Modulators of gene amplification alter evolution of antibiotic resistance in Staphylococcus aureus

Kalinga Pavan T. Silva; Anthony M. Martini; Anupama Khare

doi:10.1371/journal.pgen.1012011

Abstract

Gene amplification is thought to be common in bacterial populations, providing a rapid and reversible mode of adaptation to diverse stresses, including the acquisition of antibiotic resistance. We previously showed that the opportunistic pathogen Staphylococcus aureus evolves resistance to the dual-targeting fluoroquinolone delafloxacin (DLX) that inhibits both the DNA gyrase and DNA topoisomerase IV via amplification of an efflux pump encoding gene sdrM. However, the pathways that control gene amplification, and consequently adaptive trajectories, remain understudied, especially in gram-positive bacteria like S. aureus. Here, we show that specific DNA repair and chromosomal separation proteins alter the frequency of gene amplification and selection of amplified regions in S. aureus. Through a screen of 40 mutants deficient in various DNA processes, we determined that while sdrM amplification was still the almost universal path to DLX resistance, other mutations that increased sdrM expression reduced the selection frequency of sdrM amplification, demonstrating the critical role of sdrM in DLX resistance. We found that similar to other bacteria, both sdrM amplification and loss of amplified gene copies required a functional RecA recombinase, but multiple other mutants in pathways required for amplification in other species still exhibited frequent sdrM amplification, suggesting that S. aureus may have alternate routes of gene amplification. Finally, loss of function mutants of the tyrosine recombinase XerC, that is known to play a role in chromosomal separation, were deficient for sdrM amplification, indicating that XerC is a novel modulator of gene amplification, or the maintenance or selection of amplified gene copies. Thus, our work sheds light on genetic factors that alter gene amplification-mediated evolutionary trajectories to antibiotic resistance in S. aureus and can potentially unlock mechanisms by which such evolution of resistance can be inhibited.

Author summary

Bacteria can become resistant to antibiotics by making tandem copies of segments of their DNA that contain resistance genes in a process called gene amplification. We had previously shown that the pathogen Staphylococcus aureus becomes resistant to the clinically relevant antibiotic delafloxacin (DLX) by gene amplification of sdrM, that codes for an efflux pump. Additional copies of sdrM increase the amount of DLX removed from the cell, resulting in S. aureus cells resistant to DLX. However, the mechanisms of gene amplification in S. aureus are not known. In this study, we used experimental evolution of a targeted panel of mutants to determine that two recombinase enzymes that mediate exchange of DNA, RecA and XerC, are critical for gene amplification in S. aureus. While RecA is required for gene amplification in other bacteria, XerC had not been previously implicated in gene amplification. We further found that mutations that increase expression of sdrM provide another path to DLX resistance, highlighting the importance of sdrM expression for DLX resistance in S. aureus. Our work sheds light on the vital determinants of gene amplification and can aid in the development of therapeutics against the emergence of antibiotic resistance in S. aureus.

Citation: Silva KPT, Martini AM, Khare A (2025) Modulators of gene amplification alter evolution of antibiotic resistance in Staphylococcus aureus. PLoS Genet 21(12): e1012011. https://doi.org/10.1371/journal.pgen.1012011

Editor: Sean Crosson, Michigan State University, UNITED STATES OF AMERICA

Received: November 5, 2025; Accepted: December 22, 2025; Published: December 31, 2025

This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

Data Availability: The whole genome sequencing (WGS) data have been deposited at NCBI Short Read Archive (SRA) under the bioproject PRJNA1242835. Raw data for figures is either in specific Supporting Tables, or in S1 Data.

Funding: This work was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research. The contributions of the NIH authors are considered Works of the United States Government. The findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

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

Introduction

Many antibiotics used to treat bacterial infections are growing increasingly ineffective as clinical and environmental bacteria acquire resistance via horizontal gene transfer or de novo mutation-mediated evolution [1,2]. Such antimicrobial resistance is a major threat to global public health [3,4]. Staphylococcus aureus, especially methicillin-resistant S. aureus (MRSA), commonly acquires resistance to a broad range of antibiotics, and is a major cause of antibiotic resistant infections world-wide [5,6].

While many instances of mutation-driven evolution of antibiotic resistance involve sequence alterations, gene amplification can also lead to antibiotic resistance via a variety of mechanisms [7,8]. Gene amplification has been previously implicated in resistance to multiple antibiotics in S. aureus. Duplication of the gene encoding the efflux pump norA led to ciprofloxacin resistance [9], amplification of the SCCmec cassette caused oxacillin resistance [10], and amplification of other genomic regions resulted in resistance against macrolides [11] and intermediate resistance against vancomycin [12].

Delafloxacin (DLX) is a dual-targeting antibiotic capable of inhibiting the activities of the DNA gyrase and DNA topoisomerase IV in multiple bacterial species, including S. aureus [8,13]. Such antibiotics that target two or more components in the cell should theoretically lead to a lower frequency of resistance evolution, as cells would need mutations in multiple targets for resistance [14,15]. In an earlier study, we identified two distinct mechanisms leading to high DLX resistance in S. aureus. The more rapid and prevalent evolutionary trajectory was via gene amplification of the gene encoding an efflux pump, SdrM, that increased DLX efflux from the cell, while the second adaptive path, that we observed in the absence of functional sdrM, was through mutations in both the canonical DLX targets, DNA gyrase and DNA topoisomerase IV [16].

Previous work has shown a significant reduction in duplication and amplification frequency in mutants deficient for the DNA recombinase RecA in multiple bacterial species upon exposure to selective pressures [17,18]. Apart from RecA, other proteins involved in DNA double-strand break (DSB) repair and some involved in other DNA metabolic processes are critical for gene amplification in E. coli [19]. However, factors that may alter the frequency of gene amplification, or maintenance and selection of amplified genomic loci, and thus evolutionary trajectories of antibiotic resistance, have not been characterized in gram-positive bacteria including S. aureus.

Here, we investigate the genes and pathways involved in sdrM gene amplification in S. aureus. We find that RecA is required for gene amplification in S. aureus. In the absence of functional RecA, S. aureus requires two steps to evolve DLX resistance, where the initial mutations partially compensate for the DNA damage sensitivity in the recA mutant, and are followed by mutations in sdrM or the canonical DLX targets. RecA also regulates the maintenance of amplified sdrM copies, as these are more stable in the absence of functional RecA. Further, through a targeted screen of mutants in various DNA repair and metabolism pathways, we observe that sdrM amplification remained the pervasive mechanism of DLX resistance, even in the absence of proteins involved in the DSB repair pathway that is required for gene amplification in E. coli. However, the evolution of point mutations that increase expression of sdrM selects against sdrM amplification, indicating that albeit less frequent, specific sequence alterations that achieve increased sdrM expression can lead to a lower selective advantage of sdrM amplification, and highlighting the importance of SdrM function in conferring DLX resistance. Finally, we find that the widely conserved tyrosine recombinase XerC, involved in chromosomal separation, is a novel effector of sdrM amplification.

Results

Resistance evolution via gene amplification requires the recombinase RecA

Gene amplification is thought to be a two-step process where an initial duplication event is mediated by either homologous or non-homologous recombination, and is followed by higher order amplification via RecA-mediated homologous recombination [8]. Thus, while the initial duplication may be RecA-dependent or independent, RecA is thought to be essential for the subsequent gene amplification [20]. To test whether this is also valid for gene amplification in S. aureus and determine how the absence of RecA alters evolutionary trajectories of DLX resistance, we evolved the S. aureus JE2 recA::Tn mutant from the Nebraska Transposon Mutant Library (NTML) [21] for DLX resistance. The recA::Tn mutant is ~ 56-fold more sensitive to DLX compared to the WT (S1 Fig). Fluoroquinolones cause DNA DSBs [22] and DLX is a dual-targeting fluoroquinolone that targets both the DNA gyrase and topoisomerase IV enzymes. Through a TUNEL assay, which allows for the detection for 3’-OH ends of fragmented DNA, we confirmed that DLX exposure also leads to DNA damage (S2 Fig). This DNA damage likely underlies the significant DLX hypersensitivity of the DNA repair defective recA::Tn mutant. We passaged three independent populations of recA::Tn in increasing concentrations of DLX, starting from an initial DLX concentration ~ 0.5x the MIC of the recA::Tn mutant (the MIC is ~ 0.006 µg/mL) and increasing the DLX concentration during subsequent passaging until the concentration was at least 8 µg/mL (see Methods for details). We performed whole genome sequencing (WGS) on populations from select intermediate passages as well as the terminal passage.

None of the three recA::Tn populations showed sdrM gene amplification in any of the sequenced passages (Fig 1A). Instead, mutations in the canonical targets DNA gyrase (gyrA, gyrB), and topoisomerase IV (parC, parE), and in and upstream of sdrM were prevalent in the later passages (Fig 1A and S1 Table). Interestingly, these mutations were selected during the evolution only once the evolving populations had reached a DLX resistance level similar to that of the wild type (WT), while mutations in genes associated with DNA repair were observed in the early passages. Individual isolates from each population (R1, R2, and R3 from populations 1, 2, and 3 respectively) from a passage prior to the emergence of the canonical mutations had DLX resistance similar to that of the WT (Fig 1B). Further, complementing RecA back into these isolates resulted in only an up to ~2-fold increase in DLX resistance, compared to the ~ 38-fold increase seen in the recA::Tn mutant (Fig 1C), suggesting that these isolates may have recouped the intrinsic DLX resistance lost due to the nonfunctional RecA via mutations in genes different from the canonical targets or sdrM.

Download:

Fig 1. RecA is required for sdrM gene amplification.

(A) Common mutations and sdrM amplification seen in three independently evolved populations of the recA::Tn mutant. Within each population, earlier to later passages are shown from left to right. Blue squares show the presence of a mutation in the loci listed on the y-axis, or sdrM amplification, as indicated. Dashed red lines indicate when the populations reached a resistance level similar to the WT. The populations from which the ‘intermediate isolates’ were selected are shown by black arrows. (B) DLX MICs of WT, the recA::Tn mutant, and the three intermediate isolates. (C) DLX MICs of WT, the recA::Tn mutant, and the three intermediate isolates, each carrying either an empty pKK30 plasmid, or one with the recA gene. (D) Presence of mutations (as shown by the blue squares) in the DNA gyrase and topoisomerase IV targets, sdrM, and DNA repair associated loci, as well as sdrM amplification, in three independently evolved populations each of the R1 intermediate isolate with either the pKK30 empty vector or one carrying recA. For each population, passages are shown chronologically from left to right. (B, C) Data shown are the mean ± standard deviation of three independent biological replicates. Significance is shown for comparison to (B) the recA::Tn mutant and (C) the respective empty vector strain as tested by Brown-Forsythe and Welch ANOVA tests followed by the Dunnett’s T3 multiple comparison’s test (* p < 0.05, ** p < 0.01, # p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.g001

To further clarify if RecA is required for gene amplification, we evolved three populations of the intermediate isolate R1 complemented with RecA, in parallel to three populations of R1 with an empty vector control. Given that the DLX MIC for R1 was similar to the WT, and it did not have any of the canonical mutations, we reasoned that if RecA was necessary for amplification, re-introducing a functional RecA and selecting for DLX resistance should lead to sdrM amplification. One of the three RecA-complemented populations showed sdrM gene amplification, while none of the empty vector control populations did (Fig 1D and S2 Table), further validating that RecA is required for sdrM amplification upon selection for DLX resistance.

Absence of functional RecA results in a two-step evolution of DLX resistance

RecA plays two important roles in DNA repair, first by activating the SOS response induced by DNA damaging agents, and second by acting as a critical effector of homologous recombination-mediated DSB repair [23,24]. Given that DLX also leads to DNA damage (S2 Fig), we hypothesized that the mutations in the DNA repair associated genes seen in the intermediate isolates may be reducing the DNA damage susceptibility of the recA::Tn mutant. Therefore, we tested the susceptibility of the intermediate isolates against multiple DNA damaging agents. We found that the intermediate isolates not only showed lower DLX susceptibility compared to the recA::Tn mutant, but also had reduced susceptibility against three other DNA damaging agents, a different fluoroquinolone ciprofloxacin (CPX) [25,26], the anti-cancer drug doxorubicin (DOXO) [27] and the chemotherapeutic mitomycin C (MMC) [28], compared to the hyper-sensitive recA::Tn mutant (Fig 2A).

Download:

Fig 2. Mutations in recA::Tn intermediate isolates reduce DLX and DNA damage sensitivity.

(A) MICs for ciprofloxacin (CPX), doxorubicin (DOXO), and mitomycin C (MMC) of WT, the recA::Tn mutant, and the three intermediate isolates. (B) The position of a G to A mutation (shown in red) in the putative LexA-binding site (shown in black) located between the genes B7H15_RS12090 and B7H15_RS12095 (P_090/095). (C) DLX MICs for WT and the indicated allele-replacement strains in the WT background. (D) DLX MICs for the recA::Tn mutant and the indicated allele-replacement strains in the recA::Tn background. (E) MICs for ciprofloxacin (CPX), doxorubicin (DOXO), and mitomycin C (MMC) of the recA::Tn mutant and the indicated allele-replacement strains in the recA::Tn background. (A, C-E) Data shown are the mean ± standard deviation of three independent biological replicates. Significance is shown for comparison to (A) the recA::Tn mutant, (C-E) the respective parental strain as tested by a one-way ANOVA with Dunnett’s test for multiple comparisons (* p < 0.05, ** p < 0.01, *** p < 0.001, # p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.g002

Given that the intermediate isolates showed reduced susceptibility to DLX and DNA damaging agents, but do not have canonical DLX resistance mutations, we investigated common mutations seen in these isolates, which were also seen in the early passages of the recA::Tn evolution (Fig 1A). All three isolates had coding sequence mutations in the genes encoding the helicase RexB, and a RecD2-like helicase, as well as an intergenic mutation upstream of the genes B7H15_RS12090 and B7H15_RS12095 (P_090/095) (Table 1). RexB is involved in the processing of double stranded breaks (DSBs) during DSB repair, similar to the activity of RecBCD in most gram-negative bacteria [23]. The two mutations observed in RexB, E48K and G576D, are both in residues that are conserved across species (S3A Fig), and the E48 residue in the Bacillus subtilis homolog AddB has been implicated in binding to DNA [29], indicating that mutations in these residues likely alter RexB function. While the role of the RecD2 helicase has not been characterized in S. aureus, RecD2 is present in other bacteria that lack RecBC and is thought to be involved in varied DNA repair processes [30–32]. In the intermediate isolates, we observed two different mutations, G360A and A227E, but the evolving populations showed numerous other coding sequence mutations, including A346D, T367P, T368T, T420M, D449G, Q480H, and N629K, which are all in conserved residues (S1 Table and S3B Fig). Given the diversity of residues, the effect of these mutations on DNA repair is unclear.

Download:

Table 1. Common mutations in the three recA::Tn evolved intermediate isolates.

https://doi.org/10.1371/journal.pgen.1012011.t001

The third mutation we observed in all three intermediate isolates was an identical G to A substitution in the intergenic region between the adjacent divergently encoded genes, B7H15_RS12090 and B7H15_RS12095 (P_090/095). B7H15_RS12090 is in the SOS regulon of S. aureus along with the two immediately downstream genes located in the same operon (B7H15_RS12085 and trhA) [23]. The intergenic mutation (P_090/095^G→A) lies within the predicted LexA binding site in this region [25] (Fig 2B) and may thus alter the regulation of these three genes.

To test the role of each mutation in conferring reduced DLX susceptibility in the recA::Tn background, we transferred rexB^G576D, and recD2^A227E, which were both found in two independently evolved intermediate isolates each, as well as the P_090/095^G→A mutation, individually to both the WT and recA::Tn backgrounds. We observed that while only the P_090/095^G→A mutation conferred a ~ 2-fold increase in DLX resistance in the WT background, (Fig 2C), all three mutations individually led to reduced DLX susceptibility in the DLX-hypersensitive recA::Tn mutant background (Fig 2D), accounting for their selection during the evolution.

We also tested the role of the individual mutations in alleviating DNA damage sensitivity. We found that in the recA::Tn background, while the recD2^A227E allele-replacement strain was less susceptible to all three DNA damaging agents, and the P_090/095^G→A mutation led to reduced susceptibility against DOXO and MMC, rexB^G576D did not significantly alter resistance against any of the agents, indicating a specific effect on DLX susceptibility in the absence of functional RecA (Fig 2E). Thus, in the absence of sdrM amplification, the recA::Tn mutant evolves DLX resistance via a two-step process: initial mutations associated with DNA repair genes that partially compensate for its DNA damage sensitivity, followed by point mutations in the DNA gyrase, DNA topoisomerase IV, and sdrM.

