Journal of Antimicrobial Chemotherapy

JAC Advance Access originally published online on December 4, 2006
Journal of Antimicrobial Chemotherapy 2007 59(6):1247-1260; doi:10.1093/jac/dkl460

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 59/6/1247    most recent dkl460v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Request Permissions Disclaimer Google Scholar Articles by Stavri, M. Articles by Gibbons, S. Search for Related Content PubMed PubMed Citation Articles by Stavri, M. Articles by Gibbons, S. Social Bookmarking          What's this?

© The Author 2006. 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

Special section: Efflux

Bacterial efflux pump inhibitors from natural sources

Michael Stavri¹, Laura J. V. Piddock² and Simon Gibbons^1,*

¹ Centre for Pharmacognosy and Phytotherapy, The School of Pharmacy, University of London 29-39 Brunswick Square, London WC1N 1AX, UK ² Antimicrobial Agents Research Group, Division of Immunity and Infection, The Medical School, University of Birmingham Birmingham B15 2TT, UK

---

^*Corresponding author. Tel: +44-207-7535913; Fax: +44-207-7535909; E-mail: simon.gibbons{at}pharmacy.ac.uk

	Abstract

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
The rapid spread of bacteria expressing multidrug resistance (MDR) has necessitated the discovery of new antibacterials and resistance-modifying agents. Since the initial discovery of bacterial efflux pumps in the 1980s, many have been characterized in community- and hospital-acquired Gram-positive and Gram-negative pathogens, such as Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli and, more recently, in mycobacteria. Efflux pumps are able to extrude structurally diverse compounds, including antibiotics used in a clinical setting; the latter are rendered therapeutically ineffective. Antibiotic resistance can develop rapidly through changes in the expression of efflux pumps, including changes to some antibiotics considered to be drugs of last resort. It is therefore imperative that new antibiotics, resistance-modifying agents and, more specifically, efflux pump inhibitors (EPIs) are characterized. The use of bacterial resistance modifiers such as EPIs could facilitate the re-introduction of therapeutically ineffective antibiotics back into clinical use such as ciprofloxacin and might even suppress the emergence of MDR strains. Here we review the literature on bacterial EPIs derived from natural sources, primarily those from plants. The resistance-modifying activities of many new chemical classes of EPIs warrant further studies to assess their potential as leads for clinical development.

Keywords: MDR , MRSA , Staphylococcus aureus , NorA, efflux , Mycobacterium , Pseudomonas , Escherichia , modulators

	Introduction

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
Staphylococcus aureus is an important community- and major hospital-acquired pathogen.¹,² This organism is cause for considerable concern due to its ability to acquire resistance towards the newest antibacterial drugs currently on the market. In the UK, the number of death certificates citing methicillin-resistant S. aureus (MRSA) trebled from 398 in 1998 to 1168 in 2004.³ These figures are believed to be a conservative estimate of the actual number of deaths as a result of MRSA. Similar data are available for many developed countries. Fluoroquinolones were thought to be useful anti-staphylococcal agents, but resistance quickly emerged.⁴–⁶ Resistance to vancomycin, once thought of as the drug of last resort for the treatment of MRSA, has now been described in a strain in the USA.⁷,⁸ MRSA resistant to linezolid, an anti-staphylococcal oxazolidinone have already been reported.⁹ This leaves the streptogramin mixture quinupristin/dalfopristin and the cyclic lipopeptide daptomycin as the drugs for the treatment of methicillin-susceptible S. aureus (MSSA) and MRSA, although resistance has been reported to these drugs as well. Should widespread resistance to these agents emerge, substances that can increase susceptibility to currently licensed agents would be a very attractive option.

Gram-negative bacteria and mycobacteria both possess thick outer membranes that are highly hydrophobic, providing these organisms with a permeability barrier¹⁰ especially towards hydrophilic compounds such as macrolide antibiotics like erythromycin.¹¹ This in part explains the greater resistance observed by Gram-negative bacteria as opposed to Gram-positive organisms. Various mechanisms provide bacteria with resistance to antibiotics; these include target-site modification and antibiotic inactivation. Resistance to macrolides,¹² vancomycin,¹³ ß-lactams,¹⁴,¹⁵ fluoroquinolones¹⁶ and aminoglycosides¹⁷ has been achieved by target-site alteration, while antibiotic inactivation mechanisms have accounted for resistance towards ß-lactams,¹⁸ aminoglycosides¹⁹ and chloramphenicol.²⁰ These mechanisms are important to bacteria to provide resistance, but do so to only a single class of compound. To become multidrug resistant a bacterium must acquire multiple mechanisms, and whilst many species have done so the spectra of resistances vary. Many efflux pumps are encoded chromosomally and their presence enhances resistance mediated by these individual mechanisms. Problems also now exist with the prevalence of human pathogenic bacteria overexpressing pumps and conferring multidrug resistance (MDR). A single pump can provide bacteria with resistance to a wide array of chemically and structurally diverse compounds. The problems of resistant Gram-positive and Gram-negative bacteria highlight the urgent need for new drugs with new modes of action and/or combination therapy to treat infections caused by resistant human pathogens such as S. aureus, Pseudomonas aeruginosa and Mycobacterium tuberculosis.

Here we review the literature concerning bacterial resistance modulators from natural sources and will highlight plants with the potential of discovering new bacterial efflux pump inhibitors (EPIs). The majority of EPIs discussed are putative inhibitors of efflux pumps of the highly problematic human pathogen S. aureus. This review also explores the problems posed by Gram-negative bacteria and mycobacteria possessing efflux mechanisms as a mechanism of resistance towards clinically relevant drugs and how EPIs may be one way of tackling this mechanism of resistance.

