Molecular mechanisms of antimicrobial tolerance and resistance in bacterial and fungal biofilms

•Protection against oxidative stress provides microbial biofilms with tolerance to cidal drugs.•In bacteria, the antibiotic-induced expression of efflux systems is linked to biofilm resistance.•The biofilm matrix and cell envelope-associated polysaccharides contribute to tolerance.•There is evidence that suggests these mechanisms are not independent. The formation of microbial biofilms is an important reason for failure of antimicrobial therapy. However, the molecular mechanisms underlying the survival of biofilm cells are still not completely understood. In this review we discuss three mechanisms that play an important role in biofilm survival: (i) biofilm-specific protection against oxidative stress; (ii) biofilm-specific expression of efflux pumps; and (iii) protection provided by matrix polysaccharides. We demonstrate that these mechanisms are found both in bacterial and fungal biofilms and are often surprisingly similar between distantly related organisms. In addition, we give an overview of the data that suggests that these mechanisms may not be independent. The formation of microbial biofilms is an important reason for failure of antimicrobial therapy. However, the molecular mechanisms underlying the survival of biofilm cells are still not completely understood. In this review we discuss three mechanisms that play an important role in biofilm survival: (i) biofilm-specific protection against oxidative stress; (ii) biofilm-specific expression of efflux pumps; and (iii) protection provided by matrix polysaccharides. We demonstrate that these mechanisms are found both in bacterial and fungal biofilms and are often surprisingly similar between distantly related organisms. In addition, we give an overview of the data that suggests that these mechanisms may not be independent. Biofilms are defined as consortia of microorganisms that are attached to a biotic or abiotic surface. Biofilm formation is a multistage process in which microbial cells adhere to the surface (initial reversible attachment), whereas the subsequent production of an extracellular matrix (containing polysaccharides, proteins, and DNA) results in a firmer attachment [1Stoodley P. et al.Biofilms as complex differentiated communities.Annu. Rev. Microbiol. 2002; 56: 187-209Crossref PubMed Scopus (2282) Google Scholar, 2Bjarnsholt T. et al.The in vivo biofilm.Trends Microbiol. 2013; 21: 466-474Abstract Full Text Full Text PDF PubMed Scopus (496) Google Scholar]. Cells in a biofilm (sessile cells) are phenotypically and physiologically different from non-adhered (planktonic) cells and one of the typical properties of cells in a mature biofilm is that much higher concentrations of antimicrobial drugs are required to kill sessile cells compared to planktonic cells [3Stewart P.S. Costerton J.W. Antibiotic resistance of bacteria in biofilms.Lancet. 2001; 358: 135-138Abstract Full Text Full Text PDF PubMed Scopus (3484) Google Scholar]. Biofilm formation is often considered to be the underlying reason as to why treatment with an antimicrobial agent fails and, because an estimated 65–80% of all infections are thought to be biofilm-related, this presents a serious challenge [4Hall-Stoodley L. et al.Bacterial biofilms: from the natural environment to infectious diseases.Nat. Rev. Microbiol. 2004; 2: 95-108Crossref PubMed Scopus (5091) Google Scholar]. Several mechanisms are thought to be involved in biofilm tolerance and resistance, including slow penetration of the antimicrobial agent through the biofilm, changes in the chemical microenvironment within the biofilm (leading to zones of slow or no growth), adaptive stress responses, and presence of a small population of extremely tolerant ‘persister’ cells [3Stewart P.S. Costerton J.W. Antibiotic resistance of bacteria in biofilms.Lancet. 2001; 358: 135-138Abstract Full Text Full Text PDF PubMed Scopus (3484) Google Scholar]. These persisters can tolerate certain antimicrobial agents (i.e., they are not killed) and can be considered as specialized survivor cells [5Lewis K. Persister cells.Annu. Rev. Microbiol. 