RecA is required for loss of amplified copies of sdrM

To test whether RecA plays a role in the instability of the amplified sdrM locus, we introduced the recA::Tn mutation to an evolved strain from our previous study, 1.7a, which had sdrM amplification leading to a ~ 140-fold increase in DLX resistance compared to the WT [16]. The 1.7a recA::Tn strain had a ~ 18-fold reduction in DLX resistance compared to the parental 1.7a strain, and a ~ 94-fold higher DLX resistance compared to the recA::Tn strain (Fig 3A). We passaged this strain in different concentrations of DLX, and found that unlike what we had previously seen in the 1.7a mutant [16] (shown here in the left panel of Fig 3B), the copy number of sdrM did not show significant variability in the 1.7a recA::Tn mutant (Fig 3B), suggesting that RecA likely plays a role in the expansion of the sdrM amplification. Further, we passaged 1.7a and 1.7a recA::Tn in DLX-free media for five passages and found that the normalized sdrM copy number compared to the initial passage remained significantly higher in the 1.7a recA::Tn strain compared to in 1.7a, where the copy number reached ~1 indicating an almost complete loss of the amplified copies of the locus (Fig 3C).These data validate that RecA is critical for the loss of amplified copies of sdrM. Given the stability of the amplified sdrM locus in the 1.7a recA::Tn strain, we tested its growth in DLX-free media to determine the effect of amplified sdrM copies on fitness. We found that while 1.7a showed similar growth compared to the WT, and the recA::Tn mutant grew slower than those strains, the 1.7a recA::Tn mutant showed substantially reduced growth even compared to the recA::Tn mutant, indicating a significant fitness cost of the amplified locus, at least in the absence of functional RecA (Fig 3D and S4 Fig).

Download:

Fig 3. Amplified segments are stabilized in the absence of RecA and have a fitness cost.

(A) DLX MICs for WT, and 1.7a, as well as recA::Tn mutants in both backgrounds. Data shown are the mean ± standard deviation of three independent biological replicates. (B) Normalized sdrM copy number (normalized to the copy number in the initial culture) for 1.7a (left) and 1.7a recA::Tn (right) upon passaging in either no DLX, or in two different DLX concentrations, corresponding to ~0.25x and ~0.5x the respective MICs. The data for 1.7a are from our previous study [16] licensed under the Creative Commons Attribution 4.0 International License and are shown here as sdrM copy number normalized to the initial value (instead of absolute sdrM copy number as in the original figure). Data shown are the mean ± standard deviation of two independent passaging experiments. (C) Normalized sdrM copy number (normalized to the copy number in the respective initial culture) for 1.7a and 1.7a recA::Tn upon extended passaging without DLX. The dotted lines denote an absolute copy number of 1 for the respective samples. Data from two independent passaging experiments are shown. (D) Growth curves shown as OD₆₀₀ measurements of WT, and 1.7a, as well as recA::Tn mutants in both backgrounds. Data shown are the mean ± standard error of the mean of three independent biological replicates. Significance is shown for comparison to (A) the 1.7a recA::Tn strain, and (C) the respective 1.7a passage as tested by (A) Brown-Forsythe and Welch ANOVA tests followed by the Dunnett’s T3 multiple comparison’s test or (C) a two-way ANOVA with Sidak’s multiple comparisons test (* p < 0.05, *** p < 0.001, # p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.g003

Mutant library screen shows that evolution of DLX resistance leads to prevalent sdrM amplification with diverse junctions

The absence of functional SdrM alters DLX resistance trajectories by requiring dual target mutations in the DNA gyrase and topoisomerase IV [16], while the absence of RecA leads to a two-step evolution of DLX resistance (Figs 1 and 2). To identify other genes that modulate the formation or selection of an amplified sdrM locus, and thereby the evolutionary trajectory of DLX resistance, we considered genes involved in DNA repair and recombination, the SOS regulon, and chromosomal separation, and those located in the genomic neighborhood of sdrM. We identified 39 genes of interest that had available loss of function transposon mutants in the NTML [21] (S3 Table). We also constructed two mutants that had lexA alleles (lexA^S130A and lexA^G94E) previously reported to encode uninducible versions of lexA, thereby inhibiting the SOS response [25,33,34]. We determined the DLX MICs of the corresponding mutants (S1 Fig) and for a majority of mutants, as well as a WT control, we evolved three independent populations in increasing DLX concentrations, starting from ~0.5x the respective MIC. For some mutants with very low MICs (xerC::Tn and recG::Tn), we evolved one population until its MIC was similar to the WT and then split that intermediate population into three independent populations for further evolution. We initially evolved the populations up to a DLX concentration of 2 µg/mL and tested for the presence of amplified sdrM using WGS. For the populations without sdrM gene amplification, the evolution was continued until a DLX concentration between 8–16 µg/mL.

Given the large numbers of populations evolved, we first analyzed the general trends for DLX resistance in S. aureus. Of the 126 independent populations evolved, 108 showed sdrM amplification, reinforcing the finding that sdrM amplification comprises a major evolutionary path to DLX resistance. 121 distinct genomic segments around sdrM were amplified, each found in only one evolved population, with the exception of two amplified segments that were found in two different independently evolved populations each (S4 Table). The amplified segments were diverse both in size and exact genomic coordinates of their ends (Fig 4A-4C). While most of the upstream ends of the fragments were located in a ~ 4.7kb kb region starting ~50 bp upstream of sdrM, all the downstream ends were located in a ~ 6 kb region starting ~3.2 kb away from sdrM (Fig 4D). These downstream ends were located almost exclusively in a tRNA-rRNA cluster, located four genes downstream of sdrM, resulting in these four genes also being present in most amplified segments. Our previous work had showed that of all the genes present in the amplified segments, DLX resistance depended solely on the overexpression of sdrM [16]. The prevalence of amplicon ends in the tRNA-rRNA cluster, skipping the intermediate genes, thus indicates that the tRNA-rRNA cluster may be a hot-spot for the initiation of amplification. Further, a synteny analysis of 28 genomes from different Staphylococcus species and 2 genomes from closely related Mammaliicoccus species (S5 Table) revealed that in all genomes, sdrM homologs were located between ~2–5 kb away from rRNA loci (S5 Fig), indicating that this proximity is conserved. Finally, the ends of the amplified segments had limited homology (Fig 4E), indicating a role of non-homologous recombination in the formation of these junctions, as observed in the WT in our previous study [16].

Download:

Fig 4. Amplified fragments containing sdrM are highly prevalent, diverse, and non-homologous, and originate in an rRNA-tRNA locus.

(A) All distinct amplified fragments containing sdrM seen in our evolved populations. The terminus downstream of sdrM in almost all the fragments is in the tRNA-rRNA locus. (B) Two examples of sdrM amplification each showing two copies of the amplified fragment (bordered by the brown lines). The one on the left is among the shortest amplified fragments, whereas the one on the right is significantly longer. (C) Length distribution of the amplified fragments containing sdrM. (D) Distribution of the distances of the downstream and upstream ends of the amplified fragments from the coding sequence of sdrM. The downstream ends are significantly further away as they are concentrated in the rRNA-tRNA locus, while the upstream ends start immediately upstream of sdrM. (E) Distribution of the length of homology between the two ends of the amplified fragments indicates a lack of significant homology. (F) Relative frequency of novel junctions representing a duplication (compared to rpoC levels), as measured by qPCR, in genomic DNA from successive passages of two independently evolved WT populations. For both populations, the junctions were detected in the passage containing 0.5 µg/mL DLX (shown by the black arrow).

https://doi.org/10.1371/journal.pgen.1012011.g004

Given the numerous distinct amplified segments we found, we wanted to determine whether these could be detected in the unselected parental strains. We first measured the detection limit for two novel junctions that were selected in one WT population each to be 1 in ~10⁴-10⁵ cells (S6A Fig). Further, we found that in these two WT populations, the junctions were first detected in populations being passaged at 0.5 µg/mL DLX, which is ~ 2x the WT MIC (Fig 4F), but not in the previous passages. Thus, if these junctions, which represent at least a duplication, were present in the unselected parental strain, they must exist at a frequency lower than 1 in 10⁴-10⁵ cells. Similarly, we found that the detection limit of two novel junctions that were found in one population each of two transposon mutants (spoIIIE::Tn and recQ2::Tn) was also 1 in ~10⁴-10⁵ cells (S6 B, C Fig). These junctions were also not detected in the respective parental strains, indicating a comparably low frequency of 1 in 10⁴-10⁵ cells, if they existed in the preselected parental strain. To identify any other junctions in this region in the unselected WT strain, we used an inverse PCR-like strategy, in conjunction with TOPO TA cloning (S7A Fig). We identified three different junctions from three independently grown WT cultures (S7B Fig), neither of which were present in any of our evolved populations. However, we could not detect them using junction-specific primers, indicating that if they existed in the WT strain, their frequency in the population was less than 1 in ~10⁴-10⁵.

Mutations increasing sdrM expression are associated with a lack of sdrM amplification

We next examined the mutations observed during the evolution, focusing on common mutations seen in more than one mutant background (S6 Table). Apart from amplification of sdrM, evolved populations from almost all mutants also showed increased coverage of a ~ 43 kb genomic region that contains multiple predicted phage proteins, as well as two toxin-antitoxin system toxins, and one anti-phage defense protein (S8A Fig). Using PHASTER, an online tool that identifies and annotates prophages [35,36], we determined that this region encodes an intact prophage (S8B Fig), suggesting that exposure to increasing DLX concentrations led to induction of this prophage.

Similar to our previous study [16], many evolved populations contained mutations in the canonical DNA gyrase and topoisomerase IV targets and in sdrM (Fig 5A). Apart from the previously identified coding sequence mutations in sdrM (A268S and Y363H) that increase DLX efflux and resistance [16], we also observed additional sdrM mutations that emerged including T180I, Y363F, Y363N, K390I, K390N, K390T, and an in-frame 3 bp deletion resulting in removal of the threonine at position 8 (S6 Table), which may be novel alleles of sdrM that confer DLX resistance. A few other common mutations were coding sequence and intergenic upstream alterations in genes associated with DNA damage and repair (recD2, dinG) [32,37], bacterial transcription (nusA) [38] and antibiotic resistance (norA) [9], which may play a role in DLX resistance at least in the specific mutant backgrounds where they were selected.

Download:

Fig 5. Amplification of sdrM is prevalent upon selection for DLX resistance but negatively associated with mutations increasing sdrM expression.

(A) Relative MICs (normalized to the WT) and the number of independently evolved populations that showed the mutations in the indicated loci, as well as sdrM amplification and increased prophage coverage, for WT and the screened mutants (a complete list of mutants is in S3 Table). (B) Common mutations (indicated by the blue squares) seen in the independently evolved populations that did not contain amplified sdrM. (C) Percentage of populations with or without sdrM amplification that had mutations in the indicated loci. (D) Common mutations seen in three independently evolved populations of an allele-replacement strain carrying the P_090/095^G→A mutation. Presence of mutations in the corresponding loci or sdrM amplification is indicated by a blue square. For each population, passages are shown from left to right in chronological order. (E) Normalized expression (compared to rpoC) of the indicated genes in the WT and the allele replacement strain carrying the P_090/095^G→A mutation. (F) DLX MICs for WT, the P_090/095^G→A mutant, and the sdrM::Tn mutant in both backgrounds. (E, F) Data shown are the mean ± standard deviation of three independent biological replicates. Significance is shown for comparison to (E) the respective WT sample as tested by ratio paired t-tests, and (F) the WT as tested by a one-way ANOVA with Dunnett’s test for multiple comparisons (* p < 0.05, ** p < 0.01, # p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.g005

Of the 126 evolved populations, 18 did not show sdrM amplification. In these populations, we observed mutations in the DLX targets (DNA gyrase and topoisomerase IV), coding and upstream mutations in sdrM, and mutations in the intergenic region upstream of the genes B7H15_RS12090 and B7H15_RS12095 (P_090/095) (Fig 5B). For the intergenic region, most populations had the same G to A substitution (P_090/095^G→A) in the LexA-binding site seen in the recA::Tn populations (Fig 2B), while one population had a single base-pair deletion at the same position. While the frequency of all these mutations was higher in the populations without sdrM amplification (Fig 5C), mutations upstream of sdrM, and in P_090/095 were almost exclusively seen in evolved populations without sdrM amplification. Additionally, in an allele-replacement strain carrying the P_090/095^G→A mutation, only one of three populations selected for DLX resistance showed sdrM amplification. Further, the amplified fragments emerged transiently and disappeared, which was a rare observation across any of our previous evolutions in other strains (Fig 5D and S7 Table). Thus, the presence of the P_090/095^G→A mutation likely reduces the selection of amplified sdrM fragments upon DLX exposure.

Three genes adjacent to this intergenic region are thought to be in the SOS regulon, and their expression is likely to be affected by mutations in the LexA binding site [25]. We found that the expression of these genes as well as the adjacent genes sdrM and B7H15_RS12095 is significantly higher in the presence of the P_090/095^G→A mutation, compared to the WT strain (Fig 5E). Further, while a P_090/095^G→A mutant has higher DLX resistance (Fig 2C), when we introduced the sdrM::Tn mutation into that background, the DLX resistance level decreased to that of an sdrM::Tn mutant, indicating that the increased resistance in the P_090/095^G→A mutant is likely solely due to sdrM overexpression (Fig 5F).

Additionally, in a mutant containing an sdrM intergenic mutation at the -164 position along with the A268S coding sequence mutation that we had previously studied [16], sdrM expression was significantly higher compared to a mutant containing only the A268S mutation (S9 Fig). The mutations we see in the sdrM intergenic region in our evolved populations are mostly clustered between positions -145 to -148 and -161 to -164 (S6 Table) and accordingly may also confer DLX resistance by increasing sdrM expression. Thus, mutations that increase sdrM expression likely reduce the selective advantage of sdrM amplification, providing alternate trajectories of DLX resistance evolution.

XerC, but not the DSB repair pathway, is a major modulator of sdrM gene amplification

Of the 41 mutants tested, 39 showed sdrM amplification in at least two of the three evolved populations (Fig 5A), suggesting that the proteins encoded by the genes that are mutated in these strains may not play a major role in amplification. However, minor roles cannot be excluded given the small number of independently evolved populations for each mutant. These proteins include RexA and RexB, which together play a role in DSB repair similar to RecBCD in E. coli, that has been implicated in gene amplification [19]. This suggests that the mechanisms of gene amplification in S. aureus may be different from those in E. coli.

We also tested mutants in multiple DNA polymerases, helicases, and nucleases that are thought to be involved in DNA repair and recombination [23,39], and most of these mutants amplified sdrM in at least two of the three evolved populations. Further, the two uninducible LexA alleles did not alter the frequency of gene amplification, indicating that the SOS response does not play a significant role in gene amplification, similar to what has been reported in E. coli [40]. However, two mutants, xerC::Tn, and B7H15_RS12060::Tn had only one independent population each with amplified sdrM, indicating a potential defect in gene amplification.

XerC is a widely conserved tyrosine recombinase that is critical for the separation of chromosomes [41], and B7H15_RS12060 encodes a putative P-Loop NTPase that is located in proximity to sdrM. To further test whether these genes played a role in gene amplification, we conducted a secondary screen. We evolved 12 independent populations of each mutant in a 96-deep well plate for 14 days, where each population was passaged daily in a serial dilution of DLX, and the population growing in the highest DLX concentration was propagated to the next passage (S10 Fig). As controls, we also performed a similar evolution for the WT, and for a mutant (B7H15_RS12085::Tn) that showed sdrM amplification in two out of three populations in our initial tube evolution (S6 Table).

For both the WT and the control mutant, nine out of the 12 evolved populations contained amplified sdrM (Fig 6A and S8 Table), indicating that mutants that amplified sdrM in two out of three evolved populations in the original tube evolution may not have major defects in gene amplification. For the B7H15_RS12060::Tn strain, only five out of twelve populations amplified sdrM, while none of the xerC::Tn mutant populations showed gene amplification (Fig 6A). Further, in this resistance evolution protocol, most evolved populations without sdrM amplification did not attain significant resistance (Fig 6B).

