There is no standardized protocol to predict the concentration levels of microbicides that are left on surfaces as a result of the use of these products, and there is no standardized method to predict the potential risk that such levels pose to emerging antibacterial resistance. The ability to distinguish between selection and adaption processes for antimicrobial resistance in bacteria and the impact of different concentrations of microbicide exposure have not been fully investigated to date. This study considers the effect of exposure to a low concentration of chlorhexidine digluconate (CHX) on selected phenotypes of Escherichia coli and relates the findings to the risk of emerging antimicrobial resistance.
KEYWORDS: Escherichia coli, chlorhexidine, cross-resistance, metabolome, microbicide, resistance
ABSTRACT
There is no standardized protocol to predict the concentration levels of microbicides that are left on surfaces as a result of the use of these products, and there is no standardized method to predict the potential risk that such levels pose to emerging antibacterial resistance. The ability to distinguish between selection and adaption processes for antimicrobial resistance in bacteria and the impact of different concentrations of microbicide exposure have not been fully investigated to date. This study considers the effect of exposure to a low concentration of chlorhexidine digluconate (CHX) on selected phenotypes of Escherichia coli and relates the findings to the risk of emerging antimicrobial resistance. A concentration of 0.006 mg/ml CHX is a realistic “during use” exposure concentration measured on surfaces. At this concentration, it was possible for CHX-susceptible bacteria to survive, adapt through metabolic alterations, exhibit a transient decrease in antimicrobial susceptibility, and express stable clinical cross-resistance to front-line antibiotics. Efflux activity was present naturally in tested isolates, and it increased in the presence of 0.00005 mg/ml CHX but ceased with 0.002 mg/ml CHX. Phenotypic microarray assays highlighted a difference in metabolic regulation at 0.00005 mg/ml and 0.002 mg/ml CHX; more changes occurred after growth with the latter concentration. Metabolic phenotype changes were observed for substrates involved with the metabolism of some amino acids, cofactors, and secondary metabolites. It was possible for one isolate to continue transferring ampicillin resistance in the presence of 0.00005 mg/ml CHX, whilst 0.002 mg/ml CHX prevented conjugative transfer. In conclusion, E. coli phenotype responses to CHX exposure are concentration dependent, with realistic residual CHX concentrations resulting in stable clinical cross-resistance to antibiotics.
INTRODUCTION
The European biocidal product regulation (1) aims to standardize and monitor the introduction to the market and use of biocidal products (BP). Included in the “conditions for granting an authorization” section (1), it is stipulated that the responsible regulatory body must be notified if the product manufacturer is aware or becomes aware that there is potential for the development of resistance to the active substance. It is also mandatory that the “biocidal product has no unacceptable effects on the target organisms,” in particular, “unacceptable resistance or cross-resistance,” and that the “chemical diversity of the active substances is adequate to minimize the occurrence of resistance in the target harmful organism.” Despite these clear requirements, there is not as yet a standardized method to evaluate and predict the risk of the development of resistance to microbicides. There is also not an adequate understanding of how environmental concentrations of microbicides remaining after disinfection affect surviving target organisms. Knapp and colleagues (2) proposed a decision tree-based method that utilizes MICs and minimal bactericidal concentration (MBCs) combined with antibiotic susceptibility profiling and stability testing to predict potential microbicide-driven resistance when applied in situ. The method focused on “in-use” concentrations of chlorhexidine and benzalkonium chloride and investigated surviving bacteria for any changes in the susceptibility profile to antimicrobial agents, including the microbicide or biocidal formulation (2).
Biocidal formulated products are generally applied with the active ingredient at a concentration considerably higher than necessary to kill targeted microorganisms. The concentration of the active ingredient that is applied at the point of initial decontamination is termed “in-use” concentration (3). Biocidal product activity depends upon a number of factors such as concentration of actives, contact time, formulation excipients, pH, organic challenge, temperature, and target microorganisms (4). Product dilution upon usage and surface will contribute to a lower residual concentration on surfaces, which may result in concentrations below the MIC for given microorganisms (5).
Chlorhexidine digluconate (CHX) is typically incorporated into surface disinfectant products at 20 mg/ml and is often marketed with claims of residual activity of up to 6 h and even 48 h when applied to the skin surface (6). The concentration of chlorhexidine in products typically varies from 0.1 mg/ml in ophthalmic preparations to 40 mg/ml in surgical scrubs (7). The Scientific Committee on Emerging and Newly Identified Health Risks (8) stated, “Despite the regulatory requirements to study the environmental stability of individual products, data on the fate and concentrations of microbicides in the environment are sparse.” Although CHX may provide a degree of antimicrobial activity at a low concentration, there is little information available on the CHX concentration remaining on surfaces following application and on the effect of residual microbicide concentrations on emerging resistance in bacteria.
This investigation aims to expose a range of previously characterized Escherichia coli strains to a concentration of CHX typically found as a residual on surfaces after application and to measure any changes in antimicrobial susceptibility profile that arise, while providing some understanding of the associated bacterial mechanisms.
RESULTS
Measurements of any changes in antimicrobial susceptibility following exposure to CHX.
There was an average 99.97% decrease in CHX concentration once the solution was removed from the glass surface at time zero. The average CHX concentration recovered over time was recorded as 0.006 ± 0.002 mg/ml (Fig. 1). Baseline MIC and MBC data showed that the CHX concentration recovered from the glass surface was within the range of the CHX MIC and MBC for the test strains (see Table S1 in the supplemental material). There was no statistical correlation between drying time and CHX concentration (P = 0.62; Pearson’s correlation analysis, r2 = 0.07).
FIG 1.
Concentration of CHX (initially set at 20 mg/m) recovered after drying this solution directly on a glass surface. Error bars are standard deviations from the means. Dashed lines depict the range of MIC and MBC values for all E. coli isolates tested (see Table S1 in the supplemental material). Note the abscissa distance for the histogram is not proportional to increasing drying time (data based on 3 biological repeats).
When exposed to CHX 20 mg/ml for 30 s in a suspension test, no viable bacteria were recovered (Fig. 2). However, exposure to lower concentrations (0.002 or 0.007 mg/ml) resulted only in a 2- to 3-log10 reduction in E. coli 13P5 or IB2 viability regardless of the contact time (up to 5 min) (Fig. 2). There was no significant difference (P ≥ 0.0001) in bacterial killing following exposure to 0.002 or 0.007 mg/ml CHX. Preexposing isolates to 0.0075 mg/ml CHX did not affect (P ≥ 0.0001) their susceptibility to the different CHX concentrations tested (Fig. 2).