2010; 64: 357-372Crossref PubMed Scopus (1431) Google Scholar]. In addition, rates of horizontal gene transfer are typically higher in biofilms than in planktonic cultures [6Madsen J.S. et al.The interconnection between biofilm formation and horizontal gene transfer.FEMS Immunol. Med. Microbiol. 2012; 65: 183-195Crossref PubMed Scopus (403) Google Scholar], and increased transfer of antibiotic resistance determinants on mobile genetic elements was noted in biofilms of various organisms, including Staphylococcus aureus [7Savage V.J. et al.Staphylococcus aureus biofilms promote horizontal transfer of antibiotic resistance.Antimicrob. Agents Chemother. 2013; 57: 1968-1970Crossref PubMed Scopus (234) Google Scholar]. Although there are several excellent reviews on antimicrobial resistance in biofilms [2Bjarnsholt T. et al.The in vivo biofilm.Trends Microbiol. 2013; 21: 466-474Abstract Full Text Full Text PDF PubMed Scopus (496) Google Scholar, 3Stewart P.S. Costerton J.W. Antibiotic resistance of bacteria in biofilms.Lancet. 2001; 358: 135-138Abstract Full Text Full Text PDF PubMed Scopus (3484) Google Scholar, 4Hall-Stoodley L. et al.Bacterial biofilms: from the natural environment to infectious diseases.Nat. Rev. Microbiol. 2004; 2: 95-108Crossref PubMed Scopus (5091) Google Scholar, 5Lewis K. Persister cells.Annu. Rev. Microbiol. 2010; 64: 357-372Crossref PubMed Scopus (1431) Google Scholar, 8Soto S.M. Role of efflux pumps in the antibiotic resistance of bacteria embedded in a biofilm.Virulence. 2013; 4: 223-229Crossref PubMed Scopus (276) Google Scholar, 9Mah T.F. Biofilm-specific antibiotic resistance.Future Microbiol. 2012; 7: 1061-1072Crossref PubMed Scopus (299) Google Scholar, 10Coenye T. Response of sessile cells to stress: from changes in gene expression to phenotypic adaptation.FEMS Immunol. Med. Microbiol. 2010; 59: 239-252PubMed Google Scholar, 11Fux C.A. et al.Survival strategies of infectious biofilms.Trends Microbiol. 2005; 13: 34-40Abstract Full Text Full Text PDF PubMed Scopus (1423) Google Scholar], most of these focus on a specific organism, a specific mechanism, or both. In addition, more often than not it is unclear whether the mechanisms discussed are more important in biofilms than in planktonic cells. In this review we focus on mechanisms that are found both in bacterial and fungal biofilms, that is, protection against drug-induced oxidative stress, drug efflux, and shielding by the biofilm matrix. We will highlight the research data that indicate that these mechanisms are important in biofilms and discuss the potential relation between the different mechanisms. With a series of elegant experiments, James Collins and coworkers [12Dwyer D.J. et al.Gyrase inhibitors induce an oxidative damage cellular death pathway in Escherichia coli.Mol. Syst. Biol. 2007; 3: 91Crossref PubMed Scopus (347) Google Scholar, 13Dwyer D.J. et al.Role of reactive oxygen species in antibiotic action and resistance.Curr. Opin. Microbiol. 2009; 12: 482-489Crossref PubMed Scopus (335) Google Scholar, 14Kohanski M.A. et al.A common mechanism of cellular death induced by bactericidal antibiotics.Cell. 2007; 130: 797-810Abstract Full Text Full Text PDF PubMed Scopus (2036) Google Scholar, 15Kohanski M.A. et al.Mistranslation of membrane proteins and two-component system activation trigger antibiotic-mediated cell death.Cell. 2008; 135: 679-690Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar] demonstrated that bactericidal antibiotics (including β-lactams, fluoroquinolones, and aminoglycosides) induce hydroxyl radical formation in Escherichia coli and S. aureus, indicating that the biochemical changes that are the consequence of treatment with lethal doses of these bactericidal antibiotics create an intracellular environment that promotes formation of highly deleterious reactive oxygen species (ROS). The current working model is that the primary drug–target interaction stimulates the oxidation of NADH via the electron transport chain [13Dwyer D.J. et al.Role of reactive oxygen species in antibiotic action and resistance.Curr. Opin. Microbiol. 