Download:

Fig 6. XerC is a novel determinant of amplification.

(A) Percentage of 12 independently evolved populations of the WT, RS12085::Tn, RS12060::Tn, and xerC::Tn that showed sdrM amplification, putative prophage induction, or mutations in the indicated genes. (B) Maximum DLX concentration that showed growth at each passage (shown as the respective number of generations) for the 12 independently evolved populations of WT, RS12085::Tn, RS12060::Tn, and xerC::Tn. Orange lines represent populations with sdrM amplification, while grey lines represent populations without sdrM amplification. (C) Percentage of 12 independently evolved populations of the xerC::Tn and RS12060::Tn strains complemented with either an empty vector or one containing the respective gene, that showed sdrM amplification, putative prophage induction, or mutations in the indicated loci. (D) Normalized sdrM copy number (normalized to the copy number in the initial culture) upon extended passaging without DLX for an evolved population each of WT, RS12060::Tn, and xerC::Tn showing sdrM amplification. Dotted lines denote the absolute copy number for the respective samples. Data shown are the mean ± standard deviation of two independent passaging experiments.

https://doi.org/10.1371/journal.pgen.1012011.g006

The B7H15_RS12060::Tn and xerC::Tn mutants have transposon insertions in the respective genes, but the insertions could possibly cause polar effects, especially for xerC, which is the first gene in an operon that contains hslUV, encoding a heat-shock responsive quality control protease [42], and codY, encoding a master regulator of metabolism and virulence [43]. To therefore test whether xerC and B7H15_RS12060 were involved in amplification, we complemented the respective genes back on a plasmid and repeated the deep well evolution (S8 Table). We observed no change in the frequency of gene amplification upon complementation of B7H15_RS12060, but an increased frequency upon xerC complementation, demonstrating that xerC is a determinant of gene amplification (Fig 6C). Finally, we passaged an evolved population containing amplified sdrM from each mutant background (from the tube evolutions) in DLX-free medium and observed no significant difference in the rate of loss of amplified copies of sdrM between the three strains (Fig 6D). However, while the absolute copy number in the WT and B7H15_RS12060::Tn backgrounds reached a value of 1 or less, indicating a complete loss of amplified copies of sdrM, the absolute copy number remained high in the xerC::Tn background, indicating a possible role of XerC in the maintenance of sdrM amplicons (Fig 6D).

The role of XerC in gene amplification is independent of RecA expression

To further investigate the role of XerC, we tested multiple hypotheses for the amplification defect in the xerC::Tn mutant. We first measured the growth characteristics of the xerC::Tn mutant and found that, similar to several other mutants that had low DLX MICs (S1 Fig), the xerC::Tn mutant had a slower growth rate compared to the WT even in the absence of DLX (Fig 7A). However, given that we saw prevalent sdrM amplification in all the other slow-growing mutants tested (rexB::Tn, recG::Tn, and recF::Tn), a reduced growth rate does not correlate with a deficiency in gene amplification.

Download:

Fig 7. XerC promotes gene amplification in a RecA-independent manner.

(A) The indicated strains were diluted 1:100 in fresh medium, and OD₆₀₀ was measured every 30 minutes. Data shown are the mean ± SEM for three independent biological replicates. (B) Percentage of 12 independently evolved populations of the recA::Tn strain complemented with either an empty vector or one containing P_recA::recA, and the ΔxerC mutant complemented with either an empty vector, or one containing P_xerC::xerC or P_tufA::recA, that showed sdrM amplification, putative prophage induction, or mutations in the indicated loci. (C) Maximum DLX concentration that showed growth at each passage (shown as the respective number of generations) for the 12 independently evolved populations of the indicated strains. Orange lines represent populations that showed sdrM amplification, while grey lines represent populations without sdrM amplification. (D) The relative carrying capacity (normalized to the carrying capacity of the no DLX control condition), which represents the maximum value of reporter induction (normalized fluorescence), of the recA promoter reporter is shown for both the WT and the ΔxerC mutant at DLX concentrations lower than the respective IC-40 where the strains showed robust growth. The rate of increase (k) of the relative carrying capacity is similar for both strains. The carrying capacity was determined from the induction curves shown in S12C Fig using Growthcurver [44] for fitting and analysis. Data shown are the mean ± standard deviation of four independent biological replicates.

https://doi.org/10.1371/journal.pgen.1012011.g007

Next, to eliminate any polar effects of the xerC::Tn mutation, we generated a clean deletion mutant ΔxerC. The ΔxerC mutant had a similar growth rate to the xerC::Tn mutant, which could be complemented with a XerC expressing plasmid (S11A Fig). Further, complementation also reversed the DLX sensitivity of the ΔxerC mutant (S11B Fig), indicating that both the growth defect and DLX hyper-sensitivity of the ΔxerC mutant were due to the lack of functional XerC. To directly test whether XerC was important for gene amplification, we repeated our deep-well evolution with the ΔxerC mutant containing either an empty vector, or one expressing xerC. As a control, we also evolved the recA::Tn mutant containing either an empty vector, or one expressing recA. Given the slower growth rates of both the recA::Tn (Fig 3D) and ΔxerC (S11A Fig) mutants, we allowed each passage to grow for 48 hours (instead of the 24 hours we had done previously). Similar to what we observed in our earlier experiments, none of the populations containing the empty vector for either mutant showed sdrM amplification, while upon complementation, 9 out of 12 recA::Tn populations, and 7 out of 12 ΔxerC populations amplified sdrM (Fig 7B, C). These results further validate that RecA and XerC are important for sdrM gene amplification, and that the amplification defect seen in both mutants is likely not due to their slower growth rates.

A recent study suggested that the xerC::Tn mutant shows lower levels of recA expression and SOS response induction upon exposure to multiple DNA damaging agents [45]. Given the requirement of RecA for sdrM amplification upon DLX exposure, we hypothesized that the amplification defect in the xerC mutants could be due to lower expression of recA. Thus, we tested fluorescence of a recA::dsRed promoter reporter in the WT and ΔxerC backgrounds upon exposure to DLX. Given the higher DLX sensitivity of a ΔxerC mutant, we first established the respective DLX concentrations that had similar growth inhibitory effects for both strains (see Methods for details). For each strain, we determined the time at which the strain first reached maximum growth in the absence of DLX. We then identified the lowest DLX concentration where the growth (OD₆₀₀) at that time point was below 60% of the maximum growth (inhibited by at least 40%) seen in the absence of DLX. We define this as the DLX IC-40 (Inhibitory concentration of 40%) for the respective strain (S12 A, B Fig).

We tested the recA promoter reporter at 2-fold DLX concentrations (from 0.5x to 0.0675x the respective IC-40) and found that the induction of the reporter was similar for both strains at most concentrations, with the ΔxerC mutant showing slightly higher induction at the lower DLX concentrations (S12C Fig). At the highest DLX concentration tested (0.5x the respective IC-40), the WT had a significantly higher induction compared to the ΔxerC mutant. However, the growth of the ΔxerC mutant was considerably inhibited at this DLX concentration (S12B Fig), which may lead to non-specific defects in gene expression. Further, the normalized rates of induction of recA, as measured by the slope of the fit for the relative carrying capacity of the induction (normalized to the no treatment condition) over the DLX concentration, were similar for both the WT and the ΔxerC mutant (Fig 7D). Thus, in our experimental conditions, recA promoter induction is not significantly inhibited in a ΔxerC mutant upon exposure to DLX.

To further test whether the defect in sdrM amplification seen in the ΔxerC mutant was due to a deficiency in recA expression, we introduced plasmids expressing recA either under its native promoter or under the constitutive tufA promoter [46] into the ΔxerC mutant. Overexpression of recA under either of these promoters increased DLX MICs back to WT levels in a recA::Tn mutant but had no effect on DLX MICs in the ΔxerC mutant, indicating that the DLX sensitivity of the ΔxerC mutant cannot be compensated by RecA overexpression (S12D Fig). Further, we repeated our deep-well evolution with a ΔxerC mutant strain containing a plasmid expressing recA under the tufA promoter (to bypass any native recA regulation). None of the evolved populations showed sdrM gene amplification, indicating that the amplification defect in a ΔxerC mutant is independent of recA levels, and cannot be reversed by recA overexpression (Fig 7B, C).

Discussion

Gene amplification can lead to resistance against many distinct antibiotics in S. aureus [8], but modulators of amplification in S. aureus have not been described. In this study we used adaptive evolution to investigate the genetic effectors involved in sdrM gene amplification, and the selection and maintenance of amplified segments, that lead to DLX resistance in MRSA (Fig 8). Similar to other bacterial species, we determined that functional RecA was required for sdrM gene amplification in S. aureus. Given the lack of homology between the ends of amplified fragments, it is likely that RecA is not required for the initial duplication, but for the subsequent steps of higher order amplification. Additionally, we found that RecA was also involved in the expansion and contraction of amplified segments (Fig 3B and 3C). Stable amplified copies of sdrM in the recA::Tn background led to a growth defect (Fig 3D, S4 Fig), indicating that in the absence of the selective pressure of DLX, maintenance of amplified sdrM has a fitness cost, at least without functional RecA. We also found that mutants in other genes and pathways required for amplification in E. coli still showed sdrM amplification indicating that S. aureus may have alternate or redundant mechanisms for gene amplification. Finally, we identified that XerC, which has not been previously implicated in gene amplification, is a novel effector of gene amplification in MRSA, as mutants lacking functional XerC showed reduced rates of sdrM amplification and evolution of DLX resistance.

Download:

Fig 8. Overexpression of sdrM is a critical determinant of DLX resistance in S. aureus.

Upon exposure to increasing concentrations of DLX, S. aureus evolves DLX resistance via overexpression of the gene encoding the SdrM efflux pump. This results either from frequent gene amplification of sdrM mediated by the RecA and XerC recombinases, or due to less frequent intergenic point mutations upstream of sdrM. Frequency of occurrence is denoted by the thickness of the arrows.

https://doi.org/10.1371/journal.pgen.1012011.g008

Mutations in several DNA repair-associated genes were observed when the recA::Tn strain was evolved both in the initial tube evolution (Fig 1A) and in the deep well evolution (Fig 7B). The rexB gene, which encodes a subunit of the AddAB helicase-nuclease that functions with RecA in DSB repair [23], was mutated almost exclusively in the recA::Tn background. While the effect of the specific RexB mutations will be the subject of future studies, it is possible that the mutations seen lead to altered enzyme activity that provide a selective advantage only when DSB repair is impaired in the absence of RecA.

Mutations in recD2, which encodes a helicase that has been implicated in various DNA repair functions in other species [30–32], were observed almost exclusively in mutants lacking recA, rexA, or rexB (S1, S6, S8 Tables), all of which are involved in DSB repair. Given the diversity of mutations seen in recD2, several of which are frameshift mutations, it is likely that the recD2 mutations seen may be loss of function alleles. The specific role of RecD2 in S. aureus has not been characterized, but our data suggest that while the normal function of RecD2 may be in DNA repair, upon exposure to DLX, RecD2 activity may have a fitness cost in the absence of DSB repair. This is supported by the fact that a recD2^A227E allele leads to DLX hypersensitivity in the WT background (Fig 2C) but alleviates the sensitivity to DLX and other DNA damaging agents in the absence of a functional RecA (Fig 2D, E).

Another common mutation seen in the early steps of the recA::Tn evolution, was the intergenic mutation located in the predicted LexA binding site [25] between the B7H15_RS12090 and B7H15_RS12095 genes (P_090/095) (Figs 1A, 2B). Three of the 16 previously reported SOS regulon genes in S. aureus are thought to be controlled by that LexA binding site [25], and we see increased expression of those genes in the P_090/095^G→A mutant (Fig 5E). While the function of these genes in DNA repair has not been investigated, and bioinformatic annotations do not reveal an obvious connection to DNA repair, their regulation by the SOS response, and reduced DNA damage sensitivity in a mutant that overexpresses these genes suggests a putative role in DNA repair. The P_090/095^G→A mutant also showed elevated expression of sdrM, which was immediately downstream of the SOS operon. Further work is necessary to determine whether sdrM is in the SOS regulon, and whether the transcriptional overexpression seen in this mutant is only due to disruption of LexA binding. In a previous study, a mutation in the same LexA-binding site was seen in isolates of the S. aureus COL strain that were selected for resistance against oxadiazoles, a new class of β-lactams [47]. SOS induction has been reported in S. aureus [33] and other species [48] upon exposure to β-lactams, indicating a role of this mutation and the mis-regulated genes in adaptation against diverse stresses.

In our evolved populations, we observed a negative association between sdrM amplification, and the LexA-binding site mutation or mutations in the upstream intergenic region of sdrM, both of which increase sdrM expression (Fig 5C). Further, a mutant carrying the P_090/095^G→A mutation showed only a transient amplification in one out of three populations upon selection for DLX resistance (Fig 5D). Such transient amplification was rare in most other strains where amplified sdrM copies became fixed in the population once they arose (S6 Table), indicating that sdrM amplification may not have a significant selective advantage when sdrM is already overexpressed. Further, while complementing RecA back to the parental recA::Tn mutant led to an amplification frequency similar to that seen for the WT (Fig 7B, C), complementing RecA to the recA::Tn intermediate isolate R1 did not lead to sdrM amplification in two of the three evolved populations (Fig 1D). This is likely because R1 had the P_090/095^G→A mutation and already expressed sdrM at a higher level. Thus, these sdrM overexpressing mutations represent an alternate route of DLX resistance to sdrM amplification. Notably, both these evolutionary trajectories of DLX resistance center around increased sdrM expression, underscoring the critical role of sdrM function in DLX resistance.

Of the 126 populations that were evolved for DLX resistance (Fig 5A), 108 populations showed gene amplification, while only 13 populations had at least one of the sdrM upstream or P_090/095 mutations (of which 2 also showed sdrM amplification). Thus, sdrM amplification is selected at a significantly higher frequency compared to the specific point mutations, consistent with what has been reported before for the relative frequencies of these two types of mutations [7].

The varied amplified sdrM fragments in our evolved populations showed limited homology between the fragment ends (Fig 4E). Similar lack of homology in amplified fragments has been reported in other species such as Acinetobacter sp., E. coli, and Salmonella, upon exposure to diverse selection pressures [18,19,49]. Almost all of our amplified fragments had one end nearly 4 kb downstream of sdrM in the tRNA-rRNA loci, while the other end was in a region starting immediately upstream of sdrM. This ~4 kb region contains four genes – lmrS and sepA, encoding two other efflux pumps, and genes encoding a P-loop NTPase, and a 61 amino acid hypothetical protein. We had previously shown that neither lmrS or sepA, nor a combination of all the four genes contributes to DLX resistance [16]. Further, here we show that mutants with transposon insertions in sepA or the P-loop NTPase encoding gene still show prevalent gene amplification (Fig 5A, 6A-C).These data indicate that amplification of this region is selected solely due to the presence of sdrM, and that the amplified fragments terminating in the tRNA-rRNA loci, and thus including the ~ 4 kb region, is a property of the amplification process. We postulate there are two, non-exclusive underlying reasons for this observation. First, amplified fragment ends in the tRNA-rRNA locus could have a higher selective advantage as amplified copies of this locus would lie upstream of sdrM and could lead to higher sdrM expression due to potential read-through transcription, similar to what has been seen in Streptococcus pneumoniae [50]. Second, the highly transcribed tRNA-rRNA clusters are thought to be hotspots for codirectional replication-transcription conflicts [51], and any resulting DNA damage may lead to a higher prevalence of duplications ending in this cluster. Future studies will focus on the mechanistic investigation of these possibilities. The proximity of sdrM and its homologs to rRNA-tRNA loci is conserved in many related species (S5 Fig), raising the possibility that amplification of these genes may be a phenomenon seen across species upon exposure to the appropriate selective pressures.

Almost all our evolved populations showed increased coverage of an intact prophage (Fig 5A, S8 Fig), likely indicating prophage induction due to DNA damage [52]. This prophage encodes a homolog of RecT, that has been implicated in recombination without homology or between short homologous sequences [53,54]. However, we did not detect significant prophage induction in a recT::Tn mutant (Fig 5A), but still saw sdrM amplification, indicating that while RecT likely plays a role in the induction of this prophage, sdrM amplification can occur in the absence of RecT and prophage induction. Further, there were multiple populations, e.g., in the recX::Tn, rarA::Tn, and ruvX::Tn mutants (S6 Table), where we saw prophage induction, but no sdrM amplification, suggesting these two events are unlinked.

