Antibiotics 13 00094

antibiotics
Review
Laboratory Evolution of Antimicrobial Resistance in Bacteria to
Develop Rational Treatment Strategies
Tomoya Maeda 1,2, * and Chikara Furusawa 2,3
1 Laboratory of Microbial Physiology, Research Faculty of Agriculture, Hokkaido University, Kita 9, Nishi 9,
Kita-ku, Sapporo 060-8589, Japan
2 Center for Biosystems Dynamics Research, RIKEN, 6-2-3 Furuedai, Suita 565-0874, Japan;
chikara.furusawa@riken.jp
3 Universal Biology Institute, The University of Tokyo, 7-3-1 Hongo, Tokyo 113-0033, Japan
* Correspondence: tomoya.maeda@agr.hokudai.ac.jp; Tel.: +81-11-706-4115
Abstract: Laboratory evolution studies, particularly with Escherichia coli, have yielded invaluable
insights into the mechanisms of antimicrobial resistance (AMR). Recent investigations have illu-
minated that, with repetitive antibiotic exposures, bacterial populations will adapt and eventually
become tolerant and resistant to the drugs. Through intensive analyses, these inquiries have unveiled
instances of convergent evolution across diverse antibiotics, the pleiotropic effects of resistance muta-
tions, and the role played by loss-of-function mutations in the evolutionary landscape. Moreover,
a quantitative analysis of multidrug combinations has shed light on collateral sensitivity, reveal-
ing specific drug combinations capable of suppressing the acquisition of resistance. This review
article introduces the methodologies employed in the laboratory evolution of AMR in bacteria and
presents recent discoveries concerning AMR mechanisms derived from laboratory evolution. Addi-
tionally, the review outlines the application of laboratory evolution in endeavors to formulate rational
treatment strategies.
Keywords: antimicrobial resistance; laboratory evolution; collateral sensitivity
Citation: Maeda, T.; Furusawa, C.

Laboratory Evolution of
Antimicrobial Resistance in Bacteria to 1. Introduction
Develop Rational Treatment Antibiotic overuse in medical and agricultural domains precipitates the rapid emer-
Strategies. Antibiotics 2024, 13, 94. gence of antimicrobial resistance (AMR) and multidrug-resistant bacterial “superbugs”
https://doi.org/10.3390/ globally. AMR has been considered one of the top 10 global public health threats by the
antibiotics13010094 World Health Organization (WHO) [1–3]. In 2019, Escherichia coli, followed by Staphylo-
Academic Editors: Viviana Cafiso coccus aureus, Klebsiella pneumoniae, Streptococcus pneumoniae, Acinetobacter baumannii, and
and Floriana Campanile Pseudomonas aeruginosa, accounted for approximately 1 million deaths attributable to AMR
and 3.6 million deaths associated with AMR worldwide [3]. AMR occurs naturally through
Received: 25 December 2023
genetic changes, either via de novo mutations or the acquisition of resistance genes through
Revised: 12 January 2024
horizontal gene transfer [4]. Cells resistant to antibiotics exhibit an elevation in the mini-
Accepted: 16 January 2024
mal inhibitory concentrations (MICs), which denotes the lowest antibiotic concentration
Published: 18 January 2024
needed to hinder bacterial replication. In contrast, antibiotic persister cells and tolerant cells
demonstrate enhanced survival under antibiotic exposure without a concurrent rise in the
MIC [5]. Tolerance is the population-level ability to survive a longer duration of transient
Copyright: © 2024 by the authors. drug treatment, while persistence is the subpopulation-level tolerance phenomenon [5]. In
Licensee MDPI, Basel, Switzerland. some cases, a subpopulation among susceptible cells exhibits higher MICs, a phenomenon
This article is an open access article distinct from persistence, defined as heteroresistance [6,7]. Similar to the resistance develop-
distributed under the terms and ments, higher tolerance, persistence, and heteroresistance abilities can be acquired through
conditions of the Creative Commons genetic mutations selected over time with antibiotic doses [8,9]. Given that both resistance
Attribution (CC BY) license (https:// and tolerance contribute significantly to the failure of antibiotic treatments, understanding
creativecommons.org/licenses/by/ the mechanisms of their evolution becomes imperative.
4.0/).
Antibiotics 2024, 13, 94. https://doi.org/10.3390/antibiotics13010094 https://www.mdpi.com/journal/antibiotics

Antibiotics 2024, 13, 94 2 of 24
Investigations into mutations conferring AMR and persistence have been conducted
through laboratory evolution and genome analysis of clinically resistant isolates. These
studies have unveiled the causal relationship between mutations and the acquisition of
resistance, including target modification, drug inactivation, and drug transport [10–13].
However, both clinical isolates and laboratory-evolved strains often carry mutations in
genes beyond those classified into the three broad categories [14]. These results imply
a complex relationship between genetic changes and AMR or persistence, owing to the
multifaceted effects of antibiotics on the biomolecular network. To unravel the complexity
of AMR or persistence mechanisms and evolution, the quantitative analysis of phenotypic
and genotypic changes faces limitations when using clinical isolates due to the absence of
nearest ancestral strains and the presence of numerous neutral mutations. Additionally,
mutations conferring AMR, which are infrequently identified, are known to be challenging
to detect through genome analysis of clinical isolates [14]. Consequently, to overcome these
limitations, laboratory evolution, followed by whole-genome sequencing and phenotyping
assays, has emerged as a promising methodology.
In light of the substantial upfront investment and prolonged clinical trials required
for the development of novel drugs, an alternative approach to combat AMR is urgently
needed. Recognizing that the emergence of AMR is rooted in evolutionary dynamics, the
application of evolutionary trade-offs to suppress further resistance emerges as a promising
strategy [15–18]. The development of AMR against a primary drug may lead to a dual
outcome, inducing either heightened or reduced susceptibility to other drugs concurrently.
This phenomenon is known as collateral sensitivity, wherein
Antibiotics 2024, 13, x FOR PEER REVIEW 3 of 26 resistance to a particular drug
increases susceptibility to another, or cross-resistance, wherein resistance to one drug am-

plifies resistance
lower-dimensional to another
representation (Figure
helps to bridge the1)gap
[19]. Antibiotic
between the intricatepairs
detailsexhibiting
of collateral sensitivities
and their
cellular genetic
components determinants
and the essential biologicalhave beensuch
information, extensively
as growth rate,explored
adapt- through laboratory evolu-
ability, and survival. The dimension reduction implies that phenotypic changes are con-
tion [20–27]. These studies have identified new collateral-sensitivity and cross-resistance
strained to a lower-dimensional subspace, meaning that certain key traits change more
interactions,
significantly proposing
than others novel,
[30,31,35]. Once these rational treatment
constraints are strategies
acquired through evolution,to exploit collateral sensitivity.
subsequent adaptations to new environments tend to follow the same restricted paths
This review article elucidates the laboratory evolution
[35]. These concepts are crucial for understanding how microbes adapt and evolve in re-
methodologies employed to com-
prehend
sponse AMRlike
to challenges mechanisms and
antibiotic exposure. The evolution in pathogenic
focus on lower-dimensional bacteria and presents successful
represen-
tations helps to identify key factors that play major roles in adaptation, making it easier
examples of studies on designing rational treatment
to predict and potentially control the evolutionary dynamics of AMR.
strategies.
Figure 1. Cross-resistance and collateral sensitivity. Cross-resistance refers to the phenomenon

Figure
where 1. Cross-resistance
bacteria, andtocollateral
having developed resistance sensitivity.
a drug A, concurrently Cross-resistance
acquire resistance to another refers to the phenomenon where
drug, B. On the
bacteria, other hand,
having collateral sensitivity
developed resistanceis observed
to a when
drugbacteria, having developed
A, concurrently re-
acquire resistance to another drug, B.
sistance to drug C, show an increased susceptibility to another drug, A. When collateral sensitivity
isOn the other
established, hand, collateral
it is anticipated sensitivity
that the simultaneous is observed
development of resistancewhen bacteria,
to both drug A and having developed resistance to
drug C is challenging. This expectation suggests that combinations of these drugs could potentially
drug C, show an increased susceptibility to another drug, A. When collateral sensitivity is established,
hinder the development of bacterial resistance.
it is anticipated that the simultaneous development of resistance to both drug A and drug C is
2. Methodologies of Laboratory Evolution
challenging. This expectation suggests that combinations of these drugs could potentially hinder the
Analyzing the mechanisms underlying antibiotic resistance and/or tolerance can be
development
accomplished of bacterial
through resistance.
laboratory evolution, employing a serial transfer approach in the
presence of high antibiotic concentrations. These methods involve regularly transferring
a portion of the culture to a fresh medium at specific intervals, fostering drug resistance
The “landscape” metaphor in evolutionary biology refers to a visual representation
or tolerance (Figure 2A,B). To assess the reliability and stochastic nature of evolutionary
of how
changes, different
multiple traits
independent or characteristics
culture lines are typically runof organisms
concurrently (in
for each this case, microbial resistance to
exper-
iment. In experiments of this nature, enhancing the selection pressure during laboratory
antibiotics) change over time [28,29]. It is dynamic because the conditions that microbes
evolution often involves incrementally raising the antibiotic concentration alongside the
development of antibiotic resistance in the evolving cells (Figure 2A). For instance, Lázár
et al. conducted laboratory evolution of E. coli under 12 antibiotics, progressively elevating
the dosage (by 1.5-fold) at every fourth transfer [21]. An alternative approach involves
culturing cells in media containing drug gradients, where cells capable of growth at the
highest drug concentrations are transferred to subsequent passages (Figure 2B) [22,27,36].
In previous studies, we conducted high-throughput laboratory evolution of E. coli using
a 2-fold drug gradient across 10 dilutions [22] or a 20.25-fold gradient across 22 dilutions
face, such as exposure to antibiotics or other environmental factors, are constantly shifting.
Just as a landscape can have peaks and valleys, the dynamic landscape of resistance
illustrates how microbial traits might rise or fall in response to various selection pressures.
The concept of the “dimensionality of phenotypic states” in the context of AMR
evolution discusses how traits or characteristics of microbes (like their ability to resist
drugs) can be expressed in different ways [30,31]. This entails exploring how these traits
can change and adapt over time in response to factors like drug exposure. There are
numerous components and features in microbial cells, often measured as omics data such
as transcriptome (the set of all RNA transcripts), proteome (the entire set of proteins),
metabolome (the complete set of intracellular small-molecule chemicals), and genomic
mutations (the loci and the type of mutations) [22,27,32–34]. These cellular components
contribute to a large number of degrees of freedom, forming high-dimensional data. Despite
this complexity, changes related to adaptation and evolution were observed as being
effectively limited to a lower-dimensional subspace [22,27]. This implies that, even though
there are many variables, only a few of them are crucial for expressing traits that impact the
microbe’s ability to evolve and adapt, especially in the context of AMR. Therefore, certain
traits related to AMR evolution can be represented or changed in simpler, more predictable
ways during evolution by focusing on a few key aspects rather than the entire complexity
of the microbe’s characteristics. Simplifying this high-dimensional data into a lower-
dimensional representation helps to bridge the gap between the intricate details of cellular
components and the essential biological information, such as growth rate, adaptability, and
survival. The dimension reduction implies that phenotypic changes are constrained to
a lower-dimensional subspace, meaning that certain key traits change more significantly
than others [30,31,35]. Once these constraints are acquired through evolution, subsequent
adaptations to new environments tend to follow the same restricted paths [35]. These
concepts are crucial for understanding how microbes adapt and evolve in response to
challenges like antibiotic exposure. The focus on lower-dimensional representations helps
to identify key factors that play major roles in adaptation, making it easier to predict and
potentially control the evolutionary dynamics of AMR.
2. Methodologies of Laboratory Evolution

