Skip to main content
NCBI home page
iScience logoLink to iScience
. 2020 Oct 28;23(11):101740. doi: 10.1016/j.isci.2020.101740

N-Acetyl-D-Glucosamine Acts as Adjuvant that Re-Sensitizes Starvation-Induced Antibiotic-Tolerant Population of E. Coli to β-Lactam

Miaomiao Wang 1,2, Edward Wai Chi Chan 1, Chen Yang 1,2, Kaichao Chen 2, Pui-kin So 1, Sheng Chen 1,2,3,∗
PMCID: PMC7662850  PMID: 33225246

Summary

Bacterial tolerance to antibiotics causes reduction in efficacy in antimicrobial treatment of chronic and recurrent infections. Nutrient availability is one major factor that determines the degree of phenotypic antibiotic tolerance. In an attempt to test if specific nutrients can reverse phenotypic tolerance, we identified N-acetyl-D-glucosamine (GlcNAc) as a potent tolerance-suppressing agent and showed that it could strongly re-sensitize a tolerant population of E. coli to ampicillin. Such re-sensitization effect was attributable to two physiology-modulating effects of GlcNAc. First, uptake of GlcNAc by the tolerant population triggers formation of the peptidoglycan precursor UDP-N-acetyl-D-glucosamine (UDP-GlcNAc) and subsequently re-activates the peptidoglycan biosynthesis process, rendering the organism susceptible to β-lactam antibiotics. Second, activation of glycolysis by-products of GlcNAc catabolism drives the re-sensitization process. Our findings imply that GlcNAc may serve as a non-toxic β-lactam adjuvant that enhances the efficacy of treatment of otherwise hard-to-treat bacterial infections due to phenotypic antibiotic tolerance.

Subject Areas: Biological Sciences, Microbiology, Microorganism, Molecular Microbiology

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Exogenous GlcNAc renders antibiotic-tolerant E. coli susceptible to β-lactams

  • •

    Exogenous GlcNAc re-activates peptidoglycan synthesis under starvation

  • •

    GlcNAc-elicited peptidoglycan synthesis also involves activation of glycolysis

  • •

    GlcNAc re-sensitization effect is not associated with cAMP and ROS production

Biological Sciences; Microbiology; Microorganism; Molecular Microbiology

Introduction

Current antimicrobial regimens are often ineffective in completely eradicating the infecting agents in treatment of bacterial infections, predisposing occurrence of relapsed and chronic infections. Antibiotic-tolerant sub-populations of non-resistant strains are recalcitrant to the bactericidal effects of even a high dose of antimicrobials, and are often responsible for failure of antimicrobial therapy, for example 26.6% women experienced recurrence of urinary tract infection of E. coli within 6 months of the initial infection (Fisher et al., 2017; Foxman, 1990; Russo et al., 1995). Antibiotic tolerance is a reversible phenotype initiated stochastically or by environmental stresses and is characterized by a slow or non-growing state, which renders antimicrobial action ineffective; yet a tolerant bacterial sub-population can rapidly resume active growth when environmental conditions become favorable (Bernier et al., 2013; de Steenwinkel et al., 2010; Horne and Tomasz, 1977; Yu et al., 2019).

Based on the current understanding of cellular mechanisms that underlie tolerance formation, several strategies have been proposed to eradicate antibiotic-tolerant bacterial cells. The first approach involves direct killing of metabolically inactive tolerant cell. Two synthetic retinoid compounds, CD437 and CD1530, were found to exhibit killing effects on S. aureus (methicillin-resistant S. aureus [MRSA])-tolerant cells that are unresponsive to methicillin by disrupting the membrane bilayers (Kim et al., 2018). Yet these compounds, which were previously used as anti-tumor agents, are highly toxic and hence unsuitable for treatment of patients who suffer from serious bacterial infections (Han et al., 2016; Valli et al., 2008). Another compound, ADEP4, which renders Clp protease ATP independent, was found to promote protein digestion and kill tolerant M. tuberculosis and S. aureus organisms (Conlon et al., 2013). The second approach aims to resuscitate tolerant cells or re-sensitize them to antibiotics. Cis-2-decenoic acid, a fatty acid signaling molecule, was found to revert the dormant cells to a metabolically active state, which is susceptible to antimicrobial agents (Marques et al., 2014). Metabolites such as glucose, fructose, pyruvate, and mannitol were also reported to sensitize tolerant cells to gentamicin by promoting the generation of a proton motive force (PMF), which is required for aminoglycoside uptake (Allison et al., 2011; Barraud et al., 2013; Meylan et al., 2017). The combined usage of several conventional antibiotics also exerts synergistic bactericidal effect on tolerant cells' effect. For example, daptomycin boosts the efficacy of eradication of Borrelia burgdorferi-tolerant cells by β-lactams (Feng et al., 2015). The third approach is to interfere with the tolerance formation process. It was reported that nitric oxide perturbs tolerance formation and renders bacterial cells susceptible to ampicillin or ofloxacin by inhibiting respiration (Orman and Brynildsen, 2016). Benzamide-benzamidazole-type compounds enable tolerant cells to be killed by meropenem, ciprofloxacin, and amikacin by inhibiting quorum sensing (Starkey et al., 2014). PKUMDL-LTQ series compounds, which are inhibitors of the toxin HipA, were reported to exhibit an effect of reducing the level of bacterial tolerance to ampicillin and kanamycin (Li et al., 2016). Likewise, phage-encoded expression of LexA3 inhibits SOS response-mediated DNA repair, rendering both planktonic and biofilm-associated tolerant cells susceptible to quinolones, β-lactams, and aminoglycosides (Lu and Collins, 2009).

Clinical application of the aforementioned methods for eradication of bacterial tolerant sub-population has been limited by drawbacks of each of these methods. For example, compounds that mediate killing of tolerant cells are often toxic to mammalian cells or are not suitable for clinical application (Kim et al., 2018). In this work, we screened for nutrients that can suppress starvation-induced tolerance response, so as to develop a non-toxic and clinically feasible anti-tolerance therapeutic approach. We found that N-acetyl-D-glucosamine (GlcNAc) and D-glucosamine (GlcN) could strongly re-sensitize bacterial tolerant sub-population that emerged during nutrient starvation stress to β-lactams. GlcN·HCl was previously reported to exhibit a strong antimicrobial effect on Staphylococcus saprophytics and Micrococcus luteus (Malik et al., 2013). GlcNAc is also known to induce death of tolerant cells in Candida albicans by triggering reactive oxygen species (ROS) (Du et al., 2015). GlcNAc is a major component of the bacterial cell wall, a matrix composed of alternating GlcNAc and N-acetylmuramic acid (MurNAc) molecules cross-linked by peptide bridges and are the target of β-lactams. Both GlcNAc and GlcN are converted to glucosamine- 6-phosphate (GlcN6P) upon transportation into the cytosol (Mukhija and Erni, 1996; Souza et al., 1997). GlcN6P could enter the hexosamine pathway and become converted to UDP- N-acetyl-D-glucosamine (UDP-GlcNAc), the precursor for synthesis of peptidoglycan; it may also be catabolized to Fru6P, a substrate of glycolysis (Midelfort and Rose, 1977; Rodriguez-Diaz et al., 2012).

