Mutants of Salmonella Enteritidis selected following exposure to the sanitizer chlorine or to the preservatives sodium nitrite, sodium benzoate or acetic acid show resistance to multiple antibiotics (tetracycline, chloramphenicol, nalidixic acid, and ciprofloxacin). Complementation experiments with a functional marR restored antibiotic susceptibility of selected mutants to levels similar to wild-type strains, suggesting that mar mutation was responsible for resistance. The multiple antibiotic resistance (mar) operon is a global regulator controlling intrinsic resistance towards structurally and functionally unrelated antibiotics and other noxious agents. Mutants selected after exposure to an inducing agent maintained elevated antibiotic resistance after serial subculture in media void of the inducing agent. Results highlight the importance of monitoring the use of antimicrobial agents to ensure that concentrations capable of inactivating target pathogens are used.
Multiple antibiotic resistance
Salmonellosis is an important foodborne disease of public health significance worldwide. Foodborne bacteria resistant to a range of antimicrobial agents are a major health concern. Generally, resistance to antibiotics is highlighted since this has a significant and real impact on treatment of human illness. Fresh fruits and vegetables prior to delivery to retail markets or during processing for packaging are often sanitized in solutions containing up to 150 ppm chlorine. Chlorine is the sanitizer of choice for use on equipment in many food processing facilities. A wide range of preservatives including sodium nitrite, acetic acid, and sodium benzoate are used in foods. Therefore, the development of resistance in foodborne pathogens to antibiotics and other antimicrobial agents (sanitizers and food preservatives), is a concern.
In a recent study, Salmonella cultures (n=502) isolated by FDA laboratories from domestic and imported food products were screened for antibiotic resistance . Approximately 49% (247 of 502 isolates) of isolates were resistant to one or more antibiotics and 25 isolates (10.1%) were resistant to three or more antibiotics. These results are in agreement with research on antibiotic resistance of Salmonella isolates collected in Europe . Antimicrobial resistance of Salmonella is associated with recombination and transfer of resistance genes. Resistance due to intrinsic mechanisms is often overlooked. This not only holds true for antibiotics, but for other antimicrobial agents as well.
The mar operon, located on the chromosome of Enterobacteriaceae, is one intrinsic mechanism of antimicrobial resistance [3–8]. The mar locus is widespread among enteric bacteria, including Escherichia coli, Salmonella, Shigella, Klebsiella, Citrobacter, Hafnia, and Enterobacter species . Antibiotic resistance of mar mutants is in part associated with a decreasing influx and increasing efflux of toxic agents . Altered influx is associated with down-regulating the synthesis of membrane proteins including OmpF . In Salmonella, increased efflux is achieved primarily by increased synthesis of the AcrAB efflux pump . MarA regulates the expression of more than 60 chromosomal genes .
The Salmonella mar locus is structurally and functionally similar to marRAB in E. coli. Initially, researchers reported that within the genus Salmonella, S. Typhimurium was positive for the locus whereas S. Arizonae was negative, raising the question as to whether the operon is conserved in Salmonella. Using a marRAB gene probe from S. Enteritidis rather than S. Typhimurium, a second group of researchers found the operon in all Salmonella tested . Using S. Typhimurium DT104 researchers demonstrated that low-level antibiotic resistance, cyclohexane resistance, and modulation of outer membrane proteins can occur in both a mar-dependent and mar-independent manner .
Previous studies have demonstrated resistance and cross-resistance of mar mutants to a range of antimicrobial compounds (reviewed by ). These studies have not addressed how induction of the mar phenotype confers antimicrobial resistance to the foodborne pathogen S. Enteritidis. Specifically, research has not assessed how exposure to chlorine or food preservatives commonly used in the food industry may provide cross-resistance to antibiotics used in human medicine or animal therapy. The aim of this work was to investigate the role of mar in conferring resistance to antibiotics following exposure to chlorine or preservatives in the foodborne pathogen S. Enteritidis.