Analyzing the mechanisms underlying antibiotic resistance and/or tolerance can be
accomplished through laboratory evolution, employing a serial transfer approach in the
presence of high antibiotic concentrations. These methods involve regularly transferring a
portion of the culture to a fresh medium at specific intervals, fostering drug resistance or tol-
erance (Figure 2A,B). To assess the reliability and stochastic nature of evolutionary changes,
multiple independent culture lines are typically run concurrently for each experiment. In
experiments of this nature, enhancing the selection pressure during laboratory evolution
often involves incrementally raising the antibiotic concentration alongside the development
of antibiotic resistance in the evolving cells (Figure 2A). For instance, Lázár et al. conducted
laboratory evolution of E. coli under 12 antibiotics, progressively elevating the dosage (by
1.5-fold) at every fourth transfer [21]. An alternative approach involves culturing cells
in media containing drug gradients, where cells capable of growth at the highest drug
concentrations are transferred to subsequent passages (Figure 2B) [22,27,36]. In previous
studies, we conducted high-throughput laboratory evolution of E. coli using a 2-fold drug
gradient across 10 dilutions [22] or a 20.25 -fold gradient across 22 dilutions [27]. Compared
to the method where the drug dosage is gradually increased at fixed intervals, using drug
gradients enables a broader range of selection pressure, providing a more nuanced selection
process. This favors cells that exhibit adaptive responses to a range of drug concentra-
tions, allowing for a continuous and more natural selection process. These serial transfer
experiments are often laborious due to the requirements of numerous culture conditions
and independent culture lines as replicates for the evaluation of reproducibility. Therefore,
these experiments are frequently conducted using laboratory automation systems, such
as automatic pipetting robots equipped with robot arms, which are capable of precisely
aspirating and dispensing sample liquids into microplates and tubes. Some of the robots
are equipped with robot arms to transfer plates to equipment used to monitor cell growth,
like microplate readers and shaker incubators [37]. It is essential to note that the influence
of selection pressure on AMR evolution was assessed by both the drug increment approach
and drug gradient approach [36,38]. The drug increment approach suggested that rapidly
changing environments limit mutational opportunities by reducing population size and
eliminating specific sets of mutations as viable evolutionary options [38]. The findings
emphasize the importance of understanding how the rate of environmental change affects
both demographic and genetic aspects of evolutionary rescue, especially in the context
of unprecedented anthropogenic environmental alterations [38]. On the other hand, the
drug gradient approach found that bacterial populations evolving under strong selection
developed high levels of cross-resistance against multiple antibiotics, while those evolving
under milder selection exhibited weaker cross-resistance [36]. This study also observed that
strongly selected populations acquired a higher number of mutations in genes specific to
the target pathways of the drugs used for selection [36]. These mutations were more diverse
in strongly selected populations compared to those under mild selection [36]. Strongly
selected populations often acquired mutations in essential genes with higher fitness costs,
Antibiotics 2024, 13, xwhile mildly
FOR PEER REVIEW selected populations tended to acquire mutations in genes5 with
of 26 lower fitness
costs [36].
Methods
Figure 2. Figure of laboratory
2. Methods evolution.
of laboratory evolution. (A)
(A) Drug Drug approach.
increment increment approach.
During During laboratory
laboratory evolu-
tion, drug concentrations are regularly increased, e.g., 1.5-fold at every fourth transfer. (B) Drug
evolution, drug concentrations are regularly increased, e.g., 1.5-fold at every fourth transfer. (B) Drug
gradient approach. Cells are grown in microwell plates containing drug gradients, and the culture
that can grow at the highest drug concentration is selected and transferred to fresh drug gradient
media. (C) Repeating batch culturing method between antibiotic treatment and the regrowth cycle
for the acquisition of drug tolerance or persistence. (D) Morbidostat [39]. (E) Agar plate method
under a drug gradient [40]. (F) MEGA-plate [23].
Instead of employing the serial transfer approach, continuous culturing is also appli-
cable to the laboratory evolution of drug resistance (Figure 2D). Toprak et al. devised the
gradient approach. Cells are grown in microwell plates containing drug gradients, and the culture
that can grow at the highest drug concentration is selected and transferred to fresh drug gradient
media. (C) Repeating batch culturing method between antibiotic treatment and the regrowth cycle
for the acquisition of drug tolerance or persistence. (D) Morbidostat [39]. (E) Agar plate method
under a drug gradient [40]. (F) MEGA-plate [23].
Instead of employing the serial transfer approach, continuous culturing is also ap-
plicable to the laboratory evolution of drug resistance (Figure 2D). Toprak et al. devised
the morbidostat system, a tool capable of constantly monitoring cell growth and adjusting
drug concentrations within the culture vials to maintain the selection pressure by using
feedback control [39]. This system comprises several components, including a flat-bottom
glass vial, a magnetic stirrer to mix the culture, an optical density (OD) detection system,
and a computer-controlled set of peristaltic pumps used for liquid transfer [39,41]. The
distinctive advantage of the morbidostat lies in its ability to dynamically adapt antibiotic
concentrations within culture vials, ensuring a sustained drug-induced inhibition, thereby
fine-tuning the selection pressure according to the evolving adaptation rate [41]. Moreover,
the morbidostat keeps the bacterial population at low densities, preventing nutrient limi-
tations in growth, and modulates the growth rate to align with the fixed dilution rate by
adjusting the antibiotic concentration [41]. However, while continuous culturing offers
significant advantages, it may not be as suitable for high-throughput experiments in con-
trast to serial transfer methods using automated systems, as it is constrained by the limited
amount of equipment available. Furthermore, the continuous culturing process might
promote the formation of biofilms during laboratory evolution, potentially necessitating
the replacement of culture vials to avoid biofilm formation [42].
For laboratory evolution toward drug persistence or tolerance acquisition, bacterial
populations are intermittently exposed to higher concentrations of antibiotics than MIC
(Figure 2C). A typical method is repeating batch culturing between antibiotic treatment and
the regrowth cycle [43–45]. To select tolerant or persister cells, stationary cultures are treated
with antibiotics at 100–200-fold MIC for certain durations (usually ~5 h) [45]. By conducting
such cycling experiments, Van den Bergh et al. obtained 40 evolved E. coli strains showing
1000- to 10,000-fold increases in persistence against aminoglycoside antibiotics [45]. The
increased persistence and/or tolerance of laboratory-evolved cells were also observed in S.
aureus [46] and other ESKAPE pathogens (Enterococcus faecium, S. aureus, K. pneumoniae, A.
baumannii, P. aeruginosa, and Enterobacter species) [47], indicating the generality of the cyclic
antibiotic treatment protocols in bacterial acquisition of antibiotic persistence and tolerance.
The colony transfer method serves as an alternative experimental setup for cells
forming severe aggregates in liquid media (Figure 2E). For example, Mycobacterium forming
aggregates in anti-tuberculosis drugs underwent laboratory evolution on agar plates with
drug gradients [40]. In this experiment, resistant populations were repeatedly selected
from cells at the border of the growth inhibitory zone, created by a filter paper soaked with
a drug solution [40]. Microbial evolutionary dynamics towards antibiotics can be directly
visualized using soft agar, allowing bacteria to swim and spread (Figure 2F). The microbial
evolution and growth arena (MEGA)-plate, consisting of successive areas with different
concentrations of antibiotics overlaid by thin, soft agar, was developed [23]. Time-lapse
imaging of the MEGA-plate captures real-time processes of mutations and selections in
the presence of high dosages of antibiotics. The MEGA-plate stands as a valuable tool for
directly visualizing and understanding evolutionary dynamics in bacterial populations [23].
3. Insights into AMR Mechanisms from Laboratory Evolution

Laboratory evolution studies provide invaluable insights into the dynamic landscape
of resistance, revealing intricate details about the evolutionary trajectories, population
dynamics, and metabolic adaptations shaping bacterial responses to drug exposure. The
complex interplay of cross-resistance and collateral sensitivity in the evolution of AMR has
been intensively analyzed using E. coli as a model bacterium [21,22,27]. These studies em-
ploy a comprehensive approach, integrating laboratory evolution and genomic sequencing

to map cross-resistance and collateral-sensitivity interactions in E. coli.
3.1. Genotype-Phenotype Relationships in AMR Evolution

Lázár et al. conducted laboratory evolution of E. coli by exposing cells to increasing
dosages of 12 antibiotics [21]. This study revealed convergent molecular evolution across
antibiotics, mutations conferring resistance enhancing sensitivity to other drugs, and the
identification of chemogenomic profile similarity as a predictor of cross-resistance [21].
Similar results were observed in other laboratory evolution studies [22,27]. Despite diverse
initial conditions, similar mutations and resistance mechanisms emerged under antibiotic
selection. Mutations that confer resistance to a specific antibiotic often simultaneously
enhance sensitivity to many other drugs, suggesting that resistance evolution is driven
by mutations with broad consequence [21,22,27]. Approximately 27% of the observed
mutations resulted in proteins with compromised or no activities, highlighting the role
of loss-of-function mutations in antibiotic adaptation [21]. Cross-resistance patterns are
strongly influenced by chemogenomic profile similarity between antibiotics, indicating
that exposure to a single antibiotic often leads to multidrug resistance. Interestingly, cross-
resistance between two antibiotics is largely independent of whether they show synergistic
effects in combination [21].
To explore the complex relationship between genotype and phenotype in antibiotic-
resistant E. coli strains, Suzuki et al. also conducted laboratory evolution of E. coli’s resis-
tance to 11 antibiotics using a serial transfer method by selecting a population showing the
growth under the highest drug concentration possible [22]. Through 90-day laboratory evo-
lution and integrated transcriptome–genome analyses, the study revealed that resistance
acquisition to one drug significantly alters susceptibility to others. They demonstrated that
a simple linear model can quantitatively predict the resistance of 25 drugs via the expres-
sion changes of only eight genes [22]. The eight genes were acrB, encoding a subunit of a
multi-drug efflux pump; ompF, encoding an outer membrane porin protein; cyoC, encoding
the subunit of cytochrome bo3 terminal oxidase; pps, encoding the phosphoenolpyruvate
synthase; tsx, encoding a nucleoside channel, a receptor of phage T6 and colicin K; oppA,
encoding an oligopeptide transporter; folA, encoding the dihydrofolate reductase; and pntB,
encoding the beta subunit of pyridine nucleotide transhydrogenase [22].
To understand the constraints shaping the evolution of AMR, a systematic investiga-
tion of evolutionary constraints through high-throughput laboratory evolution of E. coli was
conducted with the addition of 95 antibacterial chemicals covering a wide range of action
mechanisms [27]. Machine learning techniques were employed to analyze phenotype–
genotype data, revealing low-dimensional phenotypic states and trade-off relationships
associated with drug resistance. The findings provide insights into the intricate interplay
between genomic, transcriptomic, and resistance profiles, contributing to a comprehensive
understanding of evolutionary constraints in adaptive evolution [27]. Since E. coli’s evo-
lutionary dynamics were attributable to a relatively small number of intracellular states,
this indicates that E. coli is equipped with only a limited number of strategies for antibiotic
resistance [27]. This study also quantified collateral responses to other antibiotics and found
that 336 and 157 pairs of drugs within the 2162 combinations exhibited cross-resistance and
collateral sensitivity, respectively [27]. Heteroresistance, a prevalent phenotype in clinically
isolated strains to date [7,48], was observed in 15 out of 33 pairs in the laboratory-evolved
strains, suggesting its frequent occurrence in the process of acquiring resistance [27]. It
was suggested that the perturbation of metabolic activity by metabolic inhibitors, reactive
oxygen species generation, and alteration of cytoplasmic membrane permeabilization by
cationic peptides could serve as possible strategies to suppress antibiotic resistance [27].
Laboratory evolution also uncovered the often-overlooked area of metabolic adap-
tations in response to antibiotic treatment, shedding light on how alterations in cellular
metabolism can contribute to antibiotic resistance [14]. This study emphasized the impor-
tance of population-level analyses in understanding the evolutionary landscape in response
to drug treatment. This study aimed to maximize metabolic adaptation rather than growth
adaptation in laboratory evolution protocols [14]. This shift in dynamics allowed for a
more comprehensive view of antibiotic-specific metabolic variants, revealing underappreci-
ated genes related to central carbon and energy metabolism [14]. Mutations in metabolic
genes that arose in response to antibiotic treatment were found in multiple independent
populations and responses to different drugs [14]. Particularly, a mutation in the SucA
enzyme, part of the 2-oxoglutarate dehydrogenase complex, increased antibiotic resistance
by preventing the antibiotic-mediated induction of tricarboxylic acid (TCA) cycle activity,
avoiding metabolic toxicity, and minimizing lethality [14]. These metabolic mutations were
identified in clinical E. coli pathogens at levels comparable to known resistance mutations,
indicating their clinical relevance [14]. It should be noted that TCA cycle activity is known
to influence bacterial susceptibility to antibiotics, and its modulation has been linked to
both antibiotic resistance and tolerance. The inactivation of the TCA cycle resulted in
enhancing persister cell formation in S. aureus [49], and down-regulation of the TCA cycle
was related to levofloxacin resistance in Vibrio alginolyticus [50]. In addition, increasing the
TCA cycle flux can promote aminoglycoside uptake, thereby eliminating the drug-resistant
Gram-negative bacteria [51,52].
3.2. Identified Key Genes Conferring Cross-Resistance and Collateral Sensitivity in E. coli
Our comprehensive high-throughput laboratory evolution of E. coli systematically
investigated the underlying mechanisms of cross-resistance and collateral sensitivity [27].
The key genes associated with these phenomena in E. coli are cataloged in Table 1, drawing
upon insights from our previous study. Notably, the study emphasized the pivotal role
of mutations in genes governing transporters and porins in mediating antibiotic resis-
tance in E. coli [27]. Perturbations in uptake and efflux activities emerged as principal
mechanisms governing cross-genetic resistance and heteroresistance [27]. Specifically, the
study illuminated the significance of the overexpression of efflux pumps AcrAB/TolC and
EmrAB/TolC, coupled with the inactivation of their repressors, in conferring resistance to a
spectrum of antibiotics [27]. Furthermore, the investigation pinpointed the involvement of
YcbZ, a putative protease implicated in translation and ribosome biogenesis, in mediating
cross-resistance against multiple antibiotics [27]. Mutations in PrlF, associated with the
PrlF-YhaV toxin–antitoxin system, were found to be linked to resistance against specific
antibiotics, such as aztreonam and carbenicillin [27]. The observed cross-resistance was
partly ascribed to the diminished expression of OmpF, underscoring the intricate interplay
of genes in the stress response [27]. Additionally, the study uncovered collateral sensitiv-
ities tied to prlF-mediated resistance, revealing trade-offs between acquired resistances
and susceptibility to specific drugs, including rifampicin [27]. It was postulated that these
collateral sensitivities might be linked to the global mRNA destabilization effect induced
by increased RNase activity of PrlF [27]. Moreover, the derepression of an alternative sigma
factor, RpoS, resulting from a mutation in the regular RssB, conferred both cross-resistance
and collateral sensitivity to various drugs [27]. The proposed mechanism for such collateral
sensitivity involved the competition between RpoS and the housekeeping sigma factor
RpoD, leading to decreased carbon source availabilities and diminished competitiveness for
low concentrations of nutrients [27]. The subsequent Section 3.5 elaborates on the trade-off
mechanism of aminoglycoside resistance and drugs unrelated to aminoglycosides.
Table 1. Catalog of key genes conferring cross-resistance and collateral sensitivity in E. coli.
Mutation Cross-Resistance Collateral Sensitivity

