 Jun;25(6):456–466. Epub 2017 Jan 12. pmid:28089288.
  25. 25. Slupphaug G, Kavli B, Krokan HE. The interacting pathways for prevention and repair of oxidative DNA damage. Mutat Res. 2003 Oct 29;531(1–2):231–51. pmid:14637258.
  26. 26. Ji J, Kan S, Lee J, Lysakowski S. Antibiotic Tolerance in Escherichia coli Under Stringent Response Correlates to Increased Catalase Activity. Journal of Experimental Microbiology and Immunology (JEMI). 2013 Apr;17:40–5.
  27. 27. Chang DE, Smalley DJ, Conway T. Gene expression profiling of Escherichia coli growth transitions: an expanded stringent response model. Mol Microbiol. 2002 Jul;45(2):289–306. pmid:12123445.
  28. 28. Hobbs JK, Boraston AB. (p)ppGpp and the Stringent Response: An Emerging Threat to Antibiotic Therapy. ACS Infect Dis. 2019 Sep 13;5(9):1505–1517. Epub 2019 Jul 22. pmid:31287287.
  29. 29. Bokinsky G, Baidoo EE, Akella S, Burd H, Weaver D, Alonso-Gutierrez J, et al. HipA-triggered growth arrest and β-lactam tolerance in Escherichia coli are mediated by RelA-dependent ppGpp synthesis. J Bacteriol. 2013 Jul;195(14):3173–82. Epub 2013 May 10. pmid:23667235.
  30. 30. Kobayashi S, Valentine MR, Pham P, O’Donnell M, Goodman MF. Fidelity of Escherichia coli DNA polymerase IV. Preferential generation of small deletion mutations by dNTP-stabilized misalignment. J Biol Chem. 2002 Sep 13;277(37):34198–207. Epub 2002 Jul 3. pmid:12097328.
  31. 31. Seaver LC, Imlay JA. Hydrogen peroxide fluxes and compartmentalization inside growing Escherichia coli. J Bacteriol. 2001 Dec;183(24):7182–9. pmid:11717277.
  32. 32. Baba T, Ara T, Hasegawa M, Takai Y, Okumura Y, Baba M, et al. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol Syst Biol. 2006;2:2006.0008. Epub 2006 Feb 21. pmid:16738554.
  33. 33. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A. 2000 Jun 6;97(12):6640–5. pmid:10829079.
  34. 34. Schuurmans JM, Nuri Hayali AS, Koenders BB, ter Kuile BH. Variations in MIC value caused by differences in experimental protocol. J Microbiol Methods. 2009 Oct;79(1):44–7. Epub 2009 Jul 25. pmid:19635505.
  35. 35. van der Horst MA, Schuurmans JM, Smid MC, Koenders BB, ter Kuile BH. De novo acquisition of resistance to three antibiotics by Escherichia coli. Microb Drug Resist. 2011 Jun;17(2):141–7. Epub 2011 Jan 16. pmid:21235391.
  36. 36. Di Tommaso P, Chatzou M, Floden EW, Barja PP, Palumbo E, Notredame C. Nextflow enables reproducible computational workflows. Nat Biotechnol. 2017 Apr 11;35(4):316–319. pmid:28398311.
  37. 37. Ewels P, Magnusson M, Lundin S, Käller M. MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics. 2016 Oct 1;32(19):3047–8. Epub 2016 Jun 16. pmid:27312411.
  38. 38. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014 Apr 1;30(7):923–30. Epub 2013 Nov 13. pmid:24227677.
  39. 39. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Statist. Soc. 1995;57:289–300.
  40. 40. Deatherage DE, Barrick JE. Identification of mutations in laboratory-evolved microbes from next-generation sequencing data using breseq. Methods Mol Biol. 2014;1151:165–88. pmid:24838886.
  41. 41. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021 Feb 16;10(2):giab008. pmid:33590861.