	Bacterial efflux families

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
The phenomenon of microbial multidrug efflux was first reported by Ball et al.²¹ and McMurry et al.²² for the efflux of tetracycline in Escherichia coli. This resistance was transferable between strains and was encoded by tet (tetracycline) determinants, which were encoded either on plasmids or transposons.²³,²⁴ Since this initial discovery, further efflux systems have been identified in Gram-positive and Gram-negative bacteria and, more recently, in mycobacteria. Bacterial efflux transporters can be divided into five main families primarily based on amino acid sequence homology.²⁵ These are the major facilitator (MF) superfamily, the resistance-nodulation-division (RND) family, the small MDR (SMR) family, the ATP binding cassette (ABC) family and the multiple antibiotic and toxin extrusion (MATE) family. The first three families achieve the energy required to extrude a drug out of the cell via the proton motive force in a proton-drug antiport system, whilst the MATE family is driven by the exchange of either proton or sodium ions. In contrast, the ABC family couples drug extrusion with the hydrolysis of ATP.²⁵ Efflux of drugs from Gram-positive bacteria is mediated by a single cytoplasmic membrane-located transporter of the MF, SMR or ABC families.²⁵ The efflux pumps in Gram-negative bacteria are more complex due to the presence of an outer membrane; they form a tripartite protein channel, which requires a protein that traverses the periplasm known as the membrane fusion protein (MFP) and an outer membrane efflux protein (OEP) along with the cytoplasmic membrane-located transporter. It is not uncommon for an organism to code for more than one efflux pump, which may either be expressed constitutively or induced in direct response to the presence of a substrate. P. aeruginosa constitutively expresses the MexAB-OprM multidrug efflux pump, which is the main member of the RND family in this organism.²⁶ However, P. aeruginosa also have the MexXY-OprM pump, which is inducible in the presence of any of its substrates, such as aminoglycosides.²⁷ Multidrug efflux pumps therefore contribute to the intrinsic resistance of P. aeruginosa.²⁶

Bacterial efflux pumps offer potential targets to combat problematic infectious diseases such as those caused by MRSA, E. coli and P. aeruginosa. In Gram-positive organisms, the pumps studied in greatest detail include the NorA, Tet(K) and Msr(A) transporters.²⁸ In Gram-negative bacteria, studies have focused on the tripartite AcrAB-TolC and MexAB-OprM efflux pumps of E. coli and P. aeruginosa, respectively,²⁸ and also the FloR efflux pump of Salmonella enterica serovar Typhimurium.²⁹ The Tap and DrrAB efflux proteins of mycobacteria, including M. tuberculosis, can extrude a range of chemically diverse compounds.²⁵,²⁸

A genetic approach to determine the consequences of inhibiting the efflux pumps of P. aeruginosa was undertaken by Lomovskaya et al.;³⁰ inhibition significantly decreased MICs for both antibiotic-susceptible and -resistant bacteria, reversed acquired resistance, and resulted in a decreased frequency of emergence of P. aeruginosa mutants highly resistant to fluoroquinolones. It is therefore imperative to identify agents that can block efflux and, in so doing, extend the life of existing antibacterial drugs. Currently there are no EPIs on the market that can be used in combination with a drug that is a pump substrate to recover its clinical utility. However, the concept of using a compound that inhibits resistance together with a conventional antibiotic is well proven and co-amoxiclav is an important example.³¹

	Screening for EPIs

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
It has been known for many years that some antibiotics exert synergy when used together and the chequerboard assay³² has been used to identify such agents. Variations of this method have been applied to identify potential inhibitors of efflux pumps.

The modulation assay is a quick and easy method to identify potential EPIs in Gram-positive and Gram-negative bacteria. An initial study of the antibacterial activity of an extract or crude fractions is necessary to guard against false-positive results. A concentration, normally 4-fold lower than the MIC, is chosen when performing a potentiation assay. The modulation assay³³,³⁴ requires a sub-inhibitory concentration of a crude fraction or pure compound to be dissolved in dimethylsulphoxide (DMSO) and diluted in Mueller–Hinton broth. Serial doubling dilutions of a drug known to be a substrate for an efflux transporter, such as norfloxacin for the NorA protein, is added and microtitre plates are then interpreted in the same manner as MIC determinations. All samples are tested in duplicate.

The chequerboard assay identifies synergic combinations of antimicrobial agents and has been used to screen for potential EPIs. Serial 2-fold dilutions of a pump substrate, such as norfloxacin or berberine for the NorA pump, as well as 2-fold dilutions of a fraction or test compound will result in microtitre wells with a different combination of pump substrate and fraction or test compound concentration.³⁵,³⁶

The berberine uptake assay³⁷ has also been used in bioassay-guided isolation of MDR inhibitors. Crude extracts and fractions are tested in the presence and absence of a sub-inhibitory concentration of this antibacterial alkaloid. Bacterial growth in the absence of berberine and no growth in its presence can be taken as an indicator of the presence of an MDR inhibitor in the extract.³⁸ This is an important screening tool enabling many fractions to be tested quickly and easily.

The ethidium bromide efflux inhibition assay³⁹,⁴⁰ is a more detailed study of the potentiation activity of a test compound. Ethidium bromide is a substrate for a number of MDR efflux pumps. The activity of putative inhibitors can be measured fluorometrically due to the retention of fluorescence over time if efflux is reduced. Similar efflux assays can be performed with acriflavine or pyronin Y.⁴⁰

Accumulation studies have been used to identify potential EPIs. Various substrates that have been used in accumulation studies include ethidium bromide,²⁶,³⁹ norfloxacin,⁴¹,⁴² berberine³⁷,⁴³ and novobiocin.⁴⁴ Assays can be performed in a number of ways to determine the effect of a potential efflux inhibitor on a bacterial strain possessing an efflux pump. One method is the incorporation of an efflux inhibitor midway through a time-course assay in order to detect a difference in fluorescence. Another method is to run two separate time-course assays, one in the absence and one in the presence of an inhibitor to determine any effect a test compound may have as a potential inhibitor. An increase in drug accumulation only in the presence of an inhibitor indicates that the inhibitor is a blocker of an efflux mechanism.⁴⁵

	EPIs from plant sources

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
Work on staphylococcal efflux pumps has really only advanced in the past 2 years, with the identification of MDR transporters in addition to NorA. All data prior to this time on staphylococci and EPIs are either genetic or microbiological with little or no biochemistry. S. aureus has multiple transporters encoded by its genome, some of which are now known to transport antimicrobials, and there is poor evidence that the EPIs described to date interact solely with the NorA protein; reversal of resistance in NorA overexpressing strains is indicative but not conclusive; so, too, is the absence of activity versus strains lacking NorA. Overexpression, disruption or deletion of one pump can affect the expression of other MDR transporters.⁴⁶ A microbiological assay and even an efflux assay only gives the total phenotype, which is the sum of all transporter activity present in the bacterial cell at that time. Only crystal structure data of the purified protein and the inhibitor co-crystallized can really provide detailed information about binding. Experiments with pure protein are those that provide conclusive evidence.