Despite the previous findings, the mechanisms by which GlcNAc or GlcN inhibits bacterial antibiotic tolerance remain poorly defined. We hypothesize that GlcNAc re-sensitizes tolerant bacterial cells to β-lactams through the following mechanisms. First, GlcNAc triggers activation of peptidoglycan synthesis by generation of UDP-GlcNAc, so that the target of β-lactams becomes functionally active. Hurdle et al. proposed that the bacterial membrane bilayer and membrane proteins, which are integral to membrane function, were ideal targets of potential tolerance-inhibitory agents, as even tolerant cells need to maintain an integrated membrane and cell wall to remain viable (Hurdle et al., 2011). Bacteriophage-derived lysine CF-301 is known to exhibit pronounced MRSA killing effect by causing cell lysis by eliciting peptidoglycan degradation (Schuch et al., 2014). GlcNAc does not cause peptidoglycan degradation; instead it activates the peptidoglycan synthesis processes in tolerant cells, thereby re-sensitizing them to β-lactams. Second, GlcNAc catabolism produces Fru6P, which in turn promotes onset of glycolysis, revives oxidative phosphorylation and ATP production, and subsequently reverts the dormant cells to at least a partially metabolically active and β-lactam-sensitive state. Based on findings in this work, we believe that GlcNAc-mediated sensitization of tolerant cells to β-lactam is a highly effective and feasible tolerant cell eradication approach.

Results

Exogenous GlcNAc Drastically Enhances Susceptibility of Starvation-Induced Antibiotic-Tolerant Population of E. Coli to β-Lactam Both In Vitro and In Vivo

Bacteria become highly tolerant to multiple antibiotics upon complete deprivation of all essential nutrients (Fung et al., 2010; Nguyen et al., 2011). We hypothesize that such starvation-mediated tolerance phenotype is not solely due to shutdown of metabolic activities but also involved in the activation of cellular mechanisms that enhance the capability of bacteria to survive environmental stresses, and that starvation-induced protective mechanisms can be suppressed by supplementation of specific nutrients. To test these hypotheses, we screened for compounds that could suppress phenotypic tolerance by re-sensitizing the bacterial sub-population to ampicillin. Among a total of 95 carbon sources in the Phenotype MicroArrays (PM plates, Biolog) tested (Figure S1), we identified GlcNAc as the compound that exhibited the highest potential in re-sensitizing the tolerant sub-population to ampicillin. In the presence of this compound, ampicillin was able to eradicate a significant proportion of tolerant sub-population, so that only a small number of bacteria in such population were able to survive and re-grow in the screening assay (Figure S1). GlcNAc was non-toxic, as its presence did not affect bacterial growth in LB broth (Figure S2A). We also determined if GlcNAc could support re-growth of a starvation-induced antibiotic-tolerant population. In the presence of GlcNAc, the size of the bacterial population that had been subjected to starvation for 24 h by re-suspending in saline increased only about 2-fold in a 24-h period (∼1.2 × 108 CFU/mL), when compared with the negative control in which GlcNAc was not supplemented (∼5.5 × 107 CFU/mL). Considering the large original cell population, the effect of GlcNAc on bacterial proliferation is negligible (Figure S2B). We then further assessed the potential of GlcNAc to re-sensitize a starvation-induced tolerant population to ampicillin. Treatment with 100 μg/mL ampicillin for 24 h was only able to eradicate a very small proportion of the test population, with ∼5 × 107 CFU/mL of the original population of ∼108 CFU/mL being able to survive the treatment, indicating that a large tolerant sub-population formed during nutrient starvation. In the presence of GlcNAc, however, the survival fitness of this tolerant sub-population was found to decrease drastically upon ampicillin treatment. When compared with the sample treated with ampicillin alone, the survival rate of the tolerant sub-population decreased ∼4 logs upon treatment with ampicillin in the presence of 1 mM GlcNAc (Figure 1A). With an increasing GlcNAc concentration or increasing treatment time, tolerant cells could be almost completely eradicated (Figures 1A and 1C). To further confirm the effect of GlcNAc, we also tested whether GlcN, which can be assimilated by bacteria in a manner similar to that of GlcNAc, could re-sensitize tolerant sub-population. The re-sensitization effect of GlcN was found to be similar to that of GlcNAc (Figure S2C).

Figure 1.

Figure 1

Open in a new tab

Effect of Supplementation of Exogenous GlcNAc on Susceptibility of Starvation-Induced Tolerant E coli Sub-population to β-Lactams both In Vitro and In Vivo

(A) Starved E. coli were incubated with increasing concentrations of GlcNAc for 24 h in the presence and absence of ampicillin (AMP,100 μg/mL).

(B) Starved E. coli were incubated with different β-lactams for 24 h in the presence or absence of GlcNAc (1 mM); cells re-suspended in saline were included as negative control. CTX, cefotaxime (100 μg/mL); MEM, meropenem (40 μg/mL); FOS, fosfomycin (40 μg/mL).

(C) Starved E. coli were incubated with ampicillin for up to 33 h in the presence or absence of GlcNAc (1 mM); cells re-suspended in saline were included as negative control.

(D) Stationary phase E. coli cells were re-suspended in saline and then incubated with ampicillin for 24 h in the presence of GlcNAc (1 mM) or GlcN (1 mM); cells re-suspended in saline were included as negative control.

(E) E. coli mice deep-seeded thigh infection model. 1×107 cells were injected to the two thighs of each test mouse. At 24 h post-infection, the mice were subjected to the indicated dosage of antibacterial treatment (intraperitoneally) every 12 h for 72 h. Mice were euthanized and the thighs were aseptically excised, homogenized in saline, and the number of bacteria that survived determined (expressed in CFU). Results are displayed as mean ± SD, and the degree of significance was determined by unpaired, two-sided Student's test (ns, no significance; ∗∗p < 0.01; ∗∗∗∗p < 0.0001). Results of animal studies were analyzed by one-way ANOVA and post-hoc Tukey test (∗∗p< 0.01). Three biological repeats were carried out.

See also Figures S1 and S2.