In our broad screen for genetic determinants of gene amplification, of the 40 transposon mutants we evolved, only recA::Tn and xerC::Tn were significantly impaired in sdrM gene amplification (or maintenance or selection of amplified segments). DNA repair pathways in S. aureus have been only partly characterized, and the roles of several DNA repair proteins as well as redundancies present in various repair pathways have not been fully dissected [23,39]. The list of screened mutants included those with transposon insertions in genes involved in DSB repair, nucleotide excision repair, base excision repair, Holliday junction resolution, and cell division (S3 Table) [23,39]. In E. coli, it has been shown that mutants in DSB repair (recA, recB, ruvC) are necessary for amplification [19], while in Salmonella enterica Typhimurium, individual mutations in recB and recF did not significantly alter the frequency of gene duplication on a plasmid (higher order amplification was not tested), while a recA mutant and a recB recF double mutant showed significantly lower rates of duplication [18]. Our data show that unlike E. coli, mutants in RexA or RexB, which are components of the RexAB (AddAB) helicase/nuclease that performs a similar function as RecBCD in DSB repair, still show gene amplification. Given the requirement of RecA, this indicates that other helicase/nuclease combinations may function with RecA in homologous recombination during amplification. Further, while DNA polymerase I (PolA) is required for amplification in E. coli [19,55], the S. aureus polA::Tn mutant showed prevalent amplification. Similar to E. coli [19], mutations in DNA polymerases IV and V (DinB and UmuC) still allowed for amplification, as did mutations in another DNA polymerase encoded by RS08105 indicating that in S. aureus other polymerases may play a role in amplification, or there may be redundancies in polymerase function. Thus, gene amplification in S. aureus does not require DSB repair and is likely mediated by novel determinants.

One such determinant that we identified is the tyrosine recombinase XerC. We found that the lack of a functional XerC significantly reduced the frequency of sdrM amplification. XerC is a widely conserved protein involved in the segregation of daughter chromosomes during cell division, where it functions with the related recombinase XerD to bind dif sites near the chromosomal terminus, and with the DNA translocase FtsK [41,56]. In S. aureus, the xerC::Tn mutant has increased sensitivity to DNA gyrase inhibiting and cell-wall synthesis targeting antibiotics, and was reported to be deficient in the induction of recA [45]. However, any role of XerC in gene amplification remains unknown. The S. aureus chromosomal terminus is located around the genomic coordinate ~1.47 Mb [57], while sdrM is located around ~2.30 Mb, indicating that the S. aureus dif sites are not located in the vicinity of sdrM. However, the presence of cryptic dif-like or other XerC target sites near the sdrM locus cannot be ruled out and will be explored further. In S. aureus, XerD is an essential protein, unlike in other species, suggesting that it may have additional functions [56]. We found that the ΔxerC mutant did not have a defect in recA induction upon exposure to DLX in our experimental conditions, and that overexpression of RecA did not complement the ΔxerC defect in gene amplification. Further, mutants in other genes that are critical for chromosomal separation, noc::Tn and scpB::Tn [58–60], still showed prevalent amplification (Fig 5A), indicating that a defect in chromosomal separation was not sufficient to inhibit gene amplification. Thus, XerC plays an unknown, likely RecA-independent, role in gene amplification, possibly via promoting recombination or resolution of recombination intermediates which may be mediated by cryptic dif-like or other XerC target sites, and future work will investigate this mechanism further.

In our tube evolutions, one out of the three xerC::Tn populations amplified sdrM albeit in the later passages (S6 Table), while none of the 30 total populations of the xerC::Tn or ΔxerC mutants evolved in deep-well plates showed sdrM amplification. This is likely due to the difference in selective pressures in the two evolution schemes. In the tube evolution, populations are passaged in increasing concentrations of DLX, allowing for longer times for growth if required, leading to a high selective advantage of, and thus the selection of, DLX resistance mutations even if they arise at a very low frequency or have significant fitness defects. However, in the deep-well evolution scheme, at every passage, populations are grown in a two-fold serial dilution series of DLX and can thus continue to be passaged at the same DLX concentration without gaining additional DLX resistance. Further, due to the higher media volume in the tubes, population sizes are larger in the tube evolution, possibly allowing for the selection of highly infrequent mutation or amplification events.

Gene amplification can lead to rapid evolution of resistance against many classes of antibiotics, including ones that have multiple targets in the cell [8,16]. Pathways required for such amplification can thus be putative targets in combination with antibiotics, to lower the frequency of resistance evolution. We show that evolution of DLX resistance is inhibited in S. aureus in the absence of XerC, and further work is required to evaluate its potential as a therapeutic target. Additionally, our data suggest that, apart from RecA and XerC, genes with redundant function may play a role in gene amplification in S. aureus, and that such genes and pathways may differ between bacterial species. Combinatorial mutagenesis, and genome-wide screens may thus be critical to identify additional determinants of gene amplification.

Methods

Strains and growth conditions

All strains and plasmids used in this study are listed in S9 Table.

For experiments in liquid media, bacteria were grown at 37^oC, shaking at 300 rpm in modified M63 media (13.6 g/L KH₂PO₄, 2g/L (NH₄)₂SO₄, 0.4 μM ferric citrate, 1mM MgSO₄; pH adjusted to 7.0 with KOH) supplemented with 0.3% glucose, 0.1 μg/mL biotin, 2 μg/mL nicotinic acid, 1 × Supplement EZ (Teknova) and 1 × ACGU solution (Teknova) [16]. For experiments with 96-well or deep well plates, the plates were incubated in a Titramax 1000 (Heidolph) incubator at 37^oC shaking at 900 rpm. To construct and mutate plasmids and strains, cells were grown in LB liquid medium (10 g/L bacto-tryptone, 5 g/L yeast extract, 10 g/L NaCl) at 25^oC, 30^oC, or 37^oC as indicated, shaking at 300 rpm, or on LB plates (15 g/L agar) supplemented with the appropriate antibiotics (10 μg/mL chloramphenicol, 50 μg/mL kanamycin, 10 μg/mL trimethoprim, 10 μg/mL erythromycin) or 0.4% para-chlorophenylalanine (PCPA).

Minimum inhibitory concentration measurements

MICs were tested as described previously in modified M63 media [16]. Briefly, a 96-well flat clear bottom plate (Corning) was used to prepare two-fold serial dilutions with eight concentrations of DLX. Cells grown overnight in modified M63 were diluted 1:5000 in fresh modified M63 and added at a 1:1 ratio to the 96-well plate with the DLX dilutions such that the final dilution of the cells was 1:10000. Cells were grown for 24 hours in a Titramax 1000 (Heidolph) incubator at 37^oC, shaking at 900 rpm. The recA::Tn, 1.7a recA::Tn, rexA::Tn, and xerC::Tn strains in Fig 3A and S1 Fig, were grown for an additional 24 hours to compensate for their slower growth. Following the growth, OD₆₀₀ was measured using a microplate reader (Biotek Synergy H1) and the MIC values were determined by fitting the OD₆₀₀ vs. DLX concentration to a modified Gompertz function [61].

Growth curves and analysis

Cells were grown overnight in tubes with 2 mL of modified M63, diluted 1:100 in 100μL or 200μL modified M63 and added to a COSTAR clear bottom 96-well plate. OD₆₀₀ was measured using a microplate reader (Biotek Synergy H1) every 30 min for the indicated time, while shaking at 800 rpm at 37°C. The growth curves were analyzed with the R package Growthcurver to determine growth rate, doubling time, the time to reach the maximum growth rate, and carrying capacity [44].

Experimental evolution of DLX resistance in tubes

Evolution of S. aureus JE2 cells in tubes was carried out as described previously [16]. Briefly, for most strains, three independent colonies were picked and cultured in 2 mL of modified M63. After overnight growth, cells were transferred to new tubes containing 2 mL modified M63 with DLX at a concentration ~ 0.5x the respective MIC. After 24 h growth, 40 μL cells were transferred to tubes containing twice the concentration of DLX as the prior passage in 2 mL modified M63 and this was repeated up to a DLX concentration of 2 μg/mL. If sufficient growth was not visually apparent after 24 hours, cells were given an additional 24–48 hours to grow. If growth was still not seen, the evolution was reset to the previous passage and the DLX increment was adjusted depending on the growth. If gene amplification was not detected in the final passage (2 μg/mL), evolutions were continued similarly and carried out until a terminal concentration of 8 or 16 μg/mL.

For mutants (xerC::Tn and recG::Tn) which had very low MICs (<10-fold the WT MIC), a single population was evolved until the DLX concentration in the evolution approached the JE2 WT MIC (~0.22 μg/mL). For the next passage, cells were cultured with a dilution of 1:50 into three independent tubes containing twice the previous DLX concentration, and subsequently these three populations were independently evolved as mentioned above. For the recA::Tn mutant, we sequenced intermittent passages beginning from those that grew in a DLX concentration one-tenth of the WT MIC, i.e., 0.025 µg/mL DLX. For the other tube evolutions, we sequenced the passage that grew at 2 µg/mL, and if subsequent evolutions were performed, we sequenced intermittent passages, as well as the terminal one.

Experimental deep well evolution of DLX resistance

Mutants that needed to be tested further for amplification frequency were streaked out on selective plates, and 12 independent colonies per strain were cultured in 400 μL of modified M63 in deep well plates (Nunc 96 DeepWell Polystyrene Plates) for 24 h with shaking at 37^oC. For the xerC::Tn mutant, the original evolved population (prior to splitting the evolution into three independent populations as described above), that had an MIC similar to the WT, was streaked out on a plate. A single colony (XC5_7) was re-streaked out, and 12 colonies from this plate were cultured to start the 12 independent populations. 4 μL cells were then transferred to a fresh deep well plate containing 400 μL of modified M63 with 8 serially diluted DLX concentrations, where the maximum DLX concentration was set at 1–2 μg/mL for the data shown in Fig 6A-C, and for the complemented versions of recA::Tn and ΔxerC in Fig 7B, C, and at 0.03125-0.125 μg/mL for the other strains shown in Fig 7B, C. This meant that each independent population was cultured in all the 8 DLX concentrations. For the subsequent passages, 8μL of cells in the well at the highest DLX concentration that showed growth (defined by OD₆₀₀ > 0.400 as measured in a Biotek Synergy H1 microplate reader) were transferred into a new deep well plate containing the same DLX concentration profile and grown for 24 h. The maximum DLX concentration for the subsequent passages was increased if there was growth in the 2^nd well of the previous passage. The evolution was continued for 14 passages in total for the strains in Fig 6A-C and 11 passages for the strains in Fig 7B,C, and the genomic DNA of the terminal passage was sent out for whole genome sequencing.

Whole genome sequencing

Genomic DNA was extracted from cultures using the Qiagen DNeasy Blood and Tissue kit. Sequencing libraries were prepared using the Illumina Nextera XT DNA Library Preparation kit and sequenced (75 cycle or 100 cycle) on the Illumina NextSeq 550 (75 cycle) or the NextSeq 2000 (100 cycle) instruments to obtain single-end reads. The FASTQ files were processed by removing adapters and trimming using fastp v0.23.2. Sequences were aligned, variants called, and novel junctions identified using breseq v0.37.17 [62]. The mutations were marked as present in the respective figures (Figs 1A, 1D, 5A-D) if at least 30% of the population contained the particular mutation. Copy number variation in the efflux pump sdrM was identified as described previously [16]. Briefly, we used the BAM2COV command in breseq to determine the read coverage depth across the whole genome. The sdrM copy number was determined by dividing the average coverage depth of sdrM with the average coverage depth of the whole genome (both normalized to length). The threshold used to define sdrM amplification was the presence of a novel junction around sdrM and an sdrM copy number greater than 1.3x the WT sdrM copy number. Gene maps shown in Fig 4 and S7 Fig were drawn using gggenes [63].

Construction of allele replacement, overexpression, and reporter strains in S. aureus JE2

Construction of the allele replacement mutants and overexpression strains in JE2 was done as described previously [16]. Briefly, for the allele replacement, the new allele with flanking homology, or upstream and downstream regions with flanking homology for the ΔxerC mutant, were cloned into the pIMAY* plasmid [64], all of which had been PCR amplified. The plasmid was then electroporated into S. aureus JE2. Strains carrying the integrated plasmid were selected at 37^oC (the selective temperature), the plasmids were subsequently excised via PCPA counter-selection, and plasmid loss was confirmed via testing for chloramphenicol sensitivity. The final mutant was confirmed by flanking PCR and Sanger sequencing.

To construct the overexpression strains, the appropriate genes with their native promoters were cloned into the pKK30 vector [65] and electroporated into S. aureus JE2. To bypass the native regulation of recA and constitutively express it, recA was expressed from a tufA promoter [46] fused to the gyrA leader sequence. A gene fragment containing the tufA promoter (289 bp region ending 45 bp upstream of the start codon) fused to the regulatory region of gyrA (38 bp immediately upstream of the start codon) was ordered from IDT (sequence provided by the lab of Shaun Brinsmade, Georgetown University). The native recA promoter from pKK30-recA was replaced by this gene fragment.

To construct pKK15A, the backbone was derived from pKK30 [65], removing the native R6K origin and the P_sarA1::dfrA (trimethoprim resistance marker). The kanamycin resistance cassette (aphA3) was derived from the pKAN plasmid from the NTML library toolbox [66]. The p15A origin was derived from pIMAY* [64]. Column-purified PCR fragments were concatenated using a three-fragment Gibson assembly and transformed into E. coli IM08B to generate pKK15A. The P_sarA1::dsRed3.T3 construct was amplified from pKM16 [67] and inserted into pKK15A to generate pKK15A P_sarA1::dsRed3.T3, and P_sarA1 was subsequently replaced by P_recA to generate the recA promoter reporter (pKK15A P_recA::dsRed3.T3).

Primers used for constructing all mutants and overexpression strains are listed in S10 Table.

Mutagenesis of lexA gene

The Agilent QuikChange Lightning Site-Directed Mutagenesis Kit was used as per the manufacturer’s instructions to generate the lexA mutant alleles. Mutagenic primers were created utilizing the Agilent web-based QuikChange Primer Design tool and are listed in S10 Table.

Phage transduction

The donor strains (recA::Tn or sdrM::Tn) were grown overnight in LB, diluted 100-fold in fresh LB the following day, and grown at 30^oC for 1.5-2 h. Then, 1 mL of 10 mg/mL CaCl₂ and 10 μL of 10¹⁰ pfu/mL bacteriophage 85 (phi85) suspension was added to this culture which was incubated at room temperature for 30 min without shaking and then rotated slowly (80 rpm) at 37^oC for 5–6 h or overnight. Afterwards, the culture was filtered using a 0.45 μm filter and the phage-containing filtrate was stored at 4^oC.

To transduce the transposon mutations to the appropriate JE2 backgrounds, we grew the recipient cells overnight in 2 mL LB, and collected the cell pellets the following day by centrifugation. Cell pellets were resuspended in 300 μL of LB broth with 5 mM CaCl₂ and 700 μL of the donor phi85 phages were added. This mixture was incubated at 37^oC without shaking for 20 min. Cells were then immediately washed two times with 1mL of 40 mM sodium citrate and finally plated on LB + erythromycin plates containing 200 μM sodium citrate. The transposon insertion was confirmed via PCR and in the case of the P_090/095^G→A mutant with the sdrM::Tn mutation, a second PCR was conducted to confirm the presence of the P_090/095^G→A mutation to rule out potential transduction-based replacement due to the proximity of this mutation to the chromosomal position of sdrM.

Synteny analysis

The protein sequence of SdrM was analyzed by BLASTP [68] against the ClusteredNR database (nr_cluster_seq). The Constraint-based Multiple Alignment Tool was selected for the top 100 searches under default parameters, and for each strain cluster the top candidate with an assembled genome was chosen for the alignment. The corresponding genomes were identified, and using the NCBI nucleotide database, the distance between the closest rRNA/tRNA locus and the start codon of sdrM was manually determined.