The antihypertensive plant alkaloid reserpine (1) (Figure 1) was first isolated from the roots of Rauwolfia vomitoria Afz.⁴⁷ Its EPI activity was originally demonstrated against the Bmr efflux pump, which mediates tetracycline efflux in Bacillus subtilis.⁴⁸ Klyachko et al.⁴⁹ demonstrated that reserpine interacts directly with the Bmr protein at amino acids phenylalanine-143, valine-286 and phenylalanine-306, which form a reserpine-binding site. Reserpine also potentiated the activity of tetracycline (a 4-fold reduction in MIC) in two clinical isolates of MRSA, IS-58 and XU212, which possessed the Tet(K) efflux protein.⁵⁰ Reserpine also reversed NorA-conferred MDR,⁵¹ and Kaatz and Seo⁵² showed that it enhanced the activity of norfloxacin against S. aureus. NorA is one of the major MDR transporters in S. aureus and causes a significant decrease in susceptibility towards fluoroquinolones. In a study carried out by Schmitz et al.,⁵³ the MICs of the fluoroquinolones ciprofloxacin, moxifloxacin and sparfloxacin in the presence of reserpine were lowered by as much as 4-fold in 48, 21 and 11 of 102 S. aureus isolates tested, respectively. Fluoroquinolone resistance in Streptococcus pneumoniae has also been negated in the presence of reserpine.⁵⁴,⁵⁵ In 1999 Gill et al.⁵⁶ identified PmrA, which has 43% amino acid similarity with NorA. Subsequent reduction of norfloxacin MIC against a norfloxacin-resistant construct (R6N) in the presence of reserpine led the authors to interpret that reserpine inhibited the PmrA protein. However, this interaction has not been conclusively proven. Fluoroquinolone accumulation studies using strain R6N as compared with R6 (wild-type) resulted in a decrease in the accumulation of these drugs.⁵⁷ However, the addition of reserpine at a concentration used in potentiation assays failed to increase fluoroquinolone accumulation suggesting that this compound interacts with another protein other than PmrA.⁵⁷ An ABC transporter associated with ciprofloxacin resistance has recently been identified by Marrer et al.⁵⁸ and subsequent deletion of this transporter resulted in these S. pneumoniae strains being conferred multidrug susceptible.⁵⁹

View larger version (9K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 1. Chemical structures of compounds 1–4.

 
A number of MDR pump inhibitors against the human pathogen S. aureus have been described by the Lewis group. Berberine (2), isolated from Berberis fremontii, is an alkaloid with only weak antibacterial activity (MIC = 256 mg/L) against a wild-type strain of S. aureus.⁶⁰ However, the isolation of the flavonolignan 5'-methoxyhydnocarpin-D (5'-MHC-D) (3) and a synergistic study between these two compounds led to a 16-fold increase in the antibacterial activity of berberine (MIC = 16 mg/L).⁶⁰ 5'-MHC-D also had a synergistic effect with several other NorA substrates, including norfloxacin.

The isolation of the porphyrin pheophorbide a (4) from Berberis species and a further flavonolignan, silybin (5) (Figure 2) a diastereomeric mixture of 5a and 5b, from Milk thistle (Silybum marianum)³⁸,⁶¹ also demonstrated synergistic activity against S. aureus.

View larger version (11K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 2. Chemical structures of compounds 5–10.

 
There have also been a number of methoxylated flavones⁶² and isoflavones⁶³ described as putative inhibitors of the MDR pump NorA in the presence of subinhibitory concentrations of berberine and the fluoroquinolone norfloxacin. The flavones chrysosplenol-D (6) (MIC = 25 mg/L with 30 mg/L berberine) and chrysoplenetin (7) (MIC = 6.25 mg/L with 30 mg/L berberine), isolated from Artemisia annua (Asteraceae), were shown previously to potentiate the activity of the antimalarial artemisinin against Plasmodium falciparum.⁶⁴ Both 6 and 7 are weakly antibacterial against S. aureus and this is not an uncommon feature for a potentiator to also exert a direct antibacterial effect. Care must be taken when interpreting results to ensure that activity is solely due to potentiation and not by direct inhibition. It is likely that these compounds act against a MDR efflux pump, to which artemisinin is a substrate, in P. falciparum, in a similar manner as in S. aureus. The active isoflavones from Lupinus argenteus, genistein (8), orobol (9) and biochanin A (10) also reduced the MIC of berberine (16-fold) and norfloxacin (4-fold).

A study of popular horticultural taxa such as Geranium has led to the isolation of putative inhibitors of S. aureus NorA, these included the polyacylated neohesperidosides (11 and 12) (Figure 3) from G. caespitosum.⁶⁵ The pentaester, 12, increased the activity of berberine, ciprofloxacin, norfloxacin and rhein, an antibacterial component of rhubarb.

View larger version (10K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 3. Chemical structures of compounds 11–14.

 
An investigation of Dalea versicolor (Fabaceae) ‘mountain delight’ resulted in the isolation of phenolic metabolites that enhanced the activity of berberine, erythromycin and tetracycline against S. aureus.⁶⁶ The chalcone (13) enhanced the activity such that the MICs were comparable to those for a mutant lacking NorA, suggesting that this agent is an inhibitor of NorA. This compound, along with the stilbene (14), also increased the activity of these antibiotics against Bacillus cereus with activity being the greatest in combination with berberine. This alkaloid is a substrate for the NorA efflux pump and a genomic comparison has revealed the presence of a Bmr homologue in this species, which is a homologue of NorA. Together with the activity data recorded against S. aureus, it is suggested that agent 13 is a putative EPI of the NorA efflux pump.

A new arylbenzofuran aldehyde (spinosan A) (15), a known pterocarpan (16) and isoflavone (17) (Figure 4) were isolated from another Dalea species, the ‘smoke tree’, Dalea spinosa, which exerted a potentiation activity in the presence of berberine.³⁵ All three compounds enhanced berberine activity against the wild-type strain of S. aureus, lowering the MIC 4- to 8-fold. These compounds also caused an increased activity of berberine against an isogenic NorA mutant lowering the MIC 2- to 15-fold, but none was active against the NorA overexpressing mutant. It is likely that agents 15–17 cause inhibition of an efflux pump other than NorA.

View larger version (5K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 4. Chemical structures of compounds 15–17.

