SGM Journals
PATHOGENS AND PATHOGENICITY

The Lon protease of Pseudomonas aeruginosa is induced by aminoglycosides and is involved in biofilm formation and motility

Alexandra K. Marr, Joerg Overhage, Manjeet Bains, Robert E. W. Hancock

  • Centre for Microbial Diseases and Immunity Research, University of British Columbia, 2259 Lower Mall, Vancouver, British Columbia V6T 1Z4, Canada
  • Correspondence
    Robert E. W. Hancock
    bob{at}cmdr.ubc.ca

    • Microbiology 2007; 153(2):474–482 · https://doi.org/10.1099/mic.0.2006/002519-0

    View at publisher PubMed

    Abstract

    Pseudomonas aeruginosa is an important nosocomial opportunistic human pathogen and a major cause of chronic lung infections in individuals with cystic fibrosis. Serious infections by this organism are often treated with a combination of aminoglycosides and semi-synthetic penicillins. Subinhibitory concentrations of antibiotics are now being recognized for their role in microbial persistence and the development of antimicrobial resistance, two very important clinical phenomena. An extensive screen of a P. aeruginosa PAO1 luciferase gene fusion library was performed to identify genes that were differentially regulated during exposure to subinhibitory gentamicin. It was demonstrated that subinhibitory concentrations of gentamicin and tobramycin induced a set of genes that are likely to affect the interaction of P. aeruginosa with host cells, including the gene encoding Lon protease, which is known to play a major role in protein quality control. Studies with a lon mutant compared to its parent and a complemented strain indicated that this protein was essential for biofilm formation and motility in P. aeruginosa.

    • A supplementary table of primers is available with the online version of this paper.

    • ↵†Present address: Department of Zoology, University of British Columbia, 6270 University Boulevard, Vancouver, British Columbia V6T 1Z4, Canada.

    Edited by: M. A. Kertesz

    INTRODUCTION

    Pseudomonas aeruginosa, a Gram-negative rod, causes rapidly progressing infections in hospitalized individuals, especially those with a compromised immune system. In addition, it is the predominant respiratory pathogen in cystic fibrosis patients, causing chronic lung infections that once established are impossible to eradicate. Therapy of serious infections caused by P. aeruginosa often consists of the combination of a semi-synthetic penicillin and an aminoglycoside, since cross-resistance between major classes of anti-pseudomonal antibiotics is less likely to develop (Hancock & Speert, 2000). Inhaled antimicrobial agents used in patients with cystic fibrosis have included carbenicillin, gentamicin, cephaloridine, tobramycin, colistin, polymyxin B and amikacin alone or in combination with parenteral antibiotics (Hancock & Speert, 2000). However, optimization of the current therapy of P. aeruginosa infections presents an intimidating problem for medical practitioners since these bacteria show intrinsic resistance to a wide range of antimicrobial agents. In addition, the ability of P. aeruginosa to develop further resistance during antibiotic therapy is a major problem for cystic fibrosis patients, due to the repeated therapeutic courses required to control chronic infection.

    Intrinsic resistance in P. aeruginosa has been shown to be due to restricted antibiotic uptake through the outer membrane coupled to secondary mechanisms such as active efflux and inducible antibiotic-degrading enzymes (Hancock, 1997). For aminoglycosides, uptake across the outer membrane has been shown to involve the self-promoted uptake pathway and is opposed by active efflux through the MexXY RND efflux pump in collaboration with a variety of outer-membrane channels (Aires et al., 1999; Jo et al., 2003; Vogne et al., 2004). P. aeruginosa also demonstrates the ability to develop adaptive resistance to aminoglycosides (Karlowsky et al., 1997) that is reversible upon removal of selective pressure. Little is known about the mechanisms of adaptive resistance in P. aeruginosa, although Macfarlane et al. (2000) demonstrated that susceptibility to aminoglycosides could be regulated by the two-component regulatory system PhoPQ, which responds to divalent cation concentrations. Furthermore it was shown that the adaptive resistance process requires the rapid production of the multidrug efflux pump MexXY and the interaction of these proteins with the constitutively produced outer-membrane component OprM (Hocquet et al., 2003). Recently it was shown that subinhibitory concentrations of aminoglycoside antibiotics induce biofilm formation, which contributes to bacterial persistence in chronic infections (Hoffman et al., 2005).