Fluorescent promoter-reporter assay

Cells were grown overnight in LB media with 50 μg/mL kanamycin to maintain the pKK15a-dsRed plasmid. To minimize the formation of cell clumps that will disrupt the fluorescent readings LB media was chosen. The following day an 8-well 2-fold dilution series of DLX was made in LB + Kanamycin, in a 96-well Black clear flat bottom plate (COSTAR), and blank controls were also included. Overnight cells were diluted 1:100 into this plate and grown in a microplate reader (Synergy) at 37^oC for 24 hours. At each 20-minute interval the fluorescence (excitation 560 nm and emission 587 nm) and optical density (600nm) were measured. The first 200 minutes of the data points were discarded because of high noise due to lag phase growth. The growth for each concentration tested for each mutant was fitted to a non-logistic curve using the R package Growthcurver [44]. The fitted values for the carrying capacity were averaged across the four independent biological replicates for the no-DLX condition (for WT and the ΔxerC mutant) and the time at which the optical density of the mean growth for the no-DLX condition reached the mean carrying capacity was determined. We set a threshold to define robust growth as an OD₆₀₀ value at this time-point that was ≥ 60% of the mean carrying capacity for the no-DLX condition. The lowest DLX concentration that inhibited growth more than 40% at this time point was defined as the IC-40, i.e., the lowest DLX concentration whose OD₆₀₀ at this time-point was less than the threshold. All treatment conditions that showed a higher optical density than the threshold at this time-point were deemed as having robust growth and considered for subsequent analysis.

The ratio of the background subtracted fluorescence to the corresponding optical density at each time-point was defined as the normalized fluorescence. The average normalized fluorescence for all the selected DLX concentrations and the no-DLX condition across all four independent biological replicates were fit to logistical growth curves using the R package Growthcurver [44], and the average carrying capacity for each condition, which signifies the maximum normalized fluorescence reached, was determined. To compare the two strains directly, the relative carrying capacity was determined where the carrying capacity for each DLX concentration was normalized to the respective no-DLX control. The slope (k) of the datapoints in Fig 7D was calculated by a linear fit of the relative carrying capacity of the normalized fluorescence using the exponential growth with log(population) equation for nonlinear regression analysis in GraphPad Prism.

Quantitative PCR (qPCR)

For quantifying sdrM copy number, qPCR was done as described previously [16]. Briefly, gDNA samples diluted to 10 ng/µL were mixed with appropriate qPCR primers and Applied Biosystems Power SYBR Green PCR Master Mix (Thermo Scientific) in a Microamp EnduraPlate Optical 384 Well Clear Reaction Plate (Thermo Fisher Scientific) and the qPCR was carried out in a QuantStudio 5 real-time PCR machine (Thermo Fisher Scientific). We used rpoC as the control housekeeping gene. To determine the limit of detection for the junctions, 1 μL gDNA of the evolved population (10 ng/µL) was serially diluted 10-fold in 9 μL of genomic DNA from the parental WT strain (10 ng/µL) seven consecutive times. Then the qPCR was carried out as described above.

Reverse transcription quantitative PCR (RT-qPCR)

To determine the expression level of genes neighboring the lexA binding site including sdrM, cells grown overnight in tubes with 2 mL modified M63 were diluted 1:100 in 10 mL of modified M63 in a flask and grown for an additional 3–4 hours to an OD₆₀₀ of 0.3-0.4. For the sdrM allele-replacement strains, we similarly grew the cells to an OD₆₀₀ of ~0.5 and also took a sample of overnight grown cells. 2x volume of RNAprotect Bacteria Reagent (QIAGEN) was added to the culture and after 10 min incubation at room temperature, cells were centrifuged at 4000 rpm, the supernatant removed, and the pellets stored at -80°C. RNA was extracted using the NORGEN BIOTEK Total RNA purification kit and genomic DNA was removed with the TURBO DNA-free Kit (Invitrogen). Absence of genomic DNA was confirmed via PCR. cDNA was synthesized utilizing the Superscript III Reverse Transcriptase (Thermo Fisher Scientific) with random primers. The subsequent qPCR was carried out as described in the previous section.

TOPO TA cloning

Genomic DNA used as a template for the PCR reactions was concentrated to ~200 ng/µL and 2 µL of it was added to each 23 µL PCR reaction. PCR reactions were performed using the OneTaq or LongAmp polymerases, purified with the Zymo Clean and Concentrator-5 kit, and cloned into the pCR4-TOPO TA plasmid using the TOPO TA cloning kit for Sequencing, with One Shot TOP10 Chemically Competent E. coli (Invitrogen). Blue-white selection was used to identify colonies carrying plasmids with inserts by supplementing the LB + Kanamycin plates with 40 μg/mL of X-gal (Invitrogen). Plasmids were extracted using the Qiaprep Spin Miniprep kit (Qiagen) and sequenced via whole plasmid sequencing to identify the inserted fragments.

Determination of amplification stability

For the amplification stability experiment with 1.7a (recA::Tn), cells were inoculated into modified M63 from the corresponding frozen stock and grown overnight. For the amplification loss experiment, which was done for five passages, cells were grown from frozen stocks into modified M63 containing DLX at ~0.25x the corresponding MICs, in order to stabilize the amplified fragments prior to the treatment-free passaging. The cultures were then diluted 1:1000 in 2 mL of modified M63 (with or without DLX) and grown for 24 h. Genomic DNA was extracted from the remaining culture. This procedure was repeated for the appropriate numbers of passages, and WGS was performed on the genomic DNA from each passage. Copy number for sdrM was determined as described above.

TUNEL assay

We followed the terminal deoxynucleotide transferase dUTP nick end labeling (TUNEL) assay as described previously for bacteria [69]. Briefly, cells were grown overnight in 2 mL of modified M63, diluted 1:50 in 15 mL of modified M63, and grown to an OD₆₀₀ of 0.3-0.4. Next, 2 mL of cells were transferred to three falcon tubes, and two of the tubes were treated with the appropriate concentration of DLX or MMC respectively. The remaining tube was kept as a treatment-free control, and all three tubes were grown for an additional 3 h. From these cultures, 1 mL was transferred to microcentrifuge tubes, washed once with ice cold 1x Phosphate buffered saline (PBS,1X, Thermo Scientific), and fixed by treatment with 1mL of 4% Paraformaldehyde (PFA) on ice for 30 min. Cells were then washed with ice cold PBS once, and resuspended in 250 μL of PBS and 1mL of ice cold 70% ethanol, and incubated 12–16 hours at -20^oC.

After incubation, the tubes were centrifuged, the supernatants removed, and the cells labeled with the Apo-Direct Kit (BD Bioscience). Fluorescence output was analyzed using an Apogee MicroPLUS flow cytometer (ApogeeFlow Systems Inc), with excitation and emission wavelengths of 488 nm and 515 nm respectively. A region of interest (ROI) was drawn around the WT cell events for medium vs large light scatter angle, and this ROI was used as a gate for subsequent analysis of other strains. The proportion of the population with DNA damage was quantified as the Overton positive percentage using FlowJo (v10.1) [70].

Statistics

Statistical comparison of datasets was performed using GraphPad Prism 9. Tests used and significance thresholds are mentioned in the respective figure legends.

Supporting information

S1 Fig. Selected mutants show a wide range of DLX sensitivities.

DLX MICs of WT and the indicated mutants were measured, and the mutant MICs were normalized to the WT MIC. Data shown are the mean ± standard deviation for three independent biological replicates. Significance is shown for comparison to a value of 1, as tested by one-sample t-tests (* p < 0.05, ** p < 0.01, *** p < 0.001, # p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.s001

(TIF)

S2 Fig. DLX causes DNA damage.

(A) Representative fluorescence from TUNEL staining of WT cells treated with no antibiotic, 1 µg/mL MMC, or 5 µg/mL DLX, or a no-dye control. (B) The Overton positive percentage (representing the percentage of the population that has increased fluorescence compared to the control no antibiotic population) for the DLX and MMC treated samples. Significance is shown for comparison to a value of 0, as tested by one-sample t-tests (* p < 0.05).

https://doi.org/10.1371/journal.pgen.1012011.s002

(TIFF)

S3 Fig. Mutations seen in RexB and RecD2 in the initial recA::Tn evolved populations are in relatively conserved residues.

(A, B) Multiple sequence alignments of (A) RexB and (B) RecD2 from the Conserved Domain Database [71,72], where residues in red are highly conserved, and those in blue are not. Arrows denote the position of mutations, which are also highlighted, seen in the initial recA::Tn evolved populations.

https://doi.org/10.1371/journal.pgen.1012011.s003

(TIFF)

S4 Fig. Amplified segments containing sdrM have a fitness cost.

(A) Growth rates, (B) doubling times, and (C) time to reach the peak growth rate are shown for the WT, and 1.7a, and recA::Tn mutants in both backgrounds. These parameters were obtained from growth curves shown in Fig. 3D using Growthcurver [44] for fitting and analysis. Data shown are the means ± standard deviation for three independent biological replicates. Significance is shown for comparison to the 1.7a recA::Tn strain, as tested by a one-way ANOVA with Dunnett’s test for multiple comparisons (* p < 0.05, ** p < 0.01, # p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.s004

(TIF)

S5 Fig. Proximity of sdrM homologs to the rRNA-tRNA locus is conserved in related species.

Homologs of sdrM were identified using BLASTP [68] against the ClusteredNR database on NCBI using default settings. Shown are the distances of the sdrM homologs to the nearest rRNA-tRNA cluster.

https://doi.org/10.1371/journal.pgen.1012011.s005

(TIF)

S6 Fig. Novel junctions corresponding to the sdrM amplification can be detected at a 10^-4 – 10^-5 dilution.

qPCR C_T values for the control housekeeping gene rpoC and the respective novel junction present in two different evolved DLX resistant populations from (A) WT, (B) spoIIIE::Tn, and (C) recQ2::Tn, measured in a 10-fold dilution series where genomic DNA from the evolved population was diluted in genomic DNA from the parental strain.

https://doi.org/10.1371/journal.pgen.1012011.s006

(TIF)

S7 Fig. Rare novel junctions around sdrM may be present in the unselected WT strain.

(A) Inverse-PCR like strategy to detect novel junctions present around sdrM. Shown are two hypothetical amplified fragments (outlined by the brown lines). We designed outward facing primers (Primer 1 and Primer 2) binding to sdrM and B7H15_RS12060. If a duplicated or amplified fragment is present in that region, the novel junction should get amplified by the PCR primers. (B) Three putative duplicated or amplified fragments identified in the WT strain using the inverse PCR like strategy.

https://doi.org/10.1371/journal.pgen.1012011.s007

(TIFF)

S8 Fig. The second region with increased coverage in several DLX resistant evolved populations encodes for a putative intact prophage.

(A) Genomic coverage from whole-genome sequencing of a WT evolved population showing increased coverage for the 43 kb prophage region as well as the sdrM locus. (B) The JE2 genome was analyzed using PHASTER [35] to identify prophage regions, and the PHASTER output is shown. Region 4 (in green) encodes an intact prophage whose coordinates match the region that shows increased sequencing coverage depth in the evolved DLX resistant populations.

https://doi.org/10.1371/journal.pgen.1012011.s008

(TIFF)

S9 Fig. An intergenic mutation upstream of sdrM leads to increased sdrM expression.

Expression of sdrM was measured in log phase (OD₆₀₀ = 0.5) and stationary phase (overnight cultures) of the previously reported sdrM2* and sdrM3* allele-replacement mutants [16], using qPCR where the rpoC gene was used as the control housekeeping gene. The sdrM2* mutant has the A268S coding sequence mutation, while the sdrM3* mutant has the A268S mutation as well as a C to G change at position -164 (upstream of sdrM). Data shown are the means ± standard deviation for three independent biological replicates. Significance is shown for comparison to 1, as tested by a one-sample t-test (** p < 0.01).

https://doi.org/10.1371/journal.pgen.1012011.s009

(TIF)

S10 Fig. Scheme for the deep well evolution assay.

12 independent populations of the strain to be evolved were inoculated in a 2-fold dilution series of DLX and grown for 24 or 48 hours depending on the specific strains being tested. Populations from the well with the highest DLX concentration that showed growth were transferred to the next passage. Wells showing growth are denoted in yellow, and of those, the ones in the highest DLX concentration are marked with the red circles. Populations from these red-circled wells were propagated to the next passage. After the specified number of passages, the terminal passages from the wells with the highest DLX concentration that showed growth were sent for WGS.

https://doi.org/10.1371/journal.pgen.1012011.s010

(TIFF)

S11 Fig. Growth and DLX resistance defects in the ΔxerC mutant can be complemented by XerC expression.

(A) Overnight cultures of the indicated strains were diluted 1:100 in modified M63 medium and OD₆₀₀ was measured every 30 minutes. Data shown are mean ± SEM of three independent biological replicates. (B) DLX MICs of the ΔxerC mutant containing either the empty pKK30 plasmid or one expressing xerC from its native promoter. Data shown are mean ± standard deviation for three independent biological replicates. Significance is shown for comparison to the empty vector control, as tested by an unpaired t-test (# p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.s011

(TIFF)

S12 Fig. The ΔxerC mutant does not have a defect in recA induction upon DLX exposure.

(A-C) WT and the ΔxerC mutant were grown in a dilution series of DLX, and fluorescence (excitation 560 nm; emission 587 nm) and A₆₀₀ measured every 20 minutes. Shown are the mean ± SEM of three independent biological replicates. A₆₀₀ of (A) WT and (B) the ΔxerC mutant is shown. The horizontal dotted line represents 60% of the mean carrying capacity for the no DLX condition, and the vertical dotted line represents the time at which the growth without DLX first reaches the mean carrying capacity. The growth curves in green represent the DLX concentrations below the respective IC-40, and the curves in grey represent the DLX concentrations equal to or greater than the respective IC-40. (C) Normalized fluorescence (relative to the OD₆₀₀) for WT and the ΔxerC mutant grown in 0.5x, 0.25x, 0.125x, 0.0675x the respective IC-40, or without DLX. (A-C) Data shown are the mean ± SEM of four independent biological replicates. (D) DLX MICs of the WT, recA::Tn, and ΔxerC containing either the pKK30 empty vector, or one containing P_recA::recA or P_tufA::recA (for recA::Tn, and ΔxerC). Data shown are the mean ± standard deviation of three independent biological replicates. Significance is shown for comparison to the respective empty vector control, as tested by a one-way ANOVA followed by a Tukey’s test for multiple comparisons (# p < 0.0001).

https://doi.org/10.1371/journal.pgen.1012011.s012

(TIFF)

S1 Table. Mutations seen in the recA::Tn evolved populations.

https://doi.org/10.1371/journal.pgen.1012011.s013

(XLSX)

S2 Table. Mutations seen in the evolved populations of recA::Tn intermediate isolate R1 complemented with either an empty pKK30 plasmid or one carrying recA.

https://doi.org/10.1371/journal.pgen.1012011.s014

(XLSX)

S3 Table. List of strains evolved for DLX resistance.

https://doi.org/10.1371/journal.pgen.1012011.s015

(XLSX)

S4 Table. The distinct amplified fragments observed in the evolved populations from the tube evolutions of transposon and allele-replacement mutants.

https://doi.org/10.1371/journal.pgen.1012011.s016

(XLSX)

S5 Table. Distances between sdrM homologs and the nearest rRNA locus in multiple Staphylococcus and Mammaliicoccus species.

https://doi.org/10.1371/journal.pgen.1012011.s017

(XLSX)

S6 Table. Mutations seen in the evolved populations from the tube evolutions of transposon and allele-replacement mutants.

https://doi.org/10.1371/journal.pgen.1012011.s018

(XLSX)

S7 Table. Mutations seen in the evolved populations of the P_090/095^G→A allele-replacement strain.

https://doi.org/10.1371/journal.pgen.1012011.s019

(XLSX)

S8 Table. Mutations seen in the evolved populations from the deep-well evolutions.

https://doi.org/10.1371/journal.pgen.1012011.s020

(XLSX)

S9 Table. Strains and plasmids used in this study.

https://doi.org/10.1371/journal.pgen.1012011.s021

(PDF)

S10 Table. Primers used in this study.

https://doi.org/10.1371/journal.pgen.1012011.s022

(PDF)

S1 Data. Numerical data underlying graphs shown in figures.

https://doi.org/10.1371/journal.pgen.1012011.s023

(XLSX)

Acknowledgments

We would like to acknowledge the Center for Cancer Research (CCR) Genomics Core for whole-genome sequencing. This work used the computational resources of the NIH High Performance Computing Biowulf Cluster (http://hpc.nih.gov). We thank Francois De Mets and Shaun Brinsmade (Georgetown University) for the sequence of the tufA promoter fused to the gyrA leader, Susan Gottesman and Gisela Storz for comments on the manuscript, and members of the Khare, Gottesman, and Ramamurthi labs for helpful discussion and feedback.

