Efflux-mediated response of Staphylococcus aureus exposedto ethidium bromide

Isabel Couto1,2, Sofia Santos Costa1, Miguel Viveiros1, Marta Martins1,3 and Leonard Amaral1,3*

1Unidade de Micobacterias, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa (UNL),

Portugal; 2Centro de Recursos Microbiologicos (CREM), Faculdade de Ciencias e Tecnologia, UNL, Portugal;3UPMM, Instituto de Higiene e Medicina Tropical, UNL, Portugal

Received 21 January 2008; returned 27 February 2008; revised 17 April 2008; accepted 21 April 2008

Objectives: By adapting an antibiotic-susceptible Staphylococcus aureus strain to increasingconcentrations of ethidium bromide, a known substrate of efflux pumps (EPs), and by phenotypicallyand genotypically analysing the resulting progeny, we characterized the molecular mechanisms ofS. aureus adaptation to ethidium bromide.

Methods: S. aureus ATCC 25923 was grown in increasing concentrations of ethidium bromide. TheMICs of representatives of eight classes of antibiotics, eight biocides and two dyes against ATCC25923 and its ethidium bromide-resistant progeny ATCC 25923EtBr were determined with or without sixefflux pump inhibitors (EPIs). Efflux activity in the presence/absence of EPIs was evaluated by real-time fluorometry. The presence and expression of eight EP genes were assayed by PCR and quantitativeRT–PCR (qRT–PCR), respectively. Mutations in grlA, gyrA and norA promoter regions were screenedby DNA sequencing.

Results: Compared with its parental strain, ATCC 25923EtBr was 32-fold more resistant to ethidiumbromide and also more resistant to biocides and hydrophilic fluoroquinolones. Resistance to thesecould be reduced by the EPIs chlorpromazine, thioridazine and reserpine. Increased efflux of ethidiumbromide by ATCC 25923EtBr could be inhibited by the same EPIs. qRT–PCR showed that norA was35-fold over-expressed in ATCC 25923EtBr, whereas the remaining EP genes showed no significantincrease in their expression. Sequencing of the norA promoter region revealed a 70 bp deletion inATCC 25923EtBr.

Conclusions: Exposure of S. aureus to quaternary compounds such as ethidium bromide results indecreased susceptibility of the organism to a wide variety of compounds, including quinolones andbiocides through an efflux-mediated response, which for strain ATCC 25923 is mainly NorA-mediated.This altered expression may result from alterations in the norA promoter region.

Keywords: staphylococci, efflux pumps, NorA, overexpression

Introduction

Staphylococcus aureus is one of the most common humanpathogens, being responsible for a wide variety of infections,many of which can be life-threatening. Besides its pathogenicpotential, this bacterium also shows many different mechanismsof resistance towards antibiotics. Several of these mechanismsare well known, and have been characterized: resistance tob-lactam antibiotics mediated by PBP2a, encoded by the mecAgene, or resistance to fluoroquinolones resulting from mutations

in either topoisomerase IV or gyrase genes.1 On the other hand,antibiotic resistance based on efflux systems capable of extrud-ing the drug or other noxious agents from the cell is less wellcharacterized for these bacteria.

To date, more than 10 efflux pumps (EPs) have beendescribed for S. aureus.2 Most of these pumps belong to themajor facilitator superfamily, namely the chromosomallyencoded NorA, NorB, NorC, MdeA and SdrM as well as theplasmid-encoded QacA/B pumps.3 – 8 Other types of pumps havealso been described for S. aureus such as MepA, a member of

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

*Correspondence address. Unidade de Micobacterias, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa,Rua da Junqueira 96, 1349-008 Lisbon, Portugal. Tel: þ351-21-3652652; Fax: þ351-21-3632105; E-mail: lamaral@ihmt.unl.pt

Journal of Antimicrobial Chemotherapy (2008) 62, 504–513

doi:10.1093/jac/dkn217

Advance Access publication 28 May 2008

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

504

# The Author 2008. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.

For Permissions, please e-mail: journals.permissions@oxfordjournals.org

the multidrug and toxic compound extrusion family, as well asSmr, which belongs to the small multidrug resistance (SMR)family, and SepA.9 – 11 Although these pumps show differentsubstrate specificity, most of them are capable of extrudingcompounds of different chemical classes, thus providing the cellwith the means to develop a multidrug resistance (MDR) pheno-type or to survive in a hostile environment.

In order to find specific inhibitors of these pumps or simplyto demonstrate the presence of efflux activity, different com-pounds have been tested for their ability to hinder the activity ofS. aureus EPs, including the proton motive force uncoupler car-bonyl cyanide m-chlorophenylhydrazone (CCCP), the alkaloidreserpine, as well as different members of the phenothiazines,among others.12 – 16

In this work, a fully antibiotic-susceptible S. aureus strainwas challenged with increasing concentrations of a well-knownsubstrate of EPs, ethidium bromide, and the mechanisms bywhich the cells adapted to this compound were evaluated bydifferent methods. This experimental approach allowed us tostudy for the first time the response of an S. aureus strain that isforced to cope with high concentrations of a noxious pump sub-strate, in particular, to evaluate the efflux system(s)preferentially used by the cells under such conditions.

Materials and methods

Bacterial strains

The study was conducted with S. aureus strain ATCC 25923, a clini-cal isolate collected at Seattle in 1945, fully antibiotic-susceptible,b-lactamase-negative, with an ethidium bromide MIC of 6.25 mg/L.This strain was initially grown in 5 mg/L of ethidium bromide andincubated at 37ºC. When visible growth was obtained, the culture

was transferred to new media, containing increasing concentrationsof ethidium bromide (5, 10, 20, 40, 60, 80 and 100 mg/L). Duringan 82 day period, 7 cultures were obtained, with MICs of ethidiumbromide varying between 6.25 and 200 mg/L. The final, most resist-

ant culture was named ATCC 25923EtBr. This method of adaptationof a bacterial species to a given agent has been describedpreviously.17 – 19

Growth conditions

Cultures were grown in tryptic soy broth (TSB) or agar (TSA)(Oxoid Ltd, Basingstoke, UK) at 378C. When necessary, the strainATCC 25923EtBr was grown in a medium supplemented with50 mg/L of ethidium bromide. For determination of the MICs of the

agents employed in this study, cultures were grown in a Mueller–Hinton broth (MHB), and for Kirby–Bauer antibiotic susceptibilityassays, in Mueller–Hinton agar, at 378C.