FIG 2.
Inactivation kinetics of E. coli 1B2 (a) and 13P5 (b) in the presence of CHX concentrations of 0.002 (circles), 0.007 (triangles), or 20 mg/ml (squares). Isolates were preexposed to no CHX (white) or 0.0075 mg/ml CHX (black) (data based on 3 biological repeats).
When exposed directly to residual CHX concentrations remaining on surfaces (0.0047 to 0.0075 mg/ml) for 5 min, a significant (>5 log10) reduction in bacterial viability was observed (Table 1). When viable bacteria where recovered after a 5-min exposure time, changes in MIC/MBC were observed primarily from surfaces with a 24-and 168-h CHX drying time (Table 2). The highest fold increase in MIC (32-fold) was observed in E. coli ATCC 25922 and the CTX-M-14 isolates (IL3, IL4, and 1B2) on surfaces dried for 168 h, while the highest MBC increase (62-fold) was observed with 1L3 and 1B2 on surfaces dried for 168 h.
TABLE 1.
Bacterial reduction in viability following a 5-min contact time with CHX concentrations found on surfaces after drying for different lengths of timea
| E. coli strain | Drying time (h) (CHX concn [mg/ml])b
|
|||
|---|---|---|---|---|
| 0 (0.0049) | 6 (0.0097) | 24 (0.0047) | 168 (0.0075) | |
| ATCC 25922 (n = 3)c | >5.82 ± 0.15 | >5.82 ± 0.15 | >5.82 ±0.15 | >5.83 ±0.08 |
| CTX-M-14 | ||||
| 1L3 (n = 3) | >5.81 ± 0.44 | >5.81 ± 0.44 | >5.81 ± 0.44 | 5.34 ± 0.49 |
| 1L4 (n = 3) | >5.93 ± 0.29 | 5.08 ± 0.41 | >5.93 ± 0.29 | >5.99 ± 0.05 |
| 1B2 (n = 5) | >6.00 ± 0.21 | >6.00 ± 0.21 | 5.62 ± 0.54 | 5.46 ±0.47 |
| CTX-M-15 | ||||
| 13P4 (n = 3) | >5.67 ± 0.40 | >5.67 ± 0.40 | 5.55 ± 0.41 | 5.49 ± 0.68 |
| 13P5 (n = 5) | >5.89 ± 0.13 | 5.76 ± 0.28 | 5.58 ± 0.44 | 5.58 ± 0.63 |
| TEM-20 | ||||
| 25P5 (n = 3) | >5.78 ± 0.26 | >5.78 ± 0.26 | >5.78 ± 0.26 | >5.81 ± 0.01 |
| 25OS1 (n = 3) | >5.80 ± 0.30 | >5.80 ± 0.30 | >5.80 ± 0.30 | >5.92 ± 0.30 |
Based on Fig. 1.
Values with “>” indicate no viable bacteria recovered. The limit of detection was 1 × 102 CFU/ml.
n, number of biological replicates.
TABLE 2.
Changes in susceptibility for bacteria surviving a 5-min exposure to CHX concentrations measured on surfaces after drying
| Strain | Baseline MIC or MBC (mg/ml) | MIC or MBC for CHX concn (mg/ml [fold increase]): |
|||
|---|---|---|---|---|---|
| 0.0049 | 0.0097 | 0.0047 | 0.0075 | ||
| For MIC determinations | |||||
| E. coli ATCC 25922 | 0.005 | 0.04 (8-fold)a | −b | 0.04 (8-fold)a | 0.16 (32-fold)a |
| CTX-M-14 | |||||
| E. coli 1L3 | 0.005 | − | 0.04 (8-fold)a | 0.04 (8-fold)a | 0.16 (32-fold)c |
| E. coli 1L4 | 0.005 | − | 0.04 (8-fold)a | − | 0.16 (32-fold)a |
| E. coli 1B2 | 0.005 | − | − | 0.04 (8-fold)c | 0.16 (32-fold)c |
| 0.01 | − | − | 0.08 (8-fold) | 0.16 (16-fold) | |
| 0.02 | − | − | 0.04 (4-fold) | 0.16 (8-fold) | |
| CTX-M-15 | |||||
| E. coli 13P4 | 0.02 | − | − | − | − |
| E. coli 13P5 | 0.01 | − | 0.04 (4-fold)a | 0.08 (8-fold)a | 0.1 (10-fold)a |
| TEM-20 | |||||
| E. coli 25P5 | 0.005 | − | − | 0.04 (4-fold)a | − |
| E. coli 25OS1 | 0.005 | − | − | − | − |
| For MBC determinations | |||||
| E. coli ATCC 25922 | 0.005 | 0.04 (8-fold)a | − | 0.08 (16-fold)a | 0.16 (32-fold)a |
| CTX-M-14 | |||||
| E. coli 1L3 | − | 0.04 (8-fold)a | 0.04 (8-fold)a | 0.16 (32-fold)a | − |
| − | − | − | 0.32 (62-fold)a | − | |
| E. coli 1L4 | 0.005 | − | 0.04 (8-fold)a | − | 0.16 (32-fold)a |
| E. coli 1B2 | 0.005 | − | - | 0.08 (16-fold)c | 0.31 (62-fold)c |
| 0.01 | − | - | 0.08 (8-fold) | 0.16 (16-fold) | |
| CTX-M-15 | |||||
| E. coli 13P4 | 0.005 | − | − | − | − |
| E. coli 13P5 | 0.01 | − | 0.04 (4-fold)a | 0.08 (8-fold)a | 0.16 (16-fold)a |
| TEM-20 | |||||
| E. coli 25P5 | 0.005 | − | − | 0.04 (4-fold)a | − |
| E. coli 25OS1 | 0.005 | − | − | − | − |
Observed in one-third of repeats.
−, no recoverable growth postexposure.
Observed in one-half of repeats.
CHX MBC increased further for E. coli 13P5 following one additional passage in CHX 0.002 mg/ml. However, MBC decreased to the baseline value following 5 passages in CHX 0.002 mg/ml and remained at baseline value after 10 passages (Fig. 3). There were no significant changes (P ≥ 0.0001) in MBC for E. coli 1B2 following passaging in CHX 0.002 mg/ml (Fig. 3). When changes in MIC were considered, there were no significant differences in MIC (P ≥ 0.0001) before and following passaging either of the isolates in CHX 0.002 mg/ml (see Fig. S1).
FIG 3.