Chloramphenicol, Rifampicin, Cefmetazole, Aztreonam, Carbenicillin,
ompF Norfloxacin, Phleomycin, DL-3-hydroxynorvaline, Mecillinam, D-Cycloserine
Tetracycline, Furaltadone, Erythromycin, Puromycin
Chloramphenicol, Rifampicin, Cefmetazole, Aztreonam, Acriflavine,
rssB Protamine Sulfate, D-Cycloserine
Carbenicillin, Phleomycin, Tetracycline, Erythromycin, Puromycin
Protamine Sulfate, D-Cycloserine,
glpT Carbenicillin, Fosfomycin, Mitomycin C, Phleomycin, Puromycin
Erythromycin
cyaA Carbenicillin, Fosfomycin, Phleomycin, DL-3-hydroxynorvaline, Vancomycin
Mecillinam, Tetracycline, Erythromycin, Puromycin
Chloramphenicol, Aztreonam, Carbenicillin, Phleomycin,
ycbZ D-Cycloserine
DL-3-hydroxynorvaline, Tetracycline, Erythromycin, Puromycin
Kanamycin, D-Cycloserine, Phleomycin, DL-3-hydroxynorvaline,
cyoE Rifampicin, Erythromycin
Puromycin
cyoA Kanamycin, Phleomycin Vancomycin
cyoB Kanamycin, D-Cycloserine, Phleomycin Sulfisoxazole, Vancomycin
Aztreonam, Carbenicillin, Phleomycin, DL-3-hydroxynorvaline,
nuoG Chloramphenicol
Tetracycline, Puromycin
Acriflavine, Mitomycin C, Phleomycin, DL-3-hydroxynorvaline,
mipA Vancomycin
ptsP Aztreonam, Kanamycin, Phleomycin, DL-3-hydroxynorvaline, Puromycin D-Cycloserine
rfe Rifampicin, Carbenicillin, Mitomycin C, Phleomycin, Mecillinam D-Cycloserine
purR Carbenicillin, Phleomycin, DL-3-hydroxynorvaline, Puromycin Sulfisoxazole, D-Cycloserine
Chloramphenicol, Carbenicillin, Phleomycin, DL-3-hydroxynorvaline,
corA Sulfisoxazole
oxyR DL-3-hydroxynorvaline Norfloxacin
apt Carbenicillin, Puromycin D-Cycloserine
sdaA Carbenicillin, Phleomycin, DL-3-hydroxynorvaline, Puromycin D-Cycloserine
Chloramphenicol, Aztreonam, Carbenicillin, Phleomycin,
nfsA DL-3-hydroxynorvaline, Mecillinam, Nitrofurantoin, Furaltadone, D-Cycloserine
Erythromycin, Puromycin
ilvL Chloramphenicol, Acriflavine, Carbenicillin, Puromycin Sulfisoxazole
Chloramphenicol, Cefmetazole, Aztreonam, Acriflavine, Carbenicillin,
gshA Phleomycin, DL-3-hydroxynorvaline, Tetracycline, Erythromycin, Sulfisoxazole
Puromycin
dacA Chloramphenicol, Acriflavine, Mitomycin C, Phleomycin Cefmetazole, Erythromycin
frlA Chloramphenicol, Acriflavine Vancomycin
Chloramphenicol, Rifampicin, Kanamycin, Acriflavine, Carbenicillin, Protamine Sulfate, D-Cycloserine,
fusA
Sulfisoxazole Vancomycin
Chloramphenicol, Aztreonam, Acriflavine, Carbenicillin, Phleomycin,
glyT Sulfisoxazole
DL-3-hydroxynorvaline, Tetracycline, Erythromycin, Puromycin
Chloramphenicol, Cefmetazole, Aztreonam, Carbenicillin, Norfloxacin,
gyrA Acriflavine, Fosfomycin, D-Cycloserine
Tetracycline, Erythromycin, Puromycin
hisS Chloramphenicol, Tetracycline D-Cycloserine
iscR Chloramphenicol, Carbenicillin, Mitomycin C, Puromycin D-Cycloserin
livM Chloramphenicol, Carbenicillin, Tetracycline\ Sulfisoxazole
Table 1. Cont.
Mutation Cross-Resistance Collateral Sensitivity

Chloramphenicol, Cefmetazole, Acriflavine, Carbenicillin, D-Cycloserine,
lon Phleomycin, DL-3-hydroxynorvaline, Mecillinam, Tetracycline, Mitomycin C
Erythromycin, Puromycin
rne Sulfisoxazole, Vancomycin
Carbenicillin, Mitomycin C, Erythromycin
Chloramphenicol, Rifampicin, Cefmetazole, Aztreonam, Carbenicillin,
rob D-Cycloserine
Norfloxacin, Mitomycin C, Tetracycline, Erythromycin, Puromycin
Chloramphenicol, Carbenicillin, Mitomycin C, Amitriptyline, Tetracycline,
rpoB D-Cycloserine
Puromycin
Chloramphenicol, Aztreonam, Carbenicillin, Norfloxacin, Phleomycin,
serA Fosfomycin, D-Cycloserine
DL-3-hydroxynorvaline, Tetracycline, Puromycin
Chloramphenicol, Aztreonam, Kanamycin, Carbenicillin, D-Cycloserine,
prlF Rifampicin
Phleomycin, Mecillinam, Puromycin
yjcO Carbenicillin, Phleomycin, Tetracycline Sulfisoxazole, D-Cycloserine
acrR Carbenicillin, Mitomycin C, Tetracycline, Promethazine, Nitrofurantoin, D-Cycloserine
Furaltadone, Erythromycin, Puromycin
3.3. Costs of AMR Evolution

The impact of evolution in the absence of antibiotics on the fitness effects of resistance
mutations in E. coli was analyzed [53]. Rifampicin-resistant and -sensitive E. coli were
subjected to experimental evolution in a drug-free environment [53]. The fitness effects of
newly acquired resistance elements were quantified under antibiotic-free conditions [53].
The results revealed that streptomycin-resistance mutations exhibited smaller fitness effects
in rifampicin-resistant genotypes adapted to antibiotic-free growth medium compared to
non-adapted genotypes [53]. This epistatic variation in the costs of resistance was observed
not only between different resistance mutations, but also between resistance elements and
beneficial mutations acquired during adaptation to drug-free conditions [53].
The rates of resistance evolution in bacteria exposed to different antibiotics were
analyzed by quantifying the distribution of fitness effects (DFE) of mutations [54]. The
DFE of mutations refers to the range and characteristics of fitness changes that result
from genetic mutations within a population. The DFE is thought to be a crucial factor
influencing the evolutionary dynamics of drug resistance in bacterial populations [54].
The DFE in the presence of eight antibiotics, representing various modes of action, was
comprehensively measured [54]. Surprisingly, the width of the DFE varied dramatically
between antibiotics, with some drugs exhibiting a lower DFE width than that in the
absence of stress [54]. This divergence in DFE width was attributed to distinct, drug-
specific dose–response characteristics [54]. This research identified resistance variability
and dose sensitivity as key drug-specific properties shaping the DFE [54]. Nitrofurantoin,
in particular, exhibited exceptionally low resistance variability, limiting the evolution
of resistance. Laboratory evolution experiments using nitrofurantoin confirmed slow
resistance evolution via reproducible mutations [54].
Metabolic constraints on AMR evolution were investigated through laboratory evolu-
tion of E. coli under three different antibiotics (ampicillin, chloramphenicol, and norfloxacin)
while growing on glucose or acetate as the sole carbon source [55]. Profiling more than
500 intracellular and extracellular metabolites in 190 evolved populations revealed that
carbon and energy metabolism strongly constrain the evolutionary trajectories of antibi-
otic resistance [55]. Both the speed and mode of resistance acquisition were influenced
by metabolic adaptations [55]. This study demonstrated that resistance evolves more
rapidly on glucose compared to acetate, indicating greater metabolic plasticity during
Antibiotics 2024, 13, 94 10 of 24
respiro-fermentative metabolism [55]. Environmental conditions played a crucial role in de-

termining the trade-off between the costs and benefits of resistance mutations, influencing
how rapidly a resistant mutant establishes itself within the population. This study identified
condition-dependent compensatory mechanisms in antibiotic-resistant populations, includ-
ing shifts in metabolic pathways [55]. For example, there was a shift from respiratory to
fermentative metabolism of glucose upon overexpression of efflux pumps, highlighting the
role of metabolic adaptations in response to antibiotic selection pressure [55]. Metabolome-
based predictions revealed emerging weaknesses in antibiotic-resistant strains, such as the
hypersensitivity to fosfomycin in ampicillin-resistant strains [55]. Additionally, monitoring
~750 intracellular metabolites in E. coli immediately after antibiotic exposure revealed the
early bacterial response to antimicrobial compounds and potential weaknesses in terms
of tolerating antibiotic therapies [56]. For example, unbalanced ammonia metabolism
contributed to increased chloramphenicol toxicity, linking it to the generation of reactive
oxygen species [56].
The transferability of ten resistance-conferring mutations and resistance genes (acrR,
envZ, fis, gyrA, mprA, ompC, phoQ, soxR, trkH, and ycbZ) between E. coli and the closely
related S. enterica subsp. serovar Typhimurium strain was investigated [57]. This study
revealed that the resistance mutations in E. coli often did not confer resistance in S. en-
terica [57]. Surprisingly, in some cases, these mutations led to drug hypersensitivity in
S. enterica [57]. In-depth analysis of a key gene involved in aminoglycoside resistance
(trkH) indicated that extreme differences in mutational effects between the two species
were attributed to preexisting mutations in other genes [57]. This study emphasized the
limited conservation of mutational effects driving collateral sensitivity and cross-resistance
phenotypes among antibiotic-resistant bacteria. The effects strongly depended on the
genetic background, and even a single resistance mutation could alter collateral responses
to other antibiotics.
3.4. Impact of Multidrug Combinations on AMR Evolution

The impact of multidrug combinations on the competitive selection between sen-
sitive and resistant bacterial populations was analyzed using doxycycline-resistant and
doxycycline-sensitive E. coli [58]. In a hyper-antagonistic class of drug combinations, a
drug can render the combined treatment selective against its resistance allele. This inverted
selection is insensitive to the underlying resistance mechanism and occurs at sublethal con-
centrations while maintaining inhibition of the wild type. The findings suggest a trade-off
between the effect of drug interactions on absolute potency and the relative competitive
selection imposed on emerging resistant populations. The study emphasizes a previously
unappreciated feature of the fitness landscape for the evolution of resistance and points
to potential avenues for designing antimicrobial combinations with improved selection
against resistance [58].
The dynamics of resistance development in E. coli when exposed to five different
single antibiotics (ciprofloxacin, amikacin, tetracycline, chloramphenicol, and piperacillin)
and all possible antibiotic drug pairs were investigated [59]. The degree of resistance
development against drug combinations was found to be linked to collateral sensitivity
and the resistance that occurred during adaptation to the component drugs [59]. It was
confirmed that drug resistance causing collateral sensitivity was suppressed in the growth
environment under the treatment of amikacin and the other drug pairs [59]. This implies
that drug combinations can limit the evolution of resistance if resistance to one drug in the
combination results in collateral sensitivity to another drug.
The evolutionary conservation of collateral responses to five clinically relevant antibi-
otics (cefepime, ciprofloxacin, gentamicin, meropenem, and tetracycline) across species,
particularly within the ESKAPE pathogens, was investigated [60]. This study identified
14 instances of universal cross-resistance that accounted for 80% of all the collateral re-
sponses and two global collateral sensitivity relationships (cefepime and gentamicin pair
and ciprofloxacin and gentamicin pair) among the evolved lineages [60]. Although the
The evolutionary conservation of collateral responses to five clinically relevant anti-
biotics (cefepime, ciprofloxacin, gentamicin, meropenem, and tetracycline) across species,
particularly within the ESKAPE pathogens, was investigated [60]. This study identified 14
instances of universal cross-resistance that accounted for 80% of all the collateral re-
Antibiotics 2024, 13, 94 11 of 24
sponses and two global collateral sensitivity relationships (cefepime and gentamicin pair
and ciprofloxacin and gentamicin pair) among the evolved lineages [60]. Although the
genetic basis for the collateral sensitivity was unclear, genomic analyses revealed diver-
genetic
gent andbasis for the evolutionary
conserved collateral sensitivity was suggesting
trajectories, unclear, genomic analyses
that collateral revealedmay
responses diver-be
gent and conserved evolutionary trajectories, suggesting
preserved across species for the selected drugs [60]. that collateral responses may be
preserved across species for the selected drugs [60].
3.5. Mechanism of the Trade-Off between AMR Evolution
3.5. Mechanism of the Trade-Off between AMR Evolution
The trade-off relationships between aminoglycoside resistance and other drugs were
The trade-off relationships between aminoglycoside resistance and other drugs were
demonstrated by several studies (Figure 3) [20,22]. Mutations in genes whose products
demonstrated by several studies (Figure 3) [20,22]. Mutations in genes whose products are
are involved in respiration, such as cytochrome bo3 oxidase and NDH-I, were commonly
involved in respiration, such as cytochrome bo3 oxidase and NDH-I, were commonly identi-
identified
fied in aminoglycoside-evolved
in aminoglycoside-evolved strainsSince
strains [20,22]. [20,22]. Since aminoglycoside
aminoglycoside uptake is knownuptake is
to re-
quire proton motive force (PMF) [61,62], decreased respiration results in a reduced PMF anda
known to require proton motive force (PMF) [61,62], decreased respiration results in
reduced PMF
inhibition and inhibitionuptake.
of aminoglycoside of aminoglycoside
On the otheruptake. On the other
hand, activation hand,
of the activation
multidrug effluxof
pump AcrAB/TolC through the inactivation of its repressor AcrR is a common multidrugis
the multidrug efflux pump AcrAB/TolC through the inactivation of its repressor AcrR
a commonmechanism
resistance multidrug resistance mechanism
against various againstexcept
antibiotics, various
forantibiotics, except for
aminoglycosides amino-
[21,22,27].
glycosides
Since [21,22,27].
AcrAB/TolC is aSince AcrAB/TolCproton
PMF-dependent is a PMF-dependent
antiporter [63,64],proton antiporter
a decrease in PMF[63,64],
leadsa
decrease in PMF leads to hyper-susceptibility to other drugs that are
to hyper-susceptibility to other drugs that are discharged by AcrAB/TolC. Therefore, the discharged by
AcrAB/TolC. Therefore, the acquisition of simultaneous resistances against
acquisition of simultaneous resistances against both aminoglycoside and other drugs that both amino-
glycoside
are and by
discharged other drugs that
AcrAB/TolC are adischarged
poses significant by AcrAB/TolC
challenge. poses a significant
This difficulty chal-
was experimen-
lenge. This difficulty was experimentally demonstrated by the laboratory
tally demonstrated by the laboratory evolution of E. coli under the simultaneous additionevolution of E.
coli
of under the
two-drug simultaneous
combinations addition of two-drug combinations [59,65].
[59,65].
Figure3.3.The
Figure Thetrade-off
trade-offmechanism
mechanism of of aminoglycoside
aminoglycoside resistance
resistance and
and drugs
drugs other
other than
than aminoglyco-
aminoglyco-
sides. Q,
sides. Q, quinone;
quinone; QH
QH22,, quinol.
quinol. E.
E. coli
coli possesses
possesses two
two isozymes
isozymes with
with different
different abilities
abilitiesto
togenerate
generate
PMFfor
PMF forNADH
NADHdehydrogenase
dehydrogenaseand andcytochrome
cytochromeoxidase
oxidaseininthe
therespiratory
respiratorychain.
chain.Aminoglycosides
Aminoglycosides
are known to be taken up into cells in a PMF-dependent manner. E. coli develops resistance to
aminoglycosides through mutations that decrease PMF generation in either NDH-I or Cytbo3 oxidase,
both of which exhibit a high ability to generate PMF. In contrast, many antibiotics other than amino-
glycosides develop resistance by enhancing the activity of PMF-dependent multidrug efflux systems
like AcrAB/TolC. However, this resistance mechanism requires a high PMF. Therefore, a trade-off
occurs between resistance to aminoglycosides and resistance to other drugs due to the contrasting
PMF requirements in these two resistance pathways.
3.6. Population Dynamics of AMR Evolution