Antibiotics

Antibiotics in powder form were purchased from different sources,as follows: nalidixic acid, erythromycin, tetracycline, rifampicin andchloramphenicol were purchased from Sigma-Aldrich (St Louis,USA); norfloxacin was purchased from ICN Biomedicals Inc. (OH,

USA) and ciprofloxacin from Fluka Chemie GmbH (Buchs,Switzerland). Kirby–Bauer discs containing nalidixic acid (30 mg),norfloxacin (10 mg), ciprofloxacin (5 mg), tetracycline (30 mg),oxacillin (1 mg), gentamicin (10 mg), cefotaxime (30 mg), penicillin(10 U) and fusidic acid (10 mg) were acquired from Oxoid Ltd.

Biocides and dyes

Ethidium bromide, benzalkonium chloride, berberine, dequalinium

chloride, 40,6-diamidino-2-phenylindole dihydrochloride, penta-midine isothionate salt and tetraphenylphosphonium bromide werepurchased from Sigma-Aldrich. The biocides chlorhexidine andcetyltrimethylammonium bromide were purchased from Fluka

Chemie GmbH and rhodamine from Merck (Darmstadt, Germany).All solutions were prepared on the day of the experiment and keptprotected from light.

Efflux pump inhibitors (EPIs)

CCCP, thioridazine, chlorpromazine, verapamil and reserpine werepurchased from Sigma-Aldrich. Ouabain was purchased from FlukaChemie GmbH. Solutions of thioridazine, chlorpromazine, verapa-mil and ouabain were prepared in deionized water; reserpine wasprepared in DMSO and CCCP in 50% methanol (v/v). All solutions

were prepared on the day of the experiment and kept in darkness.

Drug susceptibility testing

Antibiotics. MICs were determined by the broth microdilutionmethod, as per the guidelines of CLSI20 and carried out in triplicate.Results were evaluated according to the CLSI breakpoints, except

for nalidixic acid, for which there are no defined breakpoints.Kirby–Bauer antibiotic susceptibility assays were conducted as out-lined by the CLSI standards and performed in triplicate. The zonesof inhibition were measured in millimetres and analysed by the

CLSI breakpoints, except for fusidic acid, which was analysed bythe BSAC standards21 and nalidixic acid, for which there are nodefined breakpoints.

Biocides and dyes. MICs were determined by the broth microdilu-

tion method. Briefly, cultures were grown overnight in TSB, at378C, diluted in PBS and the suspension turbidity adjusted to a0.5 McFarland standard. Twenty microlitres of this suspension werethen transferred to each well of 96-well plates containing 2-folddilutions of each agent to be tested, diluted in MHB. The plates

were then incubated at 37ºC for 18 h and the MIC values, cor-responding to the lowest concentration of compound that inhibitedvisible growth, recorded. Each MIC determination was carried outin triplicate.

Efflux pump inhibitors. Each EPI employed in this study was evalu-ated for its ability to reduce or reverse antibiotic resistance to givenantibiotics and biocides, both of which are characteristics that definethe agent as an inhibitor of efflux pump activity.22 The evaluation

of an agent for EPI activity was conducted at a concentration of theagent that did not inhibit the growth of the organism in a mediumcontaining varying concentrations of an antibiotic or biocide and abacterial inoculum corresponding to the one used for MIC determi-nation. Parallel cultures were tested in media containing no EPI and

varying concentrations of the antibiotic or biocide. The cultureswere incubated for 18 h and their contents (growth) evaluated visu-ally; minimum bactericidal concentrations were determined bycounting colony-forming units. The final concentrations of the EPIsused, which correspond to half of the MIC determined for each EPI

unless otherwise stated, were: thioridazine (12.5 mg/L), chlorproma-zine (25 mg/L), verapamil (200 mg/L), reserpine (20 mg/L), ouabain(100 mg/L) and CCCP (0.18 mg/L). The MIC of each EPI wasdetermined by the broth microdilution method, as described above.All assays were performed in triplicate.

Ethidium bromide-induced efflux in S. aureus

505

Preparation of chromosomal DNA

Genomic DNA was extracted with the QIAamp DNA minikit

(QIAGEN, Hilden, Germany), with an additional step of 30 mindigestion with lysostaphin (Sigma) (200 mg/L) prior to extraction.

Macrorestriction analysis

Cultures were typed by pulsed-field gel electrophoresis (PFGE)analysis, using well-established protocols. Briefly, agarose discs

containing intact chromosomal DNA were prepared as describedpreviously23 and restricted with SmaI, according to the manufac-turer’s recommendations (New England Biolabs, Ipswich, MA,USA). Restriction fragments were then resolved by PFGE, whichwas carried out in a contour-clamped homogeneous electric field

(CHEF) apparatus (CHEF-DRII, Bio-Rad, Hercules, CA, USA), asdescribed previously.23

Preparation of plasmid DNA

The QIAprep Spin Miniprep kit (QIAGEN) was used, with the fol-lowing modification: prior to extraction, cells were incubated with

lysostaphin (35 mg/L) at 378C for 90 min, as described previously.24

RNA extraction

Total RNA was extracted from the parent ATCC 25923 strain

unexposed to ethidium bromide and the ATCC 25923EtBr straingrown in TSB supplemented with 50 mg/L of ethidium bromidewith the Rneasy Mini Kit (QIAGEN), as per the manufacturer’sinstructions. Before extraction of total RNA, the cultures were

treated with an RNAprotect bacterial reagent (QIAGEN). Theextracted RNA was treated with RNAse-free DNAse (QIAGEN) for2 h at room temperature and re-purified using the same kit.

PCR amplification of efflux pump genes

Separate fragments of DNA internal to each of eight EP genes pre-viously described for S. aureus were amplified by PCR, using theprimers described in Table 1. The reaction mixture (50 mL) con-tained 2.5 U of Taq Polymerase (Fermentas Inc., Ontario, Canada),1� Taq buffer (Fermentas), 25 pmol of each primer, 200 mM dNTP

and 1.75 mM MgCl2. The PCRs were conducted in a thermocyclerMastercycler personal 5332 (Eppendorf AG, Hamburg, Germany).Amplification conditions were as follows: DNA was denatured at948C for 4 min, followed by 35 cycles of denaturation at 948C for