Average MBC values for E. coli 13P5 and 1B2 before and after 1, 5, and 10 passages in CHX 0.002 mg/ml (gray) or in broth only (white). Isolates were originally exposed to 0.0075 mg/ml CHX (corresponding to CHX concentration recovered from surfaces after 168 h). (a) E. coli 13P5; (b) E. coli 1B2. Standard deviations from the mean are shown (data based on 3 biological repeats).
Escherichia coli ATCC 25922 was found to be susceptible to all antibiotics tested, while the isolates showed different antibiotic susceptibility pattern (see Table S2). CTX-M-14 isolates were clinically resistant to tetracycline, trimethoprim, and trimethoprim-sulfamethoxazole; CTX-M-15 and TEM-20 isolates were resistant to ampicillin, cefpodoxime, and cephalothin (Table S2). In addition, E. coli 1B2 (CTX-M-14) was resistant to streptomycin, and E. coli 13P5 (CTX-M-15) was also resistant to tetracycline, streptomycin, nalidixic acid, trimethoprim, and trimethoprim-sulfamethoxazole (Table S2).
Exposure to CHX 0.0047 or 0.0075 mg/ml resulted in amoxicillin/clavulanic acid or amoxicillin/clavulanic acid and cefoxitin resistance in E. coli 13P5 and in amoxicillin/clavulanic acid resistance in E. coli 1B2 following exposure to 0.0047 mg/ml CHX (Table 3). Clinical resistance to amoxicillin/clavulanic acid remained stable in E. coli 13P5 following passaging, but resistance to cefoxitin was lost (Table 3). In contrast, E. coli 1B2 developed additional clinical resistance to ampicillin, amoxicillin/clavulanic acid, cefpodoxime, and cephalothin following 5 passage in CHX 0.0002 mg/ml or in broth.
TABLE 3.
Clinically relevant changes from antibiotic sensitive to resistant according to EUCAST (2020) breakpoint values for E. coli before and after a 5-min initial exposure to, and passage in, CHX or broth only
| Strain | Initially exposed to CHX at:a
|
|||
|---|---|---|---|---|
| 0.0047 mg/ml, passaged in: |
0.0075 mg/ml, passaged in: |
|||
| CHX | Broth | CHX | Broth | |
| E. coli 13P5 | ||||
| Initial exposure | AMC | AMC, FOX | ||
| Passage 1 | AMC | −b | AMC | − |
| Passage 5 | AMC | AMC | AMC, IPMc | AMC |
| Passage 10 | AMC | AMC | AMC | AMC |
| E. coli 1B2 | ||||
| Initial exposure | − | AMC | ||
| Passage 1 | − | AMCc | AMPc | AMC |
| Passage 5 | − | AMP, AMCc , CIP, CPD, CF | AMP, AMC, CPD, CF | AMP, AMC, CIP, CPD, CF |
| Passage 10 | CF | AMP, AMC, CPD, CF | AMP, AMC, CPD, CF | AMP, AMC, CPD, CF |
AMP, ampicillin; AMC, amoxicillin/clavulanic acid; CPD, cefpodoxime; CF, cephalothin; CIP, ciprofloxacin; IMP, imipenem; FOX, cefoxitin.
−, no change in antibiotic susceptibility observed.
Only observed in one-half of the repeats.
Conjugation assay.
Exposure to 0.00005 mg/ml CHX did not result in a statistically significant change (P = 0.730) in transfer rates in E. coli 13P5, as measured by the rescue of the ampicillin phenotype from the initial transfer rate of 1.34 × 10−5. Exposure to 0.002 mg/ml CHX appeared to halt conjugation (see Table S3).
Effect of exposure to CHX residues on efflux.
Bacterial exposure to CHX at 0.00005 mg/ml resulted in active efflux regardless of the isolate, as indicated by no change in relative fluorescence (Fig. 4). The addition of carbonyl cyanide 3-chlorophenylhydrazone (CCCP) causes the level of fluorescence to increase (P ≤ 0.0001), confirming the fact that efflux pumps may be responsible for the extrusion of ethidium bromide (EtBr).
FIG 4.
Trace graphs showing relative fluorescence values recorded for E. coli exposed to CHX after 10 min and with or without the addition of CCCP. ○, CHX 0.00005 mg/ml; ■, CHX 0.00005 mg/ml plus CCCP. (a) E. coli ATCC 25922; (b) E. coli IB2; (c) E. coli 13P5.
Effect of CHX residues on microbial metabolism.
Overall, bacterial exposure to a concentration of 0.002 mg/ml CHX had a greater impact on metabolism than exposure to 0.005 mg/ml CHX (Table 4). Bacterial growth in the presence of CHX 0.00005 mg/ml reduced the ability of E. coli 13P5 to metabolize salicin, 6.5% (wt/vol) NaCl, and 4% (wt/vol) urea (see Fig. S2).
TABLE 4.
Summary of CHX exposures at concentrations of 0.00005 or 0.002 mg/ml on bacterial metabolisma
| Metabolism condition | Effect comparison |
|||
|---|---|---|---|---|
| No CHX compared to CHX (mg/ml) at: |
Compared to each other at CHX (mg/ml): |
|||
| 0.00005 | 0.002 | 0.00005 (mg/ml) | 0.002 (mg/ml) | |
| Salicin | ↓ | = | ↓ | ↑ |
| Methylene diphosphonic acid | ↑ | = | ↑ | ↓ |
| pH 4.5 plus l-alanine | ↑ | ↑ | = | = |
| pH 4.5 plus l-serine | = | ↑ | = | = |
| pH 9.5 plus phenylethylamine | ↑ | ↑ | = | = |
| 5.5% (wt/vol) NaCl | = | ↓ | ↑ | ↓ |
| 6% (wt/vol) NaCl | = | ↓ | ↑ | ↓ |
| 6% (wt/vol) NaCl and betaine | = | ↓ | = | = |
| 6% (wt/vol) NaCl and creatinine | = | ↓ | = | = |
| 4% (wt/vol) urea | ↓ | = | ↓ | ↑ |
| 5% (wt/vol) urea | ↓ | ↓ | = | = |
↑, increase; ↓, decrease; =, no change.
The metabolism of methylene diphosphonic acid increased in the presence of 0.00005 mg/ml CHX. Growth in the presence of CHX 0.002 mg/ml decreased the bacterial metabolism when exposed to 6% (wt/vol) NaCl and 5% (wt/vol) urea. However, the metabolism at pH 4.5 in the presence of L-serine and at pH 9.5 in the presence of L-phenylethylamine was increased (Fig. S2). Escherichia coli 13P5 was less efficient at metabolizing salicin when exposed to 0.00005 mg/ml than when exposed to 0.002 mg/ml or not exposed at all. Growth in the presence of CHX (0.00005 and 0.002 mg/ml) increased the ability of E. coli 13P5 to metabolize L-alanine at pH 4.5, L-phenylethylamine at pH 9.5, and 6% (wt/vol) NaCl and betaine compared to that by bacteria unexposed to CHX (Table 4).