2009; 12: 482-489Crossref PubMed Scopus (335) Google Scholar, 14Kohanski M.A. et al.A common mechanism of cellular death induced by bactericidal antibiotics.Cell. 2007; 130: 797-810Abstract Full Text Full Text PDF PubMed Scopus (2036) Google Scholar]. Hyperactivation of this electron transport chain, which depends on the tricarboxylic acid (TCA) cycle, stimulates superoxide (O2•−) production. This O2•− damages [Fe–S] clusters in proteins, making Fe2+ available for oxidation to Fe3+ in the Fenton reaction (Fe2+ + H2O2 → Fe3+ + HO• + OH−). This results in the production of the highly reactive hydroxyl radical (HO•) that can damage cellular macromolecules, leading to cell death [13Dwyer D.J. et al.Role of reactive oxygen species in antibiotic action and resistance.Curr. Opin. Microbiol. 2009; 12: 482-489Crossref PubMed Scopus (335) Google Scholar, 14Kohanski M.A. et al.A common mechanism of cellular death induced by bactericidal antibiotics.Cell. 2007; 130: 797-810Abstract Full Text Full Text PDF PubMed Scopus (2036) Google Scholar] (Figure 1). Although recent studies question the link between bactericidal antibiotics and ROS [16Keren I. et al.Killing by bactericidal antibiotics does not depend on reactive oxygen species.Science. 2013; 339: 1213-1216Crossref PubMed Scopus (395) Google Scholar, 17Liu Y. Imlay J.A. Cell death from antibiotics without the involvement of reactive oxygen species.Science. 2013; 339: 1210-1213Crossref PubMed Scopus (429) Google Scholar], it may very well be that differences in experimental procedures are at the basis of these divergent conclusions and that ROS induction only contributes to bacterial killing under certain experimental conditions [18Fang F.C. Antibiotic and ROS linkage questioned.Nat. Biotechnol. 2013; 31: 415-416Crossref PubMed Scopus (35) Google Scholar]. The above-mentioned studies were all carried out with planktonic bacteria, and few studies have addressed this issue specifically in the context of biofilms. In 2004, Battán et al. showed that higher concentrations of the bactericidal antibiotics ceftazidime or piperacillin were required to induce ROS production in Pseudomonas aeruginosa biofilms compared to planktonic cells [19Battán P.C. et al.Resistance to oxidative stress caused by ceftazidime and piperacillin in a biofilm of Pseudomonas.Luminescence. 2004; 19: 265-270Crossref PubMed Scopus (21) Google Scholar]. Later, Aiassa et al. showed that ciprofloxacin did not stimulate ROS production in Proteus mirabilis biofilms, in contrast to what happens in planktonic cultures, and demonstrated that superoxide dismutase (SOD) and glutathione levels were increased in biofilms [20Aiassa V. et al.Resistance to ciprofloxacin by enhancement of antioxidant defenses in biofilm and planktonic Proteus mirabilis.Biochem. Biophys. Res. Commun. 2010; 393: 84-88Crossref PubMed Scopus (44) Google Scholar]. In line with this initial observation, in a follow-up study it was found that sessile cells of P. mirabilis already have higher levels of oxidized lipids and proteins compared to planktonic cells, and that treatment of these sessile cells with ciprofloxacin did not lead to increased oxidation [21Aiassa V. et al.Macromolecular oxidation in planktonic population and biofilms of Proteus mirabilis exposed to ciprofloxacin.Cell Biochem. Biophys. 2014; 68: 49-54Crossref PubMed Scopus (4) Google Scholar]. A recent study demonstrated that ciprofloxacin does induce ROS production in P. aeruginosa biofilms but the data obtained suggest that HO• production is prevented in areas of the biofilm with low metabolic activity and/or anaerobic conditions, potentially explaining the increased ciprofloxacin tolerance of older biofilms [22Jensen P.O. et al.Formation of hydroxyl radicals contributes to the bactericidal activity of ciprofloxacin against Pseudomonas aeruginosa biofilms.Pathog. Dis. 2014; https://doi.org/10.1111/2049-632X.12120Crossref PubMed Scopus (65) Google Scholar]. This would be in line with the observation that nutrient or oxygen deprivation reduces HO• production and simultaneously reduces ciprofloxacin activity in planktonic P. aeruginosa cultures [23Brochmann R.P. et al.Bactericidal effect of colistin on planktonic Pseudomonas aeruginosa is independent of hydroxyl radical formation.Int. J. Antimicrob. Agents. 2014; 43: 140-147Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar]. In P. aeruginosa and E. coli biofilms, the stringent response (SR, a regulatory mechanism involved in response to nutrient limitation) is involved in keeping ROS levels low [24Nguyen D. et al.Active starvation responses mediate antibiotic tolerance in biofilms and nutrient-limited bacteria.Science. 