  42. 42. Brynildsrud O, Snipen LG, Bohlin J. CNOGpro: detection and quantification of CNVs in prokaryotic whole-genome sequencing data. Bioinformatics. 2015 Jun 1;31(11):1708–15. Epub 2015 Feb 1. pmid:25644268.
  43. 43. Waskom M. seaborn: statistical data visualization. J Open Source Softw. 2021 Apr 6;6(60):3021.
  44. 44. Zhou Y, Pan Q, Pires DEV, Rodrigues CHM, Ascher DB. DDMut: predicting effects of mutations on protein stability using deep learning. Nucleic Acids Res. 2023 Jul 5;51(W1):W122–W128. pmid:37283042.
  45. 45. Grundström T, Jaurin B. Overlap between ampC and frd operons on the Escherichia coli chromosome. Proc Natl Acad Sci U S A. 1982 Feb;79(4):1111–5. pmid:7041115.
  46. 46. Lindberg F, Normark S. Contribution of chromosomal beta-lactamases to beta-lactam resistance in enterobacteria. Rev Infect Dis. 1986 Jul-Aug;8 Suppl 3:S292–304. pmid:3529322.
  47. 47. Corvec S, Caroff N, Espaze E, Marraillac J, Reynaud A. -11 Mutation in the ampC promoter increasing resistance to beta-lactams in a clinical Escherichia coli strain. Antimicrob Agents Chemother. 2002 Oct;46(10):3265–7. pmid:12234856.
  48. 48. Bergström S, Lindberg FP, Olsson O, Normark S. Comparison of the overlapping frd and ampC operons of Escherichia coli with the corresponding DNA sequences in other gram-negative bacteria. J Bacteriol. 1983 Sep;155(3):1297–305. pmid:6350266.
  49. 49. Darphorn TS, Hu Y, Koenders-van Sintanneland BB, Brul S, Ter Kuile BH. Multiplication of ampC upon Exposure to a Beta-Lactam Antibiotic Results in a Transferable Transposon in Escherichia coli. Int J Mol Sci. 2021 Aug 26;22(17):9230. pmid:34502136.
  50. 50. Chen Y, Gu J, Ashworth G, Wang Z, Zhang Z, Dong C. Crystal structure of the lipopolysaccharide outer core galactosyltransferase WaaB involved in pathogenic bacterial invasion of host cells. Front Microbiol. 2023;14. pmid:37808302
  51. 51. De Buck E, Lammertyn E, Anné J. The importance of the twin-arginine translocation pathway for bacterial virulence. Trends Microbiol. 2008 Sep;16(9):442–53. Epub 2008 Aug 18. pmid:18715784.
  52. 52. Yan BB, Dong XS, Wang JP, Li XY, An L, Wang XR, et al. Glutamate-pantothenate pathway promotes antibiotic resistance of Edwardsiella tarda. Front Microbiol. 2023 Sep 13;14:1264602. pmid:37779691.
  53. 53. Schmidt O, Schuenemann VJ, Hand NJ, Silhavy TJ, Martin J, Lupas AN, et al. prlF and yhaV encode a new toxin-antitoxin system in Escherichia coli. J Mol Biol. 2007 Sep 28;372(4):894–905. Epub 2007 Jul 21. pmid:17706670.
  54. 54. Wang WF, Cheng X, Molineux IJ. Isolation and identification of fxsA, an Escherichia coli gene that can suppress F exclusion of bacteriophage T7. J Mol Biol. 1999 Sep 24;292(3):485–99. pmid:10497016.
  55. 55. Rajagopalan K, Dworkin J. Escherichia coli YegI is a novel Ser/Thr kinase lacking conserved motifs that localizes to the inner membrane. FEBS Lett. 2020 Nov;594(21):3530–3541. Epub 2020 Sep 13. pmid:32888201.
  56. 56. Bishop RE, Leskiw BK, Hodges RS, Kay CM, Weiner JH. The entericidin locus of Escherichia coli and its implications for programmed bacterial cell death. J Mol Biol. 1998 Jul 24;280(4):583–96. pmid:9677290
  57. 57. Myka KK, Hawkins M, Syeda AH, Gupta MK, Meharg C, Dillingham MS, et al. Inhibiting translation elongation can aid genome duplication in Escherichia coli. Nucleic Acids Res. 2017 Mar 17;45(5):2571–2584. pmid:27956500.