When one considers the metabolic pathway mapping of E. coli (K-12), all of the changes observed in this study were located in the domains of amino acid metabolism, carbohydrate metabolism, metabolism of cofactors and vitamins, and the biosynthesis of secondary metabolites (Fig. 5).
FIG 5.
Metabolic network of E. coli K-12 (produced with https://www.genome.jp/kegg; accessed 23 July 2020). Location of pathways that include substrates involved in observed metabolic changes (n = 2). (a) Comparison between bacteria not exposed to CHX and those exposed to 0.00005 mg/ml CHX. (b) Comparison between bacteria not exposed to CHX and those exposed to 0.002 mg/ml CHX. ↑, increased metabolism of substrate; ↓, decreased metabolism of substrate.
DISCUSSION
The amount of CHX recovered from surfaces (an average of 0.006 ± 0.002 mg/ml) represents the “during use” exposure concentration, a realistic prediction of true-to-life conditions of product use during its application (3). “During use” exposure differed from “low concentration,” which often reflects a concentration value below the MIC and which is not necessarily representative of a concentration that applies in practice. It was noticeable there was some variability in the concentrations measured at recovery (range, 0.0047 to 0.0097 mg/ml) over a 168-h period, although the increasing concentration of CHX did not appear to have an impact on the MIC/MBC or on antibiotic susceptibility. Furthermore, measured CHX residual concentrations were between the MIC/MBC and 10-fold lower than the MIC/MBC for the isolates tested. These residual CHX concentrations were associated with greater changes in CHX susceptibility profile.
Some biocidal manufacturers make claims about their formulations, including having “residual biocidal activity.” Related to this is the suggestion that a subinhibitory concentration of a biocidal product may exert a selective pressure sufficient to drive bacterial adaptation (4, 9–16). In this study, CHX concentrations similar or 10-fold lower than the MIC/MBC were capable of selecting for changes in MIC/MBC to the biocidal active as well as altering bacterial susceptibility phenotypes to one or more antimicrobial compounds. These findings support the hypothesis that residual CHX concentrations remaining on abiotic surfaces can select for antibiotic resistance (Table 2). Furthermore, the stability of these clinical antibiotic susceptibility changes further narrows the chemotherapeutic options for treatment (Table 3).
To gain a better understanding of the impact of low subinhibitory CHX concentrations in selecting for changes in bacterial susceptibility, we investigated two concentrations, 0.0005 mg/ml, which provided an environment conducive to an adaptive tolerance while having a low antibacterial impact, and 0.002 mg/ml, exerting a greater selective condition and enhanced pressure for metabolic changes. Noticeable differences were recorded when bacteria were exposed to the two concentrations tested. CHX concentrations of 0.00005 mg/ml resulted in an increase in efflux pump activity, as measured by EtBr accumulation assays. In contrast CHX 0.002 mg/ml appeared to suppress efflux, although such observation is likely linked to the bactericidal effect of this concentration (data not shown). E. coli possesses multiple efflux transporters primarily belonging to the hydrophobic and amphiphilic efflux RND (HAE-RND) family (17). Indeed, CCCP has been used to screen RND efflux pump activity, although it is a proton motive force (PMF) inhibitor and is likely to have an impact on other bacterial PMF-driven functions (18). Here, we showed that an RND pump is likely involved in E. coli in response to CHX exposure and potentially contributes to a decrease in antibiotic susceptibility (Table 3). Deletion of acrA or tolC has been shown to render E. coli cells more susceptible to aminoglycosides but also to ampicillin (18). We did not perform any MIC testing in the presence of CCCP, which would have indicated the potential impact on an RND transporter. Future study will look at the expression of specific RND transporters when E. coli is exposed to residual CHX concentration.
There was also a clear disparity in the number of metabolic changes observed following prior exposure to these two concentrations of CHX, with CHX 0.002 mg/ml appearing to exhibit a greater effect on bacterial metabolism (Table 4). All of the metabolic changes observed were located in the domains of amino acid metabolism, carbohydrate metabolism, metabolism of cofactors and vitamins, and the biosynthesis of secondary metabolites (Fig. 5). These findings agree with those reported by Condell et al. (9), where the largest number of upregulated genes after CHX exposure were involved in general cell metabolism. A concentration of 4% (wt/vol) urea and salicin were metabolized equally either with exposure to 0.002 mg/ml CHX or in the absence of it but decreased following exposure to 0.00005 mg/ml CHX. Upon exposure to 0.002 mg/ml CHX, a 1.67-log10 reduction in viability for this bacterium was recorded, which, in this case, could have resulted from a surviving subpopulation of bacteria showing the same level of salicin metabolism (19–22). The ability to utilize salicin (a secondary metabolite β-glucoside) as a carbon source was previously shown to provide a competitive advantage, referred to as a growth advantage in stationary phase (GASP) in E. coli (23, 24). The bgl operon harbors the genes that function for β-glucoside catabolism and is a silent operon that has been implicated in the upregulation of proteins associated with transport functions or enzymes involved in cellular metabolism. The presence of the bgl operon appears to activate additional metabolic functions owing to enhanced ability to access nutrient substrates (23, 24). Furthermore, increases in other carbon sources such as L-alanine, L-serine, and phenylethylamine were identified. A combination of the increased utilization of amino acids and carbohydrates suggest that CHX induces a membrane-related stress response in E. coli 13P5. It has been suggested that surviving bacterial cells are directing mechanisms primarily towards cell membrane processes, such as changes in outer membrane structure and signaling functions (9, 25). The impact of a low concentration of CHX on metabolism and a potential mechanistic link to decreased susceptibility to CHX, but also unrelated compounds, is interesting and needs to be further investigated.
It has been shown that in the case of antibiotics and microbicides, this selection process might be relevant at very low concentrations (12, 13, 25–28). Antibiotic concentration-specific outcomes selecting for mutations relating to increased resistance have been highlighted (29–31), supporting the concept of a distinct difference in the selection process deriving from low and high concentrations of antibiotics. A similar comparative study has not yet been performed for microbicides.