2011; 334: 982-986Crossref PubMed Scopus (678) Google Scholar, 25Khakimova M. et al.The stringent response controls catalases in Pseudomonas aeruginosa and is required for hydrogen peroxide and antibiotic tolerance.J. Bacteriol. 2013; 195: 2011-2020Crossref PubMed Scopus (124) Google Scholar]. The SR protects P. aeruginosa against antibiotic-induced oxidative stress by maintaining adequate levels of catalase and SOD activity and by limiting the production of 4-hydroxy-2-alkylquinolines (HAQs), known to have pro-oxidant effects [24Nguyen D. et al.Active starvation responses mediate antibiotic tolerance in biofilms and nutrient-limited bacteria.Science. 2011; 334: 982-986Crossref PubMed Scopus (678) Google Scholar, 25Khakimova M. et al.The stringent response controls catalases in Pseudomonas aeruginosa and is required for hydrogen peroxide and antibiotic tolerance.J. Bacteriol. 2013; 195: 2011-2020Crossref PubMed Scopus (124) Google Scholar]. In E. coli, which does not produce HAQs, inactivation of the SR also leads to a decrease of sessile cells surviving treatment with antibiotics, raising the possibility that control of antibiotic-induced oxidant stress by the SR is common in bacteria [24Nguyen D. et al.Active starvation responses mediate antibiotic tolerance in biofilms and nutrient-limited bacteria.Science. 2011; 334: 982-986Crossref PubMed Scopus (678) Google Scholar]. Data obtained with tobramycin-treated Burkholderia cenocepacia biofilms confirmed that avoiding exposure to antibiotic-induced ROS is a key factor in survival of persisters in these biofilms [26Van Acker H. et al.Biofilm-grown Burkholderia cepacia complex cells survive antibiotic treatment by avoiding production of reactive oxygen species.PLoS ONE. 2013; 8: e58943Crossref PubMed Scopus (103) Google Scholar]. When B. cenocepacia biofilms were treated with high concentrations of the aminoglycoside antibiotic tobramycin, ROS production was increased but approximately 0.1% of the cells survived the treatment. Comparing gene expression profiles of these surviving cells with that of untreated sessile cells revealed that several genes from the TCA cycle and genes involved in the electron transport chain were downregulated, while genes from the glyoxylate shunt were upregulated. This suggests that surviving persister cells downregulate the TCA cycle to avoid production of ROS and at the same time activate an alternative pathway, the glyoxylate shunt (Figure 1). The importance of protection against ROS was further confirmed by the low numbers of persisters in two catalase mutants, the synergistic effect of tobramycin and the SOD inhibitor N, N′-diethyldithiocarbamate (DDTC), the marked downregulation of a ferredoxin reductase (involved in recycling Fe3+ to Fe2+ and thus driving the Fenton reaction) in surviving cells, and the upregulation of various genes encoding proteins involved in ROS detoxification (including several catalases and alkyl hydroperoxidases) in surviving cells. Although it is at present unclear whether this is a biofilm-specific mechanism, the levels of persister cell surviving treatment with high levels of tobramycin or ciprofloxacin was up to 1000-fold higher in biofilms than in planktonic cultures, suggesting that the protective effect is more pronounced in biofilms. The potential role of ROS in mediating the effect of antifungal drugs is not new. In 1986, Sokol-Anderson et al. demonstrated that the fungicidal effect of amphotericin B (AmB) is co-mediated by the induction of oxidative damage [27Sokol-Anderson M.L. et al.Amphotericin B-induced oxidative damage and killing of Candida albicans.J. Infect. Dis. 1986; 154: 76-83Crossref PubMed Scopus (226) Google Scholar]. Later it was shown that the production of endogenous ROS is also increased in planktonic Candida albicans cells treated with miconazole [28Kobayashi D. et al.Endogenous reactive oxygen species is an important mediator of miconazole antifungal effect.Antimicrob. Agents Chemother. 