  58. 58. Jacoby GA. AmpC beta-lactamases. Clin Microbiol Rev. 2009 Jan;22(1):161–82, Table of Contents. pmid:19136439.
  59. 59. Händel N, Schuurmans JM, Feng Y, Brul S, ter Kuile BH. Interaction between mutations and regulation of gene expression during development of de novo antibiotic resistance. Antimicrob Agents Chemother. 2014 Aug;58(8):4371–9. Epub 2014 May 19. pmid:24841263.
  60. 60. Jørgensen RL, Nielsen JB, Friis-Møller A, Fjeldsøe-Nielsen H, Schønning K. Prevalence and molecular characterization of clinical isolates of Escherichia coli expressing an AmpC phenotype. J Antimicrob Chemother. 2010 Mar;65(3):460–4. Epub 2010 Jan 22. pmid:20097787.
  61. 61. Nelson EC, Elisha BG. Molecular basis of AmpC hyperproduction in clinical isolates of Escherichia coli. Antimicrob Agents Chemother. 1999 Apr;43(4):957–9. pmid:10103209.
  62. 62. Li GL, Duo LB, Luan Y, Wang CY, Wang WP, Zhang HG, et al. Identification of genotypes of plasmid-encoded AmpC beta-lactamases from clinical isolates and characterization of mutations in their promoter and attenuator regions. Gene Expr. 2012;15(5–6):215–23. pmid:23539899.
  63. 63. Caroff N, Espaze E, Bérard I, Richet H, Reynaud A. Mutations in the ampC promoter of Escherichia coli isolates resistant to oxyiminocephalosporins without extended spectrum beta-lactamase production. FEMS Microbiol Lett. 1999 Apr 15;173(2):459–65. pmid:10227175.
  64. 64. Kim JS, Cho DH, Heo P, Jung SC, Park M, Oh EJ, et al. Fumarate-Mediated Persistence of Escherichia coli against Antibiotics. Antimicrob Agents Chemother. 2016 Mar 25;60(4):2232–40. pmid:26810657.
  65. 65. Gross R, Yelin I, Lázár V, Datta M Sen, Kishony R. Beta-lactamase dependent and independent evolutionary paths to high-level ampicillin resistance. Nat Commun. 2024 Jun 25; 15(5383). pmid:38918379
  66. 66. Rosato RR, Fernandez R, Paz LI, Singh CR, Rosato AE. TCA cycle-mediated generation of ROS is a key mediator for HeR-MRSA survival under β-lactam antibiotic exposure. PLoS One. 2014 Jun 16;9(6):e99605. pmid:24932751.
  67. 67. Dwyer DJ, Belenky PA, Yang JH, MacDonald IC, Martell JD, Takahashi N, et al. Antibiotics induce redox-related physiological alterations as part of their lethality. Proc Natl Acad Sci U S A. 2014 May 20;111(20):E2100–9. Epub 2014 May 6. pmid:24803433.
  68. 68. Belenky P, Ye JD, Porter CB, Cohen NR, Lobritz MA, Ferrante T, et al. Bactericidal Antibiotics Induce Toxic Metabolic Perturbations that Lead to Cellular Damage. Cell Rep. 2015 Nov 3;13(5):968–80. Epub 2015 Oct 22. pmid:26565910.
  69. 69. Iverson TM, Luna-Chavez C, Cecchini G, Rees DC. Structure of the Escherichia coli fumarate reductase respiratory complex. Science. 1999 Jun 18;284(5422):1961–6. pmid:10373108.
  70. 70. Imlay JA. A metabolic enzyme that rapidly produces superoxide, fumarate reductase of Escherichia coli. J Biol Chem. 1995 Aug 25;270(34):19767–77. pmid:7649986.