2 Materials and methods
2.1 Bacterial isolates and PCR of marR
Five different S. Enteritidis isolates obtained from the New Jersey Department of Health and designated RU201, RU202, RU204, RU205, and RU210 were used in the study. All isolates were grown and maintained at 30°C in LB broth or on LB agar plates. S. Enteritidis isolates were examined using PCR to confirm the presence of the mar operon. Multiple sequence alignment of the known marR sequences from Salmonella enterica serovars Typhimurium, Enteritidis, and Dublin were conducted using the DIALIGN 2.1 method  to design primers. Sequence of the forward primer is 5′-CCA GTG ATC TGT TCA ATG AAA-′3 and reverse primer is 5′-CGT CTG GAC ATC GTC ATA C-′3. A 1-µl aliquot of S. Enteritidis whole-cell suspension was used as template for PCR. The 50-µl PCR reaction mixture consisted of 1 µM of each primer, 2 mM each of deoxynucleotide triphosphate (New England Biolabs, Inc.), 5 µl of 10× PCR buffer (Fisher Scientific), 5 U of Taq polymerase (Fisher Scientific), and the template. PCR was carried out in a Perkin-Elmer GeneAmp PCR system 2400 (Perkin-Elmer Corp.). All reactions were subjected to a hot start for 5 min at 95°C. Thirty-five cycles, each consisting of 30 s at 94°C, 45 s at 55°C, 45 s at 72°C, with an extension at 72°C for 7 min were carried out. PCR product (450 bp) was separated on 1% agarose gel by electrophoresis, stained with ethidium bromide, and visualized by UV transillumination.
2.2 Induction of mar phenotype through exposure to chlorine, acetic acid, sodium benzoate, sodium nitrite, or tetracycline
Salmonella isolates were exposed to chlorine to determine whether exposure to an omnipresent sanitizer that is commonly used in both industry and households would induce mar behavior. Bacteria were cultured overnight in LB broth at 37°C, the culture was centrifuged, supernatant decanted, and the cells resuspended in PBS (pH 6.0) to achieve a final concentration of 109 cells ml−1. Using a microtiter plate format, cells (107) were suspended in a 25-ppm chlorine solution for 10 min, after which an equal volume (100 µl) of quarter strength LB broth was added to neutralize the chlorine, and the plate incubated at 30°C for 2 h. An aliquot of the cell suspension was used to inoculate LB broth that was incubated at 30°C for 18 h for use in the microbroth dilution assay.
Isolate RU205 was used to determine whether common food preservatives would induce the mar phenotype or select for mar mutants. LB agar was supplemented with acetic acid (0.05% w/v), sodium benzoate (1.0% w/v), or sodium nitrite (1.0% w/v). Isolate RU205 was cultured overnight in LB broth. A 100-µl volume of culture was spread on the surface of each plate and incubated at 30°C for 24 h. Colonies were randomly selected and classified as preservative-induced mutants.
Tetracycline was added to LB agar tempered to 50°C to achieve a final concentration of 8 µg ml−1 and poured into plates. Salmonella isolates were grown in 5 ml of LB broth at 30°C overnight and 100 µl were spread onto the tetracycline plates. The plates were incubated at 30°C for 36 h. Colonies were picked, cultured in LB broth overnight and spread onto plates containing 12 µg ml−1 tetracycline. The plates were incubated at 30°C for 36 h and colonies randomly picked. The process was repeated using plates containing 20 µg ml−1 tetracycline. The colonies selected from these plates were designated antibiotic-induced mutants.
Following selection on an antimicrobial agent isolate nomenclature was changed to reflect the selective agent. For example, following exposure of wild-type isolate RU205 to chlorine the colony selected was designed RU205Ch. Isolates were designated ‘Ch’, chlorine; ‘T’, tetracycline; ‘SB’, sodium benzoate' ‘AA’, acetic acid; and ‘SN’, sodium nitrite.
2.3 Determination of the minimum inhibitory concentration (MIC)
Each mutant selected from the preservatives, the sanitizer, or antibiotic was tested for resistance to the antibiotics tetracycline, ciprofloxacin, chloramphenicol, and nalidixic acid. The microbroth dilution assay was performed according to procedures outlined by the National Committee for Clinical Laboratory Standards . Clinical resistance to nalidixic acid, ciprofloxacin, chloramphenicol, and tetracycline is ≥32 µg ml−1, ≥4 µg ml−1, ≥32 µg ml−1, and ≥16 µg ml−1.