In this study, we established the residual CHX concentration remaining on surfaces following the use of a solution of 20 mg/ml CHX. Importantly, bacteria surviving exposure to these residual concentrations were not less susceptible to the bactericidal effect of CHX than bacteria that were not exposed (Fig. 2). We also observed that exposure to different sub-MIC CHX concentrations (0.0005 mg/ml or 0.002 mg/ml) produced different responses in isolates and that the main two isolates, 13P5 (ST10; CTX-M-15) and 1B2 (ST1629; CTX-M-14), had different responses to these CHX sub-MICs. A decreased susceptibility (MIC and MBC) to CHX and antibiotics was observed, although a decreased CHX MIC was not stable. Although RND efflux may be involved (Fig. 4), differences in metabolism were also observed, which enforces the postulate that decreased antimicrobial susceptibility following exposure to a microbicide is multifactorial in bacteria (9, 32–34).
In conclusion, following CHX application to a surface, the residual concentration of CHX may be present in the environment at levels conducive to bacterial adaptation through metabolic changes, which may be associated with decreased CHX susceptibility and an increase in clinical resistance to antibiotics of importance to human health. The practical impact of such observations associated with CHX use would need be ascertained in situ, by monitoring environmental isolates where CHX products are used.
MATERIALS AND METHODS
Microorganisms.
This study investigated seven environmental isolates of E. coli (provided by S. Fanning; University College Dublin) and a reference strain E. coli ATCC 25922 that was used for comparison (Table 5). All strains were cultured in Müller-Hinton broth (MHB; Fisher Scientific, Loughborough, UK) and incubated at 37 ± 1°C for 18 to 24 h. When necessary, all bacterial strains were cultured on MH agar (MHA) plates and stored in the fridge at 4 to 6°C for up to 1 month. Overnight MHB bacterial cultures were centrifuged at 5,000 × g for 15 min at 20 ± 1°C. The supernatant recovered was discarded, and the bacterial pellet was then resuspended in 10 ml phosphate-buffered saline (PBS; Fisher Scientific, Loughborough, UK). Turbidity of these test suspensions was adjusted spectrophotometrically (optical density at 600 nm [OD600]) using sterile MHB to achieve a turbidity equivalent to approximately 1 × 108 to 2 × 108 CFU/ml. Enumeration of test suspensions was carried out using the drop counting method (35).
TABLE 5.
Summary of ESBL producing E. coli isolates and their resistance features
| Isolate | MLSTa | ESBLb | Size of plasmid(s) (kb) isolated from transconjugants |
|---|---|---|---|
| UCD-CFS ECP-1L3 | ST23 | CTX-M-14 | 200, 120 |
| UCD-CFS ECP-1B2 | ST1629 | CTX-M-14 | 110 |
| UCD-CFS ECP-1L4 | ST23 | CTX-M-14 | 130 |
| UCD-CFS ECP-13P5 | ST10 | CTX-M-15 | 80 |
| UCD-CFS ECP-13P4 | ST10 | CTX-M-15 | 70 |
| UCD-CFS ECP-25OS1 | ST34 (ST10 Cpix) | TEM-20 | 120, 60 |
| UCD-CFS ECP-25P5 | ST10 | TEM-20 | 110, 50 |
MLST, multilocus sequence type.
ESBL, extended-spectrum β-lactamase.
Determination of surface-dried CHX concentrations.
One milliliter of 20 mg/liter chlorhexidine digluconate (CHX; Fischer Scientific, UK) was pipetted into a flat-bottomed glass McCartney bottle. Then, 1 ml of CHX was removed with a pipette (time, 0 h), and the remaining residue was left to dry at room temperature (21°C) in a biological safety level 2 cabinet for 6, 24, 48, 96, or 168 h. After the appropriate drying time, 1 ml of sterile deionized water (diH2O) was added to the bottle, and CHX residue was resuspended using a vortex mixer and magnetic stirrer for 1 min. This solution was aspirated and dispensed into a glass autosampler vial for high-performance liquid chromatography (HPLC) analysis (Thermo Scientific, UK). The mobile phase was a 1:1 ratio of water and acetonitrile (HPLC grade; Sigma Aldrich, UK) with 0.5% (vol/vol) trifluoracetic acid (HPLC grade; Sigma Aldrich, UK). The retention time was 6 min. An initial calibration curve was made with a CHX standard stock (20 mg/ml), halving concentrations running from 0.5 mg/ml to 0.001 mg/ml.
Modified carrier test and bacterial survival after exposure.
Modified carrier tests were performed to measure bacterial cell survival after exposure to surface-dried residual concentrations of CHX. One milliliter of 20 mg/ml CHX was pipetted into a glass flat-bottomed McCartney bottle. Then, 1 ml was removed by pipette at 0 h, and the bottle was left to dry at room temperature (21°C) for 6, 24, or 168 h. After the appropriate drying time, 20 μl of standardized washed inoculum (1 × 108 to 2 × 108 CFU/ml) was added to the bottom of the McCartney bottle and left for an exposure time of either 5 min or 24 h. Following exposure, 1 ml De-Engley (DE) neutralizer (Fisher Scientific, Loughborough, UK) was added to the bottle, and the inoculum was resuspended using a vortex mixer for 1 min. The reduction in viability (log10) after exposure was determined. In addition, a 100-μl sample was removed from the test vial after neutralization, placed into 10 ml MHB, and incubated for 18 to 24 h at 37°C in order to perform additional susceptibility testing.
MIC and minimal bactericidal concentrations.
MICs and minimal bactericidal concentrations (MBCs) were measured using the British Standard (BS) European Norm (EN) International Organization for Standardization (ISO) 20776-1 (36) broth microdilution method before and after exposure to CHX. Each plate was incubated for 24 h at 37°C, and results were recorded based on positive or negative visible growth.
Antibiotic susceptibility testing.
An adjusted bacterial inoculum (1 × 104 CFU/ml) from samples taken before or after exposure to CHX was spread onto MHB agar plate. Antibiotic-containing discs (Beckton Dickinson, UK) were placed onto the agar surface, and plates were incubated for 24 h at 37°C. Zones of inhibition were recorded and breakpoints calculated in accordance with the EUCAST (2020) protocol (37).
Inactivation kinetics.
Inactivation kinetics were only investigated with E. coli 1B2 and 13P5. We tested both isolates that were preexposed to 0.0075 mg/ml CHX (corresponding to a 168-h drying time on surfaces) and isolates that were not preexposed to CHX. One milliliter of standardized bacterial culture (1× 109 CFU/ml) was mixed with 1 ml PBS. Eight milliliters of CHX at 20 mg/ml, 0.002 mg/ml, or 0.007 mg/ml was added to the bacterial suspension and vortexed for 30 s. E. coli 1B2 and 13P5 were exposed to CHX for contact times of 0.5, 1, 3, and 5 min at room temperature.