  71. 71. Guest JR. Partial replacement of succinate dehydrogenase function by phage- and plasmid-specified fumarate reductase in Escherichia coli. J Gen Microbiol. 1981 Feb;122(2):171–9. pmid:6274999.
  72. 72. Cecchini G, Schröder I, Gunsalus RP, Maklashina E. Succinate dehydrogenase and fumarate reductase from Escherichia coli. Biochim Biophys Acta. 2002 Jan 17;1553(1–2):140–57. pmid:11803023.
  73. 73. Martins D, McKay G, Sampathkumar G, Khakimova M, English AM, Nguyen D. Superoxide dismutase activity confers (p)ppGpp-mediated antibiotic tolerance to stationary-phase Pseudomonas aeruginosa. Proc Natl Acad Sci U S A. 2018 Sep 25;115(39):9797–9802. Epub 2018 Sep 10. pmid:30201715.
  74. 74. Miller PF, Gambino LF, Sulavik MC, Gracheck SJ. Genetic relationship between soxRS and mar loci in promoting multiple antibiotic resistance in Escherichia coli. Antimicrob Agents Chemother. 1994 Aug;38(8):1773–9. pmid:7986007.
  75. 75. Demple B. Redox signaling and gene control in the Escherichia coli soxRS oxidative stress regulon—a review. Gene. 1996 Nov 7;179(1):53–7. pmid:8955629.
  76. 76. Nunoshiba T, Obata F, Boss AC, Oikawa S, Mori T, Kawanishi S, et al. Role of iron and superoxide for generation of hydroxyl radical, oxidative DNA lesions, and mutagenesis in Escherichia coli. J Biol Chem. 1999 Dec 3;274(49):34832–7. pmid:10574955.
  77. 77. Prieto-Alamo MJ, Abril N, Pueyo C. Mutagenesis in Escherichia coli K-12 mutants defective in superoxide dismutase or catalase. Carcinogenesis. 1993 Feb;14(2):237–44. pmid:8382113.
  78. 78. Park J, Wang HH. Systematic dissection of σ70sequence diversity and function in bacteria. Cell Rep. 2021 Aug 24;36(8):109590. pmid:34433066.
  79. 79. Gama-Castro S, Salgado H, Santos-Zavaleta A, Ledezma-Tejeida D, Muñiz-Rascado L, García-Sotelo JS, et al. RegulonDB version 9.0: high-level integration of gene regulation, coexpression, motif clustering and beyond. Nucleic Acids Res. 2016 Jan 4;44(D1):D133–43. Epub 2015 Nov 2. pmid:26527724.
  80. 80. Grigorova IL, Phleger NJ, Mutalik VK, Gross CA. Insights into transcriptional regulation and sigma competition from an equilibrium model of RNA polymerase binding to DNA. Proc Natl Acad Sci U S A. 2006 Apr 4;103(14):5332–7. Epub 2006 Mar 27. pmid:16567622.
  81. 81. Ishihama A. Functional modulation of Escherichia coli RNA polymerase. Annu Rev Microbiol. 2000;54:499–518. pmid:11018136.
  82. 82. Alper H, Stephanopoulos G. Global transcription machinery engineering: a new approach for improving cellular phenotype. Metab Eng. 2007 May;9(3):258–67. Epub 2007 Jan 8. pmid:17292651.
  83. 83. Park J, Wang HH. Systematic dissection of σ70sequence diversity and function in bacteria. Cell Rep. 2021 Aug 24;36(8):109590. pmid:34433066.
  84. 84. Hu JC, Gross CA. Mutations in rpoD that increase expression of genes in the mal regulon of Escherichia coli K-12. J Mol Biol. 1988 Sep 5;203(1):15–27. pmid:3054119.
  85. 85. Paget MS. Bacterial Sigma Factors and Anti-Sigma Factors: Structure, Function and Distribution. Biomolecules. 2015 Jun 26;5(3):1245–65. pmid:26131973.
  86. 86. Cohen Y, Hershberg R. Rapid Adaptation Often Occurs through Mutations to the Most Highly Conserved Positions of the RNA Polymerase Core Enzyme. Genome Biol Evol. 2022 Sep 6;14(9):evac105. pmid:35876137.