 
A phytochemical investigation of Mexican Morning Glory species⁶⁷ led to the isolation of three oligosaccharides exerting a potentiation effect of norfloxacin against the NorA overexpressing S. aureus strain SA-1199B. The amphipathic orizabin XIX (18) (Figure 5) increased the activity of norfloxacin 4-fold (8 mg/L from 32 mg/L) at a concentration of 25 mg/L. Whilst orizabin IX (19) enhanced norfloxacin activity 16-fold when incorporated at a concentration of 1 mg/L. The ethidium efflux inhibition assay utilizing SA-1199B demonstrated that orizabin IX and another orizabin, orizabin XV (20), were nearly equipotent. The authors have not taken the possibility of quenching into account. It has recently been demonstrated that high intracellular concentrations of ethidium bromide result in a decrease in fluorescence due to self-quenching.⁶⁸ These oligosaccharides showed good activity at low concentrations (10 µM or less) being more active than reserpine and they could be further developed to provide more potent inhibitors of this multidrug efflux pump. The acylation of some of the free hydroxyl groups of the oligosaccharide and the lipophilic alkyl chain would seem to be important in facilitating cellular uptake to its MDR pump target.

View larger version (14K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 5. Chemical structures of compounds 18–22.

 
The catechin gallates are another group of phenolic metabolites that have provided interest, initially by Hamilton-Miller's group, due to their ability to reverse methicillin resistance in MRSA.⁶⁹–⁷¹ Modulation assays of epicatechin gallate (21) and epigallocatechin gallate (22) in 96-well microtitre plates showed these compounds reduced the MIC of norfloxacin 4-fold against SA-1199B at a concentration of 20 mg/L.⁷² As both compounds possess moderate antibacterial activity against this strain, the ethidium efflux inhibition assay was performed to verify this activity. The authors have not taken the possibility of quenching into account. Both compounds were found to weakly inhibit the NorA efflux pump, with epicatechin gallate 21 being slightly more potent. Interestingly, both compounds were reported to enhance efflux at low concentrations (Figure 6) resulting in the authors hypothesizing the presence of two binding sites on the NorA transporter, one with a high affinity for catechins and another with a low affinity.⁷² Low concentrations of catechins would result in the high-affinity binding sites being occupied preferentially and therefore enhancing efflux. However, an increased concentration of catechins would result in the low-affinity sites being bound as well resulting in the reversal of efflux enhancement to one of a mild efflux inhibitor. Further work to study the low concentration of catechins could help to understand how these pumps are controlled. Another hypothesis for these data could be that catechins interact with another pump. Epigallocatechin gallate has also been shown to enhance tetracycline activity in Tet(K) resistant staphylococci.⁷³ At a concentration of 30 mg/L epigallocatechin gallate caused a 4- and 8-fold increase in the activity of tetracycline in Staphylococcus epidermidis and S. aureus, respectively. A tetracycline uptake and release study was performed on resistant and susceptible S. epidermidis with and without pre-treatment with 22 (50 mg/L) over a 15 min time-course. The tetracycline release curve was steeper in the resistant strain, without 22, than that of the susceptible strain, indicating greater efflux of the resistant strain.⁷³ The curves were shallower when strains were pre-treated with 22 indicating a greater retention of tetracycline within the cells.⁷³ Roccaro et al.⁷³ then performed the same experiments with protoplasts of S. epidermidis cells to confirm that 22 acts against Tet(K) rather than binding to peptidoglycan. The results obtained with protoplasts were not affected under these circumstances, indicating that peptidoglycan did not affect the uptake and release of tetracycline. The tetracycline-susceptible isolates of these species showed an 8-fold reduction in the MIC of tetracycline, so whilst epigallocatechin gallate appears to be an inhibitor of the Tet(K) pump, it may also cause inhibition of an as yet undefined efflux mechanism.

View larger version (12K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 6. Inhibition of ethidium efflux by catechin gallates. Reserpine was included for comparative purposes. Filled circles, reserpine; filled triangles, epicatechin gallate (21); filled diamonds, epigallocatechin gallate (22).

 
The abietane diterpenes carnosic acid (23) and carnosol (24) (Figure 7), isolated from the popular herb Rosemary (Rosmarinus officinalis), were identified as potentiators of tetracycline and erythromycin against S. aureus strains possessing the Tet(K) and Msr(A) efflux pumps, respectively.³⁴ Both 23 and 24 enhanced tetracycline activity against S. aureus XU212, possessing the Tet(K) efflux protein, at a concentration of 10 mg/L. Carnosic acid also enhanced the activity of erythromycin, causing an 8-fold reduction in MIC (32 mg/L from 256 mg/L) against the erythromycin-resistant strain RN4220 which expresses the Msr(A) efflux protein. An ethidium efflux inhibition assay incorporating carnosic acid against strain SA-1199B identified this compound as a moderately active potentiator of norfloxacin against the NorA pump with an IC₅₀ of 50 µM (16.6 mg/L), equivalent to approximately one quarter of the MIC for the strain. The authors did not take the possibility of quenching into account.

View larger version (9K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 7. Chemical structures of compounds 23–30.

 
As part of a project to identify natural plant products with modulation activity, an extract of Lycopus europaeus (Lamiaceae) was investigated. L. europaeus is also commonly known as Gipsywort in Britain and the lipophilic extract caused a potentiation of tetracycline and erythromycin against strains IS-58 and RN4220 of S. aureus possessing multidrug efflux pumps Tet(K) and Msr(A), respectively.³³ Bioassay-guided isolation of the hexane extract led to the isolation of six metabolites that included four isopimarane diterpenes (25–28) (Figure 7) and two oxidized geranylgeranyl diterpenes (29 and 30). Various chromatographic techniques were used to isolate these compounds and the activity of each fraction was tracked by modulation assay until the active components were purified. None of these compounds was active at 512 mg/L. However, when incorporated into the Mueller–Hinton broth at a concentration of 10 mg/L each of these compounds reduced the MIC of tetracycline and erythromycin 2-fold against S. aureus IS-58 [Tet(K)] and S. aureus RN4220 [Msr(A)]. Interestingly, there were no differences in activities of these diterpenes, despite the structural differences between the isopimarane and geranylgeranyl groups. None of the compounds was able to enhance norfloxacin activity against S. aureus 1199B that overexpresses NorA. A common feature of these six compounds is that they are highly lipophilic, suggesting that this is a key factor for an inhibitor of MDR efflux pumps of Gram-positive bacteria.