30 s, annealing at 458C (norA) or 538C (norB, norC, mdeA, mepA

Table 1. Primers used in this study

Primer Sequence (50 –30) Size (bp) Reference

QacA/B_Fw GCTGCATTTATGACAATGTTTG628 45

QacA/B_Rv AATCCCACCTACTAAAGCAG

SMR_Fw ATAAGTACTGAAGTTATTGGAAGT285 24

SMR_Rv TTCCGAAAATGTTTAACGAAACTA

NorA_Fw TTCACCAAGCCATCAAAAAG620 46; this work

NorA_Rv CTTGCCTTTCTCCAGCAATA

NorA_RT(Fw) TCGTCTTAGCGTTCGGTTTA246 this work

NorA_RT(Rv) TCCAGTAACCATCGGCAATA

NorAp_Fw TGTTAAGTCTTGGTCATCTGCA449 this work

NorAp_Rv AGCAGCAACAAGTAACCCTAAA

NorB_Fw AGCGCGTTGTCTATCTTTCC213 this work

NorB_Rv GCAGGTGGTCTTGCTGATAA

NorC_Fw AATGGGTTCTAAGCGACCAA216 this work

NorC_Rv ATACCTGAAGCAACGCCAAC

MepA_Fw ATGTTGCTGCTGCTCTGTTC718 this work

MepA_Rv TCAACTGTCAAACGATCACG

MepA_RT(Fw) TGCTGCTGCTCTGTTCTTTA198 this work

MepA_RT(Rv) GCGAAGTTTCCATAATGTGC

MdeA_Fw AACGCGATACCAACCATTC677 this work

MdeA_Rv TTAGCACCAGCTATTGGACCT

MdeA_RT(Fw) GTTTATGCGATTCGAATGGTTGGT155 6

MdeA_RT(Rv) AATTAATGCAGCTGTTCCGATAGA

SepA_Fw GCAGTCGAGCATTTAATGGA103 this work

SepA_Rv ACGTTGTTGCAACTGTGTAAGA

16S_27Fw AGAGTTTGATCMTGGCTCAG492 47

16S_519Rv GWATTACCGCGGCKGCTG

GrlA_Fw TGCCAGATGTTCGTGATGGT339 48

GrlA_Rv TGGAATGAAAGAAACTGTCTC

GyrA_Fw TCGTGCATTGCCAGATGTTCG394 this work

GyrA_Rv TCGAGCAGGTAAGACTGACGG

Primers used in the qRT–PCR experiments are indicated by the RT label. Fw, forward primer; Rv, reverse primer. For norB, norC and sepA, the same set ofprimers was used for both PCR and qRT–PCR.

Couto et al.

506

and sepA) for 30 s and extension at 728C for 1 min, followed by afinal extension at 728C for 5 min. The PCRs for genes qacA/B andsmr were conducted under the following conditions: DNA wasdenatured at 958C for 1 min, followed by 30 cycles of denaturation

at 958C for 1 min, annealing at 408C (qacA/B) or 488C (smr) andextension at 728C for 1 min, followed by a step of final extension at728C for 5 min. The amplification products were visualized in 1%agarose gel electrophoresis.

Screening of mutations in the norA promoter region

A 449 bp region upstream of norA was amplified using primers

NorAp_Fw and NorAp_Rv (Table 1). The amplification conditionswere the same as used for norB amplification. Amplification pro-ducts were purified and sequenced in both strands. Sequences wereanalysed and aligned using programs BioEdit, version 7.0.8.0, andClustalW, respectively.

Screening of mutations in grlA and gyrA genes

Fragments internal to the grlA and gyrA genes were amplified usingthe primers GrlA_Fw/GrlA_Rv and GyrA_Fw/GyrA_Rv, describedin Table 1. The amplification conditions were as follows: DNA wasdenatured at 948C for 4 min, followed by 35 cycles of denaturationat 948C for 30 s, annealing at 508C for 30 s and extension at 728Cfor 1 min, followed by a final extension at 728C for 5 min.Amplification products were purified and sequenced in both strands.Sequences were analysed and aligned using the programs BioEditand ClustalW, respectively.

qRT–PCR protocol

Quantitative RT–PCR (qRT–PCR) was performed in triplicate using

a QuantiTect SYBR Green RT–PCR Kit (QIAGEN). The primersused in these assays are described in Table 1. The relative quantity ofmRNA from the eight efflux pumps was determined by the com-parative threshold cycle (CT) method25 in a Rotor-Gene 3000TM

thermocycler with real-time analysis software (Corbett Research,

Sydney, Australia). The method was applied to the analysis of threeassays conducted with three independent total RNA extractions.Negative controls and a genomic DNA contamination control wereincluded. 16S rRNA was used as an endogenous control.

Assessment of efflux activity

Ethidium bromide agar screening method. Efflux activity was first

assayed by the ethidium bromide agar screening method previouslydescribed.26 Briefly, each culture was swabbed onto TSA plates con-taining ethidium bromide concentrations ranging from 0.5 to2.5 mg/L. The plates were incubated at 378C for 16 h after whichthe minimum concentration of ethidium bromide that produced

fluorescence under UV light associated with the bacterial mass wasrecorded. The plates were then incubated at 48C for another 16 h,after which the minimum ethidium bromide concentration that pro-duced fluorescence was recorded and compared with the minimalconcentration of ethidium bromide that produced fluorescence at

378C.

Semi-automated fluorometric method. A second method, recentlydeveloped in our laboratory, was also used for the assessment ofefflux activity.27 This method is based on the accumulation of agiven EP substrate (in this case ethidium bromide) inside the cells,followed by the fluorometric measurement of its efflux

in a Rotor-Gene 3000TM thermocycler using real-time analysis soft-ware (Corbett Research).

For the accumulation of ethidium bromide, cultures were grownin TSB medium at 378C and 150 rpm to an OD600 of 0.6, centri-

fuged and washed twice in PBS. The OD600 of the cellular suspen-sion was then adjusted to 0.3 in PBS. The conditions used tomaximize accumulation were: incubation with 3 mg/L ethidiumbromide and 25 mg/L chlorpromazine, the most active EPI, at 258C,and in the absence of glucose, for a 60 min period. The ethidium

bromide and chlorpromazine concentrations were used at half MICfor the cultures in order to ensure that no effect was produced onthe viability of the cells.

For the efflux assay, the ethidium bromide-loaded cells were col-lected by centrifugation and re-suspended in PBS to an OD600 of

0.3. Glucose to a final concentration of 0.4% was added to a set ofmicrotubes containing an aliquot of 1 mL of cellular suspension.One microtube was left without glucose, to which 25 mg/L chlorpro-mazine was added. Afterwards, aliquots of 95 mL were transferred

to 0.2 mL microtubes. Aliquots of 5 mL from aqueous stock sol-utions of the EPIs chlorpromazine, thioridazine and reserpine wereadded to the PCR microtubes in order to achieve final concen-trations of: 5, 10, 15 and 25 mg/L; 5, 10 and 12.5 mg/L; and 20, 40,60 and 100 mg/L, respectively. Replicate tubes without glucose and/

or without EPI were also used as controls. The efflux assays werethen conducted in the Rotor-Gene 3000TM at a temperature of 37ºC,and the fluorescence of ethidium bromide was measured at the endof every cycle of 60 s, for a total period of 25 min. Each assay wasperformed at least in triplicate and the viability of the cells was

checked at the end of the assay by plating dilutions of the culture inTSA.