These concentrations were chosen to represent the in-use concentration (20 mg/ml), the concentration found left on a surface (0.006 ± 0.002 mg/ml), and the concentration below the MIC value (0.002 mg/ml) in order to exert a selective pressure but not necessarily kill all microbial cells. Following these contact times, 1 ml of each test suspension was added to 9 ml DE neutralizing agar and vortexed for 30 s. A volume of 100 μl of the neutralized mixture was diluted in 900 μl PBS, and surviving bacteria were enumerated in duplicates on MHB using the drop counting method. Plates were incubated at 37°C for 24 h, and CFU/ml were calculated. Inactivation kinetics were plotted using the log10 CFU/ml recovered over time.
Effect of exposure to CHX residues on efflux.
Change in efflux was only investigated with E. coli 1B2 and 13P5. Overnight bacterial cultures were adjusted to 1 × 108 CFU/ml in 20 ml sterile MHB. Suspensions were incubated at 37°C in a shaking incubator (120 rpm) until the mid-log growth phase was reached (OD600 of 0.2 to 0.3; approximately 2 to 3 h). Bacterial cells were centrifuged at 5,000 × g, and the resulting pellet washed with diH2O. Suspension was adjusted to a final OD600 of 0.4. One milliliter of each bacterial isolate suspension was removed and boiled (95°C) for 10 min to be used as a positive control. Fifty microliters of an ethidium bromide (EtBr; Sigma-Aldrich, UK) stock solution (10 mg/ml) was added to each well of a 96-well microtiter plate to give a final concentration of 0.005 mg/ml. Fifty microliters of carbonyl cyanide 3-chlorophenylhydrazone (CCCP; Sigma-Aldrich, UK), an efflux pump inhibitor (EPI), was added to appropriate wells to give a final concentration of 0.1 mM. One hundred microliters of either boiled bacterial cells or test bacterial cells was added to the plate, and 50 μl of PBS was added to those wells without EPI. The final volume in all wells was 250 μl. Plates were then scanned in a Tecan microplate reader (Tecan M200 Infinite PRO; Life Sciences, UK) at 37°C for an initial 10 min to obtain baseline fluorescence values. After 10 min, 50 μl CHX (0.00005 mg/ml) was injected to each well using the automated injector module equipped to the Tecan microplate reader. The plates were then scanned in the Tecan microplate reader for an additional 50 min. Background controls consisting of microbicides or EPI and EtBr with no bacteria were run alongside experimental sets and used to normalize data accounting for any background fluorescence omitted by the inhibitory compounds used.
Phenotype stability testing.
Phenotype stability was only investigated with E. coli 1B2 and 13P5. Stability in microbicide and antibiotic susceptibility changes observed after microbicide exposure was assessed through successive passaging of surviving bacteria in microbicide-free broth or broth supplemented with 0.002 mg/ml CHX (2). Daily (24 h) passages were performed over a 10-day period, and microbicide MIC, MBC, and antibiotic susceptibility profiles were measured after passages 1, 5, and 10.
Effect of CHX residues on microbial metabolism.
Change in metabolism were only investigated with E. coli 1B2 and 13P5. CHX (0.00005 mg/ml) was chosen to provide a low antibacterial effect while providing an environment for an adaptive tolerance, whereas 0.002 mg/ml CHX was chosen to create a less favorable environment providing a more pronounced selective condition and exerting an enhanced pressure for metabolic changes.
E. coli 13P5 was grown to its third generation with three subsequent subcultures onto MHA agar (incubated at 37°C; 16 to 18 h). One loopful of colonies was taken from an MHA plate, streaked onto a new MHA plate, and incubated at 37 ± 2°C for 18 to 24 h. Two independent biological replicates were grown for each test; the two replicates were tested on separate days (n = 8). The third generation was subcultured onto three MHA agar plates containing no CHX, 0.00005 mg/ml CHX, or 0.002 mg/ml CHX and incubated at 37 ± 2°C for 16 to 18 h. A preliminary test was then performed in order to ensure that the isolate would grow on the agar plates in the presence of these concentrations of CHX.
A range of phenotype microarray (PM) plates were selected for the assay (Table 6). Before being loaded into the PM microplates (Biolog, Inc., Hayward, CA, United States), 13P5 was incubated for 16 to 18 h on MHA plates containing no CHX, 0.00005 mg/ml CHX, or 0.002 mg/ml CHX. After incubation in the presence/absence of CHX, several colonies were selected with a sterile plastic culture loop and suspended into Inoculating Fluid-0 (IF-0; Biolog, Inc., Hayward, CA, United States) until a cell density of 42% transmittance (T42%) was reached in a turbidimeter (Biolog, Inc., Hayward, CA, United States).
TABLE 6.
Features associated with the selected phenotype microarray plates
| PM plate no. | Substrate class |
|---|---|
| PM1 | Carbon sources |
| PM2A | Carbon sources |
| PM3B | Nitrogen sources |
| PM4A | Phosphorous and sulfur sources |
| PM6 | Peptide nitrogen sources |
| PM7 | Peptide nitrogen sources |
| PM8 | Peptide nitrogen sources |
| PM9 | Osmolytes |
| PM10 | pH |
For plates PM1 and PM2, 15 ml T42% cell suspension was mixed with 75 ml of Biolog redox dye mix A (1:5 dilution) in order to create a final cell suspension of T85%. For PM3 to PM8, 680 μl 2 M sodium succinate-200 μM ferric citrate solution was added to 68 ml of the T85% cell suspension. One hundred microliters of each mixture was pipetted into each well of the appropriate microplate. All PM plates were incubated in an OmniLog reader at 37°C for 72 h. Readings were recorded every 15 min, and data were analyzed in OmniLog PM software (38, 39) (Biolog, Inc.). Each experiment was performed in duplicates on two separate days with independent bacterial cultures.
Conjugation assay.