Apart from ampicillin, GlcNAc was also found to re-sensitize starvation-induced tolerant cells to other β-lactams and fosfomycin, which target peptidoglycan synthesis. β-Lactam antibiotics target and inhibit transpeptidase, commonly known as penicillin-binding proteins, which mediate the final assembly of peptidoglycan by catalyzing cross-linkage between the neighboring stem peptides via formation of pentaglycine bridges. Fosfomycin prevents the formation of UDP-GlcNAc-3-O-enoplyruvate from UDP-GlcNAc, the first step in peptidoglycan biosynthesis (Kahan et al., 1974). In the presence of GlcNAc, the susceptibility of starvation-induced tolerant cells to cefotaxime sharply increased (Figure 1B). Cell survival in the presence of meropenem or fosfomycin plus GlcNAc was two to three orders of magnitude lower than that recorded upon treatment with meropenem or fosfomycin alone (Figure 1B). Apart from their re-sensitization effect on starvation-induced antibiotic-tolerant sub-population, GlcNAc and GlcN were also able to re-sensitize stationary phase cells to ampicillin, with both being able to cause 4 log reduction in the size of the antibiotic-tolerant stationary phase population when used in combination with ampicillin (Figure 1D). The re-sensitization effect of GlcNAc was not only observable in stationary phase cells re-suspended in saline, and also in those re-suspended in spent media of stationary phase culture. In each case, a 6 log reduction in the size of the bacterial population was recorded following treatment with GlcNAc and ampicillin (Figure S2D). We also tested the effect of GlcNAc on a clinically isolated E. coli strain that is resistant to meropenem and found that survival of this strain upon encountering nutrient starvation stress also decreased ∼4 log when treated with GlcNAc and meropenem, indicating that β-lactam-resistant strains may also be killed by this approach when they enter the starvation-induced tolerance mode (Figure S2E).

The efficacy of the GlcNAc and β-lactam combination to eradicate bacterial tolerant sub-population was further investigated in a mouse infection model. A deep-seeded thigh tolerance model using E. coli BW25113 was established, with cefotaxime being used as the test agent; the results showed that the size of the surviving bacterial population recorded after treatment with the GlcNAc and cefotaxime combination was 4.8 × 105 CFU/g, which was significantly lower than that recorded after treatment with cefotaxime alone (1.9 × 107 CFU/g) (Figure 1E). Interestingly, GlcNAc did not enhance the susceptibility of the tolerant sub-population to aminoglycosides as cell survival did not change when GlcNAc was included in treatment with gentamicin, kanamycin, or streptomycin, suggesting that the underlying mechanism of GlaNAc-mediated sensitization was different from that mediated by glucose and fructose, which rendered tolerant cells susceptible to aminoglycosides (Figure S2F). Also, phenotypic tolerance was not altered when GlcNAc was used in combination with ciprofloxacin or rifampin (Figure S2F); this observation was consistent with a former report that showed that supplementation of metabolic intermediates in the glycolysis pathway did not cause tolerant cells to become susceptible to quinolones (Allison et al., 2011).

Morphology of Ampicillin-Treated Cells in the Presence of GlcNAc or GlcN

Scanning electron microscopic (SEM) study was performed on E. coli strain that had been treated with GlcNAc or GlcN in the presence or absence of ampicillin to examine the morphological changes in tolerant cells exposed to these compounds. Compared with the untreated cells, which had integrated and smooth membrane and exhibited homogeneous plasma distribution, cells treated with the GlcNAc and ampicillin combination showed signs of extensive membrane damage including cell lysis, release of cellular contents, and lack of cellular features (Figure 2). The morphology of bacteria subjected to treatment with GlcNAc alone was identical to that of the untreated control, with integrated and smooth membrane and homogeneous plasma distribution, confirming that GlcNAc itself did not exert any cellular damages. The cell structure and membrane integrity were not altered if only ampicillin was involved in treatment, and the cellular contents were homogeneously distributed without any sign of leakage, indicating that the bacterial cells were tolerant to ampicillin under starvation stress (Figure 2). Likewise, cell lysis and leakage of cellular content were observed upon treatment with the GlcN and ampicillin combination, but not GlcN alone (Figure S3). The SEM observations are consistent with the results of phenotypic assays, which showed that GlcNAc or GlcN could effectively re-sensitize tolerant cells to β-lactams.

Figure 2.

Figure 2

Open in a new tab

SEM Images of E. Coli Cells Treated with Ampicillin, GlcNAc, or the Combination of Both

Starved cells were incubated with ampicillin only (100 μg/mL), GlcNAc only (1 mM), or GlcNAc plus ampicillin (saline as negative control) for 24 h before taking the SEM image. See also Figure S3

GlcNAc Stimulates Peptidoglycan Synthesis

We next investigated the molecular basis of the re-sensitization effect of GlcNAc. Transportation of GlcNAc into the bacterial cell is accompanied by phosphorylation and deacetylation, becoming glucosamine 6-phosphate (GlcNAc6P) and then GlcN6P (NagA; EC 3.5.1.25). Likewise, glucosamine (GlcN) is also converted to GlcN6P upon entry into the bacterial cell (Postma et al., 1993), which then enters the GlcNAc metabolic pathway (Figure 7). In this work, we chose GlcNAc as a test compound and investigated the mechanism that underlies the strong GlcNAc re-sensitization effect. First, we hypothesized that the re-sensitization effect on bacterial antibiotic tolerance exhibited by GlcNAc was due to its ability to activate the GlcNAc metabolic pathway; we investigated if knockout of the nagA gene (ΔnagA), which encodes GlcNAc6P deacetylase and plays a role in converting GlcNAc6P to GlcN6P, could diminish the strength of tolerance re-sensitization. Our data showed that the cell survival rate of the ΔnagA strain upon treatment with ampicillin did not alter when supplemented with GlcNAc, indicating that activation of the GlcNAc metabolic pathway is necessary to convert tolerant cells back to the ampicillin-susceptible mode (Figure 3A).

Figure 7.

Figure 7

Open in a new tab

Proposed Mechanism of GlcNAc-Mediated Sensitization of Starvation-Inducible Bacterial Tolerant Sub-population to β-Lactams

GlcNAc/GlcN is transported into the cell membrane and becomes phosphorylated to generate GlcN6P, which was subsequently converted to UDP-GlcNAc and Fru6P, the precursor molecules for peptidoglycan synthesis and initiation of glycolysis, respectively. ATP produced by glycolysis is required for peptidoglycan synthesis. Initiation of peptidoglycan synthesis in tolerant cells under starvation stress renders them susceptible to β-lactam. OM, outer membrane; PG, peptidoglycan; IM, inner membrane.

Figure 3.