The raw data obtained was then normalized against data fromnon-effluxing cells (cells from the control tube with 25 mg/L chlor-promazine and no glucose) at each point, which were considered to

be the maximum fluorescence values that could be obtained duringthe assay. The relative fluorescence thus corresponds to the ratio offluorescence that remains per unit of time relative to the ethidium-bromide-loaded cells.

Results

Adaptation to ethidium bromide

The parent ATCC 25923 strain, susceptible to 6.25 mg/L ethi-dium bromide, was serially cultured in stepwise increases ofethidium bromide beginning with 5 mg/L ethidium bromide.After 82 days, the resistance of this strain to ethidium bromidewas raised from an MIC of 6.25 to 200 mg/L and this ethidiumbromide-adapted culture was designated as ATCC 25923EtBr.

Contamination control

The parent ATCC 25923 and its adapted ethidium-bromide-resistant progeny, ATCC 25923EtBr, were typed byPFGE analysis. Since the SmaI macrorestriction patterns of boththe parental and the ethidium bromide-adapted cultures wereidentical (data not shown), one can rule out any contaminantbeing introduced during the process by which the parent strainbecame resistant to ethidium bromide.

Susceptibility to biocides and dyes

The MIC values of several biocides and dyes against the par-ental strain and its ethidium bromide-induced-resistant progeny,

Ethidium bromide-induced efflux in S. aureus

507

ATCC 25923EtBr, are summarized in Table 2. Briefly, the MICsagainst the ethidium bromide-adapted strain have all increased,especially those of the biocides TPP and dequalinium, for whichan 8-fold increase is evident.

Susceptibility to antibiotics

Kirby–Bauer assays. The antibiotic susceptibility of the twocultures against different classes of antibiotics was firstscreened by Kirby–Bauer assays. Whereas the initial parentstrain was susceptible to all antibiotics, the adapted strainshowed evidence of decreased susceptibility to norfloxacin,ciprofloxacin and nalidixic acid (with zones of inhibition of27, 28 and 14 mm for ATCC 25923 and 19, 21 and 11 mmfor ATCC 25923EtBr, respectively). However, the diameters ofthe zones of inhibition for ATCC 25923EtBr still fall withinthe susceptibility range.

MIC determination. The MIC of each antibiotic against the par-ental and ethidium bromide-adapted strains is shown in Table 2.According to the Kirby–Bauer antibiotic disc assay, the

ethidium bromide-adapted strain was 4-fold more resistant thanits parental strain to norfloxacin and ciprofloxacin. No alterationwas observed in the MIC of nalidixic acid or any of the otherantibiotics tested.

For ciprofloxacin, the MIC obtained for ATCC 25923EtBr

(1 mg/L) equals the breakpoint for distinguishing susceptibilityfrom intermediate resistance, according to the CLSI breakpointvalues. However, measured against EUCAST breakpoint values,the MIC of ciprofloxacin for this strain would correspond toresistance, thus further illustrating the development of an MDRphenotype in this strain.28

Effect of EPIs on MICs of antibiotics, biocides and dyes

The two strains were then tested for the effect of different EPIsat half or less of their MIC on the MIC of several efflux pumpsubstrates. MICs of the different inhibitors were determined inliquid media, and did not differ between the two cultures. MICswere as follows: CCCP, 0.375 mg/L; thioridazine, 25 mg/L;chlorpromazine, 50 mg/L; verapamil, .200 mg/L; reserpine,.100 mg/L; and ouabain, .200 mg/L.

Table 2. MICs of biocides and antibiotics in the presence and absence of EPIs

Drug Culture

MIC (mg/L) in the presence of inhibitor

no inhibitor TZ CPZ VER CCCP RES

EtBrATCC 25923 6.25 0.75 0.75 3 1.5 1.5

ATCC 25923EtBr 200 (32�) 25 12.5 100 200 100

TPPATCC 25923 12.5 1.5 3 6.25 12.5 3

ATCC 25923EtBr 100 (8�) 6.25 6.25 50 100 50

DCATCC 25923 6.25 1.5 1.5 3 3 0.75

ATCC 25923EtBr 50 (8�) 12.5 12.5 25 25 12.5

DAPIATCC 25923 3 0.75 1.5 1.5 1.5 0.75

ATCC 25923EtBr 12.5 (4�) 1.5 1.5 6.25 12.5 3

PTATCC 25923 50 12.5 6.25 12.5 50 12.5

ATCC 25923EtBr 200 (4�) 50 50 100 200 100

BACATCC 25923 0.75 0.18 0.09 0.18 0.75 0.09

ATCC 25923EtBr 3 (4�) 0.375 0.75 1.5 1.5 0.75

CHXATCC 25923 0.375 0.02 0.09 0.09 0.375 0.02

ATCC 25923EtBr 0.75 (2�) 0.375 0.375 0.375 0.75 0.375

CTABATCC 25923 0.75 0.09 0.09 0.18 0.75 0.09

ATCC 25923EtBr 3 (4�) 0.375 0.375 1.5 3 0.75

RDATCC 25923 500 125 125 250 500 250

ATCC 25923EtBr 1000 (2�) 62.5 62.5 250 1000 500

BERATCC 25923 200 100 100 50 200 50

ATCC 25923EtBr .800 (4�) 400 400 800 800 800

CIPATCC 25923 0.25 0.125 0.125 0.125 0.25 0.125

ATCC 25923EtBr 1 (4�) 0.25 0.25 0.25 1 0.25

NORATCC 25923 0.5 0.125 0.125 0.25 0.25 0.25

ATCC 25923EtBr 2 (4�) 0.25 0.25 0.5 2 0.5

Values in bold type correspond to a decrease of 4-fold or higher on the MIC values in comparison to the ones in the absence of inhibitor. Values inparentheses indicate the MIC increment relative to the one of the original culture. The concentration of each EPI used is defined in the Materials and methodssection.EtBr, ethidium bromide; TPP, tetraphenylphosphonium bromide; DC, dequalinium chloride; DAPI, 40,6-diamidino-2-phenylindole dihydrochloride; PT, penta-midine; BAC, benzalkonium chloride; CHX, chlorhexidine; CTAB, cetyltrimethylammonium bromide; RD, rhodamine; BER, berberine; NOR, norfloxacin;CIP, ciprofloxacin; CCCP, carbonyl cyanide m-chlorophenylhydrazone; TZ, thioridazine; CPZ, chlorpromazine; VER, verapamil; RES, reserpine. Additionalantibiotics tested included nalidixic acid, tetracycline, erythromycin, chloramphenicol and rifampin, with no alteration in the MICs between the two cultures.The effect of ouabain in the MICs was also tested for EtBr, TPP, DC, DAPI and PT, with no observable effect.

Couto et al.