Baicalein (31) (Figure 8), a trihydroxy flavone isolated from the leaves of the commonly used herb thyme (Thymus vulgaris), was identified as possessing a strong synergic activity when used in conjunction with tetracycline or the ß-lactam antibiotics oxacillin, cefmetazole and ampicillin against MRSA.⁷⁴ Baicalein is weakly antibacterial (MIC = 100 mg/L) but at 25 mg/L, it reduced the MIC of tetracycline from 4 to 0.12 mg/L against the MRSA clinical isolate OM481. This compound reduced the MIC of tetracycline against another MRSA isolate, OM584, this time by a factor of four. Expression of an S. aureus-derived tet(K) gene into an E. coli host led to a 16-fold increase in MIC, but in the presence of baicalein this was again reduced 4-fold. There was also a decrease in MIC detected in the E. coli strain not expressing the Tet(K) efflux protein. MRSA OM481 did not possess the Tet(K) protein indicating that baicalein (31) may inhibit another MDR pump in these MRSA isolates or have more than one mechanism of action, such as interfering with the integrity of the cell wall.

View larger version (9K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 8. Chemical structures of compounds 31–37.

 
A biological evaluation of grapefruit oil, which can be isolated from the species Citrus paradisi, has highlighted some of the components as potential modulators of efflux pumps in MRSA strains. Fractionation of the grapefruit oil led to the characterization of a coumarin derivative (32), a bergamottin epoxide derivative (33) and a coumarin epoxide derivative (34) (Figure 8) able to enhance the activity of ethidium bromide and norfloxacin.⁷⁵ Both 32 and 34 caused a 2-fold reduction in MIC of ethidium bromide whilst 33 caused a 6-fold reduction. 33 and 34 also caused a 20-fold reduction in the MIC of norfloxacin against MRSA strains but not against MSSA strains. Kristiansen et al.⁷⁶ have suggested that MRSA strains may be more susceptible than MSSA to chlorpromazine, which inhibits multidrug efflux pump activity. Whilst MICs of oxacillin were lowered in the presence of chlorpromazine complete resistance reversal was not achieved. However, these data indicate that efflux may play a role in providing MRSA strains with resistance to antibiotics such as the penicillins. This view is supported by recent work performed on 12 MRSA strains that showed an up-regulation of efflux genes including norA, erm(A) and erm(B).⁷⁷

Bioassay-guided fractionation of an extract of Jatropha elliptica (Euphorbiaceae) led to the isolation of the penta-substituted pyridine, 2,6-dimethyl-4-phenyl-pyridine-3,5-dicarboxylic acid diethyl ester (35),⁷⁸ which is not antibacterial but does augment ciprofloxacin and norfloxacin activity against S. aureus SA-1199B. The coumarin lignan propacine (36) also showed moderate activity as a modulator in combination with ciprofloxacin against this strain. The activity of this compound is proposed by Marquez et al.⁷⁸ to be due to the lignan portion of the molecule based on the flavonolignan work published by Guz et al.⁷⁹ Work on plants belonging to the family Euphorbiaceae has resulted in the isolation of inhibitors of the mammalian MDR transporter P-glycoprotein, including a jatrophane diterpene that caused a 2-fold greater inhibition of daunomycin efflux, with respect to cyclosporin A, at a concentration of 5 µM.⁸⁰–⁸² So it is unsurprising that activity towards bacterial efflux mechanisms is also being reported.

Piperine (37) (Figure 8), a major plant alkaloid within the family Piperaceae including black pepper (Piper nigrum) and long pepper (Piper longum), has recently been reported to enhance the accumulation of ciprofloxacin by S. aureus with similar results being obtained when reserpine is substituted.⁸³ At concentrations of 12.5 and 25 mg/L, 37 caused a 2-fold reduction of the MIC of ciprofloxacin. This plant alkaloid was also able to reduce the MIC of ciprofloxacin against MRSA strain 15187 from >16 to 8 mg/L. A ciprofloxacin-resistant mutant strain of S. aureus also exhibiting an increased MIC against ethidium bromide was rendered susceptible by the addition of piperine. Reserpine exerted a similar effect at an equal concentration. Both of these compounds are substrates for NorA and therefore it is plausible that piperine acts as an inhibitor of this transporter.

In an evaluation of Kuwaiti plants for bacterial resistance-modifying activity, the extracts of Prosopis juliflora (Mimosaceae) were studied. This species is known to produce piperidine alkaloids such as julifloridine, juliflorine and juliprosine with some possessing a direct antibacterial activity.⁸⁴ The methanol extract was identified to possess resistance-modifying activity by causing a reduction in MIC of norfloxacin against S. aureus 1199B. Extensive bioassay-guided fractionation led to the isolation of the active constituent, 38 (Figure 9), and accurate mass determination indicated a molecular formula of C₄₀H₇₂N₃O₂. This compound possesses both lipophilic and hydrophilic properties.⁸⁵ The large size and lipophilicity of this piperidine alkaloid is a common feature of many known active potentiators against Gram-positive bacteria. From the HMBC spectra, 38 has a pyridine ring attached to a four-membered ring system that should be highly strained and unstable.⁸⁵ Mass spectrometry also indicated the presence of two piperidine rings, one with 13 methylenes attached to it and the other with 8 methylenes attached. The two alkylated piperidine ring systems and the 6:4 heterocyclic ring system provided the correct molecular formula. However, there is ambiguity as to the placement of the alkyl chains and only single crystal X-ray structural analysis will provide a definitive answer. The presence of four-membered rings possessing a quaternary nitrogen is very rare both in nature and as synthetic compounds and those described from the literature are brominated on the four-membered ring system.⁸⁶ 38 exhibited good potentiation activity against S. aureus SA-1199B (NorA), by inhibiting ethidium efflux with an IC₅₀ of 7 µM. This is slightly better than reserpine (IC₅₀ = 10 µM) but not as active as the synthetic compound GG918.⁸⁷ The authors did not take the possibility of quenching into account. 38 also exerted an antibacterial activity against a panel of MDR MRSA strains with MIC values of 4 mg/L.

View larger version (9K): [in this window] [in a new window] [Download PowerPoint slide]   Figure 9. Chemical structures of compounds 38–41.