Figure 3

Open in a new tab

Exogenous GlcNAc Stimulates Peptidoglycan Synthesis

(A–C) (A) Population size of starved ΔnagA mutant strains incubated with ampicillin for 24 h in the presence or absence of GlcNAc (1 mM); cells re-suspended in saline were included as negative control. The intracellular amount of UDP-GlcNAc detectable in the absence (B) or presence (C) of GlcNAc was measured. Chromatograms of the liquid chromatographic-mass spectrometric (LC/MS) analysis of UDP-GlcNAc in the samples are presented (m/z of 606.2 to 385 collision energy 25V and retention time of 7 min in the HPLC column). Survival rate of tolerant cells is displayed as mean ± SD; the degree of significance in difference between the test and control samples was determined by unpaired, two-sided Student's test. Three biological repeats were carried out.

See also Figure S4

In the GlcNAc metabolic pathway, GlcNAc is eventually converted to UDP-GlcNAc, which then enters the peptidoglycan biosynthesis pathway. It is widely accepted that a switch to physiological dormancy underlies formation of bacterial antibiotic tolerance, as the targets of antibiotics are inactive in the dormant state (Harms et al., 2016; Lewis, 2007). If a specific antimicrobial target can be converted to at least a partially active state by metabolite stimulation, tolerant cells should become sensitive to antibiotics. To examine whether supplementation of GlcNAc can activate peptidoglycan synthesis in tolerant cells, we quantified the amount of intracellular UDP-GlcNAc after treating the tolerant cells with GlcNAc for 24 h. The concentration of UDP-GlcNAc in cells after starvation for 24 h was found to be 0.06 μM, but increased ∼80-fold upon treatment with GlcNAc (4.95 μM) (Figures 3B, 3C, and S4).

The sharply increased amount of UDP-GlcNAc detectable in tolerant cells upon supplementation with GlcNAc suggests that dormant tolerant cells had resumed GlcNAc metabolism and peptidoglycan biosynthesis. To further confirm that peptidoglycan was being synthesized in tolerant cells, we tested whether the fluorescent D-amino acid HADA (7-hydroxycoumarin-3-carboxylic acid-3-amino-D-alanine), which specifically binds to and stains newly synthetized peptidoglycan, could stain the cell wall of tolerant cells in the presence of GlcNAc (Kuru et al., 2012). Our data showed that none of the tolerant cells under starvation could be stained by HADA, indicating that peptidoglycan biosynthesis has grounded to a halt. However, most of the tolerant cells tested could be stained by HADA 24 h after supplementation with GlcNAc, indicating that peptidoglycan biosynthesis had resumed when GlcNAc was available (Figure 4A) and that active peptidoglycan biosynthesis elicited by GlcNAc renders tolerant cells susceptible to β-lactams.

Figure 4.

Figure 4

Open in a new tab

Confocal Microscopic Images of HADA-Labeled Newly Synthesized Peptidoglycan in E. Coli

(A and B) Wild-type (WT) E. coli (BW25113) (A) and ΔndhnuoI strain (B) were subjected to 24-h starvation in the presence or absence of GlcNAc. The left and right panels are the bright-field and fluorescence images, respectively (scale bars, 5 μm).

Exogenous GlcNAc Promotes Glycolysis

Apart from being a precursor of peptidoglycan biosynthesis, GlcNAc could also be catabolized to Fru6P by the enzyme deaminase, which is encoded by the nagB gene (Alvarez-Anorve et al., 2009) (Figure 7). Fru6P is a substrate of glycolysis. Fru6P was also found to exhibit ampicillin re-sensitization effect in the PM assay in this work (Figure S1). To determine whether supplementation of GlcNAc caused an increase in the amount of Fru6P and subsequently activated glycolysis and the tricarboxylic acid (TCA) cycle in tolerant cells, the amount of Fru6P in tolerant cells was quantified by high-performance liquid chromatography (HPLC). Upon starvation for 24 h, the amount of Fru6P in the tolerant sub-population tested was 0.02 μM; however, the level was found to increase ∼20-fold in the presence of GlcNAc (0.42 μM) (Figures 5B, 5C, and S5). This sharp increase in the amount of Fru6P in the cytosol of tolerant cells indicated that supplementation of exogenous GlcNAc stimulated its own catabolism in such cells.

Figure 5.

Figure 5

Open in a new tab

Exogenous GlcNAc Promotes Glycolysis and Significantly Enhances Sensitivity of Tolerant E. coli Sub-population to β-Lactam

(A–C) (A) Starved ΔnagB mutant strains were incubated with ampicillin for 24 h in the presence or absence of GlcNAc (1 mM); cells re-suspended in saline were included as negative control. Intracellular amount of Fru6P in the absence (B) or presence (C) of GlcNAc. Chromatograms of the LC/MS analysis of Fru6Pin the samples are presented (m/z of 259 to 79, collision energy 25 V, and retention time of 12.2 min in the HPLC column).

(D) NADH level in the presence or absence of GlcNAc.

(E) Variation in population size of E. coli BW25113, ΔnuoI, Δndh, and ΔndhnuoI. Strains were incubated with ampicillin for 24 h in the presence or absence of GlcNAc; strains treated with saline only were included as negative control. Results are displayed as mean ± SD. The degree of significance in difference between results of the test samples were determined by Student's t test (ns, no significance; ∗p < 0.1; ∗∗∗p < 0.001).

See also Figure S5

To further explore the relationship between GlcNAc catabolization and tolerant cell re-sensitization, we tested the effect of GlcNAc on the ΔnagB mutant. Unlike the wild-type cell, the survival rate of tolerant cells of the ΔnagB mutant did not decrease upon treatment with the GlcNAc and ampicillin combination, indicating that cells remained tolerant to ampicillin if GlcNAc catabolism was disturbed (Figure 5A). The GlcNAc re-sensitization effect was also not observable in the stationary phase-induced tolerant cells of the ΔnagB mutant re-suspended in spent media (Figure S2D). Although GlcNAc catabolism is essential for its ability to re-sensitize cells to ampicillin, the state of glycolysis or TCA cycle upon supplementation of GlcNAc, and whether these processes are involved in the re-sensitization effect of GlcNAc, remain unknown. NADH is the major product of glycolysis and TCA cycle and is a reducing agent directly involved in energy production by donating electrons in the electron transport chain to produce ATP (Saraste, 1999). We then tested the amount of NADH in the tolerant cells in the presence of GlcNAc and found that the level of NADH was about three times higher than that recorded in the absence of GlcNAc (Figure 5D). The increasing amount of NADH in tolerant cells indicated that glycolysis and TCA cycle could be activated by GlcNAc. We then tested whether deletion of either one or both genes that encode NADH dehydrogenase (NADH dehydrogenase Ⅰ[ΔnuoI], NADH dehydrogenase Ⅱ [Δndh], and both NADH dehydrogenases [ΔndhnuoI]), which play an essential role in ATP production, affected the tolerance-suppressing effect of GlcNAc. We found that tolerant cells of the ΔndhnuoI mutant could not be eradicated by the GlcNAc and ampicillin combination, as the size of the tolerant population of this strain remained similar to that treated by ampicillin alone (∼107 CFU/mL). However, the re-sensitization effect of GlcNAc could still be observed upon deletion of a single NADH dehydrogenase gene (Figure 5E), suggesting that activation of glycolysis and TCA cycle via enhancement of NADH production and dehydrogenation was an essential step in the GlcNAc re-sensitization process.