2.4 Complementation of strains
Mutants selected on chlorine, preservatives (acetic acid, sodium benzoate, or sodium nitrite), or tetracycline were transformed with pGEM-T+marR (contains functional marR) by electroporation. The plasmid also encodes for ampicillin resistance. Transformation of the mutant isolates was accomplished with the Bio-Rad Gene Pulser electroporator (Bio-Rad Laboratories) using 2.4 kV, a capacitance of 25 µF, and resistance of 400 Ω. The transformants were plated on LB agar supplemented with 100 µg ml−1 ampicillin and incubated at 37°C for 18–24 h. MIC of transformants was compared to wild-type and mutant isolates using the microbroth dilution assay.
2.5 Subculturing of mutant isolates
Mutant isolates selected following exposure to tetracycline, chlorine, or sodium nitrite were subcultured in LB broth free of the inducing agent at 37°C for 10 consecutive days. Following serial culture, the MIC of cells collected after the first and last subculture to tetracycline was determined using the microbroth dilution assay and compared to the wild-type strains.
3.1 Cross-resistance of chlorine-selected mutants to antibiotics
Survivors of Salmonella isolates selected following exposure to 25 ppm chlorine were tested for cross-resistance to tetracycline, nalidixic acid, and chloramphenicol (Fig. 1). All chlorine-selected mutants exhibited an increase in resistance to each antibiotic when compared to wild-type isolates. The greatest increase in resistance was to tetracycline and chloramphenicol. Isolate RU204Ch exhibited an eight-fold increase in resistance to chloramphenicol, whereas all other isolates exhibited a four-fold increase in resistance.
The marRAB operon of S. Enteritidis was induced by exposure to 25 ppm chlorine solution for 10 min and MIC to antibiotics determined. MICs to tetracycline (A), chloramphenicol (B), and nalidixic acid (C) are shown. The designation ‘Ch’ (e.g., RU201Ch) indicates mutant was selected following exposure to chlorine. Closed bar, wild-type and hatched bar, mutant.
3.2 Cross-resistance of preservative-selected mutants to tetracycline
Isolate RU205 was independently exposed to acetic acid, sodium benzoate or sodium nitrite and a single colony from each treatment was selected and designated RU205AA, RU205SB, and RU205SN, respectively. Each mutant was more resistant to tetracycline compared to the wild-type (Fig. 2). Mutants selected on sodium benzoate and sodium nitrite exhibited greater resistance to tetracycline than the mutant selected on acetic acid.
S. Enteritidis mutants were selected following a single exposure to acetic acid, sodium benzoate, or sodium nitrite and MIC to tetracycline determined. Wild-type, closed bar, acetic acid-induced, diagonal hatched bar, sodium benzoate, vertical lined bar, sodium nitrite, open bar.
3.3 Cross-resistance of tetracycline-selected mutants to other antibiotics
Each of the tetracycline-selected mutants exhibited cross-resistance to chloramphenicol, nalidixic acid, and ciprofloxacin (Fig. 3). The wild-type isolates did not grow at the concentrations of ciprofloxacin tested. All isolates selected following exposure to tetracycline maintained clinical resistance to tetracycline (≥16 µg ml−1) (data not shown).
The MIC of tetracycline-selected S. Enteritidis mutants to three antibiotics. Antibiotics screened for include chloramphenicol (A), nalidixic acid (B), and ciprofloxacin (C). The designation ‘T’ (e.g., RU201T) indicates mutant was selected following exposure to tetracycline. The wild-type did not grow on the concentrations of ciprofloxacin tested. Closed bar, wild-type and hatched bar, mutant.
3.4 Complementation with functional marR
Results of complementation of two tetracycline-selected mutants, RU201T and RU205T, with pGEM-T+marR are shown in Fig. 4. Level of susceptibility to tetracycline of the two isolates was restored to that of the wild-type, RU201 and RU205. These results specifically indicate the mutation(s) are in the mar locus.
Complementation of tetracycline-selected mutants. S. Enteritidis isolates RU201T and RU205T were transformed with pGEM-T+marR and MIC to tetracycline determine. Susceptibility to tetracycline was restored to levels similar to each wild-type isolate (RU201 and RU205). Closed bar, wild-type (RU201 and RU205), hatched bar, mutant (RU201T and RU205T), open bar, transformant (RU201TpGEM-T+marR, RU205TpGEM-T+marR).