The population dynamics of norfloxacin resistance were analyzed in a bioreactor [66].
The majority of individual evolved isolates were less resistant, with lower MICs than
Antibiotics 2024, 13, 94 12 of 24
the group MIC. Although highly resistant isolates with higher MICs than the bioreactor
concentration were rare, their presence was notable [66]. The highly resistant mutants were
often present in low abundance in the population, and their emergence preceded increases
in the group MIC [66]. The beneficial effect of the highly resistant mutants on the major
lower resistant mutants resulted from indole production [66]. The highly resistant mutants
endured a fitness cost to produce and share indole, acting as a form of altruism [66]. The
production of indole by highly resistant mutants served to turn on drug efflux pumps and
oxidative-stress-protective mechanisms, enhancing the survival of less resistant isolates [66].
This altruistic behavior, despite imposing a fitness cost on the highly resistant mutants,
allowed weaker constituents to survive antibiotic stress [66].
3.7. AMR Evolution in Spatially Structured Environments

The spatial dynamics of bacterial evolution, emphasizing the interplay between muta-
tional diversity, spatial constraints, and the ultimate fitness of the evolving population, were
investigated using a growth arena (MEGA)-plate, a large experimental device allowing
for the study of evolution in a large, spatially structured environment [23]. Mutants with
high resistance to trimethoprim or ciprofloxacin did not always lead the evolutionary front.
Highly resistant mutants could be trapped behind more sensitive lineages [23]. The physi-
cal blocking of mutants by each other was observed, reminiscent of phenomena observed in
biofilm formation [23]. Mutations that increased resistance often came at the cost of reduced
growth, which was restored by compensatory mutations [23]. Compensatory mutants were
spatially restricted, appearing in localized spots behind the evolving front [23]. These
mutants, though able to outcompete their parent in certain conditions, were constrained
from contributing to the ultimate evolutionary course of the population when spatially
restricted [23]. The fitness of the bacterial population was found to be determined not
only by the fittest mutants, but also by those that were sufficiently fit and arose sufficiently
close to the advancing front [23]. Mutants with enhanced resistance and compensatory
mutations, when appearing at the front without being physically blocked, accelerated the
adaptive process [23].
3.8. Fitness Landscape and AMR Evolution

The fitness landscape represents the relationship between genotype or phenotype and
fitness in the context of antibiotic resistance evolution [67]. Since the fitness landscape is
crucial for explaining and predicting evolutionary trajectories, the phenotype-based fitness
landscape for AMR evolution in E. coli was investigated by quantifying the multidimen-
sional phenotypic changes, i.e., time-series data of resistance for eight different drugs [67].
Because of the challenge regarding the high dimensionality of genotypic changes, this study
quantified multidimensional phenotypic changes using time-series data on resistance for
eight different drugs. This study reveals that different peaks in the phenotype-based fitness
landscape correspond to different drug resistance mechanisms [67]. This finding supports
the validity of the inferred landscape as a representation of the relationship between phe-
notype and fitness in the context of AMR [67]. A distinctive aspect of the study is the
empirical approach used to infer the fitness landscape based on antibiotic resistance profiles
rather than genotypes. The immense number of possible genotype changes associated with
AMR makes empirical fitness landscapes based on genotypes less capable of predicting and
controlling evolution. This study demonstrates that the directions of evolution predicted
by the inferred phenotype-based fitness landscape align with the observed experimental
trajectories, at least for evolution under certain antibiotics (tetracycline, kanamycin, and
norfloxacin) [67]. This consistency suggests that resistance profiles capture the internal
degrees of freedom of E. coli for predicting evolution.
3.9. Antibiotic Tolerance and Persistence Development

It has been demonstrated that conditions inducing bacterial tolerance can operate
in various non-specific ways, both in vitro and during infection [68–72]. While tolerant
Antibiotics 2024, 13, 94 13 of 24
populations exhibit high survival rates under antibiotic treatment, they suffer from im-
paired proliferation during infection. In contrast, antibiotic persistence is highlighted as a
risk-limiting strategy, enabling bacteria to survive antibiotic treatment without compro-
mising their ability to colonize the host [68]. The findings emphasize the importance of
understanding the dynamics of these recalcitrance mechanisms to improve the design of
more effective antibiotic therapies, considering the different fitness trade-offs associated
with tolerance and persistence.
Metabolic homeostasis, the balance of metabolic processes in the cell, can be perturbed
to promote antibiotic persistence [73–81]. A laboratory evolution and population-wide
sequencing identified mutations in respiratory complex I (type I NADH dehydrogenase;
NDH-I) as key contributors to increased persister formation in E. coli [45]. This finding was
consistent across both model and pathogenic E. coli strains [45]. Mutations in NDH-I that
compromise proton pumping led to significant cytoplasmic acidification upon metabolic
perturbations [45]. The proposed mechanistic model suggests that strong metabolic pertur-
bations, such as entering the stationary phase or abrupt nutrient shifts, result in cytoplasmic
acidification [45]. Mutations in NDH-I exacerbate this acidification, acting as central signal-
ing hubs connecting perturbed metabolic homeostasis with persister cell formation [45].
4. Designing Rational Treatment Strategies through Laboratory Evolution

Addressing the escalating threat of AMR, there is a growing interest in combination-
based treatments to boost efficacy. One strategy involves identifying collateral sensitivities.
It was demonstrated, through laboratory evolution and whole-genome sequencing, that
alternating drugs during treatment effectively slows resistance evolution compared to
single-drug treatments, exploiting evolutionary trade-offs [16]. Melnikov et al. reported
that amplifying the fitness cost of drug resistance, particularly in tavaborole-resistant E. coli,
fosters natural competition between resistant and susceptible cells, showcasing collateral
sensitivity [82]. Tavaborol, a novel, synthetic small-molecule inhibitor of protein synthesis,
received approval from the Food and Drug Administration in 2014 for the treatment of
onychomycosis (nail fungus) [83–86]. Tavaborole functions by targeting the editing domain
of leucyl-tRNA synthetase (LeuRS). Through covalent binding with tRNALeu, tavaborole
prevents the dissociation of tRNALeu from LeuRS [84]. It has been demonstrated that
tavabolore and its deliberatives are effective against multidrug-resistant bacterial pathogens,
including E. coli, P. aeruginosa [86], S. aureus [87], Mycobacterium tuberculosis [88–91], and
S. pneumoniae [92]. The resistant mechanism against tavaborole was elucidated through
the acquisition of a specific mutation in the LeuRS editing domain, as revealed by whole-
genome sequencing of clinical isolates of E. coli [93] and P. aeruginosa [94]. Since the tavabole-
resistant E. coli strain carrying the LeuRS mutation showed hypersensitivity to norvaline (a
chemical analog of leucine), the cost of tavabole resistance can be amplified in the presence
of norvaline, which is misused for protein synthesis by the resistant cells [82]. Indeed,
Melnikov et al. demonstrated that simultaneous tavaborole and norvaline treatment slowed
tavaborole resistance evolution [82]. It is crucial to note that complete suppression was not
achieved, indicating a potential risk of multidrug resistance [82].
4.1. Collateral Sensitivity as a Potential Strategy for Designing Rational Antibiotic Treatment
The degree to which a drug combination can reduce the evolution of resistance de-
pends on the interplay between collateral resistance and collateral sensitivity in its com-
ponent drugs. If resistance to one drug leads to collateral sensitivity to another drug, the
combination is more effective in limiting the evolution of resistance [59]. The impact of drug
combination therapy on AMR in E. coli was investigated through laboratory evolution [65].
This study revealed that certain combinations of drugs (aminoglycoside amikacin and
chloramphenicol or enoxacin) with collateral sensitivity could suppress the acquisition
of resistance [65]. The effectiveness of drug combination with collateral sensitivity in an
opportunistic human pathogen P. aeruginosa was also investigated [24]. Unlike patterns
observed in other bacterial species, collateral effects in P. aeruginosa showed interspecific
Antibiotics 2024, 13, 94 14 of 24
differences in evolutionary trade-offs [24]. Interestingly, diverse patterns of collateral sensi-

tivity and cross-resistance have emerged among replicate populations adapted to the same
drug [24]. Genomic analysis of evolved populations reveals distinct evolutionary paths, de-
termining whether bacteria become cross-resistant or collaterally sensitive [24]. The study
identified key regulatory genes (nalC, mexZ, and pmrB) associated with collateral effects,
providing insights into the mechanisms governing these responses [24]. Overall, these
studies contributed valuable insights into the development of novel antibiotic therapies
that leverage fitness trade-offs during drug resistance evolution.
4.2. Collateral Sensitivity Cycling

At the outset, antibiotic cycling was dependent on the inherent fitness cost of AMR.
A higher fitness cost meant a more rapid decrease in the resistance frequency within a
population when the selective pressure from drug treatment was absent [95]. Neverthe-
less, the stability of fitness costs linked to resistance is not guaranteed, as compensatory
mutations that diminish these costs can compromise the effectiveness of cycling [96,97].
Therefore, using a collateral sensitivity network based on the data set of E. coli AMR
evolution to 23 clinically used drugs, a new treatment framework, collateral sensitivity
cycling was proposed (Table 2) [15]. This approach involves using drugs with compatible
collateral sensitivity profiles sequentially to treat infections, selecting against drug resis-
tance development [15]. The study identified numerous drug sets, demonstrating that
cyclic deployment of antibiotics like gentamicin and cefuroxime could effectively counter
resistance [15]. Collateral sensitivity cycling, based on reciprocal collateral sensitivities,
differs fundamentally from current drug cycling methods, offering a sustainable treatment
paradigm [15]. The findings suggest its potential application in managing Gram-negative
bacterial infections and emphasize its relevance for chronic infections and cases involving
multiple pathogens [15]. Clinical studies are needed in order to validate these principles in
hospital settings. The study anticipates that collateral sensitivity cycling will contribute to
the sustainable use of drugs in clinical disease management [15].
Using the morbidostat platform, it was demonstrated that multidrug resistance could
be manipulated by administering pairs of antibiotics and cycling between them in an
ON/OFF manner [98]. Cyclin use of polymyxin B and other antibiotics completely sup-
pressed the development of resistance to one of the antibiotics [98]. This study also em-
phasized the importance of exploiting the collateral sensitivity of various antibiotics for
designing effective treatment methods that can suppress or reverse drug resistance [98].
The impact of an antibiotic cycling strategy on the prevalence of antibiotic-resistant,
Gram-negative bacteria was investigated through a cluster-randomized crossover study
conducted across eight intensive care units (ICUs) in Europe [99]. Antibiotic-resistant,
Gram-negative bacteria were operationally defined as Enterobacteriaceae with extended-
spectrum β-lactamase production or resistance to piperacillin–tazobactam, along with
Acinetobacter spp. and P. aeruginosa displaying resistance to piperacillin–tazobactam or
carbapenems within the study parameters [99]. The study centered on three antibiotic
groups—third-generation or fourth-generation cephalosporins, piperacillin–tazobactam,
and carbapenems—and evaluated the prevalence of resistant bacteria during cycling (where
a specific antibiotic class was preferentially employed for a 6-week period) and mixing
(where the antibiotic class was altered for each consecutive patient) [99]. Despite observed
variations in antibiotic utilization across ICUs, the comprehensive analysis of study an-
tibiotics’ overall volume and consumption patterns during cycling and mixing revealed
no statistically significant differences [99]. The principal examination, accounting for pa-
tients present during monthly point-prevalence surveys, disclosed a mean prevalence of
antibiotic-resistant, Gram-negative bacteria of 23% during cycling and 22% during mixing,
with no discernible statistical distinction between the two strategies [99]. The study con-
cluded that antibiotic cycling does not manifest a substantial reduction in the prevalence of
antibiotic-resistant bacteria in ICUs [99]. It is noteworthy to mention that this study was con-
ceived in 2010 [99], predating the development of collateral sensitivity cycling introduced
Antibiotics 2024, 13, 94 15 of 24
by Imamovic et al. in 2013 [15]. Consequently, there exists a positive prospect regarding
the clinical efficacy of this strategy; however, additional clinical evidence is imperative to
adjudicate the reliability of this approach. To implement collateral sensitivity cycling in real-
world scenarios, the collateral effects of clinically relevant drug combinations must exhibit
robustness across genetically diverse backgrounds. Therefore, the robustness of collateral
sensitivity in various clinical resistant strains has been intensively analyzed. For example,
the study demonstrated the robustness of the collateral sensitivity of ciprofloxacin towards
gentamicin, fosfomycin, ertapenem, and colistin in E. coli [100], and towards gentamicin,
fosfomycin, colistin, aztreonam, and tobramycin in P. aeruginosa [25,101]. The efficacy levels
of aminoglycoside and β-lactam in eradicating quinolone-resistant P. aeruginosa in patients
suffering from cystic fibrosis were also demonstrated [25].
4.3. Sequential Drug Regimens Based on Collateral Sensitivities