 
Salicylic acid (39), a simple phenolic present in many plant species, has recently been shown to induce a reduction of both the antibiotic ciprofloxacin and MDR substrate ethidium bromide for S. aureus.⁸⁸ This is the reverse of what would be expected for a putative EPI. This has been demonstrated by accumulation studies of ciprofloxacin and ethidium bromide in the presence and absence of this agent. An ethidium efflux assay in the presence of salicylic acid also resulted in an increase in efflux of the MDR substrate ethidium as compared with the assay performed in its absence. Inactivation of NorA did not alter the ability of salicylate to induce increased ciprofloxacin and ethidium resistance. This indicates that NorA is not essential for salicylate-induced MDR for S. aureus.⁸⁸

	Microbial-derived EPIs

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
EPIs derived from microbial sources have been relatively scarce to date. The ability of microorganisms to produce antimicrobial compounds as part of their ‘chemical arsenal’ needs to be combated by susceptible microbes through the evolution of drug resistance. MDR pumps are an example, with an ability to extrude a number of chemically diverse antibiotics with the expression of just a single efflux mechanism. It would therefore seem logical that, as is the case with plants,⁶⁰ microorganisms would evolve to produce a second compound that could nullify the effect of MDR pumps in a competing microorganism resulting in the accumulation of the antimicrobial compound to a level that would be static or cidal.

Screening of microbial fermentations has resulted in the characterization of two new natural product EPIs.⁸⁹ The MDR inhibitors were isolated from Streptomyces MF-EA-371-NS1, which is a new strain closely related to Streptomyces vellosus.⁸⁹ EA-371 (40) and EA-371 (41) (Figure 9) both inhibited the MDR pump MexAB-OprM of P. aeruginosa PAM1032, which overexpresses this pump. At a concentration of 0.625 mg/L both compounds caused a 4-fold reduction in the MIC of levofloxacin. An 8-fold reduction of this fluoroquinolone was effected at 1.25 and 2.5 mg/L of EA-371 and EA-371, respectively. These compounds were not active against the triple pump deletion strain PAM1626.

	Plant extracts exhibiting potentiating activity

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
The identification of plants able to inhibit efflux pumps is important as they provide a potential for lead optimization and future use with an existing antibacterial rendered ineffective due to MDR pumps in both Gram-positive and Gram-negative bacteria.

Berberis aetnensis, an endemic plant on the volcano Mount Etna,⁹⁰ has been shown to exert a synergistic interaction in combination with ciprofloxacin. The modulation activity was located in the chloroform extract, which was fractioned further into yellow and green subfractions, obtained from the leaves of this plant. These lowered the MIC of ciprofloxacin for strains of S. aureus, E. coli and P. aeruginosa. Analytical thin-layer chromatography of the chloroform extract with an authentic sample of pheophorbide a indicated the presence of this compound.⁹⁰ The results were similar to those obtained when ciprofloxacin was added to commercial pheophorbide a at a concentration of 0.5 mg/L. The authors also hypothesized on the presence of 5'-MHC-D, which was isolated previously from Berberis species.³⁸ However, it is plausible that 5'-MHC-D was not present, but rather that an additional metabolite may provide a similar synergistic activity. It is also possible that there were further compounds in this leaf extract that possessed a potentiating activity when combined with a drug such as ciprofloxacin. An interesting finding was that the weaker dilutions (23.3 and 45 mg/L) of the chloroform extracts exhibited a far greater activity than the more concentrated extracts (233 and 450 mg/L). There is the possibility that the putative EPI or EPIs in this leaf extract actually binding to a high-affinity binding site of an efflux pump causing greater inhibition only at low concentration, but at a higher leaf extract concentration would result in low-affinity binding sites being occupied therefore causing a reduction in activity. Strains of S. aureus, E. coli and P. aeruginosa were 33-fold more susceptible to ciprofloxacin at these lower concentrations. It is possible that at high concentration the efflux inhibitors may form complexes with ciprofloxacin reducing the bioavailability of the fluoroquinolone causing a reduction in MIC.⁹¹,⁹² At lower extract concentrations there would be a greater concentration of ciprofloxacin and efflux inhibitor in the free state to exert their activities. A control to assess the possible interference caused by the solvent used to dissolve the plant extracts in the potentiation assay was not stated.

Extracts of Mezoneuron benthamianum (Caesalpinaceae) and Securinega virosa (Euphorbiaceae) exerted a potentiation activity against fluoroquinolone-, tetracycline- and erythromycin-resistant strains of S. aureus. The ethanol extract of M. benthamianum and chloroform extract of S. virosa reduced the MIC of norfloxacin against S. aureus 1199B by a factor of 4.⁹³ The petroleum spirit extract of M. benthamianum also caused inhibition of the same strain but to a lesser degree and also caused a 2-fold reduction in MIC of tetracycline (64 mg/L from 128 mg/L) against S. aureus XU212, containing the Tet(K) transporter.

The methanolic extract of Punica granatum (pomegranate) caused an increase in ethidium bromide uptake in S. aureus RN-7044, containing the pWBG32 plasmid encoding for an ethidium bromide efflux mechanism.⁹⁴ This extract exhibited synergic interactions with chloramphenicol, gentamicin, ampicillin, tetracycline and oxacillin against most of the 30 MRSA and MSSA clinical isolates tested.⁹⁴ The effect of methanol alone to assess possible interference in the potentiation assay was not stated by the authors.

Extracts of Commiphora molmol, Centella asiatica, Daucus carota, Citrus aurantium and Glycyrrhiza glabra showed good activity against three strains of S. enterica serovar Typhimurium that overexpress the AcrAB-TolC efflux protein (L. J. V. Piddock and S. Gibbons, unpublished data). This is the main efflux transporter located in the Enterobacteriaceae and is a homologue of the MexAB-OprM transporter, with the capability of extruding many structurally diverse chemicals including tetracyclines, fluoroquinolones and chloramphenicol. Some enhancing activity by the extracts was also detected with the wild-type strains, but not for those that lacked components of the AcrAB-TolC efflux pump when tested with nalidixic acid, chloramphenicol or tetracycline. However, significant reductions were recorded with each plant extract in combination with any of the three drugs tested against strains overexpressing AcrAB-TolC. The reduction in MIC of these drugs ranged from 4-fold up to 32-fold. No enhancing activity was demonstrated by the extracting solvent alone. These extracts are of interest as they represent an efflux transporter that is difficult to inhibit due to the greater resistance afforded to the Gram-negative cells due to the presence of the outer membrane.