ATP is the final product of electron transport chain reaction, and it is also required for peptidoglycan synthesis (Munshi et al., 2013; Saraste, 1999). We hypothesized that the reason why GlcNAc could not re-sensitize the tolerant cells of the ΔndhnuoI mutant was that peptidoglycan synthesis was halted as energy production was limited due to the disruption of NADH dehydrogenation. Therefore, we determined whether peptidoglycan could be synthesized upon supplementation of GlcNAc in ΔndhnuoI. The fluorescence intensity in the ΔndhnuoI mutant as detected in HADA-labeling experiment was much lower than that of the wild-type in the presence of GlcNAc, indicating that GlcNAc-induced reactivation of peptidoglycan synthesis was blocked upon deletion of the ndh and nuoI genes (Figure 4B). This finding strongly suggests that ATP synthesis is required for activation of peptidoglycan synthesis, which in turn allows tolerant cells to be switched back to the β-lactam-susceptible mode. Apparently GlcNAc can serve as a substrate for glycolysis, as well as peptidoglycan synthesis.

cAMP Regulation and ROS Were Not Involved in GlcNAc Re-sensitization

Previous studies reported that GlcNAc elicited onset of the cAMP signaling pathway, which in turn induced morphological changes, ROS generation, and eventually triggered rapid cell death in C. albicans (Castilla et al., 1998; Huang et al., 2010). It was also reported that cAMP played a role in regulating the level of E. coli tolerance to β-lactams by mediating oxidative stress response and SOS-dependent DNA repair(Molina-Quiroz et al., 2018). Based on these previous findings, we hypothesized that the cAMP signal-dependent regulation and ROS might also be involved in GlcNAc re-sensitization. To test this hypothesis, we measured the cell survival rate of tolerant cells of the Δcrp, ΔcyaA, and ΔcpdA gene knockout mutants upon treatment with the GlcNAc and ampicillin combination. The cyaA gene encodes the cAMP synthase; the cpdA gene product is a cAMP-specific phosphodiesterase that hydrolyzes cAMP; the crp gene encodes the cAMP receptor protein (Berkowitz et al., 1980; Imamura et al., 1996). However, our results showed that the synergistic antimicrobial effect of the GlcNAc and ampicillin combination could still be exerted on tolerant cells of the Δcrp, ΔcyaA, and ΔcpdA mutants, indicating that cAMP-mediated regulations were not involved in the GlcNAc re-sensitization process (Figure 6A). To determine whether ROS was involved in the killing of the tolerant sub-population by ampicillin in the presence of GlcNAc, we used 3′-p-hydroxyphenyl fluorescein (HPF) to measure the level of hydroxyl radicals (·OH) before and after treatment with the GlcNAc and ampicillin combination and found that the amount of hydroxyl radicals remained unchanged upon treatment with GlcNAc and ampicillin (Figure 6B). We also performed 2′,7′-dichlorofluorescein diacetate (DCFDA) staining to detect total cellular ROS level and found that GlcNAc-treated cells exhibited the same level of ROS as the untreated tolerant cells (Figure 6C). These findings suggest that GlcNAc does not induce ROS production and that ROS is not involved in the killing of tolerant cells by the GlcNAc and ampicillin combination.

Figure 6.

Figure 6

Open in a new tab

cAMP-Mediated Regulation and ROS Production Were Not Involved in the GlcNAc Re-sensitization Process

(A) Survival rate of three gene knockout strains Δcrp, ΔcyaA, and ΔcpdA subjected to starvation by washing and re-suspending in saline and then treatment with ampicillin for 24 h in the presence or absence of GlcNAc. Strains treated with saline only were included as negative control.

(B and C) The level of intracellular ROS in the tolerant cells as measured by flow cytometry using two different fluorescence probes of different affinities to free radicals. Starved E. coli strain BW25113 was treated with ampicillin, GlcNAc, or a combination of ampicillin and GlcNAc for 24 h before staining with HPF to quantify hydroxyl radical (·OH) (B), or with DCFDA to quantify total cellular ROS (C).

Discussion

Bacterial antibiotic tolerant sub-populations, the main culprit of chronic and recurrent infections, have been identified in almost every bacterial species. Currently, research reports on tolerant cells mostly focus on identification of genes and metabolic pathways involved in tolerance formation without providing sufficient knowledge that allows design of an effective strategy for eradication of such cells (Harms et al., 2016). In addition, factors that trigger formation of tolerance are not well-defined. Nutrient starvation is one common condition known to strongly elicit onset of stress tolerance in bacteria (Fung et al., 2010). In this work, we aimed to identify non-toxic compounds that can alter the metabolome of physiologically dormant tolerant cells so that they become partially awakened and physiologically reverted back to the antibiotic-susceptible status.

In this study, we first performed high-throughput screening of known nutrients that could suppress phenotypic tolerance and found that GlcNAc and GlcN exhibited the most pronounced effect in converting tolerant cells back to the susceptibility status. Based on results of the gene knockout, HADA labeling and microscopy imaging experiments, we propose a re-sensitization mechanism: uptake of GlcNAc or GlcN by tolerant cells allows them to switch from dormancy to a metabolically active state in which peptidoglycan biosynthesizes resumes (Figure 7). Allison et al. previously reported that carbon metabolites, such as glucose, fructose, and mannitol, could significantly enhance the susceptibility of the tolerant sub-population to aminoglycosides via activation of glycolysis, generation of PMF, and subsequently, enhancement of aminoglycoside uptake (Allison et al., 2011). More recently, a phenomenon known as trehalose-catalytic shift in M. tuberculosis-tolerant cells was reported; such tolerant cells were found to utilize trehalose as an internal carbon source to synthesize central carbon metabolism intermediates involved in glycolysis and pentose phosphate pathway. Inhibitor of trehalose-catalytic shift process could be used as an adjuvant to potentiate antibiotic efficacy by interfering with these adaptive strategies (Lee et al., 2019). Apparently, shutdown of carbon metabolism and biosynthetic activities is a key step in tolerance formation and is also essential for maintenance of the tolerance phenotype. Hence interference by exogenous carbon sources can drastically alter the level of tolerance of bacteria to β-lactam drugs. As a carbon metabolite, GlcNAc can activate glycolysis and kick start the TCA cycle in tolerant cells as the levels of both Fru6P and NADH were found to have increased upon supplementation of GlaNAc. The finding that the GlcNAc re-sensitization effect was abolished if the NADH dehydrogenase genes were deleted indicates that glycolysis and assimilation of NADH by the electron transport chain are essential steps in the GlcNAc re-sensitization process or peptidoglycan synthesis (Figure 7). Our data therefore show that the underlying mechanism by which GlcNAc can re-sensitize bacterial tolerant cells to ampicillin involves eliciting glycolysis and allowing peptidoglycan biosynthesis to resume.