The potential benefits of sequential drug regimens based on collateral sensitivities,
where exposure to a first drug induces susceptibility to a second, were investigated [26].
This study suggests that sequential drug regimens derived from in vitro evolution experi-
ments may have overstated therapeutic benefits [26]. Predictions of collaterally sensitive
responses can lead to cross-resistance, where resistance to the first drug also results in
resistance to the second [26]. Using mathematical modeling parametrized with combi-
natorially complete fitness landscapes for E. coli, the likelihood of collateral sensitivities
being stochastic was quantified [26]. This study also conducted laboratory evolution with
E. coli under cefotaxime, a beta-lactam antibiotic [26]. The study involved 60 parallel
evolutionary replicates to demonstrate the extent of heterogeneity in second-line drug
sensitivity [26]. The results showed that a second drug can stochastically exhibit either
increased susceptibility or increased resistance when following a first drug [26]. Genetic
divergence was identified as the driver of this differential response through targeted and
whole-genome sequencing [26]. Collateral sensitivity was found to be rare and not univer-
sal [26]. Different mutations, representing different evolutionary trajectories, contributed
to heterogeneity in the collateral response. These results emphasized that the success of
evolutionarily informed therapies is predicated on a rigorous, probabilistic understanding
of the contingencies that arise during the evolution of drug resistance [26]. This study
proposed collateral sensitivity likelihoods as critical statistical benchmarks for the clinical
translation of sequential drug therapies [26]. The likelihoods were derived through em-
pirical observations and mathematical modeling. These findings provide insights into the
challenges and considerations involved in designing effective sequential drug therapies to
combat AMR.
4.4. Optimization of Antibiotic Treatment for Chronic Infections by Targeting Phenotypic States
In the in vitro AMR evolution process of P. aeruginosa in synthetic cystic fibrosis (CF)
sputum medium towards 24 clinically relevant antibiotics, a phenotypic convergence to-
wards distinct states associated with specific mutations in antibiotic resistance genes was
observed, exhibiting collateral sensitivity to several antibiotic classes [25]. This study sug-
gests that chronic infections could be more effectively treated by targeting these phenotypic
states linked with particular mutations [25]. Collateral sensitivity, observed even when the
organism is already resistant to certain drugs, can be exploited to design rational treatment
strategies. For example, the application of one antibiotic may lead to subsequent resistance
development and phenotypic convergence, but this could enhance the action of another
antibiotic to which collateral sensitivity has been observed [25]. This study emphasizes
the potential clinical impact of collateral sensitivity in optimizing treatment strategies for
chronic infections, particularly in conditions like CF, where patients undergo repeated
rounds of antibiotics or lifelong therapies [25]. The application of ciprofloxacin for the treat-
ment of P. aeruginosa infections was found to lead to subsequent resistance development
and phenotypic convergence [25]. Importantly, this resistance could enhance the efficacy
of tobramycin due to the collateral sensitivity observed in resistant strains [25]. Similar
Antibiotics 2024, 13, 94 16 of 24
enhancement was observed for colistin action in bacteria that developed resistance to
ciprofloxacin or aztreonam. Mutations in the nfxB gene encoding a negative transcriptional
regulator of MexCD-OprJ efflux system associated with quinolone resistance [102] were
linked to collateral sensitivity across various drug classes and phenotypic convergence [25].
Notably, nfxB mutants were eliminated during antibiotic treatment in a CF patient, empha-
sizing the potential for targeting specific mutations associated with collateral sensitivity
in optimizing chronic infection treatment [25]. This study suggests that, with monitoring,
nfxB gene mutations or MexC protein abundance could serve as biomarkers for collateral
sensitivity in clinical settings [25].
4.5. Suppression of Tolerance Acquisition by Cycling Antibiotics with Different

Metabolic Dependencies
Antibiotics exhibit varying degrees of metabolic dependencies, and those weakly de-
pendent on metabolism maintain effectiveness even when targeting dormant cells [103]. The
metabolic dependency of E. coli was assessed by examining the bactericidal effectiveness of
various antibiotics across a range of nutrient availabilities (Table 2) [103]. This investigation
revealed distinct categories of antibiotics regarding their reliance on metabolism, with
ampicillin and ciprofloxacin identified as strongly dependent, while gentamicin, halicin,
and mitomycin C were categorized as weakly dependent [103]. Notably, ampicillin and
ciprofloxacin exhibited diminished bactericidal efficacy when faced with nutrient-depleted
cells, in contrast to mitomycin C, which maintained its effectiveness under these condi-
tions [103]. The effectiveness of strongly dependent drugs exhibited a positive correlation
with increased intracellular ATP concentration (and, therefore, metabolic state). Conversely,
the effectiveness of weakly dependent drugs appeared to be largely unaffected by varia-
tions in intracellular ATP concentration [103]. During chronic infections, the pathogen’s
metabolism can be downregulated, leading to failure in eradicating pathogens due to antibi-
otic tolerance [103–106]. The rate of tolerance evolution was shown to depend on antibiotic
metabolism dependencies [107]. Tolerance evolved more readily against antibiotics which
were strongly dependent on bacterial metabolism, such as ampicillin and ciprofloxacin,
compared to those minimally affected by the metabolic state, such as gentamicin, halicin,
and mitomycin C [107]. Additionally, this study demonstrated that cycling antibiotics with
different metabolic dependencies can interrupt the evolution of tolerance, extending the
treatment efficacy duration [107]. Therefore, cycling strategies alternating between antibi-
otics which are strongly and weakly dependent on metabolism were proven to be effective
in terms of delaying tolerance evolution, offering insights into combination treatments.
This study highlights that differences in antibiotic metabolic dependencies could guide
the design of customized treatment strategies, balancing concerns of toxicity and tolerance
evolution [107].
4.6. Long-Term Clearance Efficacy of Drug Combinations

To understand how drug combinations affect bacterial long-term clearance at clinically
relevant concentrations, the survival of S. aureus during prolonged exposure to pairwise
and higher-order cidal drug combinations was systematically quantified [108]. This study
found a phenomenon of reciprocal suppression in clearance interactions. Unlike growth
inhibition and early killing, the efficacy of drug combinations decreases rather than in-
creases as more drugs are added [108]. This finding challenges conventional wisdom and
underscores the need for a nuanced understanding of drug interactions over time. This
study suggests that strong suppressive clearance interactions are suggestive of induced
persistence in the presence of specific drug combinations [108]. Adding drugs that target
non-growing persisters, such as mitomycin C and daptomycin, restored the efficacy of
otherwise suppressive drug mixes [108]. The lack of correlation between growth inhibition
interactions and long-term clearance interactions cautions against relying solely on growth
inhibition phenotypes when prescribing drug combination therapies. Drug combinations
with high inhibitory efficacy may not necessarily have increased long-term efficacy. This
Antibiotics 2024, 13, 94 17 of 24
study also highlights that adding β-lactamase inhibitors, commonly used to potentiate
treatment against β-lactam-resistant strains, counterintuitively reduces the long-term clear-
ance efficacy of drug combinations against such strains [108]. This cautions against certain
clinically prescribed combinations. Despite the challenges posed by reciprocal suppressive
interactions, this study suggests that they open up opportunities for designing treatment
regimens that are inherently selective against resistance to any one of their agents [108].
4.7. Clinical Evidence Supporting the Efficacy of Antibiotic Combination Therapy Involving
Aminoglycosides Is Substantial
The utilization of such combinations has become widespread, particularly in address-
ing severe hospital-acquired infections caused by multidrug-resistant species, owing to the
apparent effectiveness of the evolutionary trade-off relationships between aminoglycoside
resistance and other drugs [109–111]. A retrospective study focusing on bacteremia predom-
inantly induced by Enterobacter cloacae in cancer patients was conducted by Bodey et al. in
1991 [112]. The study assessed response rates, considering the eradication of all signs and
symptoms of Enterobacter infection as the endpoint. The results indicated response rates of
59% and 74% with the single use of aminoglycoside or penicillin, respectively, while simulta-
neous administration of aminoglycoside and penicillin demonstrated a higher response rate
of 78% [112]. Examining the clinical efficacy and safety of combination therapy involving
aminoglycoside antibiotic gentamicin and macrolide antibiotic azithromycin for treating
urogenital gonorrhea caused by Neisseria gonorrhoeae infection, in a study conducted
by Kirkcaldy et al. in 2014, reported promising outcomes [113]. The randomized, multi-
site, open-label, noncomparative trial revealed a 100% microbiological cure rate among
202 evaluable participants receiving gentamicin/azithromycin [113]. Nevertheless, there
exist discrepancies regarding the use of such a two-combination therapy [109]. For instance,
a randomized trial comparing β-lactam monotherapy with β-lactam-aminoglycoside com-
bination therapy for sepsis in immunocompetent patients, inclusive of various pathogens
such as S. aureus, Enterobacteriaceae, and P. aeruginosa, concluded that the addition of an
aminoglycoside to β-lactams is discouraged due to unaltered fatality rates [114]. A sub-
sequent study corroborated this conclusion; however, this study additionally concluded
a survival benefit of β-lactam-aminoglycoside combination therapy when prescribed to
children with multidrug-resistant Gram-negative bacteria [110].
Despite lingering uncertainty, the use of aminoglycosides in combination therapy
has been recommended, emphasizing risk stratification [111,115]. It is crucial to note that
prior combination therapies were designed based on the synergistic effects of antibiotic
combinations [109]. Recent laboratory evolution experiments, however, have challenged
this perspective by demonstrating that the evolvability of AMR remains independent of
whether the combinations exhibit synergistic effects [21,59,116,117]. Rather than relying
on observed drug interactions, the results revealed a discernible pattern linking genetic
trajectories to resistance evolution [116]. Therefore, future endeavors must delve into
elucidating the mechanisms of evolutionary constraints and identifying specific drug pairs
to inform the design of rational treatment strategies for combination therapy.
4.8. Application of Antimicrobial Peptides (AMPs) for Combating AMR

The emergence of drug resistance to conventional antibiotics necessitates the explo-
ration of alternative agents, such as AMPs. AMPs demonstrate broad-spectrum antimicro-
bial activity characterized by high specificity and low toxicity, derived from both synthetic
and natural sources [118]. They play a pivotal role in the host’s innate immunity against
various microorganisms, including bacteria, fungi, parasites, and viruses [118]. AMPs are
considered promising alternatives due to their capacity to act on multiple targets, combat-
ing drug-resistant bacteria [118]. Therefore, current clinical applications of AMPs primarily
focus on bacterial infections. Cationic AMPs, effective against bacterial pathogens, target
unique anionic components such as LPS in Gram-negative bacteria and lipoteichoic acid in
Gram-positive bacteria [118]. The structural diversity of AMPs and their ability to target
Antibiotics 2024, 13, 94 18 of 24
specific lipids or intracellular proteins contribute to their selectivity against bacterial species
or strains. AMPs induce membrane permeation, leading to intracellular content leakage, or
penetrate membranes for intracellular effects [118]. In the context of biofilms—microbial
aggregates forming on tissues or medical implants—AMPs offer diverse mechanisms to
target various biofilm properties [119].
A systematic examination of resistance evolution to 14 diverse cationic AMPs and
12 antibiotics in E. coli was conducted [120]. The laboratory evolution of E. coli against
these AMPs and antibiotics demonstrated that certain AMPs, such as tachyplesin II and
R8, completely suppressed resistance acquisition during approximately 120 generations
of laboratory evolution, sufficient time to allow for the acquisition of resistances to the
12 antibiotics [120]. Tachyplesins I and II, antimicrobial peptides showing potent activity
against various pathogens, exhibited no observed resistance acquisition in a mutator E. coli
strain or in pathogens including S. enterica subsp. serovar Typhimurium, K. pneumoniae
subsp. pneumoniae, and A. baumannii, indicating a very low probability of resistance
acquisition against tachyplesin II [120]. Increased hydropathicity and fewer polar and
positively charged amino acids were associated with reduced resistance during laboratory
evolution [120]. In contrast to AMR evolution, no cross-resistance to AMPs or limited
horizontal gene transfer for AMP resistance were observed [120]. The possibility of AMP
resistance through horizontal gene transfer was assessed by heterologous expression of
metagenomic DNA fragments from soil samples (1.8 million clones) in E. coli [120].
The effectiveness of combinations of AMPs on AMP evolution in S. aureus was also
explored [121]. It was demonstrated that treatment with certain AMP combinations de-
lays resistance evolution compared to individual AMPs [121]. The lowest resistance was
observed with a random AMP library containing over a million peptides [121]. The study
highlighted the correlation between resistance evolution rate, individual AMP resistance
cost, and cross-resistance. AMP-resistant strains often remain sensitive to other AMPs,
reducing concerns regarding broad-range resistance [121]. Combining AMPs, especially
those with high resistance costs, has proven effective in hindering resistance evolution,
emphasizing the potential of AMP cocktails for sustainable treatment against antibiotic-
resistant pathogens.
Table 2. Three evolutionary-based strategies to combat AMR.
Strategy Advantages Disadvantages Clinical Trial

Positive outcomes have been
Reduce the supply of reported against, e.g., E. coloacae
The overall benefits of
effective mutations. [112], N. gonorrhoear [113], and
combinations are not always
Combination of drugs The trade-off multidrug-resistant Gram-negative
evident in routine clinical outcomes
with collateral sensitivity relationship between the bacteria [110]. Discrepant results
or from single trials, necessitating a
[16,65,82] drug pair corners were also reported against various
more comprehensive synthesis of
pathogens into an pathogens, including S. aureus,
clinical data. See Section 4.7.
evolutionary dead end. Enterobacteriaceae, and P. aeruginosa
[114]
This strategy upgrades A European cluster-randomized
The potential therapeutic advantage
the previous crossover study determined that the
might be overemphasized, with
drug-cycling strategy practice of antibiotic cycling does
genetic divergence identified as the
Collateral sensitivity dependent on unreliable not lead to a reduction in the
underlying factor influencing
cycling [15,98] fitness costs. Instead, prevalence of antibiotic-resistant
diverse responses, leading to either
this strategy relies on Gram-negative bacteria carriage
heightened or diminished
limiting the evolution of among patients admitted to
resistance to subsequent drugs [26].
drug resistance. intensive care units [99].
Cycling antibiotics with This strategy can Clinical studies are essential to
To date, no relevant clinical trial has
different metabolic interrupt the evolution validate its efficacy in real-world
been reported yet.
dependencies [107] of tolerance. hospital settings.
Antibiotics 2024, 13, 94 19 of 24
5. Conclusions
Laboratory evolution studies revealed the multifaceted nature of AMR evolution,
providing crucial insights into its mechanisms. Cross-resistance and collateral sensitivity,
influenced by genomic and phenotypic changes, pose challenges and opportunities for
developing effective treatment strategies. The integration of laboratory evolution, genomics,
and phenomics is crucial for unraveling the complex landscape of AMR, providing valuable
insights for future studies and the development of innovative therapeutic approaches.
Convergent molecular evolution across antibiotics was revealed, indicating com-
mon resistance mechanisms [21,22,27]. Mutations conferring resistance often simultane-
ously enhance collateral sensitivity to other drugs, suggesting pleiotropic effects [21,22,27].
Therefore, collateral sensitivity can be exploited to design rational treatment
strategies [15,24–26,59,65]. Future research is required in order to refine strategies based on
collateral sensitivity, understand the heterogeneity in collateral responses, and explore the
long-term efficacy of drug combinations. To validate the effectiveness of collateral sensi-
tivity in real-world scenarios, an investigation of the degree to which collateral sensitivity
observed in laboratory settings translates to clinical outcomes will be highly required. In
addition, the diversity of collateral sensitivity patterns among different bacterial species
and strains needs to be explored. It has been suggested that the success of evolutionarily-
informed therapies depends on a rigorous probabilistic understanding of contingencies [26];
probabilistic understanding of collateral sensitivities is also important for the success of
such evolutionarily-informed therapies.
Recognizing that AMPs and phages serve as viable alternatives to antibiotics, effective
combinations should extend beyond antibiotics alone. The design of combinations should
not be confined solely to antibiotics. Presently, the development of an ideal drug capable
of completely suppressing resistance acquisition remains elusive. Similar to antibiotics,
reports of resistance acquisition to AMPs and phages exist [120,122]. The distinct pros and
cons associated with antibiotics, AMPs, and phages [123,124] offer endless possibilities
for effective combinations. It is important to highlight that the use of aminoglycosides in
combinations may pose a risk of notable side effects that potentially surpasses the clinical
benefits [109,114]. Consequently, there is a strong need to identify drug pairs that exhibit
evolutionary trade-offs, distinct from the reliance on aminoglycosides. To identify ideal
combinations based on their evolutionary constraints, further high-throughput laboratory
evolution will be necessary to establish a robust starting point.
Author Contributions: Conceptualization, T.M.; writing—original draft preparation, T.M.;