508

As evident from Table 2, the two phenothiazines (thiorida-zine and chlorpromazine), as well as reserpine, reduced theresistance of the ATCC 25923EtBr strain to many of the agents towhich resistance had been induced by exposure to increasingconcentrations of ethidium bromide. A reduction of the MICs toat least a quarter of their original values in the presence of theEPI was considered as indicative of the presence of effluxactivity. The resistance to the agents affected by EPIs indicatesa role of an efflux pump system.

Assessment of efflux activity

Ethidium bromide agar screening method. Both ATCC 25923and ATCC 25923EtBr were evaluated for the presence of EPactivity by the ethidium bromide agar screening method. Whilethe initial parental culture began to exhibit fluorescence at a con-centration of 0.5 mg/L ethidium bromide at 378C, the adaptedstrain began to show fluorescence at an ethidium bromide con-centration of 2 mg/L (Figure 1). Following transfer to 48C, theminimal ethidium bromide concentration-producing fluorescencefor the adapted strain decreased to 1 mg/L (Figure 1), whereasthe transfer of the parent culture to 48C did not alter theminimal concentration of ethidium bromide that caused the flu-orescence of the bacterial mass. These results suggest that theadapted strain is expressing a system (or systems) capable ofextruding ethidium bromide, a process that has been previouslyshown to be temperature-dependent.19,26

Semi-automated fluorometric method. The semi-automatedfluorometric method demonstrated that there was no appreciableefflux activity in the ATCC 25923 culture (Figure 2a), whereasthe ethidium bromide-adapted culture showed pronounced effluxactivity, which was inhibited by the three EPIs tested, in aconcentration-dependent manner (Figure 2b).

Genetic analysis

The two cultures were evaluated for: (i) the presence ofmutations in the chromosome generally associated with

fluoroquinolone resistance and (ii) the presence and expressionof genes coding for several efflux pumps.

Mutations in the grlA and gyrA genes. The two cultures werescreened for mutations in the chromosome most commonlyassociated with fluoroquinolone resistance in S. aureus, namelythe ones in the quinolone resistance-determining regions(QRDRs) of both the grlA and gyrA genes.29 – 31 None of themutations generally described for these two genes were found inany of the cultures that could be associated with the decrease influoroquinolone susceptibility observed for ATCC 25923EtBr.

Screening of genes coding for S. aureus EPs. Genes that codedfor the most common EPs in S. aureus were screened by PCR,using the primers described in Table 1. From the eight genesscreened, six (norA, norB, norC, mepA, mdeA and sepA) weredetected by PCR amplification. No amplification products wereobtained with primers for qacA/B and smr genes. The presenceof these plasmid-encoded EPs (QacA/B and Smr) was ruled outby the absence of plasmids in these cultures (data not shown).

qRT–PCR analysis of the expression level of genes coding forS. aureus EPs. The expression of the six EP genes previouslyfound for the two cultures was then evaluated by a qRT–PCRanalysis, which showed that norA is overexpressed in theadapted strain (34.38+ 3.36). The remaining genes show muchlower variation in their expression: mepA (2.32+ 1.97), norB(1.91+ 1.25), sepA (1.71+ 1.34), mdeA (1.25+ 1.20) andnorC (0.82+ 0.16).

norA promoter sequencing. In order to understand the reasonsfor the overexpression of norA of ATCC 25923EtBr, the promoterregion of this gene was sequenced in both the parental and theethidium bromide-adapted cultures. Sequencing of the norA pro-moter region for both ATCC 25923 and ATCC 25923EtBr

revealed that a 70 bp deletion occurred in ATCC 25923EtBr,immediately downstream of 210 sequence, that encompassesthe majority of the 50-UTR, the site of the so-called flqBmutation, which has been related to increased resistance to

Figure 1. Evaluation of efflux activity of ATCC 25923 original (left) and ATCC 25923EtBr (right) cultures by the ethidium bromide agar screening method.

Cultures were swabbed in TSA plates containing increasing concentrations of ethidium bromide. Following overnight incubation at 37ºC (upper panels) and

after transfer to 4ºC for 16 h (lower panels) fluorescence was detected under UV light.

Ethidium bromide-induced efflux in S. aureus

509

fluoroquinolones, and the site of the assigned transcriptioninitiation (Figure 3).32,33

Discussion

The results of this study show that continuous and prolongedexposure of the S. aureus strain ATCC 25923 to increasing con-centrations of ethidium bromide results in an impressive increasein resistance to this agent. In parallel, resistance to fluoroquino-lones as well as to other known EP substrates is also augmented,in particular, to the biocides tetraphenylphosphonium and

dequalinium. The larger increase in resistance to ethidiumbromide compared with that observed for the other NorA sub-strates may be explained by a greater capacity of NorA toextrude ethidium bromide, inasmuch as this pump has beendescribed as conferring only low-level resistance to fluoroquino-lones.12 However, this difference may also result from increasedcell adaptation to ethidium bromide during the 82 day selectionwith this drug, whereas challenge by the other drugs tested onlyoccurred during the 18 h period of the susceptibility assay. Onemay further hypothesize that cellular factors other than thepump itself may also play a role, namely the involvement ofsensor systems that are specifically activated by each class of

0

0.2

0.4

0.6

0.8

1

1.2

Time (min)

Rel

ativ

e flu

ores

cenc

e

No EPI 5 mg/L CPZ 10 mg/L CPZ

15 mg/L CPZ 25 mg/L CPZ

0

0.2

0.4

0.6

0.8

1

1.2

Time (min)

Rel

ativ

e flu

ores

cenc

e

No EPI 5 mg/L TZ

10 mg/L TZ 12.5 mg/L TZ

0

0.2

0.4

0.6

0.8

1

1.2

0 5 10 15 20 25 0 5 10 15 20 25

0 5 10 15 20 25 0 5 10 15 20 25

0 5 10 15 20 25 0 5 10 15 20 25

Time (min)

Rel

ativ

e flu

ores

cenc

e

No EPI 20 mg/L RES 40 mg/L RES

60 mg/L RES 100 mg/L RES

0

0.2

0.4

0.6

0.8

1

1.2

Time (min)

Rel

ativ

e flu

ores

cenc

e

No EPI 5 mg/L CPZ 10 mg/L CPZ

15 mg/L CPZ 25 mg/L CPZ

0

0.2

0.4

0.6

0.8

1

1.2

Time (min)

Rel

ativ

e flu

ores

cenc

e

No EPI 5 mg/L TZ

10 mg/L TZ 12.5 mg/L TZ

0

0.2

0.4

0.6

0.8

1

1.2

Time (min)

Rel

ativ

e flu

ores

cenc

e

No EPI 20 mg/L RES 40 mg/L RES

60 mg/L RES 100 mg/L RES

(b)(a)

Figure 2. Evaluation of ethidium bromide efflux activity by the semi-automated screening method for ATCC 25923 (a) and ATCC 25923EtBr (b). Assays

were run at 37ºC, in the presence of glucose, with or without the most active EPIs. The data presented were normalized against the conditions of

accumulation (no efflux), and the relative fluorescence was determined. CPZ, chlorpromazine; TZ, thioridazine; RES, reserpine.