To determine the conjugative transfer of the ampicillin (AMP) resistance determinant, the liquid mating method was according to that described by Lambrecht et al. (40). E. coli 13P5 was selected for further investigation due to its ampicillin resistance and its ability to recover and grow after exposure to CHX 0.002 mg/ml or 0.00005 mg/ml CHX for 5 min. The recipient E. coli J35R (S. Fanning; University College Dublin) was chosen for its chromosomally encoded rifampicin resistance. For each test, three independent biological replicates were performed. A single colony for each replicate was inoculated into 5 mL MHB for 16 to 18 h at 37°C. The donor, E. coli 13P5, was grown in the presence of ampicillin (100 μg/ml), and the recipient (E. coli J35R), was grown in the presence of rifampicin (100 μg/ml). Cultures were centrifuged at 5,000 × g, and the pellet was washed with PBS and resuspended in 5 ml MHB. Bacterial suspensions were diluted 10-fold in MHB with CHX to obtain an exposure concentration of 0.00005 mg/ml or 0.002 mg/ml. A control was performed with no CHX. Initial mating concentrations ranged from 2.30 × 107 CFU/ml to 7.30 × 107 CFU/ml for the donor strain (E. coli 13P5) and 1.17 × 108 CFU/ml to 7.30 × 108 CFU/ml for the recipient strain (E. coli J35R). Donor and recipient strains were mixed at a ratio of 1:5. Liquid mating was performed for 4 h at room temperature (25°C), after which, bacteria were enumerated using the spread plating technique. Enumerated mating suspensions were plated onto media containing ampicillin (100 μg/ml) for donors and transconjugants, rifampicin (100 μg/ml) for recipient transconjugants, or double selective plates containing ampicillin (100 μg/ml) and rifampicin (100 μg/ml) for transconjugants. Plates were incubated overnight 16 to 18 h at 37°C, and colonies were counted. The limit of detection for enumeration was 1 CFU/ml. The limit of quantification was ≥10 colonies/plate. Transfer ratios were calculated as the number of transconjugants divided by the number of recipients.
Statistical analysis
Pearson’s correlation analysis was used to determine the relationship between surface drying time and the concentration of CHX determined via HPLC. One-way and two-way analyses of variance (ANOVAs) were used when comparing differences between single and multiple factors, respectively.
Supplementary Material
ACKNOWLEDGMENT
This research received no specific grant.
Footnotes
Supplemental material is available online only.
REFERENCES
- 1.European Union. 2012. Regulation (EU) no. 528/2012 of the European Parliament and of the Council of 22 May 2012 concerning the making available on the market and use of biocidal products (text with EEA relevance). J Eur Union 167:1–123. [Google Scholar]
- 2.Knapp L, Amezquita A, Mcclure P, Stewart S, Maillard J-Y. 2015. Development of a protocol for predicting bacterial resistance to microbicides. Appl Environ Microbiol 81:2652. doi: 10.1128/AEM.03843-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wesgate R, Grasha P, Maillard J-Y. 2016. Use of a predictive protocol to measure the antimicrobial resistance risks associated with biocidal product usage. Am J Infect Control 44:458–464. doi: 10.1016/j.ajic.2015.11.009. [DOI] [PubMed] [Google Scholar]
- 4.Maillard J-Y. 2005. Antimicrobial microbicides in the healthcare environment: efficacy, usage, policies and perceived problems. Ther Clin Risk Manag 1:307–320. [PMC free article] [PubMed] [Google Scholar]
- 5.Siani H, Maillard J-Y. 2015. Best practice in healthcare environment decontamination. Eur J Clin Microbiol Infect Dis 34:1–11. doi: 10.1007/s10096-014-2205-9. [DOI] [PubMed] [Google Scholar]
- 6.George J, Klika AK, Higuera CA. 2017. Use of chlorhexidine preparations in total joint arthroplasty. J Bone Joint Infect 2:15–22. doi: 10.7150/jbji.16934. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rowe RC, Sheskey PJ, Quinn MA (ed). 2009. Chlorhexidine, p 162–165. Handbook of Pharmaceutical excipients, 6th edition Pharmaceutical Press, London, United Kingdom. [Google Scholar]
- 8.SCENIHR. 2010. Research strategy to address the knowledge gaps on the anitimicrobial resistance effects of biocides. https://ec.europa.eu/health/scientific_committees/emerging/docs/scenihr_o_028.pdf.
- 9.Condell O, Power KA, Händler K, Finn S, Sheridan A, Sergeant K, Renaut J, Burgess CM., Hinton JCD, Nally Jarlath E, Fanning S. 2014. Comparative analysis of Salmonella susceptibility and tolerance to the biocide chlorhexidine identifies a complex cellular defense network. Front Microbiol 5:373. doi: 10.3389/fmicb.2014.00373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Andersson D, Hughes D. 2012. Evolution of antibiotic resistance at non-lethal drug concentrations. Drug Resist Updat 15:162–172. doi: 10.1016/j.drup.2012.03.005. [DOI] [PubMed] [Google Scholar]
- 11.Gadea R, Glibota N, Perez Pulido R, Galvez A, Ortega E. 2017. Adaptation to biocides cetrimide and chlorhexidine in bacteria from organic foods: association with tolerance to other antimicrobials and physical stresses. J Agri Food Chem 65:1758–1770. doi: 10.1021/acs.jafc.6b04650. [DOI] [PubMed] [Google Scholar]
- 12.Gullberg E, Albrecht LM, Karlsson C, Sandegren L, Andersson DI. 2014. Selection of a multidrug resistance plasmid by sublethal levels of antibiotics and heavy metals. mBio 5:e01918-14. doi: 10.1128/mBio.01918-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gullberg E, Cao S, Berg OG, Ilback C, Sandegren L, Hughes D, Andersson DI. 2011. Selection of resistant bacteria at very low antibiotic concentrations. PLOS Pathog 7:e1002158. doi: 10.1371/journal.ppat.1002158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Qiu Z, Yu Y, Chen Z, Jin M, Yang D, Zhao Z, Wang J, Shen Z, Wang X, Qian D, Huang A, Zhang B, Li J-W. 2012. Nanoalumina promotes the horizontal transfer of multiresistance genes mediated by plasmids across genera. Proc Natl Acad Sci U S A 109:4944–4949. doi: 10.1073/pnas.1107254109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ramm L, Siani H, Wesgate R, Maillard J-Y. 2015. Pathogen transfer and high variability in pathogen removal by detergent wipes. Am J Infect Control 43:724–728. doi: 10.1016/j.ajic.2015.03.024. [DOI] [PubMed] [Google Scholar]