	Conclusions

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
MDR due to the expression of efflux pumps is an increasing clinical problem, rendering many antibiotics redundant. Novel antibiotics with new modes of action are urgently required to suppress the rise of MDR bacteria. An alternative approach would be to identify molecules that can interfere with the process of efflux. Currently there are no EPI/antimicrobial drug combinations on the market, although research into identifying potential EPIs is ongoing both in academic institutions as well as the pharmaceutical industry.⁷⁹,⁹⁵ Success has already been achieved in mammalian cells with resistance modifying agents of P-glycoprotein, one of the major MDR mechanisms in cancer cells.⁹⁶ Identifying EPIs from natural sources is still in its infancy and the number of research groups seeking inhibitors is small. No natural products have been taken up for further development as much of the data acquired for putative EPIs are only preliminary, resulting from potentiation assays and accumulation and efflux studies. Biochemical studies to provide conclusive evidence of an EPI/protein interaction are required as well as toxicity and small in vivo studies to ascertain the pharmacokinetic and toxicity data for potential EPIs. Reserpine is an example of a natural product that has not been further developed; it is neurotoxic at concentrations required to inhibit the efflux pump NorA of S. aureus.⁹⁷ The cost and time required to identify potential EPIs is another problem.⁴⁰ However, the chemical diversity that plants and microorganisms provide, as well as their requirement to biosynthesize such compounds to combat competitors for nutrients, should make the search for EPIs from such sources an attractive option.

This review has highlighted a number of bacterial EPIs derived from natural sources, primarily from plants. The activities of some of these compounds are appreciable and warrant further study as possible candidates for lead optimization. It is our belief that plants should be further exploited for their potential to produce compounds capable of blocking the mechanism of efflux. There is an ecological rationale for the production of natural products that modify bacterial resistance. It has been speculated that plants have evolved compounds which evade MDR mechanisms and that plant antimicrobials might be developed into broad-spectrum antibiotics in combination with inhibitors of MDR.³⁷

Some of the compounds described exerted both an antibacterial and potentiating activity. An important issue when identifying a potential EPI is to ensure that the activity being displayed is due to the interference of efflux rather than any other antibacterial activity.

Resistance-modifying compounds active against Gram-positive bacteria tend to be large, lipophilic molecules. To our knowledge, there have been no natural EPIs with enhancing activities for mycobacteria. This can partly be attributed to the fact that the characterization of efflux pumps of this genus has been reported more recently than for Gram-positive or Gram-negative bacteria. There is an urgent need to identify new anti-tuberculosis (TB) compounds due to the high incidence levels of MDR-TB cases. MDR pumps can extrude both first- and second-line drugs such as isoniazid, ethambutol, fluoroquinolones and aminoglycosides.⁹⁸ EPIs could increase the lifespan of these drugs, especially as there have been no new antimycobacterial compounds with new modes of action for over 30 years.

The vast majority of EPIs described so far are active against Gram-positive bacteria, particularly S. aureus. It is perhaps the Gram-negative species of Pseudomonas, Escherichia and Acinetobacter that will be the most problematic bacteria to treat in the future. These organisms have an intrinsic resistance afforded to them by their thick, lipophilic outer membrane. This view has been strengthened by the dearth of MDR inhibitors described against Gram-negative bacteria in this paper.

	Transparency declarations

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
None to declare.

	Acknowledgements

 
We would like to thank Dr Mark Webber for data mining the Bacillus cereus genome for homologues of Bmr.

	References

Top Abstract Introduction Bacterial efflux families Screening for EPIs EPIs from plant sources Microbial-derived EPIs Plant extracts exhibiting... Conclusions Transparency declarations References

 
1 Perl TM. The threat of vancomycin resistance. Am J Med (1999) 106:26–37S.[CrossRef]

2 Rotun SS, McMath V, Schoonmaker DJ, et al. Staphylococcus aureus with reduced susceptibility to vancomycin isolated from a patient with fatal bacteremia. Emerg Infect Dis (1999) 5:147–9.[ISI][Medline]

3 Office for National Statistics. Deaths involving MRSA: England and Wales, 2000–2004. Health Stat Q (2006) 29:63–8.[Medline]

4 Blumberg HM, Rimland D, Carroll DJ, et al. Rapid development of ciprofloxacin resistance in methicillin-susceptible and -resistant Staphylococcus aureus. J Infect Dis (1991) 163:1279–85.[ISI][Medline]

5 Harnett N, Brown S, Krishnan C. Emergence of quinolone resistance among clinical isolates of methicillin-resistant Staphylococcus aureus in Ontario, Canada. Antimicrob Agents Chemother (1991) 35:1911–13.[Abstract/Free Full Text]

6 Coronado VG, Edwards JR, Culver DH, et al. Ciprofloxacin resistance among nosocomial Pseudomonas aeruginosa and Staphylococcus aureus in the United States. Infect Control Hosp Epidemiol (1995) 16:71–5.[ISI][Medline]

7 Public Health Dispatch. Vancomycin-resistant Staphylococcus aureus. Morbid Mortal Weekly Rep (2002) 51:902.[Medline]

8 CDC. Staphylococcus aureus resistant to vancomycin. Morbid Mortal Weekly Rep (2002) 51:565–7.[Medline]

9 Tsiodras S, Gold HS, Sakoulas G, et al. Linezolid resistance in a clinical isolate of Staphylococcus aureus. Lancet (2001) 358:207–8.[CrossRef][ISI][Medline]

10 Nikaido H. Prevention of drug access to bacterial targets: permeability barriers and active efflux. Science (1994) 264:382–8.[Abstract/Free Full Text]

11 Poole K. Mechanisms of bacterial biocide and antibiotic resistance. J Appl Microbiol (2002) 92:55–64S.[CrossRef][Medline]

12 Leclercq R, Courvalin P. Bacterial resistance to macrolide, lincosamide, and streptogramin antibiotics by target modification. Antimicrob Agents Chemother (1991) 35:1267–72.[Free Full Text]

13 Cetinkaya Y, Falk P, Mayhall CG. Vancomycin-resistant enterococci. Clin Microbiol Rev (2000) 13:686–707.[Abstract/Free Full Text]

14 Chambers HF. Methicillin resistance in staphylococci: molecular and biochemical basis and clinical implications. Clin Microbiol Rev (1997) 10:781–91.[Abstract]

15 Hiramatsu K, Cui L, Kuroda M, et al. The emergence and evolution of methicillin-resistant Staphylococcus aureus. Trends Microbiol (2001) 9:486–93.[CrossRef][ISI][Medline]

16 Hooper DC. Mechanisms of action and resistance of older and newer fluoroquinolones. Clin Infect Dis (2000) 31:S24–8.[CrossRef][ISI][Medline]