Couto et al.

510

substrate recognized by MDR EPs, such as NorA. Recently,we have documented such interplay between EP activity, theirregulation and altered cell permeability to antibiotics inEscherichia coli.34

The demonstration that the ethidium bromide-resistant strainhas a more active, energy-dependent, efflux system was providedby the agar assay as well as by the semi-automated fluorometricmethod. The latter protocol has been recently developed in ourlaboratory and allows the assaying of efflux activity by a simpleand reproducible procedure, which is here applied for the firsttime to Gram-positive bacteria.

The extrusion of ethidium bromide from S. aureus cells hasbeen associated with the EPs NorA, NorB, MdeA, MepA, SepAand SdrM, whose genes are localized in the chromo-some.4,6,7,9,10,13 Other S. aureus pumps that are plasmid-encodedhave also been linked to ethidium bromide extrusion, namelyQacA/B, Smr, QacG, QacED1 and QacH.11,35–38 Since no plas-mids were detected in this work, we studied the expression of thegenes coding for the other, chromosomally located, ethidiumbromide-extruding pumps, except SdrM, due to its low extrudingcapacity for this substrate.7 NorC was also included in our studybecause of its high homology with NorB, which also extrudesethidium bromide, and because others have suggested that the twopumps act in a concerted manner.5,39 In the case of ATCC25923EtBr, qRT–PCR allowed us to correlate the higher effluxactivity of these cells with an increased expression of the genethat codes for NorA.

Many alterations have been described for the promoter regionof norA, including single base pair mutations such as the flqBmutation, as well as deletions and insertions32,39 – 42 that havebeen related to an increased transcription of norA and/orincreased stability of the mRNA molecule.33,39,42 In our study, a70 bp deletion was detected in the promoter region of ATCC25923EtBr. DeMarco et al.39 have previously described a 15 bpdeletion in this same region, which the authors propose to be

associated with alterations in the half-life of the norA mRNA. Inthe case of ATCC 25923EtBr, this 70 bp deletion contains theadenine residue, which is normally situated 7 bp downstream ofthe –10 sequence, to which the transcription start is assigned.The specific assignment of the new norA transcription start inATCC 25923EtBr strain and its relation to the increasedexpression of this gene would require additional studies.

Additional sequencing of the QRDRs of grlA and gyrA forboth ATCC 25923 and ATCC 25923EtBr revealed none of themutations most commonly associated with fluoroquinoloneresistance in S. aureus, namely the mutations D79V, S80F,S80Y, S81P and E84K in GrlA; or S84L, S85P or E88K inGyrA, thus further supporting the suggestion that the decreasedsusceptibility towards fluoroquinolones of the ethidiumbromide-adapted culture results from its extrusion from thecells.29 – 31 However, one cannot exclude the possibility thatother chromosomal mutations may have occurred during theadaptation process, which may contribute to the increasedphenotypic resistance of ATCC 25923EtBr.

Efflux-based resistance can be reversed or reduced by agentsthat are known to inhibit the activity of EPs of MDR bacteria.22

Among the several inhibitors tested, the phenothiazines as wellas reserpine, all of which have been described as inhibitors ofNorA, were the ones showing the highest effect in reducing theMICs of several drugs against ATCC 25923EtBr.

13,16,32 On theother hand, verapamil was less efficient in reducing ethidiumbromide-induced resistance towards some of the drugs, andneither ouabain nor CCCP altered the susceptibility of the ethi-dium bromide-adapted strain to any of the agents employed.Since several S. aureus EPs, including NorA, use the protonmotive force to promote the transport of their substrates, wewere expecting to observe some effect of CCCP, as notedby others, in assays of fluoroquinolone accumulation byS. aureus.3,32,43 However, the CCCP concentrations used inthose studies (100 mM) were far greater than the one used here

ATCC25923_norApromoter TAGTATGATTACTTTTTTGCAATTTCATATGATCAATCCCCTTTATTTTA 50 ATCC25923EtBr_norApromoter TAGTATGATTACTTTTTTGCAATTTCATATGATCAATCCCCTTTATTTTA 50 **************************************************

ATCC25923_norApromoter ATATGTCATTAATTATACAATTAAATGGAAAATAGTGATAATTACAAAGA 100 ATCC25923EtBr_norApromoter ATATGTCATTAATTATACAATTAAATGGAAAATAGTGATAATTACAAAGA 100 **************************************************

# *ATCC25923_norApromoter AAAAAATATTGTCAAATGTAGCAATGTTGTAATACAATATAGAAACTTTT 150 ATCC25923EtBr_norApromoter AAAAAATATTGTCAAATGTAGCAATGTTGTAATACAATA----------- 139 ***************************************

ATCC25923_norApromoter TACGAATATTTAGCATGAATTGCAATTTGTCGTGGAAAAGAAGAATAACA 200 ATCC25923EtBr_norApromoter --------------------------------------------------

ATCC25923_norApromoter GATTTAAGAATGACATGGAGAAAAAAGAGGTGAGCATATGAATAAACAGA 250 ATCC25923EtBr_norApromoter ---------ATGACATGGAGAAAAAAGAGGTGAGCATATGAATAAACAGA 180 *****************************************

ATCC25923_norApromoter TTTTTGTCTTATATTTTAATATTTTCTTGATTTTTTTAGGTATCGGTTTA 300 ATCC25923EtBr_norApromoter TTTTTGTCTTATATTTTAATATTTTCTTGATTTTTTTAGGTATCGGTTTA 230 **************************************************

01-53-

SD NorA

Figure 3. Alignment of the norA promoter sequences in ATCC 25923 and ATCC 25923EtBr. The consensus sequences 210, 235 and Shine-Dalgarno (SD)

for ATCC 25923 are indicated, as well as the initiation codon. The asterisk marks the nucleotide (T) of the flqB mutation. The cardinal marks the site of

transcription initiation assigned for ATCC 25923.

Ethidium bromide-induced efflux in S. aureus

511

(0.88 mM). Concentrations of CCCP of 100 mM have beenrecently shown to produce significant killing of bacteria.27 Nget al.32 noted that when working with 1 mM CCCP, reversal ofnorfloxacin accumulation by S. aureus cells corresponded toonly 6%–10% of that observed with 100 mM CCCP.