  87. 87. Arroyo-Mendoza M, Proctor A, Correa-Medina A, Brand MW, Rosas V, Wannemuehler MJ, et al. The E. coli pathobiont LF82 encodes a unique variant of σ70 that results in specific gene expression changes and altered phenotypes. bioRxiv [Preprint]. 2023 Feb 8:2023.02.08.523653. pmid:36798310.
  88. 88. Cho BK, Kim D, Knight EM, Zengler K, Palsson BO. Genome-scale reconstruction of the sigma factor network in Escherichia coli: topology and functional states. BMC Biol. 2014 Jan 24;12:4. pmid:24461193.
  89. 89. Farewell A, Kvint K, Nyström T. Negative regulation by RpoS: a case of sigma factor competition. Mol Microbiol. 1998 Aug;29(4):1039–51. pmid:9767572.
  90. 90. Maeda H, Fujita N, Ishihama A. Competition among seven Escherichia coli sigma subunits: relative binding affinities to the core RNA polymerase. Nucleic Acids Res. 2000 Sep 15;28(18):3497–503. pmid:10982868.
  91. 91. Weber H, Polen T, Heuveling J, Wendisch VF, Hengge R. Genome-wide analysis of the general stress response network in Escherichia coli: sigmaS-dependent genes, promoters, and sigma factor selectivity. J Bacteriol. 2005 Mar;187(5):1591–603. pmid:15716429.
  92. 92. Battesti A, Majdalani N, Gottesman S. The RpoS-mediated general stress response in Escherichia coli. Annu Rev Microbiol. 2011;65:189–213. pmid:21639793.
  93. 93. Baharoglu Z, Krin E, Mazel D. RpoS plays a central role in the SOS induction by sub-lethal aminoglycoside concentrations in Vibrio cholerae. PLoS Genet. 2013;9(4):e1003421. Epub 2013 Apr 11. pmid:23613664.
  94. 94. Maeda T, Iwasawa J, Kotani H, Sakata N, Kawada M, Horinouchi T, et al. High-throughput laboratory evolution reveals evolutionary constraints in Escherichia coli. Nat Commun. 2020 Nov 24;11(1):5970. pmid:33235191.
  95. 95. Harder KJ, Nikaido H, Matsuhashi M. Mutants of Escherichia coli that are resistant to certain beta-lactam compounds lack the ompF porin. Antimicrob Agents Chemother. 1981 Oct;20(4):549–52. pmid:7044293.
  96. 96. Jaffe A, Chabbert YA, Semonin O. Role of porin proteins OmpF and OmpC in the permeation of beta-lactams. Antimicrob Agents Chemother. 1982 Dec;22(6):942–8. pmid:6760806.
  97. 97. Choi W, Yamaguchi Y, Lee JW, Jang KM, Inouye M, Kim SG, et al. Translation-dependent mRNA cleavage by YhaV in Escherichia coli. FEBS Lett. 2017 Jul;591(13):1853–1861. Epub 2017 Jun 26. pmid:28573789.
  98. 98. Culviner PH, Nocedal I, Fortune SM, Laub MT. Global Analysis of the Specificities and Targets of Endoribonucleases from Escherichia coli Toxin-Antitoxin Systems. mBio. 2021 Oct 26;12(5):e0201221. Epub 2021 Sep 21. pmid:34544284.
  99. 99. Lee AJ, Wang S, Meredith HR, Zhuang B, Dai Z, You L. Robust, linear correlations between growth rates and β-lactam-mediated lysis rates. Proc Natl Acad Sci U S A. 2018 Apr 17;115(16):4069–4074. Epub 2018 Apr 2. pmid:29610312.
  100. 100. Zapun A, Contreras-Martel C, Vernet T. Penicillin-binding proteins and beta-lactam resistance. FEMS Microbiol Rev. 2008 Mar;32(2):361–85. Epub 2008 Jan 29. pmid:18248419.