    In this work global gentamicin-induced gene regulation was studied in P. aeruginosa. Genome-wide screening approaches to elucidate mechanisms of antibiotic action and resistance are often approached by DNA-microarray analysis (Lin et al., 2005; reviewed by Brazas & Hancock, 2005) and mutant library screening (Goh et al., 2002; Lewenza et al., 2005). The screening of mutant libraries is of special interest and value because it combines elements of both transcriptome and proteome approaches since mutants can be generated with additional properties such as promoterless reporter genes (e.g. lux) to generate transcriptional or translational fusions to the inactivated gene (Lewenza et al., 2005). We demonstrate that under the tested conditions a substantial set of genes are induced in P. aeruginosa that are favourable for this pathogen to adapt to the human host. This approach led to the discovery that the Lon protease of P. aeruginosa is induced by subinhibitory concentrations of aminoglycosides, and is also essential for biofilm formation and motility in this pathogen.

    METHODS

    Bacterial strains and media.

    P. aeruginosa PAO1 and derived Tn5 mutants (Lewenza et al., 2005) were used. All bacterial strains were grown at 37 °C in Luria–Bertani (LB) broth. Tn5 insertions mutants were maintained on 50 μg tetracycline ml−1.

    Mutant library screening.

    A random pool of mutants (9408 individual colonies in all) from a mini-Tn5-luxCDABE mutant library in P. aeruginosa (Lewenza et al., 2005) was screened for alterations in luminescence when grown on LB agar containing 0.25 μg gentamicin ml−1 (corresponding to 0.25 of the MIC on LB agar plates) compared to LB agar without antibiotic. Mutants were grown overnight in 96-well microtitre trays containing LB broth, diluted 1 : 1000 in LB broth and transferred with a custom 96-well pin device onto the surface of LB agar plates containing 1.5 % (w/v) Difco agar (Becton-Dickinson) with or without gentamicin. Gene expression was analysed by measuring luminescence as described previously (Lewenza et al., 2005). The luminescence produced by the ATP-dependent luciferase system was detected and quantified using a ChemiGenius Bio-Imaging System (Syngene).

    Cloning of the lon gene.

    To permit complementation of the lon : : lux mutant, the ORF of the lon gene PA1803 was PCR-amplified from genomic DNA of PAO1 using the primers PA1803UHind and PA1803DXba (see Table S1, available with the online version of this paper, for details of primers). This amplicon was then cloned into TopoBlunt (Invitrogen). The HindIII–XbaI fragment containing the lon gene was ligated into the broad-host-range vector pBBR1-MCS5 (Kovach et al., 1995), resulting in pBBR1-MCS5 : : lon+, which was transformed into Escherichia coli XL-1 Blue and subsequently into E. coli S17-1 for mobilization into P. aeruginosa 74_D9 (lon : : lux) through conjugation. This resulted in the complemented strain lon : : lux/pBBR1-MCS5 : : lon+, hereafter called clon : : lux.

    Motility and biofilm assays.

    Swimming motility was evaluated on BM2 glucose plates [62 mM potassium phosphate buffer pH 7, 7 mM (NH4)2SO4, 2 mM MgSO4, 10 μM FeSO4, 0.4 % (w/v) glucose] containing 0.3 % Difco agar, and swarming on modified BM2 [with 0.5 % Casamino acids substituted for 7 mM (NH4)2SO4] glucose plates containing 0.5 % Difco agar. Swimming and swarming colony formation were evaluated by seeding with a 1 μl aliquot of an overnight culture grown on LB, and plates were incubated for 20 h at 37 °C. Twitching was measured after 24–48 h incubation at 37 °C on LB broth plates containing 1 % agar (Kohler et al., 2000). For biofilm formation assays, 5×104 bacteria were inoculated into each well of a polystyrene 96-well microtitre plate containing 100 μl LB broth in each well. The plate was incubated for 20 h at 37 °C without shaking. Biofilms were stained with 1 % (w/v) crystal violet and absorbance measurements were taken at 595 nm (as described by Friedman & Kolter, 2004).

    Growth experiment.

    Cells of P. aeruginosa PAO1 and lon : : lux were grown overnight in 100 μl LB. Then 1 μl of the overnight culture was inoculated into 100 μl LB broth and growth at 37 °C was monitored using a TECAN Spectrafluor Plus by measuring the OD595 every 20 min without shaking.