2002; 46: 3113-3117Crossref PubMed Scopus (265) Google Scholar]. It has been suggested that the increase in endogenous ROS levels correlates with the antifungal activity: although miconazole and nystatin (as well as AmB) are fungicidal drugs that increase ROS levels, this is not the case for the fungistatic drugs clotrimazole and fluconazole [29François I.E. et al.Azoles: mode of antifungal action and resistance development. Effect of miconazole on endogenous reactive oxygen species production in Candida albicans.Curr. Med. Chem. 2006; 5: 1-11Google Scholar]. Using a systems biology approach, Collins’ group identified an oxidative damage cellular death pathway that is involved in mediating fungicidal activity of AmB, miconazole, and ciclopirox against C. albicans and Saccharomyces cerevisiae [30Belenky P. et al.Fungicidal drugs induce a common oxidative-damage cellular death pathway.Cell Rep. 2013; 3: 350-358Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar]. Cellular changes induced by the interaction of antifungals with their primary target are sensed by the RAS/protein kinase A (PKA) signaling pathway. This signaling cascade induces mitochondrial activity (induction of the TCA cycle and respiratory activity), leading to increased production of ROS and cell death [30Belenky P. et al.Fungicidal drugs induce a common oxidative-damage cellular death pathway.Cell Rep. 2013; 3: 350-358Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar]. However, treatment of C. albicans biofilms with ROS-inducing antifungals such as miconazole [31Vandenbosch D. et al.Fungicidal activity of miconazole against Candida spp. biofilms.J. Antimicrob. Chemother. 2010; 65: 694-700Crossref PubMed Scopus (88) Google Scholar, 32Bink A. et al.Superoxide dismutases are involved in Candida albicans biofilm persistence against miconazole.Antimicrob. Agents Chemother. 2011; 55: 4033-4037Crossref PubMed Scopus (95) Google Scholar] or AmB [33De Brucker K. et al.Potentiation of antibiofilm activity of amphotericin B by superoxide dismutase inhibition.Oxid. Med. Cell. Longev. 2013; 2013: 704654Crossref PubMed Scopus (20) Google Scholar] does not lead to complete killing of all sessile cells, leaving a population of highly tolerant persister cells unaffected. Unlike in many bacteria, C. albicans persisters are only found in biofilms and are not present in planktonic cultures [34LaFleur M.D. et al.Candida albicans biofilms produce antifungal-tolerant persister cells.Antimicrob. Agents Chemother. 2006; 50: 3839-3846Crossref PubMed Scopus (384) Google Scholar]. The mechanism by which fungal persister cells survive treatment with these fungicidal drugs is still unresolved, but recent research points to the involvement of natural defenses against ROS. Indeed, it was shown that inhibiting SOD with DDTC drastically increased ROS levels and led to a concomitant 18- to 200-fold reduction in the number of C. albicans biofilm cells surviving treatment with miconazole [32Bink A. et al.Superoxide dismutases are involved in Candida albicans biofilm persistence against miconazole.Antimicrob. Agents Chemother. 2011; 55: 4033-4037Crossref PubMed Scopus (95) Google Scholar]. The expression levels of most SOD genes were increased upon treatment with miconazole and in a Δsod4 Δsod5 double mutant increased ROS levels and a lower fraction of persister cells were found, confirming the role of SOD in protection of C. albicans biofilms against drug-induced, ROS-mediated, cell death. Similarly, combining AmB with dicyclohexylcarbodiimide or the alternative SOD inhibitor ammonium tetrathiomolybdate increased AmB activity against C. albicans biofilm cells, although this effect was not biofilm specific [33De Brucker K. et al.Potentiation of antibiofilm activity of amphotericin B by superoxide dismutase inhibition.Oxid. Med. Cell. Longev. 