In the United States salmonellosis is estimated to affect 1.4 million people each year, of which 95% of the cases are estimated to be foodborne . In addition, it is estimated that salmonellosis is responsible for approximately 31% of deaths associated with foodborne illness in the United States. In the farm to fork continuum bacteria are exposed to a range of conditions and chemical compounds (antimicrobial agents) that may ultimately influence survival of the pathogen. For example, exposure to chlorine in a food-processing plant may result in cross-resistance to antibiotics used in human therapy or in agriculture. This scenario led to the basis of the current study to determine whether exposure to a common sanitizing agent or widely used food preservatives would confer antibiotic resistance to foodborne pathogens through induction of the mar phenotype.
Induction of the marRAB operon in the six S. Enteritidis isolates used in the present study was accomplished by exposure to 25 ppm chlorine. The concentration of chlorine used is in the range that may be employed for sanitizing of equipment or produce . When tested for cross-resistance to the antibiotics chloramphenicol, tetracycline, nalidixic acid, and ciprofloxacin increase in resistance levels to each of the antibiotic was observed for each isolate. In general, isolates exhibited a similar level of increased resistance to each antibiotic. These results are noteworthy, coupled with additional mechanisms of antimicrobial resistance unrelated to mar that may also be induced as the pathogen journeys along the farm to fork continuum, ultimately impacting treatment regimes used for humans or animals infected with Salmonella.
The possibility that commonly used food preservatives could induce mar was tested using S. Enteritidis isolate RU205. Many preservatives used in food for human consumption, are also used in animal feed. Therefore, it would not be unexpected that Salmonella associated with a case of foodborne illness had been previously exposed to a food preservative. In this study, we were not concerned that exposure to a food preservative may confer cross-resistance to other preservatives, but rather cross-resistance to antibiotics. Results show that a single exposure to acetic acid, sodium benzoate, or sodium nitrite can result in cross-resistance to tetracycline (Fig. 2). Although clinical resistance was not achieved, as was the case when RU205 was exposed to tetracycline, level of resistance was increased by up to four-fold. Hence, these results underscore the prudent use of antimicrobial compounds.
In previous studies induction of the marRAB operon of Salmonella was accomplished following passage with tetracycline or chloramphenicol [7,13]. Our results confirm and support results of previous studies; where transfer of each Salmonella isolate on tetracycline gave rise to increased resistance to tetracycline, chloramphenicol, nalidixic acid, and ciprofloxacin. Clinical level of resistance to tetracycline (≥16 µg ml−1) was achieved; resistance to other antibiotics generally increased several fold, but remained well below the level of clinical resistance . None of the wild-type isolates exhibited innately high resistance to the antibiotics tested.
PCR analysis demonstrated that all isolates were positive for marR based on visualization of the expected 450 bp product (Fig. 5). Complementation of mutant isolates selected on chlorine, preservatives, or tetracycline with a functional marR completely restored susceptibility to antibiotics strongly suggesting that mar mutation was responsible for increased resistance. All isolates selected following exposure to a given antimicrobial agent; tetracycline, chlorine or preservative were subsequently serial cultured 10 times in media free of the inducing agent. Based on MIC following the last subculture, each isolate maintained an elevated level of tetracycline resistance (data not shown).
Amplification of marR from S. Enteritidis isolates. The expected 450-bp product was amplified from all isolates indicating the presence of marR. Lanes: 1, molecular mass marker (1000, 750, 500, 300, 150, 50 bp); 2, RU201; 3, RU202; 4, RU204; 5, RU205; 6, RU210, and 7, negative control.
Sanitizers and preservatives are commonly used to enhance the safety of foods, such as liquid eggs, poultry, and produce. The levels of these compounds used are generally sufficient to inactivate target pathogens of concern. However, situations can and do exist where a pathogen may not be exposed to a lethal dose of a given antimicrobial agent which may result in the selection of resistant mutants. Previous studies have shown that exposure to one antimicrobial agent can confer cross-resistance to other structurally and functionally different antimicrobial agents [7,18]. We demonstrate that single exposure of the foodborne pathogen S. Enteritidis to chlorine, sodium nitrite, acetic acid, or sodium benzoate can induce the marRAB operon and confer cross-resistance to antibiotics such as chloramphenicol, tetracycline, ciprofloxacin, and nalidixic acid. These results emphasize the importance of monitoring the use of antimicrobial agents to ensure that appropriate concentrations are used to inactivate target pathogens.
This work was supported by funding from the New Jersey Agricultural Experiment Station and C.J.P. was supported by a USDA Undergraduate Research fellowship.