    Minimal inhibitory concentrations (MICs).

    MICs were assessed using standard broth microdilution procedures in LB broth (McPhee et al., 2003). Growth was scored after 24 h incubation at 37 °C.

    Real-time PCR.

    Total RNA was isolated from mid-exponential-phase P. aeruginosa grown in liquid LB medium with or without 0.25 MIC gentamicin (1.25 μg ml−1) using RNeasy midi columns (Qiagen). Four micrograms of total RNA was combined with 750 ng random primers [random decamer (NS)5, Invitrogen] and heated for 10 min at 70 °C, followed by 10 min incubation at 25 °C. Reverse transcription was done by adding 0.5 mM dNTPs, 500 U SuperRNase (Ambion) ml−1, 10 μM DTT, 10 000 U SuperScriptII (Invitrogen) ml−1 in 1× reaction buffer and incubating at 37 °C for 1 h followed by 2 h at 42 °C. The cDNA was centrifuged at 14 000 r.p.m. for 1 min and 500 U SuperRNase (Ambion) ml−1 was added before incubation at 37 °C for 45 min. The RNA was destroyed by the addition of 170 mM NaOH and incubation at 65 °C for 10 min. The reaction was neutralized by addition of 170 mM HCl. The cDNA was then used as template for real-time PCR using SYBR Green PCR Master Mix (Applied Biosystems) in an ABI Prism 7000 (Applied Biosystems). The real-time PCR was carried out in 1× SYBR Green Master Mix, with 200 nM forward and reverse primers (see Table S1) and 1 μl cDNA. All reactions were normalized to the rpsL gene encoding the 30S ribosomal protein S12.

    Phage PO4 sensitivity test.

    Bacteria were grown overnight in LB broth at 37 °C. Cells were harvested, washed with sterile LB and diluted to OD600 0.05 in LB soft agar. Ten microlitres of phage PO4 suspension was added to 2 ml of these cultures, mixed gently, and overlaid on 1.5 % (w/v) LB agar plates. Plates were incubated overnight at 37 °C and phage PO4 sensitivity was detected by plaque formation on the lawn of bacteria.

    RESULTS

    Identification of gentamicin-induced genes in P. aeruginosa

    Upon treatment with tobramycin and gentamicin in aerosol form, a fraction of the infecting bacteria may inevitably be exposed to subinhibitory levels of antibiotic that select for subpopulations exhibiting adaptive resistance, especially since it is known that the exact same strain can persist for years in a single patient despite apparent therapeutic success (Doring et al., 2000). To study the influence of sub-MIC gentamicin on gene transcription in P. aeruginosa PAO1, we screened a mini-Tn5-luxCDABE fusion library (Lewenza et al., 2005) for increased luminescence when grown in the presence of 0.25 μg ml−1 gentamicin (0.25 MIC) compared to luminescence in the absence of antibiotic. The luminescence of 37 of the 9408 tested lux fusions was up-regulated (⩾2-fold) in the presence of sub-MIC gentamicin. The insertion sites of all gentamicin-regulated lux fusions were mapped using an earlier described protocol (Lewenza et al., 2005) and are listed in Table 1⇓. Two fusions were obtained more than once, namely PA0605 and PA4554 (pilY1), and in three other instances we obtained gentamicin-inducible fusions to two genes in the same operon (PA0603/5, pchEF and flgEK). Among the genes in which we obtained gentamicin-induced lux fusions were ten genes encoding membrane and transport proteins, seven separate motility genes, predominantly influencing pili and flagella expression, one gene involved in the PQS quorum-sensing system, three genes responsible for carbon, fatty acid and energy metabolism, five genes encoding transcription regulators, three genes involved in RNA modification, adaptation and degradation, a putative enzyme, a gene from large phage-like pyocin cluster and three genes with unknown functions. Taken together the majority of the gentamicin-induced lux fusions mapped to genes and/or operons (based on computational prediction using operon finding software V1.2; ) that are involved in adaptation to the environment, as well as genes involved in gene regulation and energy metabolism. Activation of gene regulation during growth on LB in the presence or absence of 0.25 MIC gentamicin was independently confirmed using real-time PCR on 12 selected genes (Table 1⇓).

    Table 1.