- 16.Tuladhar E, Hazeleger WC, Koopmans M, Zwietering MH, Beumer RR, Duizer E. 2012. Residual viral and bacterial contamination of surfaces after cleaning and disinfection. Appl Environ Microbiol 78:7769–7775. doi: 10.1128/AEM.02144-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Anes J, McCusker M, Fanning S, Martins M. 2015. The ins and outs of RND efflux pumps in Escherichia coli. Front Microbiol 10:587. doi: 10.3389/fmicb.2015.00587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Alian S, Qazi U, Sou J. 2013. AcrA and TolC are important efflux components in the development of low level adaptive aminoglycoside resistance in Escherichia coli K-12 following sub-inhibitory kanamycin pre-treatment. J Exp Microbiol Immunol 17:1–7. [Google Scholar]
- 19.Brauner A, Fridman O, Gefen O, Balaban NQ. 2016. Distinguishing between resistance, tolerance and persistence to antibiotic treatment. Nature Rev Microbiol 14:320–330. doi: 10.1038/nrmicro.2016.34. [DOI] [PubMed] [Google Scholar]
- 20.Vogwill T, Comfort AC, Furió V, MacLean RC. 2016. Persistence and resistance as complementary bacterial adaptations to antibiotics. J Evol Biol 29:1223–1233. doi: 10.1111/jeb.12864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Trastoy R, Manso T, Fernandez-Garcia L, Blasco L, Ambroa A, Perez Del Molino ML, Bou G, Garcia-Contreras R, Wood TK, Tomas M. 2018. Mechanisms of bacterial tolerance and persistence in the gastrointestinal and respiratory environments. Clin Microbiol Rev 31:e00023-18. doi: 10.1128/CMR.00023-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Levin-Reisman I, Gefen O, Fridman O, Ronin I, Shwa D, Sheftel H, Balaban NQ. 2019. Automated imaging with ScanLag reveals previously undetectable bacterial growth phenotypes. Nature Methods 7:737. doi: 10.1038/nmeth.1485. [DOI] [PubMed] [Google Scholar]
- 23.Harwani D, Zangoui P, Mahadevan S. 2012. The β-glucoside (bgl) operon of Escherichia coli is involved in the regulation of oppA, encoding an oligopeptide transporter. J Bacteriol 194:90–99. doi: 10.1128/JB.05837-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Madan R, Kolter R, Mahadevan S. 2005. Mutations that activate the silent operon of Escherichia coli confer a growth advantage in stationary phase. J Bacteriol 187:7912. doi: 10.1128/JB.187.23.7912-7917.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bury-Mone S, Nomane Y, Reymond N, Barbet R, Jacqet E, Imbeaud S, Jacq A, Bouloc P. 2009. Global analysis of extracytoplasmic stress signaling in Escherichia coli. PLOS Gen 5:e1000651. doi: 10.1371/journal.pgen.1000651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Liu A, Fong A, Becket E, Yuan J, Tamae C, Medrano L, Maiz M, Wahba C, Lee C, Lee K, Tran KP, Yang H, Hoffman RM, Salih A, Miller JH. 2011. Selective advantage of resistant strains at trace levels of antibiotics: a simple and ultrasensitive color test for detection of antibiotics and genotoxic agents. Antimicrob Agents Chemother 55:1204–1210. doi: 10.1128/AAC.01182-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sandegren L. 2014. Selection of antibiotic resistance at very low antibiotic concentrations. Upsala J Med Sci 119:103–107. doi: 10.3109/03009734.2014.904457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Khan S, Beattie TK, Knapp CW. 2017. The use of minimum selectable concentrations (MSCs) for determining the selection of antimicrobial resistant bacteria. Ecotoxicology 26:283–292. doi: 10.1007/s10646-017-1762-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Gill AE, Amyes SGB. 2004. The contribution of a novel ribosomal S12 mutation to aminoglycoside resistance of Escherichia coli mutants. J Chemother 16:347–349. doi: 10.1179/joc.2004.16.4.347. [DOI] [PubMed] [Google Scholar]
- 30.Spagnolo F, Rinaldi C, Sajorda DR, Dykhuizen DE. 2015b. Evolution of resistance to continuously increasing streptomycin concentrations in populations of Escherichia coli. Antimicrob Agents Chemother 60:1336–1342. doi: 10.1128/AAC.01359-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wistrand-Yuen E, Knopp M, Hjort K, Kosikiniemi S, Berg OG, Andersson DI. 2018. Evolution of high-level resistance during low-level antibiotic exposure. Nature Comm 9:1599. doi: 10.1038/s41467-018-04059-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Maillard J-Y, Denyer SP. 2009. Emerging bacterial resistance following biocide exposure: should we be concerned? Chim Oggi 27:26–28. [Google Scholar]
- 33.Maillard J-Y. 2018. Resistance of bacteria to biocides, p 109–126. In Schwarz S, Cavaco L, Shen J (ed), Antimicrobial resistance in bacteria from livestock and companion animals. ASM Press, Washington, DC. doi: 10.1128/microbiolspec.ARBA-0006-2017. [DOI] [Google Scholar]
- 34.Webber MA, Coldham NG, Woodward MJ, Piddock LJV. 2008. Proteomic analysis of triclosan resistance in Salmonella enterica serovar Typhimurium. J Antimicrob Chemother 62:92–97. doi: 10.1093/jac/dkn138. [DOI] [PubMed] [Google Scholar]
- 35.Miles AA, Misra SS, Irwin JO. 1938. The estimation of the bactericidal power of the blood. J Hyg (Lond) 38:732–749. doi: 10.1017/s002217240001158x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.ISO. 2019. Susceptibility testing of infectious agents and evaluation of performance of antimicrobial susceptibility test devices — Part 1: broth micro-dilution reference method for testing the in vitro activity of antimicrobial agents against rapidly growing aerobic bacteria involved in infectious diseases. ISO 20776-1 https://www.iso.org/standard/70464.html. Accessed 23 July 2020.
- 37.The European Committee on Antimicrobial Susceptibility Testing. 2020. Breakpoint tables for interpretation of MICs and zone diameters, version 10.0, 2020. https://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/Breakpoint_tables/v_10.0_Breakpoint_Tables.pdf. Accessed 23 July 2020.
- 38.Viti C, Decorosi F, Marchi E, Galardini M, Giovannetti L. 2015. High-throughput phenomics. Meth Mol Biol 1231:99–123. doi: 10.1007/978-1-4939-1720-4_7. [DOI] [PubMed] [Google Scholar]
- 39.Mackie AM, Hassan KA, Paulsen IT, Tetu SG. 2014. Biolog phenotype microarrays for phenomic characterization of microbial cells. Meth Mol Biol 1096:123–130. doi: 10.1007/978-1-62703-712-9_10. [DOI] [PubMed] [Google Scholar]
- 40.Lambrecht E, Van Meervenne E, Boon N, Van de Miele T, Wattiau P, Herman L, Hevndrickx M, Van Collie E. 2018. Characterization of cefotaxime and ciprofloxacin resistant commensal Escherichia coli originating from Belgian farm animals indicates high antibiotic resistance transfer rates. Microb Drug Resist 24:707–717. doi: 10.1089/mdr.2017.0226. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.