GlcNAc is known to induce cell death in C. albicans by inducing alteration in mitochondrial metabolism and activating the cAMP-regulated pathways, which in turn leads to accumulation of ROS and cell death (Du et al., 2015). Our data showed that the underlying mechanism of GlcNAc-mediated tolerant cell killing in E. coli differs from that in C. albicans, as GlcNAc alone can induce cell death in C. albicans due to its ability to cause ROS accumulation, whereas it cannot do so in E. coli. GlcNAc can also trigger an increase in cAMP level in C. albicans (Castilla et al., 1998). The signaling molecule cAMP plays an important role in regulation of the nag regulon, which contains genes whose products play a role in GlcNAc and GlcN metabolism in E. coli, and was reported to be involved in tolerance formation (Castilla et al., 1998; Molina-Quiroz et al., 2018; Plumbridge, 1990). However, we found that the GlcNAc re-sensitization effect was not affected by deletion of the crp, cyaA, and cpdA genes, suggesting that cAMP signaling is not involved in GlcNAc-re-sensitization of the tolerant sub-population to ampicillin, and that the bactericidal/re-sensitization effects of GlcNAc on bacteria and fungus involve activation of different cellular mechanisms.

GlcNAc and GlcN are known to exhibit beneficial pharmacological effects in treatment of osteoarthritis (Kubomura et al., 2017). The therapeutic potential of GlcNAc is mainly attributed to its anti-inflammatory and chondroprotective effects (Dalirfardouei et al., 2016). GlcNAc also shows promising therapeutic efficacy in treatment of chronic inflammatory bowel disease by restoring the intestinal matrix and improving the epithelial morphology (Salvatore et al., 2000). As GlcNAc is non-toxic, this compound should be further explored to determine its potential of being used as an antibiotic adjuvant for treatment of chronic and recurrent infections in immunocompromised patients caused by members of Enterobacteriaceae.

The experimental system reported here demonstrates a novel approach to restore the antibiotic susceptibility phenotype of bacterial tolerant cells by using GlcNAc as a β-lactam adjuvant. Findings in this work broaden the current understanding of tolerance physiology by demonstrating that activating specific metabolic pathways of tolerant cells can significantly enhance the susceptibility of tolerant cells to specific antibiotic.

Limitations of the Study

Our study demonstrated that supplementation of exogenous GlcNAc could effectively re-sensitize starvation-induced tolerant bacterial sub-population to β-lactams but not aminoglycosides and fluoroquinolones, indicating that the re-sensitization effect of GlcNAc is confined to activation of peptidoglycan synthesis. The cellular basis of this phenomenon needs to be investigated. At this stage we have only screened carbon-based nutrients. It remains to be seen whether supplementation of other types of nutrients, such as those of nitrogen and phosphorus sources, may also affect bacterial tolerance, and to what extent. On the other hand, the re-sensitization effect of GlcNAc has to be further tested in animal models as there may be factors impeding penetration of the compound into tolerant bacterial cells, especially those residing in phagocytes and those under conditions such as hypoxic stress, which may affect peptidoglycan synthesis and glycolysis. In other words, the clinical application potential of GlcNAc has to be evaluated.

Resource Availability

Lead Contact

Further information and requests for resources should be directed to and will be fulfilled by the Lead Contact, Sheng Chen (shechen@cityu.edu.hk).

Materials Availability

This study did not generate new unique reagents.

Data and Code Availability

All the data are available within the article.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

This study was funded by the Collaborative Research Fund from the Research Grant Council of the Government of Hong Kong SAR (C5026-16G) and the Research Impact Fund (R5011-18F).

Author Contributions

M.M.W. designed and performed the study; E.W.C.C. designed the study and edited the manuscript; Y.C. and K.C.C. helped with animal experiments; P.-k.S. helped with mass spectrum analysis; S.C. designed the study, supervised the whole project, and wrote the manuscript.

Declaration of Interests

The authors declare no competing interests.

Published: November 20, 2020

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2020.101740.

Supplemental Information

Document S1. Transparent Methods and Figures S1–S5
mmc1.pdf (430.4KB, pdf)