References

1. Huemer M, Mairpady Shambat S, Brugger SD, Zinkernagel AS. Antibiotic resistance and persistence-Implications for human health and treatment perspectives. EMBO Rep. 2020;21(12):e51034. pmid:33400359
- View Article
- PubMed/NCBI
- Google Scholar
2. Larsson DGJ, Flach C-F. Antibiotic resistance in the environment. Nat Rev Microbiol. 2022;20(5):257–69. pmid:34737424
- View Article
- PubMed/NCBI
- Google Scholar
3. Murray CJL, Ikuta KS, Sharara F, Swetschinski L, Robles Aguilar G, Gray A, et al. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022;399(10325):629–55.
- View Article
- Google Scholar
4. Naghavi M, Vollset SE, Ikuta KS, Swetschinski LR, Gray AP, Wool EE, et al. Global burden of bacterial antimicrobial resistance 1990–2021: a systematic analysis with forecasts to 2050. Lancet. 2024;404(10459):1199–226.
- View Article
- Google Scholar
5. Chambers HF, Deleo FR. Waves of resistance: Staphylococcus aureus in the antibiotic era. Nat Rev Microbiol. 2009;7(9):629–41. pmid:19680247
- View Article
- PubMed/NCBI
- Google Scholar
6. Vestergaard M, Frees D, Ingmer H. Antibiotic Resistance and the MRSA Problem. Microbiol Spectr. 2019;7(2):10.1128/microbiolspec.gpp3-0057-2018. pmid:30900543
- View Article
- PubMed/NCBI
- Google Scholar
7. Sandegren L, Andersson DI. Bacterial gene amplification: implications for the evolution of antibiotic resistance. Nat Rev Microbiol. 2009;7(8):578–88. pmid:19609259
- View Article
- PubMed/NCBI
- Google Scholar
8. Silva KPT, Khare A. Antibiotic resistance mediated by gene amplifications. NPJ Antimicrob Resist. 2024;2(1):35. pmid:39843582
- View Article
- PubMed/NCBI
- Google Scholar
9. Papkou A, Hedge J, Kapel N, Young B, MacLean RC. Efflux pump activity potentiates the evolution of antibiotic resistance across S. aureus isolates. Nat Commun. 2020;11(1):3970. pmid:32769975
- View Article
- PubMed/NCBI
- Google Scholar
10. Gallagher LA, Coughlan S, Black NS, Lalor P, Waters EM, Wee B, et al. Tandem Amplification of the Staphylococcal Cassette Chromosome mec Element Can Drive High-Level Methicillin Resistance in Methicillin-Resistant Staphylococcus aureus. Antimicrob Agents Chemother. 2017;61(9):e00869-17. pmid:28717043
- View Article
- PubMed/NCBI
- Google Scholar
11. Yoon E-J, Kwon A-R, Min Y-H, Choi E-C. Foggy D-shaped zone of inhibition in Staphylococcus aureus owing to a dual character of both inducible and constitutive resistance to macrolide-lincosamide-streptogramin B. J Antimicrob Chemother. 2008;61(3):533–40. pmid:18230691
- View Article
- PubMed/NCBI
- Google Scholar
12. Gao W, Monk IR, Tobias NJ, Gladman SL, Seemann T, Stinear TP, et al. Large tandem chromosome expansions facilitate niche adaptation during persistent infection with drug-resistant Staphylococcus aureus. Microb Genom. 2015;1(2):e000026. pmid:28348811
- View Article
- PubMed/NCBI
- Google Scholar
13. Candel FJ, Peñuelas M. Delafloxacin: design, development and potential place in therapy. Drug Des Devel Ther. 2017;11:881–91. pmid:28356714
- View Article
- PubMed/NCBI
- Google Scholar
14. Oldfield E, Feng X. Resistance-resistant antibiotics. Trends Pharmacol Sci. 2014;35(12):664–74. pmid:25458541
- View Article
- PubMed/NCBI
- Google Scholar
15. Nyerges A, Tomašič T, Durcik M, Revesz T, Szili P, Draskovits G, et al. Rational design of balanced dual-targeting antibiotics with limited resistance. PLoS Biol. 2020;18(10):e3000819. pmid:33017402
- View Article
- PubMed/NCBI
- Google Scholar
16. Silva KPT, Sundar G, Khare A. Efflux pump gene amplifications bypass necessity of multiple target mutations for resistance against dual-targeting antibiotic. Nat Commun. 2023;14(1):3402. pmid:37296157
- View Article
- PubMed/NCBI
- Google Scholar
17. Reams AB, Roth JR. Mechanisms of gene duplication and amplification. Cold Spring Harb Perspect Biol. 2015;7(2):a016592. pmid:25646380
- View Article
- PubMed/NCBI
- Google Scholar
18. Reams AB, Kofoid E, Kugelberg E, Roth JR. Multiple pathways of duplication formation with and without recombination (RecA) in Salmonella enterica. Genetics. 2012;192(2):397–415. pmid:22865732
- View Article
- PubMed/NCBI
- Google Scholar
19. Slack A, Thornton PC, Magner DB, Rosenberg SM, Hastings PJ. On the mechanism of gene amplification induced under stress in Escherichia coli. PLoS Genet. 2006;2(4):e48. pmid:16604155
- View Article
- PubMed/NCBI
- Google Scholar
20. Andersson DI, Hughes D. Gene amplification and adaptive evolution in bacteria. Annu Rev Genet. 2009;43:167–95. pmid:19686082
- View Article
- PubMed/NCBI
- Google Scholar
21. Fey PD, Endres JL, Yajjala VK, Widhelm TJ, Boissy RJ, Bose JL, et al. A genetic resource for rapid and comprehensive phenotype screening of nonessential Staphylococcus aureus genes. mBio. 2013;4(1):e00537-12. pmid:23404398
- View Article
- PubMed/NCBI
- Google Scholar
22. Aldred KJ, Kerns RJ, Osheroff N. Mechanism of quinolone action and resistance. Biochemistry. 2014;53(10):1565–74. pmid:24576155
- View Article
- PubMed/NCBI
- Google Scholar
23. Ha KP, Edwards AM. DNA Repair in Staphylococcus aureus. Microbiol Mol Biol Rev. 2021;85(4):e0009121. pmid:34523959
- View Article
- PubMed/NCBI
- Google Scholar
24. Kiran K, Patil KN. Expression and Characterization of the Staphylococcus aureus RecA protein: A mapping of canonical functions. Protein Expr Purif. 2022;189:105967. pmid:34481085
- View Article
- PubMed/NCBI
- Google Scholar
25. Cirz RT, Jones MB, Gingles NA, Minogue TD, Jarrahi B, Peterson SN, et al. Complete and SOS-mediated response of Staphylococcus aureus to the antibiotic ciprofloxacin. J Bacteriol. 2007;189(2):531–9. pmid:17085555
- View Article
- PubMed/NCBI
- Google Scholar
26. Shariati A, Arshadi M, Khosrojerdi MA, Abedinzadeh M, Ganjalishahi M, Maleki A, et al. The resistance mechanisms of bacteria against ciprofloxacin and new approaches for enhancing the efficacy of this antibiotic. Front Public Health. 2022;10:1025633. pmid:36620240
- View Article
- PubMed/NCBI
- Google Scholar
27. Gewirtz DA. A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacol. 1999;57(7):727–41. pmid:10075079
- View Article
- PubMed/NCBI
- Google Scholar
28. Tomasz M. Mitomycin C: small, fast and deadly (but very selective). Chem Biol. 1995;2(9):575–9. pmid:9383461
- View Article
- PubMed/NCBI
- Google Scholar
29. Krajewski WW, Fu X, Wilkinson M, Cronin NB, Dillingham MS, Wigley DB. Structural basis for translocation by AddAB helicase-nuclease and its arrest at χ sites. Nature. 2014;508(7496):416–9. pmid:24670664
- View Article
- PubMed/NCBI
- Google Scholar
30. Montague M, Barnes C, Smith HO, Chuang R-Y, Vashee S. The evolution of RecD outside of the RecBCD complex. J Mol Evol. 2009;69(4):360–71. pmid:19841849
- View Article
- PubMed/NCBI
- Google Scholar
31. Rocha EPC, Cornet E, Michel B. Comparative and evolutionary analysis of the bacterial homologous recombination systems. PLoS Genet. 2005;1(2):e15. pmid:16132081
- View Article
- PubMed/NCBI
- Google Scholar
32. Walsh BW, Bolz SA, Wessel SR, Schroeder JW, Keck JL, Simmons LA. RecD2 helicase limits replication fork stress in Bacillus subtilis. J Bacteriol. 2014;196(7):1359–68. pmid:24443534
- View Article
- PubMed/NCBI
- Google Scholar
33. Maiques E, Ubeda C, Campoy S, Salvador N, Lasa I, Novick RP, et al. beta-lactam antibiotics induce the SOS response and horizontal transfer of virulence factors in Staphylococcus aureus. J Bacteriol. 2006;188(7):2726–9. pmid:16547063
- View Article
- PubMed/NCBI
- Google Scholar
34. Cuirolo A, Plata K, Rosato AE. Development of homogeneous expression of resistance in methicillin-resistant Staphylococcus aureus clinical strains is functionally associated with a beta-lactam-mediated SOS response. J Antimicrob Chemother. 2009;64(1):37–45. pmid:19457930
- View Article
- PubMed/NCBI
- Google Scholar
35. Arndt D, Grant JR, Marcu A, Sajed T, Pon A, Liang Y, et al. PHASTER: a better, faster version of the PHAST phage search tool. Nucleic Acids Res. 2016;44(W1):W16–21. pmid:27141966
- View Article
- PubMed/NCBI
- Google Scholar
36. Zhou Y, Liang Y, Lynch KH, Dennis JJ, Wishart DS. PHAST: a fast phage search tool. Nucleic Acids Res. 2011;39(Web Server issue):W347–52. pmid:21672955
- View Article
- PubMed/NCBI
- Google Scholar
37. McRobbie A-M, Meyer B, Rouillon C, Petrovic-Stojanovska B, Liu H, White MF. Staphylococcus aureus DinG, a helicase that has evolved into a nuclease. Biochem J. 2012;442(1):77–84. pmid:22166102
- View Article
- PubMed/NCBI
- Google Scholar
38. Cohen SE, Lewis CA, Mooney RA, Kohanski MA, Collins JJ, Landick R, et al. Roles for the transcription elongation factor NusA in both DNA repair and damage tolerance pathways in Escherichia coli. Proc Natl Acad Sci U S A. 2010;107(35):15517–22. pmid:20696893
- View Article
- PubMed/NCBI
- Google Scholar
39. Cheng K, Sun Y, Yu H, Hu Y, He Y, Shen Y. Staphylococcus aureus SOS response: Activation, impact, and drug targets. mLife. 2024;3(3):343–66. pmid:39359682
- View Article
- PubMed/NCBI
- Google Scholar
40. McKenzie GJ, Lee PL, Lombardo MJ, Hastings PJ, Rosenberg SM. SOS mutator DNA polymerase IV functions in adaptive mutation and not adaptive amplification. Mol Cell. 2001;7(3):571–9. pmid:11463382
- View Article
- PubMed/NCBI
- Google Scholar
41. Midonet C, Barre F-X. Xer Site-Specific Recombination: Promoting Vertical and Horizontal Transmission of Genetic Information. Microbiol Spectr. 2014;2(6):10.1128/microbiolspec.mdna3-0056-2014. pmid:26104463
- View Article
- PubMed/NCBI
- Google Scholar
42. Frees D, Thomsen LE, Ingmer H. Staphylococcus aureus ClpYQ plays a minor role in stress survival. Arch Microbiol. 2005;183(4):286–91. pmid:15843987
- View Article
- PubMed/NCBI
- Google Scholar
43. Brinsmade SR. CodY, a master integrator of metabolism and virulence in Gram-positive bacteria. Curr Genet. 2017;63(3):417–25. pmid:27744611
- View Article
- PubMed/NCBI
- Google Scholar
44. Sprouffske K, Wagner A. Growthcurver: an R package for obtaining interpretable metrics from microbial growth curves. BMC Bioinformatics. 2016;17:172. pmid:27094401
- View Article
- PubMed/NCBI
- Google Scholar
45. Ledger EVK, Lau K, Tate EW, Edwards AM. XerC Is Required for the Repair of Antibiotic- and Immune-Mediated DNA Damage in Staphylococcus aureus. Antimicrob Agents Chemother. 2023;67(3):e0120622. pmid:36802166
- View Article
- PubMed/NCBI
- Google Scholar
46. Augagneur Y, King AN, Germain-Amiot N, Sassi M, Fitzgerald JW, Sahukhal GS, et al. Analysis of the CodY RNome reveals RsaD as a stress-responsive riboregulator of overflow metabolism in Staphylococcus aureus. Mol Microbiol. 2020;113(2):309–25. pmid:31696578
- View Article
- PubMed/NCBI
- Google Scholar
47. Xiao Q, Vakulenko S, Chang M, Mobashery S. Mutations in mmpL and in the cell wall stress stimulon contribute to resistance to oxadiazole antibiotics in methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother. 2014;58(10):5841–7. pmid:25049248
- View Article
- PubMed/NCBI
- Google Scholar
48. Miller C, Thomsen LE, Gaggero C, Mosseri R, Ingmer H, Cohen SN. SOS response induction by beta-lactams and bacterial defense against antibiotic lethality. Science. 2004;305(5690):1629–31. pmid:15308764
- View Article
- PubMed/NCBI
- Google Scholar
49. Reams AB, Neidle EL. Gene amplification involves site-specific short homology-independent illegitimate recombination in Acinetobacter sp. strain ADP1. J Mol Biol. 2004;338(4):643–56. pmid:15099734
- View Article
- PubMed/NCBI
- Google Scholar
50. Baylay AJ, Ivens A, Piddock LJV. A novel gene amplification causes upregulation of the PatAB ABC transporter and fluoroquinolone resistance in Streptococcus pneumoniae. Antimicrob Agents Chemother. 2015;59(6):3098–108. pmid:25779578
- View Article
- PubMed/NCBI
- Google Scholar
51. Merrikh H, Zhang Y, Grossman AD, Wang JD. Replication-transcription conflicts in bacteria. Nat Rev Microbiol. 2012;10(7):449–58. pmid:22669220
- View Article
- PubMed/NCBI
- Google Scholar
52. Silpe JE, Duddy OP, Bassler BL. Induction mechanisms and strategies underlying interprophage competition during polylysogeny. PLoS Pathog. 2023;19(5):e1011363. pmid:37200239
- View Article
- PubMed/NCBI
- Google Scholar
53. Shiraishi K, Hanada K, Iwakura Y, Ikeda H. Roles of RecJ, RecO, and RecR in RecET-mediated illegitimate recombination in Escherichia coli. J Bacteriol. 2002;184(17):4715–21. pmid:12169595
- View Article
- PubMed/NCBI
- Google Scholar
54. Ikeda H, Shiraishi K, Ogata Y. Illegitimate recombination mediated by double-strand break and end-joining in Escherichia coli. Adv Biophys. 2004;38:3–20. pmid:15493325
- View Article
- PubMed/NCBI
- Google Scholar
55. Hastings PJ, Slack A, Petrosino JF, Rosenberg SM. Adaptive amplification and point mutation are independent mechanisms: evidence for various stress-inducible mutation mechanisms. PLoS Biol. 2004;2(12):e399. pmid:15550983
- View Article
- PubMed/NCBI
- Google Scholar
56. Chalker AF, Lupas A, Ingraham K, So CY, Lunsford RD, Li T, et al. Genetic characterization of gram-positive homologs of the XerCD site-specific recombinases. J Mol Microbiol Biotechnol. 2000;2(2):225–33. pmid:10939248
- View Article
- PubMed/NCBI
- Google Scholar
57. Izquierdo-Martinez A, Schäper S, Brito AD, Liao Q, Tesseur C, Sorg M, et al. Chromosome segregation dynamics during the cell cycle of Staphylococcus aureus. Nat Commun. 2025;16(1):8626. pmid:41022742
- View Article
- PubMed/NCBI
- Google Scholar
58. Veiga H, Jorge AM, Pinho MG. Absence of nucleoid occlusion effector Noc impairs formation of orthogonal FtsZ rings during Staphylococcus aureus cell division. Mol Microbiol. 2011;80(5):1366–80. pmid:21477126
- View Article
- PubMed/NCBI
- Google Scholar
59. Ramos-León F, Anjuwon-Foster BR, Anantharaman V, Updegrove TB, Ferreira CN, Ibrahim AM, et al. PcdA promotes orthogonal division plane selection in Staphylococcus aureus. Nat Microbiol. 2024;9(11):2997–3012. pmid:39468247
- View Article
- PubMed/NCBI
- Google Scholar
60. Ramos-León F, Ramamurthi KS. Staphylococcus aureus as an emerging model to study bacterial cell division. J Biol Chem. 2025;301(7):110343. pmid:40484379
- View Article
- PubMed/NCBI
- Google Scholar
61. Lambert RJ, Pearson J. Susceptibility testing: accurate and reproducible minimum inhibitory concentration (MIC) and non-inhibitory concentration (NIC) values. J Appl Microbiol. 2000;88(5):784–90. pmid:10792538
- View Article
- PubMed/NCBI
- Google Scholar
62. Deatherage DE, Barrick JE. Identification of mutations in laboratory-evolved microbes from next-generation sequencing data using breseq. In: Sun L, Shou W, editors. In: Engineering and analyzing multicellular systems: methods and protocols. Vol. 1151. New York, NY: Humana Press; 2014. pp. 165–88. https://doi.org/10.1007/978-1-4939-0554-6_12
- View Article
- Google Scholar
63. Wilkins D. gggenes: Draw Gene Arrow Maps in ‘ggplot2’. 2023. Available from: https://wilkox.org/gggenes/
64. Schuster CF, Howard SA, Gründling A. Use of the counter selectable marker PheS* for genome engineering in Staphylococcus aureus. Microbiology (Reading). 2019;165(5):572–84. pmid:30942689
- View Article
- PubMed/NCBI
- Google Scholar
65. Krute CN, Krausz KL, Markiewicz MA, Joyner JA, Pokhrel S, Hall PR, et al. Generation of a Stable Plasmid for In Vitro and In Vivo Studies of Staphylococcus Species. Appl Environ Microbiol. 2016;82(23):6859–69. pmid:27637878
- View Article
- PubMed/NCBI
- Google Scholar
66. Bose JL, Fey PD, Bayles KW. Genetic tools to enhance the study of gene function and regulation in Staphylococcus aureus. Appl Environ Microbiol. 2013;79(7):2218–24. pmid:23354696
- View Article
- PubMed/NCBI
- Google Scholar
67. Mlynek KD, Bulock LL, Stone CJ, Curran LJ, Sadykov MR, Bayles KW, et al. Genetic and Biochemical Analysis of CodY-Mediated Cell Aggregation in Staphylococcus aureus Reveals an Interaction between Extracellular DNA and Polysaccharide in the Extracellular Matrix. J Bacteriol. 2020;202(8):e00593-19. pmid:32015143
- View Article
- PubMed/NCBI
- Google Scholar
68. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10:421. pmid:20003500
- View Article
- PubMed/NCBI
- Google Scholar
69. Dwyer DJ, Camacho DM, Kohanski MA, Callura JM, Collins JJ. Antibiotic-induced bacterial cell death exhibits physiological and biochemical hallmarks of apoptosis. Mol Cell. 2012;46(5):561–72. pmid:22633370
- View Article
- PubMed/NCBI
- Google Scholar
70. Overton WR. Modified histogram subtraction technique for analysis of flow cytometry data. Cytometry. 1988;9(6):619–26. pmid:3061754
- View Article
- PubMed/NCBI
- Google Scholar
71. Marchler-Bauer A, Bryant SH. CD-Search: protein domain annotations on the fly. Nucleic Acids Res. 2004;32(Web Server issue):W327–31. pmid:15215404
- View Article
- PubMed/NCBI
- Google Scholar
72. Wang J, Chitsaz F, Derbyshire MK, Gonzales NR, Gwadz M, Lu S, et al. The conserved domain database in 2023. Nucleic Acids Res. 2023;51(D1):D384–8. pmid:36477806
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Huemer M, Mairpady Shambat S, Brugger SD, Zinkernagel AS. Antibiotic resistance and persistence-Implications for human health and treatment perspectives. EMBO Rep. 2020;21(12):e51034. pmid:33400359
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Larsson DGJ, Flach C-F. Antibiotic resistance in the environment. Nat Rev Microbiol. 2022;20(5):257–69. pmid:34737424
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Murray CJL, Ikuta KS, Sharara F, Swetschinski L, Robles Aguilar G, Gray A, et al. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet. 2022;399(10325):629–55.
View Article
Google Scholar