2013; 2013: 704654Crossref PubMed Scopus (20) Google Scholar]. Although much remains to be discovered concerning the role of ROS in the activity of cidal antimicrobial agents, the examples discussed indicate that ROS production is involved in killing of bacteria and fungi under certain conditions. Protection against oxidative stress decreases the activity of bactericidal and fungicidal drugs and the mechanisms that provide this protection are important for survival of treated biofilms. The involvement of resistance-nodulation-division (RND) efflux pumps in antimicrobial resistance in Gram-negative bacteria is well established and is obviously not limited to biofilms [35Kumar A. Schweizer H.P. Bacterial resistance to antibiotics: active efflux and reduced uptake.Adv. Drug Deliv. Rev. 2005; 57: 1486-1513Crossref PubMed Scopus (362) Google Scholar, 36Fernández L. Hancock R.E. Adaptive and mutational resistance: role of porins and efflux pumps in drug resistance.Clin. Microbiol. Rev. 2012; 25: 661-681Crossref PubMed Scopus (528) Google Scholar]. P. aeruginosa contains at least 12 RND pump-encoding operons that are involved in the efflux of a wide range of substrates, including aminoglycosides, β-lactams, cephalosporins, chloramphenicol, fluoroquinolones, macrolides, tetracyclines, and trimethoprim [35Kumar A. Schweizer H.P. Bacterial resistance to antibiotics: active efflux and reduced uptake.Adv. Drug Deliv. Rev. 2005; 57: 1486-1513Crossref PubMed Scopus (362) Google Scholar, 36Fernández L. Hancock R.E. Adaptive and mutational resistance: role of porins and efflux pumps in drug resistance.Clin. Microbiol. Rev. 2012; 25: 661-681Crossref PubMed Scopus (528) Google Scholar]. However, surprisingly, in the first study to evaluate the contribution of four characterized efflux systems in P. aeruginosa to biofilm survival, this contribution was found to be minimal [37De Kievit T.R. et al.Multidrug efflux pumps: expression patterns and contribution to antibiotic resistance in Pseudomonas aeruginosa biofilms.Antimicrob. Agents Chemother. 2001; 45: 1761-1770Crossref PubMed Scopus (228) Google Scholar]. Indeed, although deletion or overexpression of the mexAB–oprM efflux pump resulted in increased and decreased susceptibility to various antibiotics, respectively, this was not specific for biofilm cells and the fraction of cells surviving antibiotic treatment in the deletion mutant was still very high compared to planktonic cells. Experiments in which other efflux systems (mexCD–oprJ, mexEF–oprN, and mexXY) were either overexpressed or deleted demonstrated that these pumps did not contribute to biofilm survival at all. However, at the same time it was observed that expression of mexAB–oprM and mexCD–oprJ was heterogeneous, with the highest levels observed in cells closest to the substratum. This suggests that (some) efflux pumps may be important for the survival of particular subpopulations in the biofilm. While looking for potential biofilm-specific defense mechanisms against a challenge with azithromycin, Gillis et al. observed that mexAB–oprM and mexCD–oprJ are essential for P. aeruginosa biofilm formation in the presence of this macrolide antibiotic because a mutant in which both systems are inactivated is not capable of forming biofilms (in contrast to mutants in which a single efflux pump is inactivated) [38Gillis R.J. et al.Molecular basis of azithromycin-resistant Pseudomonas aeruginosa biofilms.Antimicrob. Agents Chemother. 2005; 49: 3858-3867Crossref PubMed Scopus (148) Google Scholar]. Interestingly, mexC expression was limited to sessile cells exposed to azithromycin, and was not observed in planktonic cells and non-induced sessile cells, whereas mexA was constitutively expressed in planktonic cells, and in exposed and unexposed sessile cells. This strongly suggests that the MexCD–OprJ pump is a biofilm-specific defense mechanism against azithromycin in P. aeruginosa [38Gillis R.J. et al.Molecular basis of azithromycin-resistant Pseudomonas aeruginosa biofilms.Antimicrob. Agents Chemother. 2005; 49: 3858-3867Crossref PubMed Scopus (148) Google Scholar]. Similarly, while investigating tolerance to colistin, Pamp et al. observed that a P. aeruginosa mexAB–oprM mutant exhibited a significant decrease in tolerance to colistin when grown in a biofilm, whereas inherent resistance as measured by determining the minimal inhibitory concentration (MIC) for planktonic cells was unaltered, suggesting that the MexAB–OprM pump is a biofilm-specific defense mechanism against colistin in P. aeruginosa [39Pamp S.J. et al.Tolerance to the antimicrobial peptide colistin in Pseudomonas aeruginosa biofilms is linked to metabolically active cells, and depends on the pmr and mexAB-oprM genes.Mol. Microbiol. 