writing—review and editing, T.M. and C.F.; supervision, T.M. and C.F. All authors have read and
agreed to the published version of the manuscript.
Funding: Authors are supported by a Grant-in-Aid for Scientific Research on Innovative Areas
(22H04876) from the Japan Society for the Promotion of Science (JSPS), and a Grant-in-Aid for
Scientific Research (C) (21K06142) from the JSPS to T.M., a Grant-in-Aid for Scientific Research (S)
(19H05626), from JSPS and the Japan Science and Technology Agency (JST) ERATO (JPMJER1902)
to C.F.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1. O’Neill, J. Tackling Drug-Resistant Infections Globally: Final Report and Recommendations; Wellcome Trust and HM Government:
London, UK, 2016.
2. Roope, L.S.J.; Smith, R.D.; Pouwels, K.B.; Buchanan, J.; Abel, L.; Eibich, P.; Butler, C.C.; Tan, P.S.; Sarah Walker, A.; Robotham,
J.V.; et al. The Challenge of Antimicrobial Resistance: What Economics Can Contribute. Science 2019, 364, eaau4679. [CrossRef]
[PubMed]
3. Murray, C.J.; Ikuta, K.S.; Sharara, F.; Swetschinski, L.; Robles Aguilar, G.; Gray, A.; Han, C.; Bisignano, C.; Rao, P.; Wool, E.;
et al. Global Burden of Bacterial Antimicrobial Resistance in 2019: A Systematic Analysis. Lancet 2022, 399, 629–655. [CrossRef]
[PubMed]
4. Schrader, S.M.; Vaubourgeix, J.; Nathan, C. Biology of Antimicrobial Resistance and Approaches to Combat It. Sci. Transl. Med.
2020, 12, aaz6992. [CrossRef] [PubMed]
Antibiotics 2024, 13, 94 20 of 24
5. Balaban, N.Q.; Helaine, S.; Lewis, K.; Ackermann, M.; Aldridge, B.; Andersson, D.I.; Brynildsen, M.P.; Bumann, D.; Camilli,
A.; Collins, J.J.; et al. Definitions and Guidelines for Research on Antibiotic Persistence. Nat. Rev. Microbiol. 2019, 17, 441–448.
[CrossRef] [PubMed]
6. Andersson, D.I.; Nicoloff, H.; Hjort, K. Mechanisms and Clinical Relevance of Bacterial Heteroresistance. Nat. Rev. Microbiol.
2019, 17, 479–496. [CrossRef] [PubMed]
7. Band, V.I.; Hufnagel, D.A.; Jaggavarapu, S.; Sherman, E.X.; Wozniak, J.E.; Satola, S.W.; Farley, M.M.; Jacob, J.T.; Burd, E.M.; Weiss,
D.S. Antibiotic Combinations That Exploit Heteroresistance to Multiple Drugs Effectively Control Infection. Nat. Microbiol. 2019,
4, 1627–1635. [CrossRef]
8. LaFleur, M.D.; Qi, Q.; Lewis, K. Patients with Long-Term Oral Carriage Harbor High-Persister Mutants of Candida Albicans.
Antimicrob. Agents Chemother. 2010, 54, 39–44. [CrossRef]
9. Vulin, C.; Leimer, N.; Huemer, M.; Ackermann, M.; Zinkernagel, A.S. Prolonged Bacterial Lag Time Results in Small Colony
Variants That Represent a Sub-Population of Persisters. Nat. Commun. 2018, 9, 4074. [CrossRef]
10. Woodford, N.; Ellington, M.J. The Emergence of Antibiotic Resistance by Mutation. Clin. Microbiol. Infect. 2007, 13, 5–18.
[CrossRef]
11. Blair, J.M.A.; Webber, M.A.; Baylay, A.J.; Ogbolu, D.O.; Piddock, L.J.V. Molecular Mechanisms of Antibiotic Resistance. Nat. Rev.
Microbiol. 2015, 13, 42–51. [CrossRef]
12. Furusawa, C.; Horinouchi, T.; Maeda, T. Toward Prediction and Control of Antibiotic-Resistance Evolution. Curr. Opin. Biotechnol.
2018, 54, 45–49. [CrossRef]
13. Yelin, I.; Kishony, R. Antibiotic Resistance. Cell 2018, 172, 1136–1136.e1. [CrossRef] [PubMed]
14. Lopatkin, A.J.; Bening, S.C.; Manson, A.L.; Stokes, J.M.; Kohanski, M.A.; Badran, A.H.; Earl, A.M.; Cheney, N.J.; Yang, J.H.;
Collins, J.J. Clinically Relevant Mutations in Core Metabolic Genes Confer Antibiotic Resistance. Science 2021, 371, eaba0862.
[CrossRef] [PubMed]
15. Imamovic, L.; Sommer, M.O.A. Use of Collateral Sensitivity Networks to Design Drug Cycling Protocols That Avoid Resistance
Development. Sci. Transl. Med. 2013, 5, 204ra132. [CrossRef] [PubMed]
16. Kim, S.; Lieberman, T.D.; Kishony, R. Alternating Antibiotic Treatments Constrain Evolutionary Paths to Multidrug Resistance.
Proc. Natl. Acad. Sci. USA 2014, 111, 14494–14499. [CrossRef] [PubMed]
17. Nichol, D.; Jeavons, P.; Fletcher, A.G.; Bonomo, R.A.; Maini, P.K.; Paul, J.L.; Gatenby, R.A.; Anderson, A.R.A.; Scott, J.G. Steering
Evolution with Sequential Therapy to Prevent the Emergence of Bacterial Antibiotic Resistance. PLoS Comput. Biol. 2015,
11, e1004493. [CrossRef] [PubMed]
18. Fuentes-Hernandez, A.; Plucain, J.; Gori, F.; Pena-Miller, R.; Reding, C.; Jansen, G.; Schulenburg, H.; Gudelj, I.; Beardmore, R.
Using a Sequential Regimen to Eliminate Bacteria at Sublethal Antibiotic Dosages. PLoS Biol. 2015, 13, e1002104. [CrossRef]
[PubMed]
19. Pál, C.; Papp, B.; Lázár, V. Collateral Sensitivity of Antibiotic-Resistant Microbes. Trends Microbiol. 2015, 23, 401–407. [CrossRef]
20. Lázár, V.; Pal Singh, G.; Spohn, R.; Nagy, I.; Horváth, B.; Hrtyan, M.; Busa-Fekete, R.; Bogos, B.; Méhi, O.; Csörgo, B.; et al.
Bacterial Evolution of Antibiotic Hypersensitivity. Mol. Syst. Biol. 2013, 9, 700. [CrossRef]
21. Lázár, V.; Nagy, I.; Spohn, R.; Csörgő, B.; Györkei, Á.; Nyerges, Á.; Horváth, B.; Vörös, A.; Busa-Fekete, R.; Hrtyan, M.; et al.
Genome-Wide Analysis Captures the Determinants of the Antibiotic Cross-Resistance Interaction Network. Nat. Commun. 2014,
5, 4352. [CrossRef]
22. Suzuki, S.; Horinouchi, T.; Furusawa, C. Prediction of Antibiotic Resistance by Gene Expression Profiles. Nat. Commun. 2014,
5, 5792. [CrossRef]
23. Baym, M.; Lieberman, T.D.; Kelsic, E.D.; Chait, R.; Gross, R.; Yelin, I.; Kishony, R. Spatiotemporal Microbial Evolution on
Antibiotic Landscapes. Science 2016, 353, 1147–1151. [CrossRef] [PubMed]
24. Barbosa, C.; Trebosc, V.; Kemmer, C.; Rosenstiel, P.; Beardmore, R.; Schulenburg, H.; Jansen, G. Alternative Evolutionary Paths to
Bacterial Antibiotic Resistance Cause Distinct Collateral Effects. Mol. Biol. Evol. 2017, 34, 2229–2244. [CrossRef]
25. Imamovic, L.; Ellabaan, M.M.H.; Dantas Machado, A.M.; Citterio, L.; Wulff, T.; Molin, S.; Krogh Johansen, H.; Sommer, M.O.A.
Drug-Driven Phenotypic Convergence Supports Rational Treatment Strategies of Chronic Infections. Cell 2018, 172, 121–134.
[CrossRef] [PubMed]
26. Nichol, D.; Rutter, J.; Bryant, C.; Hujer, A.M.; Lek, S.; Adams, M.D.; Jeavons, P.; Anderson, A.R.A.; Bonomo, R.A.; Scott, J.G.
Antibiotic Collateral Sensitivity Is Contingent on the Repeatability of Evolution. Nat. Commun. 2019, 10, 334. [CrossRef] [PubMed]
27. Maeda, T.; Iwasawa, J.; Kotani, H.; Sakata, N.; Kawada, M.; Horinouchi, T.; Sakai, A.; Tanabe, K.; Furusawa, C. High-Throughput
Laboratory Evolution Reveals Evolutionary Constraints in Escherichia coli. Nat. Commun. 2020, 11, 5970. [CrossRef] [PubMed]
28. Wright, S. The Roles of Mutation, Inbreeding, Crossbreeding, and Selection in Evolution. Proc. Sixth Int. Congr. Genet. 1932, 1,
356–366.
29. Kauffman, S.; Levin, S. Towards a General Theory of Adaptive Walks on Rugged Landscapes Section of Ecology and Systematics,
and Ecosystems Research Center. J. Theor. Biol 1987, 128, 11–45. [CrossRef]
30. Furusawa, C.; Kaneko, K. Formation of Dominant Mode by Evolution in Biological Systems. Phys. Rev. E 2018, 97, 042410.
[CrossRef]
31. Sato, T.U.; Kaneko, K. Evolutionary Dimension Reduction in Phenotypic Space. Phys. Rev. Res. 2020, 2, 013197. [CrossRef]
Antibiotics 2024, 13, 94 21 of 24
32. Marguerat, S.; Schmidt, A.; Codlin, S.; Chen, W.; Aebersold, R.; Bähler, J. Quantitative Analysis of Fission Yeast Transcriptomes
and Proteomes in Proliferating and Quiescent Cells. Cell 2012, 151, 671–683. [CrossRef]
33. Matsumoto, Y.; Murakami, Y.; Tsuru, S.; Ying, B.W.; Yomo, T. Growth Rate-Coordinated Transcriptome Reorganization in Bacteria.
BMC Genom. 2013, 14, 808. [CrossRef]
34. Schmidt, A.; Kochanowski, K.; Vedelaar, S.; Ahrné, E.; Volkmer, B.; Callipo, L.; Knoops, K.; Bauer, M.; Aebersold, R.; Heinemann,
M. The Quantitative and Condition-Dependent Escherichia coli Proteome. Nat. Biotechnol. 2016, 34, 104–110. [CrossRef]
35. Charlebois, D.A. Quantitative Systems-Based Prediction of Antimicrobial Resistance Evolution. NPJ Syst. Biol. Appl. 2023, 9, 40.
[CrossRef]
36. Oz, T.; Guvenek, A.; Yildiz, S.; Karaboga, E.; Tamer, Y.T.; Mumcuyan, N.; Ozan, V.B.; Senturk, G.H.; Cokol, M.; Yeh, P.; et al.
Strength of Selection Pressure Is an Important Parameter Contributing to the Complexity of Antibiotic Resistance Evolution. Mol.
Biol. Evol. 2014, 31, 2387–2401. [CrossRef] [PubMed]
37. Horinouchi, T.; Minamoto, T.; Suzuki, S.; Shimizu, H.; Furusawa, C. Development of an Automated Culture System for Laboratory
Evolution. J. Lab. Autom. 2014, 19, 478–482. [CrossRef] [PubMed]
38. Lindsey, H.A.; Gallie, J.; Taylor, S.; Kerr, B. Evolutionary Rescue from Extinction Is Contingent on a Lower Rate of Environmental
Change. Nature 2013, 494, 463–467. [CrossRef] [PubMed]
39. Toprak, E.; Veres, A.; Yildiz, S.; Pedraza, J.M.; Chait, R.; Paulsson, J.; Kishony, R. Building a Morbidostat: An Automated
Continuous-Culture Device for Studying Bacterial Drug Resistance under Dynamically Sustained Drug Inhibition. Nat. Protoc.
2013, 8, 555–567. [CrossRef]
40. Maeda, T.; Kawada, M.; Sakata, N.; Kotani, H.; Furusawa, C. Laboratory Evolution of Mycobacterium on Agar Plates for Analysis
of Resistance Acquisition and Drug Sensitivity Profiles. Sci. Rep. 2021, 11, 15136. [CrossRef]
41. Toprak, E.; Veres, A.; Michel, J.-B.; Chait, R.; Hartl, D.L.; Kishony, R. Evolutionary Paths to Antibiotic Resistance under
Dynamically Sustained Drug Selection. Nat. Genet. 2011, 44, 101–105. [CrossRef]
42. Liu, P.C.; Lee, Y.T.; Wang, C.Y.; Yang, Y.T. Design and Use of a Low Cost, Automated Morbidostat for Adaptive Evolution of
Bacteria under Antibiotic Drug Selection. J. Vis. Exp. 2016, 2016, e54426. [CrossRef]
43. Fridman, O.; Goldberg, A.; Ronin, I.; Shoresh, N.; Balaban, N.Q. Optimization of Lag Time Underlies Antibiotic Tolerance in
Evolved Bacterial Populations. Nature 2014, 513, 418–421. [CrossRef] [PubMed]
44. Van Den Bergh, B.; Michiels, J.E.; Wenseleers, T.; Windels, E.M.; Vanden Boer, P.; Kestemont, D.; Meester, L.d.; Verstrepen, K.J.;
Verstraeten, N.; Fauvart, M.; et al. Frequency of Antibiotic Application Drives Rapid Evolutionary Adaptation of Escherichia coli
Persistence. Nat. Microbiol. 2016, 1, 16020. [CrossRef] [PubMed]
45. Van den Bergh, B.; Schramke, H.; Michiels, J.E.; Kimkes, T.E.P.; Radzikowski, J.L.; Schimpf, J.; Vedelaar, S.R.; Burschel, S.;
Dewachter, L.; Lončar, N.; et al. Mutations in Respiratory Complex I Promote Antibiotic Persistence through Alterations in
Intracellular Acidity and Protein Synthesis. Nat. Commun. 2022, 13, 1–18. [CrossRef]
46. Mechler, L.; Herbig, A.; Paprotka, K.; Fraunholz, M.; Nieselt, K.; Bertram, R. A Novel Point Mutation Promotes Growth Phase-
Dependent Daptomycin Tolerance in Staphylococcus Aureus. Antimicrob. Agents Chemother. 2015, 59, 5366–5376. [CrossRef]
[PubMed]
47. Michiels, J.E.; Bergh, B.V.D.; Verstraeten, N.; Fauvart, M.; Michiels, J. In Vitro Emergence of High Persistence upon Periodic
Aminoglycoside Challenge in the ESKAPE Pathogens. Antimicrob. Agents Chemother. 2016, 60, 4630–4637. [CrossRef] [PubMed]
48. Nicoloff, H.; Hjort, K.; Levin, B.R.; Andersson, D.I. The High Prevalence of Antibiotic Heteroresistance in Pathogenic Bacteria Is
Mainly Caused by Gene Amplification. Nat. Microbiol. 2019, 4, 504–514. [CrossRef]
49. Wang, Y.; Bojer, M.S.; George, S.E.; Wang, Z.; Jensen, P.R.; Wolz, C.; Ingmer, H. Inactivation of TCA Cycle Enhances Staphylococcus
Aureus Persister Cell Formation in Stationary Phase. Sci. Rep. 2018, 8, 10849. [CrossRef]
50. Cheng, Z.-X.; Yang, M.J.; Peng, B.; Peng, X.; Lin, X.; Li, H. The Depressed Central Carbon and Energy Metabolisms Is Associated
to the Acquisition of Levofloxacin Resistance in Vibrio Alginolyticus. J. Proteomics 2018, 181, 83–91. [CrossRef]
51. Su, Y.-B.; Peng, B.; Li, H.; Cheng, Z.; Zhang, T.; Zhu, J.; Li, D.; Li, M.; Ye, J.; Du, C.; et al. Pyruvate Cycle Increases Aminoglycoside
Efficacy and Provides Respiratory Energy in Bacteria. Proc. Natl. Acad. Sci. USA 2018, 115, E1578–E1587. [CrossRef]
52. Peng, B.; Su, Y.-B.; Li, H.; Han, Y.; Guo, C.; Tian, Y.M.; Peng, X.X. Exogenous Alanine and/or Glucose plus Kanamycin Kills
Antibiotic-Resistant Bacteria. Cell Metab. 2015, 21, 249–262. [CrossRef] [PubMed]
53. Angst, D.C.; Hall, A.R. The Cost of Antibiotic Resistance Depends on Evolutionary History in Escherichia coli. BMC Evol. Biol.
2013, 13, 163. [CrossRef] [PubMed]
54. Chevereau, G.; Dravecká, M.; Batur, T.; Guvenek, A.; Ayhan, D.H.; Toprak, E.; Bollenbach, T. Quantifying the Determinants of
Evolutionary Dynamics Leading to Drug Resistance. PLoS Biol. 2015, 13, e1002299. [CrossRef] [PubMed]
55. Zampieri, M.; Enke, T.; Chubukov, V.; Ricci, V.; Piddock, L.; Sauer, U. Metabolic Constraints on the Evolution of Antibiotic
Resistance. Mol. Syst. Biol. 2017, 13, 917. [CrossRef]
56. Zampieri, M.; Zimmermann, M.; Claassen, M.; Sauer, U. Nontargeted Metabolomics Reveals the Multilevel Response to Antibiotic
Perturbations. Cell Rep. 2017, 19, 1214–1228. [CrossRef]
57. Apjok, G.; Boross, G.; Nyerges, Á.; Fekete, G.; Lázár, V.; Papp, B.; Pál, C.; Csörgo, B.; Barlow, M. Limited Evolutionary Conservation
of the Phenotypic Effects of Antibiotic Resistance Mutations. Mol. Biol. Evol. 2019, 36, 1601–1611. [CrossRef]
58. Chait, R.; Craney, A.; Kishony, R. Antibiotic Interactions That Select against Resistance. Nature 2007, 446, 668–671. [CrossRef]
Antibiotics 2024, 13, 94 22 of 24
59. Munck, C.; Gumpert, H.K.; Wallin, A.I.N.; Wang, H.H.; Sommer, M.O.A. Prediction of Resistance Development against Drug
Combinations by Collateral Responses to Component Drugs. Sci. Transl. Med. 2014, 6, 262ra156. [CrossRef]
60. Rodriguez de Evgrafov, M.C.; Faza, M.; Asimakopoulos, K.; Sommer, M.O.A. Systematic Investigation of Resistance Evolution to
Common Antibiotics Reveals Conserved Collateral Responses across Common Human Pathogens. Antimicrob. Agents Chemother.
2021, 16, e01273-20. [CrossRef]
61. Taber, H.W.; Mueller, J.P.; Miller, P.F.; Arrow, A.S. Bacterial Uptake of Aminoglycoside Antibiotics. Microbiol. Rev. 1987, 51,
439–457. [CrossRef]
62. Allison, K.R.; Brynildsen, M.P.; Collins, J.J. Metabolite-Enabled Eradication of Bacterial Persisters by Aminoglycosides. Nature
2011, 473, 216–220. [CrossRef] [PubMed]
63. Paulsen, I.T.; Brown, M.H.; Skurray, R.A. Proton-Dependent Multidrug Efflux Systems. Microbiol. Rev. 1996, 60, 575–608.