References

  1. Allison K.R., Brynildsen M.P., Collins J.J. Metabolite-enabled eradication of bacterial persisters by aminoglycosides. Nature. 2011;473:216–220. doi: 10.1038/nature10069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alvarez-Anorve L.I., Bustos-Jaimes I., Calcagno M.L., Plumbridge J. Allosteric regulation of glucosamine-6-phosphate deaminase (NagB) and growth of Escherichia coli on glucosamine. J. Bacteriol. 2009;191:6401–6407. doi: 10.1128/JB.00633-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barraud N., Buson A., Jarolimek W., Rice S.A. Mannitol enhances antibiotic sensitivity of persister bacteria in Pseudomonas aeruginosa biofilms. PLoS One. 2013;8:e84220. doi: 10.1371/journal.pone.0084220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Berkowitz S.A., Goldhammer A.R., Hewlett E.L., Wolff J. Activation of prokaryotic adenylate cyclase by calmodulin. Ann. N Y Acad. Sci. 1980;356:360. doi: 10.1111/j.1749-6632.1980.tb29626.x. [DOI] [PubMed] [Google Scholar]
  5. Bernier S.P., Lebeaux D., DeFrancesco A.S., Valomon A., Soubigou G., Coppee J.Y., Ghigo J.M., Beloin C. Starvation, together with the SOS response, mediates high biofilm-specific tolerance to the fluoroquinolone ofloxacin. Plos Genet. 2013;9:e1003144. doi: 10.1371/journal.pgen.1003144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Castilla R., Passeron S., Cantore M.L. N-acetyl-D-glucosamine induces germination in Candida albicans through a mechanism sensitive to inhibitors of cAMP-dependent protein kinase. Cell Signal. 1998;10:713–719. doi: 10.1016/s0898-6568(98)00015-1. [DOI] [PubMed] [Google Scholar]
  7. Conlon B.P., Nakayasu E.S., Fleck L.E., LaFleur M.D., Isabella V.M., Coleman K., Leonard S.N., Smith R.D., Adkins J.N., Lewis K. Activated ClpP kills persisters and eradicates a chronic biofilm infection. Nature. 2013;503:365–370. doi: 10.1038/nature12790. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dalirfardouei R., Karimi G., Jamialahmadi K. Molecular mechanisms and biomedical applications of glucosamine as a potential multifunctional therapeutic agent. Life Sci. 2016;152:21–29. doi: 10.1016/j.lfs.2016.03.028. [DOI] [PubMed] [Google Scholar]
  9. Du H., Guan G., Li X., Gulati M., Tao L., Cao C., Johnson A.D., Nobile C.J., Huang G. N-Acetylglucosamine-Induced cell death in Candida albicans and its implications for adaptive mechanisms of nutrient sensing in yeasts. MBio. 2015;6 doi: 10.1128/mBio.01376-15. e01376–01315. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Feng J., Auwaerter P.G., Zhang Y. Drug combinations against Borrelia burgdorferi persisters in vitro: eradication achieved by using daptomycin, cefoperazone and doxycycline. PLoS One. 2015;10:e0117207. doi: 10.1371/journal.pone.0117207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fisher R.A., Gollan B., Helaine S. Persistent bacterial infections and persister cells. Nat. Rev. Microbiol. 2017;15:453–464. doi: 10.1038/nrmicro.2017.42. [DOI] [PubMed] [Google Scholar]
  12. Foxman B. Recurring urinary tract infection: incidence and risk factors. Am. J. Public Health. 1990;80:331–333. doi: 10.2105/ajph.80.3.331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fung D.K., Chan E.W., Chin M.L., Chan R.C. Delineation of a bacterial starvation stress response network which can mediate antibiotic tolerance development. Antimicrob. Agents Chemother. 2010;54:1082–1093. doi: 10.1128/AAC.01218-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Han T., Goralski M., Capota E., Padrick S.B., Kim J., Xie Y., Nijhawan D. The antitumor toxin CD437 is a direct inhibitor of DNA polymerase alpha. Nat. Chem. Biol. 2016;12:511–515. doi: 10.1038/nchembio.2082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Harms A., Maisonneuve E., Gerdes K. Mechanisms of bacterial persistence during stress and antibiotic exposure. Science. 2016;354:aaf4268. doi: 10.1126/science.aaf4268. [DOI] [PubMed] [Google Scholar]
  16. Horne D., Tomasz A. Tolerant response of Streptococcus sanguis to beta-lactams and other cell wall inhibitors. Antimicrob. Agents Chemother. 1977;11:888–896. doi: 10.1128/aac.11.5.888. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huang G., Yi S., Sahni N., Daniels K.J., Srikantha T., Soll D.R. N-acetylglucosamine induces white to opaque switching, a mating prerequisite in Candida albicans. Plos Pathog. 2010;6:e1000806. doi: 10.1371/journal.ppat.1000806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hurdle J.G., O'Neill A.J., Chopra I., Lee R.E. Targeting bacterial membrane function: an underexploited mechanism for treating persistent infections. Nat. Rev. Microbiol. 2011;9:62–75. doi: 10.1038/nrmicro2474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Imamura R., Yamanaka K., Ogura T., Hiraga S., Fujita N., Ishihama A., Niki H. Identification of the cpdA gene encoding cyclic 3',5'-adenosine monophosphate phosphodiesterase in Escherichia coli. J. Biol. Chem. 1996;271:25423–25429. doi: 10.1074/jbc.271.41.25423. [DOI] [PubMed] [Google Scholar]
  20. Kahan F.M., Kahan J.S., Cassidy P.J., Kropp H. The mechanism of action of fosfomycin (phosphonomycin) Ann. N Y Acad. Sci. 1974;235:364–386. doi: 10.1111/j.1749-6632.1974.tb43277.x. [DOI] [PubMed] [Google Scholar]
  21. Kim W., Zhu W., Hendricks G.L., Van Tyne D., Steele A.D., Keohane C.E., Fricke N., Conery A.L., Shen S., Pan W. A new class of synthetic retinoid antibiotics effective against bacterial persisters. Nature. 2018;556:103–107. doi: 10.1038/nature26157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kubomura D., Ueno T., Yamada M., Tomonaga A., Nagaoka I. Effect of N-acetylglucosamine administration on cartilage metabolism and safety in healthy subjects without symptoms of arthritis: a case report. Exp. Ther. Med. 2017;13:1614–1621. doi: 10.3892/etm.2017.4140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kuru E., Hughes H.V., Brown P.J., Hall E., Tekkam S., Cava F., de Pedro M.A., Brun Y.V., VanNieuwenhze M.S. In Situ probing of newly synthesized peptidoglycan in live bacteria with fluorescent D-amino acids. Angew. Chem. Int. Ed. Engl. 2012;51:12519–12523. doi: 10.1002/anie.201206749. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lee J.J., Lee S.K., Song N., Nathan T.O., Swarts B.M., Eum S.Y., Ehrt S., Cho S.N., Eoh H. Transient drug-tolerance and permanent drug-resistance rely on the trehalose-catalytic shift in Mycobacterium tuberculosis. Nat. Commun. 2019;10:2928. doi: 10.1038/s41467-019-10975-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lewis K. Persister cells, dormancy and infectious disease. Nat. Rev. Microbiol. 2007;5:48–56. doi: 10.1038/nrmicro1557. [DOI] [PubMed] [Google Scholar]