2008; 68: 223-240Crossref PubMed Scopus (362) Google Scholar]. Nevertheless, it was later shown that colistin induced expression of a PmexA–gfp transcriptional fusion reporter in planktonic cells as well [40Chiang W.C. et al.The metabolically active subpopulation in Pseudomonas aeruginosa biofilms survives exposure to membrane-targeting antimicrobials via distinct molecular mechanisms.FEMS Immunol. Med. Microbiol. 2012; 65: 245-256Crossref PubMed Scopus (50) Google Scholar]. Whereas the above-mentioned studies focused on the known RND transporters of P. aeruginosa, screening of approximately 4000 P. aeruginosa transposon mutants led to the identification of a novel efflux system potentially involved in biofilm-specific resistance to antibiotics [41Zhang L. Mah T.F. Involvement of a novel efflux system in biofilm-specific resistance to antibiotics.J. Bacteriol. 2008; 190: 4447-4452Crossref PubMed Scopus (261) Google Scholar]. Deletion of (parts of) the PA1874–1877 operon did not affect the susceptibility of planktonic cells to tobramycin but led to a fourfold decrease in minimal bactericidal concentration for biofilms (MBC-B), as well as in reduced MBC-B values for gentamicin and ciprofloxacin. In line with these observations, the expression of PA1874 was approximately tenfold higher in biofilms than in planktonic cells, and overexpression of PA1875–1877 in planktonic cells decreased susceptibility to tobramycin, gentamicin, and ciprofloxacin. However, the overexpression of particular genes as a means to confer antimicrobial resistance comes at a cost. Indeed, Mulet et al. demonstrated that overexpression of mexCD–oprJ (driven by inactivation of the negative regulator NfxB) leads to reduced expression of OprM and thus a reduced number of functional MexAB–OprM and MexXY–OprM pumps [42Mulet X. et al.Antagonistic interactions of Pseudomonas aeruginosa antibiotic resistance mechanisms in planktonic but not biofilm growth.Antimicrob. Agents Chemother. 2011; 55: 4560-4568Crossref PubMed Scopus (52) Google Scholar]. In addition, accumulation of MexCD–OprJ leads to an altered outer membrane physiology, which results in leakage of the imipenem-inducible AmpC β-lactamase from the periplasm [42Mulet X. et al.Antagonistic interactions of Pseudomonas aeruginosa antibiotic resistance mechanisms in planktonic but not biofilm growth.Antimicrob. Agents Chemother. 2011; 55: 4560-4568Crossref PubMed Scopus (52) Google Scholar]. As expected, this leads to drastically reduced MIC values for imipenem but contrary to expectations, not to imipenem hypersusceptibility in biofilms. This can be explained by the accumulation of secreted β-lactamase in the biofilm matrix, as has been observed in previous studies (see for example [43Ciofu O. et al.Chromosomal beta-lactamase is packaged into membrane vesicles and secreted from Pseudomonas aeruginosa.J. Antimicrob. Chemother. 2000; 45: 9-13Crossref PubMed Google Scholar, 44Bagge N. et al.Dynamics and spatial distribution of beta-lactamase expression in Pseudomonas aeruginosa biofilms.Antimicrob. Agents Chemother. 2004; 48: 1168-1174Crossref PubMed Scopus (139) Google Scholar]). Moreover, imipenem treatment of P. aeruginosa biofilms actually increases alginate production, leading to increased biofilm thickness and biomass and increased potential of β-lactamase accumulation in the biofilm matrix [45Bagge N. et al.Pseudomonas aeruginosa biofilms exposed to imipenem exhibit changes in global gene expression and beta-lactamase and alginate production.Antimicrob. Agents Chemother. 2004; 48: 1175-1187Crossref PubMed Scopus (254) Google Scholar]. In B. cenocepacia, RND pumps also play an important role in efflux. The B. cenocepacia J2315 genome encodes 22 RND efflux systems that contribute to resistance to disinfectants and antibiotics [46Bazzini S. et al.Deciphering the role of RND efflux transporters in Burkholderia cenocepacia.PLoS ONE. 2011; 6: e18902Crossref PubMed Scopus (58) Google Scholar, 47Guglierame P. et al.Efflux pump genes of the resistance-nodulation-division family in Burkholderia cenocepacia genome.BMC Microbiol. 2006; 6: 66Crossref PubMed Scopus (77) Google Scholar, 48Buroni S. et al.Assessment of three Resistance-Nodulation-Cell Division drug efflux transporters of Burkholderia cenocepacia in intrinsic antibiotic resistance.BMC Microbiol. 2009; 9: 200Crossref PubMed Scopus (64) Google Scholar, 49Rushton L. et al.Key role for