[CrossRef] [PubMed]
64. Murakami, S.; Yamaguchi, A. Crystal Structure of Bacterial Multidrug Efflux Transporter AcrB. Tanpakushitsu Kakusan Koso. 2003,
48, 26–32. [CrossRef] [PubMed]
65. Suzuki, S.; Horinouchi, T.; Furusawa, C. Acceleration and Suppression of Resistance Development by Antibiotic Combinations.
BMC Genom. 2017, 18, 328. [CrossRef]
66. Lee, H.H.; Molla, M.N.; Cantor, C.R.; Collins, J.J. Bacterial Charity Work Leads to Population-Wide Resistance. Nature 2010, 467,
82–85. [CrossRef]
67. Iwasawa, J.; Maeda, T.; Shibai, A.; Kotani, H.; Kawada, M.; Furusawa, C. Analysis of the Evolution of Resistance to Multiple
Antibiotics Enables Prediction of the Escherichia coli Phenotype-Based Fitness Landscape. PLoS Biol. 2022, 20, 18–20. [CrossRef]
68. Michaux, C.; Ronneau, S.; Giorgio, R.T.; Helaine, S. Antibiotic Tolerance and Persistence Have Distinct Fitness Trade-Offs. PLoS
Pathog. 2022, 18, e1010963. [CrossRef]
69. Meylan, S.; Andrews, I.W.; Collins, J.J. Targeting Antibiotic Tolerance, Pathogen by Pathogen. Cell 2018, 172, 1228–1238. [CrossRef]
70. Windels, E.M.; Michiels, J.E.; van den Bergh, B.; Fauvart, M.; Michiels, J. Antibiotics: Combatting Tolerance to Stop Resistance.
MBio 2019, 10, e02095-19. [CrossRef]
71. Liu, Y.; Yang, K.; Zhang, H.; Jia, Y.; Wang, Z. Combating Antibiotic Tolerance Through Activating Bacterial Metabolism. Front.
Microbiol. 2020, 11, 577564. [CrossRef]
72. Boeck, L. Antibiotic Tolerance: Targeting Bacterial Survival. Curr. Opin. Microbiol. 2023, 74, 102328. [CrossRef]
73. Balaban, N.Q.; Merrin, J.; Chait, R.; Kowalik, L.; Leibler, S. Bacterial Persistence as a Phenotypic Switch. Science 2004, 305,
1622–1625. [CrossRef] [PubMed]
74. Spoering, A.L.; Vulić, M.; Lewis, K. GlpD and PlsB Participate in Persister Cell Formation in Escherichia coli. J. Bacteriol. 2006, 188,
5136–5144. [CrossRef]
75. Hansen, S.; Lewis, K.; Vulić, M. Role of Global Regulators and Nucleotide Metabolism in Antibiotic Tolerance in Escherichia coli.
Antimicrob. Agents Chemother. 2008, 52, 2718–2726. [CrossRef] [PubMed]
76. Girgis, H.S.; Harris, K.; Tavazoie, S. Large Mutational Target Size for Rapid Emergence of Bacterial Persistence. Proc. Natl. Acad.
Sci. USA 2012, 109, 12740–12745. [CrossRef] [PubMed]
77. Orman, M.A.; Brynildsen, M.P. Dormancy Is Not Necessary or Sufficient for Bacterial Persistence. Antimicrob. Agents Chemother.
2013, 57, 3230–3239. [CrossRef]
78. Kotte, O.; Volkmer, B.; Radzikowski, J.L.; Heinemann, M. Phenotypic Bistability in Escherichia coli’s Central Carbon Metabolism.
Mol. Syst. Biol. 2014, 10, 736. [CrossRef]
79. Gutierrez, A.; Jain, S.; Bhargava, P.; Hamblin, M.; Lobritz, M.A.; Collins, J.J. Understanding and Sensitizing Density-Dependent
Persistence to Quinolone Antibiotics. Mol. Cell 2017, 68, 1147–1154.e3. [CrossRef]
80. Shan, Y.; Brown Gandt, A.; Rowe, S.E.; Deisinger, J.P.; Conlon, B.P.; Lewis, K.; Formation, A.P.; Shan, Y.; Gandt, A.B.; Rowe, S.E.;
et al. Crossm Escherichia coli. MBio 2017, 8, 1–14.
81. Zalis, E.A.; Nuxoll, A.S.; Manuse, S.; Clair, G.; Radlinski, L.C.; Conlon, B.; Adkins, J.; Lewis, K. Stochastic Variation in Expression
of the Tricarboxylic Acid. Am. Soc. Microbiol. 2019, 10, 01930-19.
82. Melnikov, S.V.; Stevens, D.L.; Fu, X.; Kwok, H.S.; Zhang, J.T.; Shen, Y.; Sabina, J.; Lee, K.; Lee, H.; Söll, D. Exploiting Evolutionary
Trade-Offs for Posttreatment Management of Drug-Resistant Populations. Proc. Natl. Acad. Sci. USA 2020, 117, 17924–17931.
[CrossRef] [PubMed]
83. Baker, S.J.; Zhang, Y.; Akama, T.; Lau, A.; Zhou, H.; Hernandez, V.; Mao, W.; Alley, M.R.K.; Sanders, V.; Plattner, J.J. Discovery of
a New Boron-Containing Antifungal Agent, 5-Fluoro-1,3-Dihydro-1-Hydroxy-2,1-Benzoxaborole (AN2690), for the Potential
Treatment of Onychomycosis Stephen. Society 2006, 49, 4447–4450.
84. Rock, F.L.; Mao, W.; Yaremchuk, A.; Tukalo, M.; Crépin, T.; Zhou, H.; Zhang, Y.K.; Hernandez, V.; Akama, T.; Baker, S.J.; et al. An
Antifungal Agent Inhibits an Aminoacyl-TRNA Synthetase by Trapping TRNA in the Editing Site. Science 2007, 316, 1759–1761.
[CrossRef]
85. Jinna, S.; Finch, J. Spotlight on Tavaborole for the Treatment of Onychomycosis. Drug Des. Devel. Ther. 2015, 9, 6185–6190.
[CrossRef] [PubMed]
86. Zhang, P.; Ma, S. Recent Development of Leucyl-TRNA Synthetase Inhibitors as Antimicrobial Agents. Medchemcomm 2019, 10,
1329–1341. [CrossRef]
Antibiotics 2024, 13, 94 23 of 24
87. Si, Y.; Basak, S.; Li, Y.; Merino, J.; Iuliano, J.N.; Walker, S.G.; Tonge, P.J. Antibacterial Activity and Mode of Action of a
Sulfonamide-Based Class of Oxaborole Leucyl-TRNA-Synthetase Inhibitors. ACS Infect. Dis. 2019, 5, 1231–1238. [CrossRef]
88. Palencia, A.; Li, X.; Bu, W.; Choi, W.; Ding, C.Z.; Easom, E.E.; Feng, L.; Hernandez, V.; Houston, P.; Liu, L.; et al. Discovery of
Novel Oral Protein Synthesis Inhibitors of Mycobacterium Tuberculosis That Target Leucyl-TRNA Synthetase. Antimicrob. Agents
Chemother. 2016, 60, 6271–6280. [CrossRef]
89. Gudzera, O.I.; Golub, A.G.; Bdzhola, V.G.; Volynets, G.P.; Kovalenko, O.P.; Boyarshin, K.S.; Yaremchuk, A.D.; Protopopov, M.V.;
Yarmoluk, S.M.; Tukalo, M.A. Identification of Mycobacterium Tuberculosis Leucyl-TRNA Synthetase (LeuRS) Inhibitors among
the Derivatives of 5-Phenylamino-2H-[1,2,4]Triazin-3-One. J. Enzyme Inhib. Med. Chem. 2016, 31, 201–207. [CrossRef]
90. Gudzera, O.I.; Golub, A.G.; Bdzhola, V.G.; Volynets, G.P.; Lukashov, S.S.; Kovalenko, O.P.; Kriklivyi, I.A.; Yaremchuk, A.D.;
Starosyla, S.A.; Yarmoluk, S.M.; et al. Discovery of Potent Anti-Tuberculosis Agents Targeting Leucyl-TRNA Synthetase. Bioorganic
Med. Chem. 2016, 24, 1023–1031. [CrossRef]
91. Li, X.; Hernandez, V.; Rock, F.L.; Choi, W.; Mak, Y.S.L.; Mohan, M.; Mao, W.; Zhou, Y.; Easom, E.E.; Plattner, J.J.; et al.
Discovery of a Potent and Specific M. Tuberculosis Leucyl-TRNA Synthetase Inhibitor: (S)-3-(Aminomethyl)-4-Chloro-7-(2-
Hydroxyethoxy)Benzo[c][1,2]Oxaborol-1(3H)-Ol (GSK656). J. Med. Chem. 2017, 60, 8011–8026. [CrossRef]
92. Hu, Q.H.; Liu, R.J.; Fang, Z.P.; Zhang, J.; Ding, Y.Y.; Tan, M.; Wang, M.; Pan, W.; Zhou, H.C.; Wang, E.D. Discovery of a Potent
Benzoxaborole-Based Anti-Pneumococcal Agent Targeting Leucyl-TRNA Synthetase. Sci. Rep. 2013, 3, srep02475. [CrossRef]
93. O’Dwyer, K.; Spivak, A.T.; Ingraham, K.; Min, S.; Holmes, D.J.; Jakielaszek, C.; Rittenhouse, S.; Kwan, A.L.; Livi, G.P.; Sathe, G.;
et al. Bacterial Resistance to Leucyl-TRNA Synthetase Inhibitor GSK2251052 Develops during Treatment of Complicated Urinary
Tract Infections. Antimicrob. Agents Chemother. 2015, 59, 289–298. [CrossRef] [PubMed]
94. Gupta, A.; Monteferrante, C.; Rasina, D.; Leitis, G.; Randall, C.P.; Tomlinson, J.H.; Jirgensons, A.; Goessens, W.H.F.; Hays, J.P.;
O’Neill, A.J. A Polymorphism in LeuS Confers Reduced Susceptibility to GSK2251052 in a Clinical Isolate of Staphylococcus
Aureus. Antimicrob. Agents Chemother. 2016, 60, 3219–3221. [CrossRef] [PubMed]
95. Raymond, B. Five Rules for Resistance Management in the Antibiotic Apocalypse, a Road Map for Integrated Microbial
Management. Evol. Appl. 2019, 12, 1079–1091. [CrossRef]
96. de Vos, M.; Müller, B.; Borrell, S.; Black, P.A.; Van Helden, P.D.; Warren, R.M.; Gagneux, S.; Victor, T.C. Putative Compensatory
Mutations in the Rpoc Gene of Rifampin-Resistant Mycobacterium Tuberculosis Are Associated with Ongoing Transmission.
Antimicrob. Agents Chemother. 2013, 57, 827–832. [CrossRef]
97. McNally, A.; Oren, Y.; Kelly, D.; Pascoe, B.; Dunn, S.; Sreecharan, T.; Vehkala, M.; Välimäki, N.; Prentice, M.B.; Ashour, A.; et al.
Combined Analysis of Variation in Core, Accessory and Regulatory Genome Regions Provides a Super-Resolution View into the
Evolution of Bacterial Populations. PLoS Genet. 2016, 12, e1006280. [CrossRef]
98. Yoshida, M.; Reyes, S.G.; Tsuda, S.; Horinouchi, T.; Furusawa, C.; Cronin, L. Time-Programmable Drug Dosing Allows the
Manipulation, Suppression and Reversal of Antibiotic Drug Resistance in Vitro. Nat. Commun. 2017, 8, 15589. [CrossRef]
99. van Duijn, P.J.; Verbrugghe, W.; Jorens, P.G.; Spöhr, F.; Schedler, D.; Deja, M.; Rothbart, A.; Annane, D.; Lawrence, C.; Nguyen Van,
J.C.; et al. The Effects of Antibiotic Cycling and Mixing on Antibiotic Resistance in Intensive Care Units: A Cluster-Randomised
Crossover Trial. Lancet Infect. Dis. 2018, 18, 401–409. [CrossRef]
100. Podnecky, N.L.; Fredheim, E.G.A.; Kloos, J.; Sørum, V.; Primicerio, R.; Roberts, A.P.; Rozen, D.E.; Samuelsen, Ø.; Johnsen, P.J.
Conserved Collateral Antibiotic Susceptibility Networks in Diverse Clinical Strains of Escherichia coli. Nat. Commun. 2018, 9, 3673.
[CrossRef]
101. Hernando-Amado, S.; López-Causapé, C.; Laborda, P.; Sanz-García, F.; Oliver, A.; Martínez, J.L. Rapid Phenotypic Convergence
towards Collateral Sensitivity in Clinical Isolates of Pseudomonas Aeruginosa Presenting Different Genomic Backgrounds.
Microbiol. Spectr. 2023, 11, e0227622. [CrossRef]
102. Poole, K.; Gotoh, N.; Tsujimoto, H.; Zhao, Q.; Wada, A.; Yamasaki, T.; Neshat, S.; Yamagishi, J.I.; Li, X.Z.; Nishino, T. Overex-
pression of the MexC-MexD-OprJ Efflux Operon in NfxB-Type Multidrug-Resistant Strains of Pseudomonas Aeruginosa. Mol.
Microbiol. 1996, 21, 713–725. [CrossRef] [PubMed]
103. Zheng, E.J.; Stokes, J.M.; Collins, J.J. Eradicating Bacterial Persisters with Combinations of Strongly and Weakly Metabolism-
Dependent Antibiotics. Cell Chem. Biol. 2020, 27, 1544–1552.e3. [CrossRef] [PubMed]
104. Lopatkin, A.J.; Stokes, J.M.; Zheng, E.J.; Yang, J.H.; Takahashi, M.K.; You, L.; Collins, J.J. Bacterial Metabolic State More Accurately
Predicts Antibiotic Lethality than Growth Rate. Nat. Microbiol. 2019, 4, 2109–2117. [CrossRef]
105. Cornforth, D.M.; Dees, J.L.; Ibberson, C.B.; Huse, H.K.; Mathiesen, I.H.; Kirketerp-Møller, K.; Wolcott, R.D.; Rumbaugh, K.P.;
Bjarnsholt, T.; Whiteley, M. Pseudomonas Aeruginosa Transcriptome during Human Infection. Proc. Natl. Acad. Sci. USA 2018,
115, E5125–E5134. [CrossRef] [PubMed]
106. Kordes, A.; Preusse, M.; Willger, S.D.; Braubach, P.; Jonigk, D.; Haverich, A.; Warnecke, G.; Häussler, S. Genetically Diverse
Pseudomonas Aeruginosa Populations Display Similar Transcriptomic Profiles in a Cystic Fibrosis Explanted Lung. Nat. Commun.
2019, 10, 3397. [CrossRef] [PubMed]
107. Zheng, E.J.; Andrews, I.W.; Grote, A.T.; Manson, A.L.; Alcantar, M.A.; Earl, A.M.; Collins, J.J. Modulating the Evolutionary
Trajectory of Tolerance Using Antibiotics with Different Metabolic Dependencies. Nat. Commun. 2022, 13, 2525. [CrossRef]
108. Lázár, V.; Snitser, O.; Barkan, D.; Kishony, R. Antibiotic Combinations Reduce Staphylococcus Aureus Clearance. Nature 2022, 610,
540–546. [CrossRef] [PubMed]
Antibiotics 2024, 13, 94 24 of 24
109. Tamma, P.D.; Cosgrove, S.E.; Maragakis, L.L. Combination Therapy for Treatment of Infections with Gram-Negative Bacteria.
Clin. Microbiol. Rev. 2012, 25, 450–470. [CrossRef]
110. Sick, A.C.; Tschudin-Sutter, S.; Turnbull, A.E.; Weissman, S.J.; Tamma, P.D. Empiric Combination Therapy for Gram-Negative
Bacteremia. Pediatrics 2014, 133, e1148–e1155. [CrossRef]
111. May, A.K. An Argument for the Use of Aminoglycosides in the Empiric Treatment of Ventilator-Associated Pneumonia. Surg.
Infect. 2016, 17, 329–333. [CrossRef]
112. Bodey, G.P.; Elting, L.S.; Rodriguez, S. Bacteremia Caused by Enterobacter: 15 Years of Experience in a Cancer Hospital. Rev.
Infect. Dis. 1991, 13, 550–558. [CrossRef] [PubMed]
113. Kirkcaldy, R.D.; Weinstock, H.S.; Moore, P.C.; Philip, S.S.; Wiesenfeld, H.C.; Papp, J.R.; Kerndt, P.R.; Johnson, S.; Ghanem, K.G.;
Hook, E.W. The Efficacy and Safety of Gentamicin plus Azithromycin and Gemifloxacin plus Azithromycin as Treatment of
Uncomplicated Gonorrhea. Clin. Infect. Dis. 2014, 59, 1083–1091. [CrossRef]
114. Paul, M.; Benuri-Silbiger, I.; Soares-Weiser, K.; Leibovici, L. β Lactam Monotherapy versus β Lactam-Aminoglycoside Combina-
tion Therapy for Sepsis in Immunocompetent Patients: Systematic Review and Meta-Analysis of Randomised Trials. Br. Med. J.
2004, 328, 668. [CrossRef]
115. Wang, N.; Luo, J.; Deng, F.; Huang, Y.; Zhou, H. Antibiotic Combination Therapy: A Strategy to Overcome Bacterial Resistance to
Aminoglycoside Antibiotics. Front. Pharmacol. 2022, 13, 839808. [CrossRef] [PubMed]
116. Jahn, L.J.; Simon, D.; Jensen, M.; Bradshaw, C.; Ellabaan, M.M.H.; Sommer, M.O.A. Compatibility of Evolutionary Responses to
Constituent Antibiotics Drive Resistance Evolution to Drug Pairs. Mol. Biol. Evol. 2021, 38, 2057–2069. [CrossRef] [PubMed]
117. Lozano-Huntelman, N.A.; Bullivant, A.; Chacon-Barahona, J.; Valencia, A.; Ida, N.; Zhou, A.; Kalhori, P.; Bello, G.; Xue, C.; Boyd,
S.; et al. The Evolution of Resistance to Synergistic Multi-Drug Combinations Is More Complex than Evolving Resistance to Each
Individual Drug Component. Evol. Appl. 2023, 16, 1901–1920. [CrossRef] [PubMed]
118. Zhang, Q.Y.; Yan, Z.B.; Meng, Y.M.; Hong, X.Y.; Shao, G.; Ma, J.J.; Cheng, X.R.; Liu, J.; Kang, J.; Fu, C.Y. Antimicrobial Peptides:
Mechanism of Action, Activity and Clinical Potential. Mil. Med. Res. 2021, 8, 48. [CrossRef]
119. Lei, M.; Jayaraman, A.; Van Deventer, J.A.; Lee, K. Engineering Selectively Targeting Antimicrobial Peptides. Annu. Rev. Biomed.
Eng. 2021, 23, 339–357. [CrossRef]
120. Spohn, R.; Daruka, L.; Lázár, V.; Martins, A.; Vidovics, F.; Grézal, G.; Méhi, O.; Kintses, B.; Számel, M.; Jangir, P.K.; et al. Integrated
Evolutionary Analysis Reveals Antimicrobial Peptides with Limited Resistance. Nat. Commun. 2019, 10, 4538. [CrossRef]
121. Maron, B.; Rolff, J.; Friedman, J.; Hayouka, Z. Antimicrobial Peptide Combination Can Hinder Resistance Evolution. Microbiol.
Spectr. 2022, 10, e00973-22. [CrossRef]
122. Shaer Tamar, E.; Kishony, R. Multistep Diversification in Spatiotemporal Bacterial-Phage Coevolution. Nat. Commun. 2022,
13, 7971. [CrossRef] [PubMed]
123. Loc-Carrillo, C.; Abedon, S.T. Pros and Cons of Phage Therapy. Bacteriophage 2011, 1, 111–114. [CrossRef] [PubMed]
124. Elibe Mba, I.; Innocent Nweze, E. Antimicrobial Peptides Therapy: An Emerging Alternative for Treating Drug-Resistant Bacteria.
Yale J. Biol. Med. 2022, 95, 445–463.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.