17 Wright GD. Aminoglycoside-modifying enzymes. Curr Opin Microbiol (1999) 2:499–503.[CrossRef][ISI][Medline]

18 Thompson KS, Smith ME. Version 2000: the new ß-lactamases of Gram-negative bacteria at the dawn of the new millenium. Microbes Infect (2000) 2:1225–35.[CrossRef][ISI][Medline]

19 Wright GD, Berghuis AM, Mobashery S. Aminoglycoside antibiotics. Structures, functions, and resistance. Adv Exp Med Biol (1998) 456:27–69.[ISI][Medline]

20 Murray IA, Shaw WV. O-Acetyltransferases for chloramphenicol and other natural products. Antimicrob Agents Chemother (1997) 41:1–6.[ISI][Medline]

21 Ball PR, Shales SW, Chopra I. Plasmid-mediated tetracycline resistance in Escherichia coli involves increased efflux of the antibiotic. Biochem Biophys Res Commun (1980) 93:74–81.[CrossRef][ISI][Medline]

22 McMurry LM, Petrucci RE Jr, Levy SB. Active efflux of tetracycline encoded by four genetically different tetracycline resistance determinants in Escherichia coli. Proc Natl Acad Sci USA (1980) 77:3974–7.[Abstract/Free Full Text]

23 Chopra I, Roberts MC. Tetracycline antibiotics: mode of action, applications, molecular biology and epidemiology of bacterial resistance. Microbiol Mol Biol Rev (2001) 65:232–60.[Abstract/Free Full Text]

24 Roberts MC. Update on acquired tetracycline resistance genes. FEMS Microbiol Lett (2005) 245:195–203.[CrossRef][ISI][Medline]

25 Piddock LJV. Clinically relevant bacterial chromosomally encoded multi-drug resistance efflux pumps. Clinl Microbiol Rev (2006) 19:382–402.[CrossRef]

26 Li XZ, Poole K, Nikaido H. Contributions of MexAB-OprM and an EmrE homolog to intrinsic resistance of Pseudomonas aeruginosa to aminoglycosides and dyes. Antimicrob Agents Chemother (2003) 47:27–33.[Abstract/Free Full Text]

27 Aires JR, Kohler T, Nikaido H, et al. Involvement of an active efflux system in the natural resistance of Pseudomonas aeruginosa to aminoglycosides. Antimicrob Agents Chemother (1999) 43:2624–8.[Abstract/Free Full Text]

28 Poole K. Efflux mediated antimicrobial resistance. J Antimicrob Chemother (2005) 56:20–51.[Abstract/Free Full Text]

29 Arcangioli MA, Leroy-Setrin S, Martel JL, et al. A new chloramphenicol and florfenicol resistance gene flanked by two integron structures in Salmonella typhimurium DT104. FEMS Microbiol Lett (1999) 174:327–32.[CrossRef][ISI][Medline]

30 Lomovskaya O, Lee A, Hoshino K, et al. Use of a genetic approach to evaluate the consequences of inhibition of efflux pumps in Pseudomonas aeruginosa. Antimicrob Agents Chemother (1999) 43:1340–6.[Abstract/Free Full Text]

31 Esposito S, Noviello S. In-vitro sensitivity of amoxicillin/clavulanic acid (Augmentin) to isolated ß-lactamases and in-vitro antibacterial activity. J Chemother (1990) 2:167–70.[ISI][Medline]

32 Lorian V. Antibiotics in Laboratory Medicine (1996) Baltimore: Williams and Wilkins.

33 Gibbons S, Oluwatuyi M, Veitch N, et al. Bacterial resistance modifying agents from Lycopus europaeus. Phytochemistry (2003) 62:83–7.[CrossRef][ISI][Medline]

34 Oluwatuyi M, Kaatz GW, Gibbons S. Antibacterial and resistance modifying activity of Rosmarinus officinalis. Phytochemistry (2004) 65:3249–54.[CrossRef][ISI][Medline]

35 Belofsky G, Carreno R, Lewis K, et al. Metabolites of the ‘smoke tree’, Dalea spinosa, potentiate antibiotic activity against multidrug-resistant Staphylococcus aureus. J Nat Prod (2006) 69:261–4.[CrossRef][Medline]

36 Marchetti O, Moreillon P, Glauser MP, et al. Potent synergism of the combination of fluconazole and cyclosporine in Candida albicans. Antimicrob Agents Chemother (2000) 44:2373–81.[Abstract/Free Full Text]

37 Tegos G, Stermitz FR, Lomovskaya O, et al. Multidrug pump inhibitors uncover remarkable activity of plant antimicrobials. Antimicrob Agents Chemother (2002) 46:3133–41.[Abstract/Free Full Text]

38 Stermitz FR, Tawara-Matsuda J, Lorenz P, et al. 5'-Methoxyhydnocarpin-D and pheophorbide A: Berberis species components that potentiate berberine growth inhibition of resistant Staphylococcus aureus. J Nat Prod (2000) 63:1146–9.[CrossRef][Medline]

39 Kaatz GW, Seo SM, O'Brien L, et al. Evidence for the existence of a multidrug efflux transporter distinct from NorA in Staphylococcus aureus. Antimicrob Agents Chemother (2000) 44:1404–6.[Abstract/Free Full Text]

40 Kaatz GW, Moudgal VV, Seo SM, et al. Phenothiazines and thioxanthenes inhibit multidrug efflux pump activity in Staphylococcus aureus. Antimicrob Agents Chemother (2003) 47:719–26.[Abstract/Free Full Text]

41 Kaatz GW, Moudgal VV, Seo SM. Identification and characterization of a novel efflux-related multidrug resistance phenotype in Staphylococcus aureus. J Antimicrob Chemother (2002) 50:833–8.[Abstract/Free Full Text]

42 Kaatz GW, Seo SM, Ruble CA. Efflux-mediated fluoroquinolone resistance in Staphylococcus aureus. Antimicrob Agents Chemother (1993) 37:1086–94.[Abstract/Free Full Text]

43 Samorsorn S, Bremner JB, Ball A, et al. Synthesis of functionalised 2-aryl-5-nitro-1H-indoles and their activity as bacterial NorA efflux pump inhibitors. Bioorg Med Chem Lett (2006) 14:857–65.[CrossRef]

44 Baranova N, Nikaido H. The baeSR two-component regulatory system activates transcription of the yegMNOB (mdtABCD) transporter gene cluster in Escherichia coli and increases its resistance to novobiocin and deoxycholate. J Bacteriol (2002) 184:4168–76.[Abstract/Free Full Text]

45