The fact that thioridazine, chlorpromazine and reserpine alsohad some effect on the MICs for the original ATCC 25923culture towards the same drugs suggests basal intrinsic effluxactivity, in accordance with that reported by Kaatz and Seo,43

who described low-level expression of NorA in an S. aureusstrain not exposed to any drug. This low-level expression mayaccount for the slight decrease in fluorescence observable in thefluorometric assay for the parental strain, after 18 min incubationin the absence of EPIs (Figure 2a).

While this paper was in review, a report by Garvey andPiddock44 was published describing the selection of efflux-mediated MDR Streptococcus pneumoniae after exposure toreserpine. Their observations confirm the results in this studythat demonstrate the potential for non-antibiotic efflux substratesto select for MDR in S. aureus. Since this MDR phenotype wascorrelated at the genetic level with overexpression of NorA, weinfer that, among the several ethidium bromide chromosomallyencoded EPs, this is the one that cells use as a first-line responseto extrude this noxious agent. A recent study revealed thatamong S. aureus clinical isolates showing increased expressionof EPs, the majority over-expressed a single EP gene, which formost cases was norA, whereas the remaining strains over-expressed two or more genes, most commonly norB and norC.39

These findings support the hypothesis made here of the eventualkey role played by NorA in multidrug efflux by S. aureus.

S. aureus ATCC 25923 has several efflux systems, of whichat least one is expressed at basal levels, and when these cells arechallenged with increasing concentrations of a noxious agentthat is a substrate of these efflux systems, the cells respond byincreasing its efflux and survive, provided that the initial con-centration of the agent is sufficiently low. This process maymirror the one that eventually takes place when a patient isinfected with a bacterium and is given an antibiotic whose finalconcentration is below the one necessary to effectively kill themicroorganism. By activating and/or over-expressing the cellularsystems of drug efflux, the bacteria may be able to survive non-lethal concentrations of the drugs and become refractory totherapy.

Acknowledgements

We acknowledge the assistance of Clara Leandro in the pre-paration of the ethidium bromide-adapted culture. The authorsare also grateful to Hermınia de Lencastre (Laboratory ofMolecular Genetics, ITQB/UNL) for access to PFGE facilities.

Funding

This work was partially funded by Project Ref. Proc. 61056,from Fundacao Calouste Gulbenkian (Portugal). M. M. was sup-ported by grant SFRH/BD/14319/2003 from Fundacao para aCiencia e a Tecnologia (FCT, Portugal).

Transparency declarations

None to declare.

References

1. Lowy FD. Antimicrobial resistance: the example of

Staphylococcus aureus. J Clin Invest 2003; 111: 1265–73.

2. Poole K. Efflux pumps as antimicrobial resistance mechanisms.

Ann Med 2007; 39: 162–76.

3. Yoshida H, Bogaki M, Nakamura S et al. Nucleotide sequence

and characterization of the Staphylococcus aureus norA gene, which

confers resistance to quinolones. J Bacteriol 1990; 172: 6942–9.

4. Truong-Bolduc QC, Dunman PM, Strahilevitz J et al. MgrA is a

multiple regulator of two new efflux pumps in Staphylococcus aureus.

J Bacteriol 2005; 187: 2395–405.

5. Truong-Bolduc QC, Strahilevitz J, Hooper DC. NorC a new

efflux pump regulated by MgrA of Staphylococcus aureus. Antimicrob

Agents Chemother 2006; 50: 1104–7.

6. Huang J, O’Toole P, Shen W et al. Novel chromosomally

encoded multidrug efflux transporter MdeA in Staphylococcus aureus.

Antimicrob Agents Chemother 2004; 48: 909–17.

7. Yamada Y, Hideka K, Shiota S et al. Gene cloning and charac-

terization of SdrM, a chromosomally-encoded multidrug efflux pump,

from Staphylococcus aureus. Biol Pharm Bull 2006; 29: 554–6.

8. Tennent JM, Lyon BR, Gillespie MT et al. Cloning and

expression of Staphylococcus aureus plasmid-mediated quaternary

ammonium resistance in Escherichia coli. Antimicrob Agents

Chemother 1985; 27: 79–83.

9. Kaatz GW, McAleese F, Seo SM. Multidrug resistance in

Staphylococcus aureus due to overexpression of a novel multidrug and

toxin extrusion (MATE) transport protein. Antimicrob Agents

Chemother 2005; 49: 1857–64.

10. Narui K, Noguchi N, Wakasugi K et al. Cloning and characteri-

zation of a novel chromosomal drug efflux gene in Staphylococcus

aureus. Biol Pharm Bull 2002; 25: 1533–6.

11. Paulsen IT, Brown MH, Dunstan SJ et al. Molecular characteri-

zation of the staphylococcal multidrug resistance export protein QacC.

J Bacteriol 1995; 177: 2827–33.

12. Kaatz GW, Seo SM, Ruble CA. Efflux-mediated fluoroquinolones

resistance in Staphylococcus aureus. Antimicrob Agents Chemother

1993; 37: 1086–94.

13. Neyfakh AA, Borsch CM, Kaatz GW. Fluoroquinolone resistance

protein NorA of Staphylococcus aureus is a multidrug efflux transpor-

ter. Antimicrob Agents Chemother 1993; 37: 128–9.

14. Kristiansen MM, Leandro C, Ordway D et al. Phenothiazines

alter resistance of methicillin-resistant strains of Staphylococcus

aureus (MRSA) to oxacillin in vitro. Int J Antimicrob Agents 2003; 22:

250–3.

15. Kristiansen MM, Leandro C, Ordway D et al. Thioridazine

reduces resistance of methicillin-resistant Staphylococcus aureus by

inhibiting a reserpine-sensitive efflux pump. In Vivo 2006; 20: 361–6.

16. Kaatz GW, Mougdal VV, Seo SM et al. Phenothiazines and

thioxanthenes inhibit multidrug efflux pump activity in Staphylococcus

aureus. Antimicrob Agents Chemother 2003; 47: 719–26.

17. Viveiros M, Portugal I, Bettencourt R et al. Isoniazid-induced

transient high-level resistance in Mycobacterium tuberculosis.

Antimicrob Agents Chemother 2002; 46: 2804–10.

18. Viveiros M, Jesus A, Brito M et al. Inducement and reversal of

tetracycline resistance in Escherichia coli K-12 and expression of

proton gradient-dependent multidrug efflux pump genes. Antimicrob

Agents Chemother 2005; 49: 3578–82.

Couto et al.

512

19. Martins A, Couto I, Aagaard L et al. Prolonged exposure of

methicillin-resistant Staphylococcus aureus (MRSA) COL strain to

increasing concentrations of oxacillin results in a multidrug-resistant

phenotype. Int J Antimicrob Agents 2007; 29: 302–5.