Antibiotics 13 00094

Uploaded by

Copyright:

Available Formats

Antibiotics 13 00094

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Antibiotics 13 00094

Uploaded by

Copyright:

Available Formats

antibiotics

Keywords: antimicrobial resistance; laboratory evolution; collateral sensitivity

Citation: Maeda, T.; Furusawa, C.

Antibiotics 2024, 13, 94. https://doi.org/10.3390/antibiotics13010094 https://www.mdpi.com/journal/antibiotics

increases susceptibility to another, or cross-resistance, wherein resistance to one drug am-

Figure 1. Cross-resistance and collateral sensitivity. Cross-resistance refers to the phenomenon

2. Methodologies of Laboratory Evolution

3. Insights into AMR Mechanisms from Laboratory Evolution

ploy a comprehensive approach, integrating laboratory evolution and genomic sequencing

3.1. Genotype-Phenotype Relationships in AMR Evolution

Mutation Cross-Resistance Collateral Sensitivity

Mutation Cross-Resistance Collateral Sensitivity

3.3. Costs of AMR Evolution

respiro-fermentative metabolism [55]. Environmental conditions played a crucial role in de-

3.4. Impact of Multidrug Combinations on AMR Evolution

3.6. Population Dynamics of AMR Evolution

3.7. AMR Evolution in Spatially Structured Environments

3.8. Fitness Landscape and AMR Evolution

3.9. Antibiotic Tolerance and Persistence Development

4. Designing Rational Treatment Strategies through Laboratory Evolution

differences in evolutionary trade-offs [24]. Interestingly, diverse patterns of collateral sensi-

4.2. Collateral Sensitivity Cycling

4.3. Sequential Drug Regimens Based on Collateral Sensitivities

4.5. Suppression of Tolerance Acquisition by Cycling Antibiotics with Different

4.6. Long-Term Clearance Efficacy of Drug Combinations

4.8. Application of Antimicrobial Peptides (AMPs) for Combating AMR

Table 2. Three evolutionary-based strategies to combat AMR.

Strategy Advantages Disadvantages Clinical Trial

Author Contributions: Conceptualization, T.M.; writing—original draft preparation, T.M.;

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.