  26. Li T., Yin N., Liu H., Pei J., Lai L. Novel inhibitors of toxin HipA reduce multidrug tolerant persisters. ACS Med. Chem. Lett. 2016;7:449–453. doi: 10.1021/acsmedchemlett.5b00420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lu T.K., Collins J.J. Engineered bacteriophage targeting gene networks as adjuvants for antibiotic therapy. Proc. Natl. Acad. Sci. U S A. 2009;106:4629–4634. doi: 10.1073/pnas.0800442106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Malik S., Singh M., Mathur A. Antimicrobial activity of food grade glucosamine. Int. J. Biotechnol. Bioeng. Res. 2013;4:307–312. [Google Scholar]
  29. Marques C.N., Morozov A., Planzos P., Zelaya H.M. The fatty acid signaling molecule cis-2-decenoic acid increases metabolic activity and reverts persister cells to an antimicrobial-susceptible state. Appl. Environ. Microbiol. 2014;80:6976–6991. doi: 10.1128/AEM.01576-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Meylan S., Porter C.B.M., Yang J.H., Belenky P., Gutierrez A., Lobritz M.A., Park J., Kim S.H., Moskowitz S.M., Collins J.J. Carbon sources tune antibiotic susceptibility in Pseudomonas aeruginosa via tricarboxylic acid cycle control. Cell Chem. Biol. 2017;24:195–206. doi: 10.1016/j.chembiol.2016.12.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Midelfort C.F., Rose I.A. Studies on the mechanism of Escherichia coli glucosamine-6-phosphate isomerase. Biochemistry. 1977;16:1590–1596. doi: 10.1021/bi00627a010. [DOI] [PubMed] [Google Scholar]
  32. Molina-Quiroz R.C., Silva-Valenzuela C., Brewster J., Castro-Nallar E., Levy S.B., Camilli A. Cyclic AMP regulates bacterial persistence through repression of the oxidative stress response and SOS-dependent DNA repair in uropathogenic Escherichia coli. mBio. 2018;9:e02144-17. doi: 10.1128/mBio.02144-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mukhija S., Erni B. Purification by Ni2+ affinity chromatography, and functional reconstitution of the transporter for N-acetylglucosamine of Escherichia coli. J. Biol. Chem. 1996;271:14819–14824. doi: 10.1074/jbc.271.25.14819. [DOI] [PubMed] [Google Scholar]
  34. Munshi T., Gupta A., Evangelopoulos D., Guzman J.D., Gibbons S., Keep N.H., Bhakta S. Characterisation of ATP-dependent Mur ligases involved in the biogenesis of cell wall peptidoglycan in Mycobacterium tuberculosis. PLoS One. 2013;8:e60143. doi: 10.1371/journal.pone.0060143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Nguyen D., Joshi-Datar A., Lepine F., Bauerle E., Olakanmi O., Beer K., McKay G., Siehnel R., Schafhauser J., Wang Y. Active starvation responses mediate antibiotic tolerance in biofilms and nutrient-limited bacteria. Science. 2011;334:982–986. doi: 10.1126/science.1211037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Orman M.A., Brynildsen M.P. Persister formation in Escherichia coli can be inhibited by treatment with nitric oxide. Free Radic. Biol. Med. 2016;93:145–154. doi: 10.1016/j.freeradbiomed.2016.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Plumbridge J.A. Induction of the nag regulon of Escherichia coli by N-acetylglucosamine and glucosamine: role of the cyclic AMP-catabolite activator protein complex in expression of the regulon. J. Bacteriol. 1990;172:2728–2735. doi: 10.1128/jb.172.5.2728-2735.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Postma P.W., Lengeler J.W., Jacobson G.R. Phosphoenolpyruvate:carbohydrate phosphotransferase systems of bacteria. Microbiol. Rev. 1993;57:543–594. doi: 10.1128/mr.57.3.543-594.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Rodriguez-Diaz J., Rubio-Del-Campo A., Yebra M.J. Regulatory insights into the production of UDP-N-acetylglucosamine by Lactobacillus casei. Bioengineered. 2012;3:339–342. doi: 10.4161/bioe.21271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Russo T.A., Stapleton A., Wenderoth S., Hooton T.M., Stamm W.E. Chromosomal restriction fragment length polymorphism analysis of Escherichia coli strains causing recurrent urinary tract infections in young women. J. Infect. Dis. 1995;172:440–445. doi: 10.1093/infdis/172.2.440. [DOI] [PubMed] [Google Scholar]
  41. Salvatore S., Heuschkel R., Tomlin S., Davies S.E., Edwards S., Walker-Smith J.A., French I., Murch S.H. A pilot study of N-acetyl glucosamine, a nutritional substrate for glycosaminoglycan synthesis, in paediatric chronic inflammatory bowel disease. Aliment. Pharmacol. Ther. 2000;14:1567–1579. doi: 10.1046/j.1365-2036.2000.00883.x. [DOI] [PubMed] [Google Scholar]
  42. Saraste M. Oxidative phosphorylation at the fin de siecle. Science. 1999;283:1488–1493. doi: 10.1126/science.283.5407.1488. [DOI] [PubMed] [Google Scholar]
  43. Schuch R., Lee H.M., Schneider B.C., Sauve K.L., Law C., Khan B.K., Rotolo J.A., Horiuchi Y., Couto D.E., Raz A. Combination therapy with lysin CF-301 and antibiotic is superior to antibiotic alone for treating methicillin-resistant Staphylococcus aureus-induced murine bacteremia. J. Infect. Dis. 2014;209:1469–1478. doi: 10.1093/infdis/jit637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Souza J.M., Plumbridge J.A., Calcagno M.L. N-acetylglucosamine-6-phosphate deacetylase from Escherichia coli: purification and molecular and kinetic characterization. Arch. Biochem. Biophys. 1997;340:338–346. doi: 10.1006/abbi.1997.9780. [DOI] [PubMed] [Google Scholar]
  45. Starkey M., Lepine F., Maura D., Bandyopadhaya A., Lesic B., He J., Kitao T., Righi V., Milot S., Tzika A. Identification of anti-virulence compounds that disrupt quorum-sensing regulated acute and persistent pathogenicity. Plos Pathog. 2014;10:e1004321. doi: 10.1371/journal.ppat.1004321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. de Steenwinkel J.E., de Knegt G.J., ten Kate M.T., van Belkum A., Verbrugh H.A., Kremer K., van Soolingen D., Bakker-Woudenberg I.A. Time-kill kinetics of anti-tuberculosis drugs, and emergence of resistance, in relation to metabolic activity of Mycobacterium tuberculosis. J. Antimicrob. Chemother. 2010;65:2582–2589. doi: 10.1093/jac/dkq374. [DOI] [PubMed] [Google Scholar]
  47. Valli C., Paroni G., Di Francesco A.M., Riccardi R., Tavecchio M., Erba E., Boldetti A., Gianni M., Fratelli M., Pisano C. Atypical retinoids ST1926 and CD437 are S-phase-specific agents causing DNA double-strand breaks: significance for the cytotoxic and antiproliferative activity. Mol. Cancer Ther. 2008;7:2941–2954. doi: 10.1158/1535-7163.MCT-08-0419. [DOI] [PubMed] [Google Scholar]
  48. Yu J., Liu Y., Yin H., Chang Z. Regrowth-delay body as a bacterial subcellular structure marking multidrug-tolerant persisters. Cell Discov. 2019;5:8. doi: 10.1038/s41421-019-0080-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent Methods and Figures S1–S5
mmc1.pdf (430.4KB, pdf)

Data Availability Statement

All the data are available within the article.

Articles from iScience are provided here courtesy of Elsevier

RESOURCES