20. Clinical and Laboratory Standards Institute. Performance

Standards for Antimicrobial Susceptibility Testing: Sixteenth

Informational Supplement M100-S16. CLSI, Wayne, PA, USA, 2006.

21. Andrews JM for the BSAC Working Party on Susceptibility

Testing. BSAC standardized disc testing method (version 4).

J Antimicrob Chemother 2005; 56: 60–76.

22. Marquez B. Bacterial efflux systems and efflux pumps inhibitors.

Biochimie 2005; 87: 1137–47.

23. Chung M, Lencastre H, Matthews P et al. Molecular typing of

methicillin-resistant Staphylococcus aureus by pulsed-field gel electro-

phoresis: comparison of results obtained in a multilaboratory effort

using identical protocols and MRSA strains. Microb Drug Resist 2000;

6: 189–98.

24. Bjorland J, Sunde M, Waage S. Plasmid-borne smr gene

causes resistance to quaternary ammonium compounds in bovine

Staphylococcus aureus. J Clin Microbiol 2001; 39: 3999–4004.

25. Langmann T, Mauerer R, Zahn A et al. Real-time reverse

transcription-PCR expression profiling of the complete human

ATP-binding cassette transporter superfamily in various tissues. Clin

Chem 2003; 49: 230–8.

26. Martins M, Santos B, Martins A et al. An instrument-free method

for the demonstration of efflux pump activity of bacteria. In Vivo 2006;

20: 657–64.

27. Viveiros M, Martins A, Paixao L et al. Demonstration of intrinsic

efflux activity of E. coli K-12 AG100 by an automated ethidium bromide

method. Int J Antimicrob Agents 2008; 31: 458–62.

28. EUCAST Clinical MIC breakpoints—fluoroquinolones. EUCAST

Website (v2.4). http://www.srga.org/eucastwt/MICTAB/MICquinolones.

htm (25 March 2008, date last accessed).

29. Ferrero L, Cameron B, Crouzet J. Analysis of gyrA and grlA

mutations in stepwise-selected ciprofloxacin-resistant mutants of

Staphylococcus aureus. Antimicrob Agents Chemother 1995; 39:

1554–8.

30. Horii T, Suzuki Y, Monji A et al. Detection of mutations in

quinolone resistance-determining regions in levofloxacin- and

methicillin-resistant Staphylococcus aureus: effects of the mutations on

fluoroquinolone MICs. Diagn Microbiol Infect Dis 2003; 46: 139–45.

31. Hooper DC. Mechanisms of fluoroquinolone resistance. Drug

Resist Updat 1999; 2: 38–55.

32. Ng EYW, Trucksis M, Hooper DC. Quinolone resistance

mediated by norA: physiologic characterization and relationship to flqB,

a quinolone resistance locus on the Staphylococcus aureus chromos-

some. Antimicrob Agents Chemother 1994; 38: 1345–55.

33. Fournier B, Truong-Bolduc Q, Zhang X et al. A mutation in the

50 untranslated region increases stability of norA mRNA, encoding a

multidrug resistance transporter of Staphylococcus aureus. J Bacteriol

2001; 183: 2367–71.

34. Viveiros M, Dupont M, Rodrigues L et al. Antibiotic stress,

genetic response and altered permeability of E coli. PLoS ONE 2007;

2: e365.

35. Brown MH, Skurray RA. Staphylococcal multidrug efflux protein

QacA. J Mol Microbiol Biotechnol 2001; 3: 163–70.

36. Heir E, Sundheim G, Holck AL. The qacG gene on plasmid

pST94 confers resistance to quaternary ammonium compounds in sta-

phylococci isolated from the food industry. J Appl Microbiol 1999; 86:

378–88.

37. Kazama H, Hamashima H, Sasatsu M et al. Distribution of the

antiseptic-resistance gene qacED1 in Gram-positive bacteria. FEMS

Microbiol Lett 1998; 165: 295–9.

38. Heir E, Sundheim G, Holck AL. The Staphylococcus qacH gene

product: a new member of the SMR family encoding multidrug resist-

ance. FEMS Microbiol Lett 1998; 163: 49–56.

39. DeMarco CE, Cushing LA, Frempong-Manso E et al.

Efflux-related resistance to norfloxacin, dyes, and biocides in blood-

stream isolates of Staphylococcus aureus. Antimicrob Agents

Chemother 2007; 51: 3235–9.

40. Schmitz F-J, Hertel B, Hofmann B et al. Relationship between

mutations in the coding and promoter regions of the norA genes in 42

unrelated clinical isolates of Staphylococcus aureus and the MICs of

norfloxacin for these strains. J Antimicrob Chemother 1998; 42:

561–3.

41. Tanaka M, Wang T, Onodera Y et al. Mechanism of quinolone

resistance in Staphylococcus aureus. J Infect Chemother 2000; 6:

131–9.

42. DeMarco CE, Frempong-Manso E, Seo SM et al. Multidrug

efflux-pump (MDEP) gene expression in geographically diverse

Staphylococcus aureus (SA) strains. In: Proceedings of the 47th

Interscience Conference on Antimicrobial Agents and Chemotherapy

(ICAAC), Chicago, IL, 2007. Abstract C2-192, p. 113. American

Society for Microbiology, Washington, DC, USA.

43. Kaatz GW, Seo SM. Inducible NorA-mediated multidrug resist-

ance in Staphylococcus aureus. Antimicrob Agents Chemother 1995;

39: 2650–5.

44. Garvey MI, Piddock LJV. The efflux pump inhibitor reserpine

selects multidrug resistant Streptococcus pneumoniae that over-

express the ABC transporters PatA and PatB. Antimicrob Agents

Chemother 2008; 52: 1677–85.

45. Anthonisen I-L, Sunde M, Steinum TM et al. Organization of

the antiseptic resistance gene qacA and Tn552-related b-lactamase

genes in multidrug-resistant Staphylococcus haemolyticus strains of

animal and human origins. Antimicrob Agents Chemother 2002; 46:

3606–12.

46. Sierra JM, Ruiz J, Anta MTJ et al. Prevalence of two different

genes encoding NorA in 23 clinical strains of Staphylococcus aureus.

J Antimicrob Chemother 2000; 46: 145–6.

47. Lane DJ. 16S/23S rRNA sequencing. In: Stackebrant E,

Goodfellow M, eds. Nucleic Acid Techniques in Bacterial Systematics.

London: John Wiley & Sons Ltd, 1991; 115–75.

48. Pan X, Hamlyn P, Talens-Visconti R et al. Small-colony mutants

of Staphylococcus aureus allow selection of gyrase-mediated resist-

ance to dual-target fluoroquinolones. Antimicrob Agents Chemother

2002; 46: 2498–506.

Ethidium bromide-induced efflux in S. aureus

513