[10] View Article

[11] Google Scholar

[ref4] 4. Naghavi M, Vollset SE, Ikuta KS, Swetschinski LR, Gray AP, Wool EE, et al. Global burden of bacterial antimicrobial resistance 1990–2021: a systematic analysis with forecasts to 2050. Lancet. 2024;404(10459):1199–226.
View Article
Google Scholar

[13] View Article

[14] Google Scholar

[ref5] 5. Chambers HF, Deleo FR. Waves of resistance: Staphylococcus aureus in the antibiotic era. Nat Rev Microbiol. 2009;7(9):629–41. pmid:19680247
View Article
PubMed/NCBI
Google Scholar

[16] View Article

[17] PubMed/NCBI

[18] Google Scholar

[ref6] 6. Vestergaard M, Frees D, Ingmer H. Antibiotic Resistance and the MRSA Problem. Microbiol Spectr. 2019;7(2):10.1128/microbiolspec.gpp3-0057-2018. pmid:30900543
View Article
PubMed/NCBI
Google Scholar

[20] View Article

[21] PubMed/NCBI

[22] Google Scholar

[ref7] 7. Sandegren L, Andersson DI. Bacterial gene amplification: implications for the evolution of antibiotic resistance. Nat Rev Microbiol. 2009;7(8):578–88. pmid:19609259
View Article
PubMed/NCBI
Google Scholar

[24] View Article

[25] PubMed/NCBI

[26] Google Scholar

[ref8] 8. Silva KPT, Khare A. Antibiotic resistance mediated by gene amplifications. NPJ Antimicrob Resist. 2024;2(1):35. pmid:39843582
View Article
PubMed/NCBI
Google Scholar

[28] View Article

[29] PubMed/NCBI

[30] Google Scholar

[ref9] 9. Papkou A, Hedge J, Kapel N, Young B, MacLean RC. Efflux pump activity potentiates the evolution of antibiotic resistance across S. aureus isolates. Nat Commun. 2020;11(1):3970. pmid:32769975
View Article
PubMed/NCBI
Google Scholar

[32] View Article

[33] PubMed/NCBI

[34] Google Scholar

[ref10] 10. Gallagher LA, Coughlan S, Black NS, Lalor P, Waters EM, Wee B, et al. Tandem Amplification of the Staphylococcal Cassette Chromosome mec Element Can Drive High-Level Methicillin Resistance in Methicillin-Resistant Staphylococcus aureus. Antimicrob Agents Chemother. 2017;61(9):e00869-17. pmid:28717043
View Article
PubMed/NCBI
Google Scholar

[36] View Article

[37] PubMed/NCBI

[38] Google Scholar

[ref11] 11. Yoon E-J, Kwon A-R, Min Y-H, Choi E-C. Foggy D-shaped zone of inhibition in Staphylococcus aureus owing to a dual character of both inducible and constitutive resistance to macrolide-lincosamide-streptogramin B. J Antimicrob Chemother. 2008;61(3):533–40. pmid:18230691
View Article
PubMed/NCBI
Google Scholar

[40] View Article

[41] PubMed/NCBI

[42] Google Scholar

[ref12] 12. Gao W, Monk IR, Tobias NJ, Gladman SL, Seemann T, Stinear TP, et al. Large tandem chromosome expansions facilitate niche adaptation during persistent infection with drug-resistant Staphylococcus aureus. Microb Genom. 2015;1(2):e000026. pmid:28348811
View Article
PubMed/NCBI
Google Scholar

[44] View Article

[45] PubMed/NCBI

[46] Google Scholar

[ref13] 13. Candel FJ, Peñuelas M. Delafloxacin: design, development and potential place in therapy. Drug Des Devel Ther. 2017;11:881–91. pmid:28356714
View Article
PubMed/NCBI
Google Scholar

[48] View Article

[49] PubMed/NCBI

[50] Google Scholar

[ref14] 14. Oldfield E, Feng X. Resistance-resistant antibiotics. Trends Pharmacol Sci. 2014;35(12):664–74. pmid:25458541
View Article
PubMed/NCBI
Google Scholar

[52] View Article

[53] PubMed/NCBI

[54] Google Scholar

[ref15] 15. Nyerges A, Tomašič T, Durcik M, Revesz T, Szili P, Draskovits G, et al. Rational design of balanced dual-targeting antibiotics with limited resistance. PLoS Biol. 2020;18(10):e3000819. pmid:33017402
View Article
PubMed/NCBI
Google Scholar

[56] View Article

[57] PubMed/NCBI

[58] Google Scholar

[ref16] 16. Silva KPT, Sundar G, Khare A. Efflux pump gene amplifications bypass necessity of multiple target mutations for resistance against dual-targeting antibiotic. Nat Commun. 2023;14(1):3402. pmid:37296157
View Article
PubMed/NCBI
Google Scholar

[60] View Article

[61] PubMed/NCBI

[62] Google Scholar

[ref17] 17. Reams AB, Roth JR. Mechanisms of gene duplication and amplification. Cold Spring Harb Perspect Biol. 2015;7(2):a016592. pmid:25646380
View Article
PubMed/NCBI
Google Scholar

[64] View Article

[65] PubMed/NCBI

[66] Google Scholar

[ref18] 18. Reams AB, Kofoid E, Kugelberg E, Roth JR. Multiple pathways of duplication formation with and without recombination (RecA) in Salmonella enterica. Genetics. 2012;192(2):397–415. pmid:22865732
View Article
PubMed/NCBI
Google Scholar

[68] View Article

[69] PubMed/NCBI

[70] Google Scholar

[ref19] 19. Slack A, Thornton PC, Magner DB, Rosenberg SM, Hastings PJ. On the mechanism of gene amplification induced under stress in Escherichia coli. PLoS Genet. 2006;2(4):e48. pmid:16604155
View Article
PubMed/NCBI
Google Scholar

[72] View Article

[73] PubMed/NCBI

[74] Google Scholar

[ref20] 20. Andersson DI, Hughes D. Gene amplification and adaptive evolution in bacteria. Annu Rev Genet. 2009;43:167–95. pmid:19686082
View Article
PubMed/NCBI
Google Scholar

[76] View Article

[77] PubMed/NCBI

[78] Google Scholar

[ref21] 21. Fey PD, Endres JL, Yajjala VK, Widhelm TJ, Boissy RJ, Bose JL, et al. A genetic resource for rapid and comprehensive phenotype screening of nonessential Staphylococcus aureus genes. mBio. 2013;4(1):e00537-12. pmid:23404398
View Article
PubMed/NCBI
Google Scholar

[80] View Article

[81] PubMed/NCBI

[82] Google Scholar

[ref22] 22. Aldred KJ, Kerns RJ, Osheroff N. Mechanism of quinolone action and resistance. Biochemistry. 2014;53(10):1565–74. pmid:24576155
View Article
PubMed/NCBI
Google Scholar

[84] View Article

[85] PubMed/NCBI

[86] Google Scholar

[ref23] 23. Ha KP, Edwards AM. DNA Repair in Staphylococcus aureus. Microbiol Mol Biol Rev. 2021;85(4):e0009121. pmid:34523959
View Article
PubMed/NCBI
Google Scholar

[88] View Article

[89] PubMed/NCBI

[90] Google Scholar

[ref24] 24. Kiran K, Patil KN. Expression and Characterization of the Staphylococcus aureus RecA protein: A mapping of canonical functions. Protein Expr Purif. 2022;189:105967. pmid:34481085
View Article
PubMed/NCBI
Google Scholar

[92] View Article

[93] PubMed/NCBI

[94] Google Scholar

[ref25] 25. Cirz RT, Jones MB, Gingles NA, Minogue TD, Jarrahi B, Peterson SN, et al. Complete and SOS-mediated response of Staphylococcus aureus to the antibiotic ciprofloxacin. J Bacteriol. 2007;189(2):531–9. pmid:17085555
View Article
PubMed/NCBI
Google Scholar

[96] View Article

[97] PubMed/NCBI

[98] Google Scholar

[ref26] 26. Shariati A, Arshadi M, Khosrojerdi MA, Abedinzadeh M, Ganjalishahi M, Maleki A, et al. The resistance mechanisms of bacteria against ciprofloxacin and new approaches for enhancing the efficacy of this antibiotic. Front Public Health. 2022;10:1025633. pmid:36620240
View Article
PubMed/NCBI
Google Scholar

[100] View Article

[101] PubMed/NCBI

[102] Google Scholar

[ref27] 27. Gewirtz DA. A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacol. 1999;57(7):727–41. pmid:10075079
View Article
PubMed/NCBI
Google Scholar

[104] View Article

[105] PubMed/NCBI

[106] Google Scholar

[ref28] 28. Tomasz M. Mitomycin C: small, fast and deadly (but very selective). Chem Biol. 1995;2(9):575–9. pmid:9383461
View Article
PubMed/NCBI
Google Scholar

[108] View Article

[109] PubMed/NCBI

[110] Google Scholar

[ref29] 29. Krajewski WW, Fu X, Wilkinson M, Cronin NB, Dillingham MS, Wigley DB. Structural basis for translocation by AddAB helicase-nuclease and its arrest at χ sites. Nature. 2014;508(7496):416–9. pmid:24670664
View Article
PubMed/NCBI
Google Scholar

[112] View Article

[113] PubMed/NCBI

[114] Google Scholar

[ref30] 30. Montague M, Barnes C, Smith HO, Chuang R-Y, Vashee S. The evolution of RecD outside of the RecBCD complex. J Mol Evol. 2009;69(4):360–71. pmid:19841849
View Article
PubMed/NCBI
Google Scholar

[116] View Article

[117] PubMed/NCBI

[118] Google Scholar

[ref31] 31. Rocha EPC, Cornet E, Michel B. Comparative and evolutionary analysis of the bacterial homologous recombination systems. PLoS Genet. 2005;1(2):e15. pmid:16132081
View Article
PubMed/NCBI
Google Scholar

[120] View Article

[121] PubMed/NCBI

[122] Google Scholar

[ref32] 32. Walsh BW, Bolz SA, Wessel SR, Schroeder JW, Keck JL, Simmons LA. RecD2 helicase limits replication fork stress in Bacillus subtilis. J Bacteriol. 2014;196(7):1359–68. pmid:24443534
View Article
PubMed/NCBI
Google Scholar

[124] View Article

[125] PubMed/NCBI

[126] Google Scholar

[ref33] 33. Maiques E, Ubeda C, Campoy S, Salvador N, Lasa I, Novick RP, et al. beta-lactam antibiotics induce the SOS response and horizontal transfer of virulence factors in Staphylococcus aureus. J Bacteriol. 2006;188(7):2726–9. pmid:16547063
View Article
PubMed/NCBI
Google Scholar

[128] View Article

[129] PubMed/NCBI

[130] Google Scholar

[ref34] 34. Cuirolo A, Plata K, Rosato AE. Development of homogeneous expression of resistance in methicillin-resistant Staphylococcus aureus clinical strains is functionally associated with a beta-lactam-mediated SOS response. J Antimicrob Chemother. 2009;64(1):37–45. pmid:19457930
View Article
PubMed/NCBI
Google Scholar

[132] View Article

[133] PubMed/NCBI

[134] Google Scholar

[ref35] 35. Arndt D, Grant JR, Marcu A, Sajed T, Pon A, Liang Y, et al. PHASTER: a better, faster version of the PHAST phage search tool. Nucleic Acids Res. 2016;44(W1):W16–21. pmid:27141966
View Article
PubMed/NCBI
Google Scholar

[136] View Article

[137] PubMed/NCBI

[138] Google Scholar

[ref36] 36. Zhou Y, Liang Y, Lynch KH, Dennis JJ, Wishart DS. PHAST: a fast phage search tool. Nucleic Acids Res. 2011;39(Web Server issue):W347–52. pmid:21672955
View Article
PubMed/NCBI
Google Scholar

[140] View Article

[141] PubMed/NCBI

[142] Google Scholar

[ref37] 37. McRobbie A-M, Meyer B, Rouillon C, Petrovic-Stojanovska B, Liu H, White MF. Staphylococcus aureus DinG, a helicase that has evolved into a nuclease. Biochem J. 2012;442(1):77–84. pmid:22166102
View Article
PubMed/NCBI
Google Scholar

[144] View Article

[145] PubMed/NCBI

[146] Google Scholar

[ref38] 38. Cohen SE, Lewis CA, Mooney RA, Kohanski MA, Collins JJ, Landick R, et al. Roles for the transcription elongation factor NusA in both DNA repair and damage tolerance pathways in Escherichia coli. Proc Natl Acad Sci U S A. 2010;107(35):15517–22. pmid:20696893
View Article
PubMed/NCBI
Google Scholar

[148] View Article

[149] PubMed/NCBI

[150] Google Scholar

[ref39] 39. Cheng K, Sun Y, Yu H, Hu Y, He Y, Shen Y. Staphylococcus aureus SOS response: Activation, impact, and drug targets. mLife. 2024;3(3):343–66. pmid:39359682
View Article
PubMed/NCBI
Google Scholar

[152] View Article

[153] PubMed/NCBI

[154] Google Scholar

[ref40] 40. McKenzie GJ, Lee PL, Lombardo MJ, Hastings PJ, Rosenberg SM. SOS mutator DNA polymerase IV functions in adaptive mutation and not adaptive amplification. Mol Cell. 2001;7(3):571–9. pmid:11463382
View Article
PubMed/NCBI
Google Scholar

[156] View Article

[157] PubMed/NCBI

[158] Google Scholar

[ref41] 41. Midonet C, Barre F-X. Xer Site-Specific Recombination: Promoting Vertical and Horizontal Transmission of Genetic Information. Microbiol Spectr. 2014;2(6):10.1128/microbiolspec.mdna3-0056-2014. pmid:26104463
View Article
PubMed/NCBI
Google Scholar

[160] View Article

[161] PubMed/NCBI

[162] Google Scholar

[ref42] 42. Frees D, Thomsen LE, Ingmer H. Staphylococcus aureus ClpYQ plays a minor role in stress survival. Arch Microbiol. 2005;183(4):286–91. pmid:15843987
View Article
PubMed/NCBI
Google Scholar

[164] View Article

[165] PubMed/NCBI

[166] Google Scholar

[ref43] 43. Brinsmade SR. CodY, a master integrator of metabolism and virulence in Gram-positive bacteria. Curr Genet. 2017;63(3):417–25. pmid:27744611
View Article
PubMed/NCBI
Google Scholar

[168] View Article

[169] PubMed/NCBI

[170] Google Scholar

[ref44] 44. Sprouffske K, Wagner A. Growthcurver: an R package for obtaining interpretable metrics from microbial growth curves. BMC Bioinformatics. 2016;17:172. pmid:27094401
View Article
PubMed/NCBI
Google Scholar

[172] View Article

[173] PubMed/NCBI

[174] Google Scholar

[ref45] 45. Ledger EVK, Lau K, Tate EW, Edwards AM. XerC Is Required for the Repair of Antibiotic- and Immune-Mediated DNA Damage in Staphylococcus aureus. Antimicrob Agents Chemother. 2023;67(3):e0120622. pmid:36802166
View Article
PubMed/NCBI
Google Scholar

[176] View Article

[177] PubMed/NCBI

[178] Google Scholar

[ref46] 46. Augagneur Y, King AN, Germain-Amiot N, Sassi M, Fitzgerald JW, Sahukhal GS, et al. Analysis of the CodY RNome reveals RsaD as a stress-responsive riboregulator of overflow metabolism in Staphylococcus aureus. Mol Microbiol. 2020;113(2):309–25. pmid:31696578
View Article
PubMed/NCBI
Google Scholar

[180] View Article

[181] PubMed/NCBI

[182] Google Scholar

[ref47] 47. Xiao Q, Vakulenko S, Chang M, Mobashery S. Mutations in mmpL and in the cell wall stress stimulon contribute to resistance to oxadiazole antibiotics in methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother. 2014;58(10):5841–7. pmid:25049248
View Article
PubMed/NCBI
Google Scholar

[184] View Article

[185] PubMed/NCBI

[186] Google Scholar

[ref48] 48. Miller C, Thomsen LE, Gaggero C, Mosseri R, Ingmer H, Cohen SN. SOS response induction by beta-lactams and bacterial defense against antibiotic lethality. Science. 2004;305(5690):1629–31. pmid:15308764
View Article
PubMed/NCBI
Google Scholar

[188] View Article

[189] PubMed/NCBI

[190] Google Scholar

[ref49] 49. Reams AB, Neidle EL. Gene amplification involves site-specific short homology-independent illegitimate recombination in Acinetobacter sp. strain ADP1. J Mol Biol. 2004;338(4):643–56. pmid:15099734
View Article
PubMed/NCBI
Google Scholar

[192] View Article

[193] PubMed/NCBI

[194] Google Scholar

[ref50] 50. Baylay AJ, Ivens A, Piddock LJV. A novel gene amplification causes upregulation of the PatAB ABC transporter and fluoroquinolone resistance in Streptococcus pneumoniae. Antimicrob Agents Chemother. 2015;59(6):3098–108. pmid:25779578
View Article
PubMed/NCBI
Google Scholar

[196] View Article

[197] PubMed/NCBI

[198] Google Scholar

[ref51] 51. Merrikh H, Zhang Y, Grossman AD, Wang JD. Replication-transcription conflicts in bacteria. Nat Rev Microbiol. 2012;10(7):449–58. pmid:22669220
View Article
PubMed/NCBI
Google Scholar

[200] View Article

[201] PubMed/NCBI

[202] Google Scholar

[ref52] 52. Silpe JE, Duddy OP, Bassler BL. Induction mechanisms and strategies underlying interprophage competition during polylysogeny. PLoS Pathog. 2023;19(5):e1011363. pmid:37200239
View Article
PubMed/NCBI
Google Scholar

[204] View Article

[205] PubMed/NCBI

[206] Google Scholar

[ref53] 53. Shiraishi K, Hanada K, Iwakura Y, Ikeda H. Roles of RecJ, RecO, and RecR in RecET-mediated illegitimate recombination in Escherichia coli. J Bacteriol. 2002;184(17):4715–21. pmid:12169595
View Article
PubMed/NCBI
Google Scholar

[208] View Article

[209] PubMed/NCBI

[210] Google Scholar

[ref54] 54. Ikeda H, Shiraishi K, Ogata Y. Illegitimate recombination mediated by double-strand break and end-joining in Escherichia coli. Adv Biophys. 2004;38:3–20. pmid:15493325
View Article
PubMed/NCBI
Google Scholar

[212] View Article

[213] PubMed/NCBI

[214] Google Scholar

[ref55] 55. Hastings PJ, Slack A, Petrosino JF, Rosenberg SM. Adaptive amplification and point mutation are independent mechanisms: evidence for various stress-inducible mutation mechanisms. PLoS Biol. 2004;2(12):e399. pmid:15550983
View Article
PubMed/NCBI
Google Scholar

[216] View Article

[217] PubMed/NCBI

[218] Google Scholar

[ref56] 56. Chalker AF, Lupas A, Ingraham K, So CY, Lunsford RD, Li T, et al. Genetic characterization of gram-positive homologs of the XerCD site-specific recombinases. J Mol Microbiol Biotechnol. 2000;2(2):225–33. pmid:10939248
View Article
PubMed/NCBI
Google Scholar

[220] View Article

[221] PubMed/NCBI

[222] Google Scholar

[ref57] 57. Izquierdo-Martinez A, Schäper S, Brito AD, Liao Q, Tesseur C, Sorg M, et al. Chromosome segregation dynamics during the cell cycle of Staphylococcus aureus. Nat Commun. 2025;16(1):8626. pmid:41022742
View Article
PubMed/NCBI
Google Scholar

[224] View Article

[225] PubMed/NCBI

[226] Google Scholar

[ref58] 58. Veiga H, Jorge AM, Pinho MG. Absence of nucleoid occlusion effector Noc impairs formation of orthogonal FtsZ rings during Staphylococcus aureus cell division. Mol Microbiol. 2011;80(5):1366–80. pmid:21477126
View Article
PubMed/NCBI
Google Scholar

[228] View Article

[229] PubMed/NCBI

[230] Google Scholar

[ref59] 59. Ramos-León F, Anjuwon-Foster BR, Anantharaman V, Updegrove TB, Ferreira CN, Ibrahim AM, et al. PcdA promotes orthogonal division plane selection in Staphylococcus aureus. Nat Microbiol. 2024;9(11):2997–3012. pmid:39468247
View Article
PubMed/NCBI
Google Scholar

[232] View Article

[233] PubMed/NCBI

[234] Google Scholar

[ref60] 60. Ramos-León F, Ramamurthi KS. Staphylococcus aureus as an emerging model to study bacterial cell division. J Biol Chem. 2025;301(7):110343. pmid:40484379
View Article
PubMed/NCBI
Google Scholar

[236] View Article

[237] PubMed/NCBI

[238] Google Scholar

[ref61] 61. Lambert RJ, Pearson J. Susceptibility testing: accurate and reproducible minimum inhibitory concentration (MIC) and non-inhibitory concentration (NIC) values. J Appl Microbiol. 2000;88(5):784–90. pmid:10792538
View Article
PubMed/NCBI
Google Scholar

[240] View Article

[241] PubMed/NCBI

[242] Google Scholar

[ref62] 62. Deatherage DE, Barrick JE. Identification of mutations in laboratory-evolved microbes from next-generation sequencing data using breseq. In: Sun L, Shou W, editors. In: Engineering and analyzing multicellular systems: methods and protocols. Vol. 1151. New York, NY: Humana Press; 2014. pp. 165–88. https://doi.org/10.1007/978-1-4939-0554-6_12
View Article
Google Scholar

[244] View Article

[245] Google Scholar

[ref63] 63. Wilkins D. gggenes: Draw Gene Arrow Maps in ‘ggplot2’. 2023. Available from: https://wilkox.org/gggenes/

[ref64] 64. Schuster CF, Howard SA, Gründling A. Use of the counter selectable marker PheS* for genome engineering in Staphylococcus aureus. Microbiology (Reading). 2019;165(5):572–84. pmid:30942689
View Article
PubMed/NCBI
Google Scholar

[248] View Article

[249] PubMed/NCBI

[250] Google Scholar

[ref65] 65. Krute CN, Krausz KL, Markiewicz MA, Joyner JA, Pokhrel S, Hall PR, et al. Generation of a Stable Plasmid for In Vitro and In Vivo Studies of Staphylococcus Species. Appl Environ Microbiol. 2016;82(23):6859–69. pmid:27637878
View Article
PubMed/NCBI
Google Scholar

[252] View Article

[253] PubMed/NCBI

[254] Google Scholar

[ref66] 66. Bose JL, Fey PD, Bayles KW. Genetic tools to enhance the study of gene function and regulation in Staphylococcus aureus. Appl Environ Microbiol. 2013;79(7):2218–24. pmid:23354696
View Article
PubMed/NCBI
Google Scholar

[256] View Article

[257] PubMed/NCBI

[258] Google Scholar

[ref67] 67. Mlynek KD, Bulock LL, Stone CJ, Curran LJ, Sadykov MR, Bayles KW, et al. Genetic and Biochemical Analysis of CodY-Mediated Cell Aggregation in Staphylococcus aureus Reveals an Interaction between Extracellular DNA and Polysaccharide in the Extracellular Matrix. J Bacteriol. 2020;202(8):e00593-19. pmid:32015143
View Article
PubMed/NCBI
Google Scholar

[260] View Article

[261] PubMed/NCBI

[262] Google Scholar

[ref68] 68. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10:421. pmid:20003500
View Article
PubMed/NCBI
Google Scholar

[264] View Article

[265] PubMed/NCBI

[266] Google Scholar

[ref69] 69. Dwyer DJ, Camacho DM, Kohanski MA, Callura JM, Collins JJ. Antibiotic-induced bacterial cell death exhibits physiological and biochemical hallmarks of apoptosis. Mol Cell. 2012;46(5):561–72. pmid:22633370
View Article
PubMed/NCBI
Google Scholar

[268] View Article

[269] PubMed/NCBI

[270] Google Scholar

[ref70] 70. Overton WR. Modified histogram subtraction technique for analysis of flow cytometry data. Cytometry. 1988;9(6):619–26. pmid:3061754
View Article
PubMed/NCBI
Google Scholar

[272] View Article

[273] PubMed/NCBI

[274] Google Scholar

[ref71] 71. Marchler-Bauer A, Bryant SH. CD-Search: protein domain annotations on the fly. Nucleic Acids Res. 2004;32(Web Server issue):W327–31. pmid:15215404
View Article
PubMed/NCBI
Google Scholar

[276] View Article

[277] PubMed/NCBI

[278] Google Scholar

[ref72] 72. Wang J, Chitsaz F, Derbyshire MK, Gonzales NR, Gwadz M, Lu S, et al. The conserved domain database in 2023. Nucleic Acids Res. 2023;51(D1):D384–8. pmid:36477806
View Article
PubMed/NCBI
Google Scholar

[280] View Article

[281] PubMed/NCBI

[282] Google Scholar

Figures

Abstract

Author summary

Introduction

Results

Resistance evolution via gene amplification requires the recombinase RecA

Absence of functional RecA results in a two-step evolution of DLX resistance

RecA is required for loss of amplified copies of sdrM

Mutant library screen shows that evolution of DLX resistance leads to prevalent sdrM amplification with diverse junctions

Mutations increasing sdrM expression are associated with a lack of sdrM amplification

XerC, but not the DSB repair pathway, is a major modulator of sdrM gene amplification

The role of XerC in gene amplification is independent of RecA expression

Discussion

Methods

Strains and growth conditions

Minimum inhibitory concentration measurements

Growth curves and analysis

Experimental evolution of DLX resistance in tubes

Experimental deep well evolution of DLX resistance

Whole genome sequencing

Construction of allele replacement, overexpression, and reporter strains in S. aureus JE2

Mutagenesis of lexA gene

Phage transduction

Synteny analysis

Fluorescent promoter-reporter assay

Quantitative PCR (qPCR)

Reverse transcription quantitative PCR (RT-qPCR)

TOPO TA cloning

Determination of amplification stability

TUNEL assay

Statistics

Supporting information

S1 Fig. Selected mutants show a wide range of DLX sensitivities.

S2 Fig. DLX causes DNA damage.

S3 Fig. Mutations seen in RexB and RecD2 in the initial recA::Tn evolved populations are in relatively conserved residues.

S4 Fig. Amplified segments containing sdrM have a fitness cost.

S5 Fig. Proximity of sdrM homologs to the rRNA-tRNA locus is conserved in related species.

S6 Fig. Novel junctions corresponding to the sdrM amplification can be detected at a 10-4 – 10-5 dilution.

S7 Fig. Rare novel junctions around sdrM may be present in the unselected WT strain.

S8 Fig. The second region with increased coverage in several DLX resistant evolved populations encodes for a putative intact prophage.

S9 Fig. An intergenic mutation upstream of sdrM leads to increased sdrM expression.

S10 Fig. Scheme for the deep well evolution assay.

S11 Fig. Growth and DLX resistance defects in the ΔxerC mutant can be complemented by XerC expression.

S12 Fig. The ΔxerC mutant does not have a defect in recA induction upon DLX exposure.

S1 Table. Mutations seen in the recA::Tn evolved populations.

S2 Table. Mutations seen in the evolved populations of recA::Tn intermediate isolate R1 complemented with either an empty pKK30 plasmid or one carrying recA.

S3 Table. List of strains evolved for DLX resistance.

S4 Table. The distinct amplified fragments observed in the evolved populations from the tube evolutions of transposon and allele-replacement mutants.

S5 Table. Distances between sdrM homologs and the nearest rRNA locus in multiple Staphylococcus and Mammaliicoccus species.

S6 Table. Mutations seen in the evolved populations from the tube evolutions of transposon and allele-replacement mutants.

S7 Table. Mutations seen in the evolved populations of the P090/095G→A allele-replacement strain.

S8 Table. Mutations seen in the evolved populations from the deep-well evolutions.

S9 Table. Strains and plasmids used in this study.

S10 Table. Primers used in this study.

S1 Data. Numerical data underlying graphs shown in figures.

Acknowledgments

References

S6 Fig. Novel junctions corresponding to the sdrM amplification can be detected at a 10^-4 – 10^-5 dilution.

S7 Table. Mutations seen in the evolved populations of the P_090/095^G→A allele-replacement strain.