    Properties of transposon fusion strains up-regulated by gentamicin and tobramycin

    Transposon insertion sites were determined using the protocol of Lewenza et al. (2005); * indicates intergenic fusion, and the mutant ID is the unique library identifier at . Pr., probable. Bioluminescence in the presence of gentamicin (Gm), tobramycin (Tb) and polymyxin B (PmB) was assessed relative to the same strain without antibiotic. The mean and standard deviation of bioluminescence was determined from three independent experiments. Shown are the ratio values comparing bioluminescence when bacteria were cultured with and without antibiotics. Selected genes were independently confirmed to demonstrate activation of gene regulation in the presence of gentamicin using real-time PCR. FC is the mean fold change of transcript upon growth in LB with 0.25 MIC gentamicin, normalized to the housekeeping gene rpsL and relative to the amount of transcript upon growth without gentamicin (comparative CT method) determined by real-time PCR. Gentamicin MIC values were determined by standard broth microdilution procedures in LB broth supplemented with different gentamicin concentrations (McPhee et al., 2003). Growth was scored after 24 h incubation at 37 °C. The most frequently observed MICs out of six independent measurements are shown.

    Gene activation of the gentamicin-induced lux fusions is largely aminoglycoside specific

    We previously demonstrated that polymyxin B, but not aminoglycosides, regulated the pmrAB (polymyxin resistance) two-component regulator operon (McPhee et al., 2003). To determine the specificity of induction, we quantified the luminescence of the 37 lux fusions when grown in the presence and absence of subinhibitory (0.25 MIC) concentrations of gentamicin, tobramycin or the positively charged antimicrobial lipopeptide polymyxin B. Table 1⇑ shows the luminescence ratios representing gene transcription during growth with and without added antibiotic. The transcription of all genes was similarly activated during growth in the presence of the two aminoglycosides tested, except for PA4961 : : lux, PA4225 : : lux and plsX : : lux, which showed stronger activation when grown in the presence of gentamicin than tobramycin. Only five lux fusions from our mutant pool (PA4961 : : lux, PA4226 : : lux, PA4225 : : lux, PA3353 : : lux and PA0621 : : lux) were slightly up-regulated in the presence of polymyxin B. These results suggest that transcriptional activation of mutants belonging to the gentamicin-induced mutant pool was predominantly aminoglycoside specific.

    Mutants carrying gentamicin-inducible lux fusions are not universally more susceptible to aminoglycosides

    Aminoglycoside resistance in P. aeruginosa typically results from drug inactivation by plasmid- or chromosomally encoded enzymes, or defects in the uptake or net accumulation (due to enhanced efflux) of the drug (Poole, 2005a, b). One possible explanation for the induction of these genes would be as a countermeasure by P. aeruginosa to prevent the lethal action of the aminoglycoside; such mutants that lack important resistance genes would be expected to be more susceptible to aminoglycosides than their parent strain, although mutants deficient in host adaptation genes would not necessarily show increased susceptibility in vitro. Indeed, as expected, the majority of the mutants had wild-type susceptibility to gentamicin (MIC 2 μg ml−1). Only two mutants PA4961 : : lux (encoding an unknown cytoplasmic membrane protein) and lon : : lux (see below) had a small but consistent lowering in MIC (Table 1⇑), showing a mean MIC value of 0.67±0.28 μg ml−1 for PA4961 : : lux and 1.00±0.00 μg ml−1 for lon : : lux in comparison to 2.00±0.00 for PAO1.

    lux fusions mapping within lon are impaired in biofilm formation

    P. aeruginosa is present in the soil, where it naturally encounters aminoglycoside-producing bacteria such as Streptomyces tenebrarius; therefore it has been suggested that adaptive resistance may be an evolutionary adaptation that occurred prior to the clinical use of antibiotics (Hoffman et al., 2005). One such adaptive response, the formation of biofilms, was recently demonstrated to be induced in P. aeruginosa when grown in sub-MIC tobramycin (Hoffman et al., 2005). Biofilms resist antibiotic treatment and contribute to bacterial persistence in chronic infections (Mah et al., 2003; Whiteley et al., 2001). Since the transcription of the majority of the genes belonging to our mutant pool was also induced in the presence of tobramycin, we screened the entire mutant pool for lux fusions showing reduced biofilm formation in comparison to the parent strain PAO1 when grown in LB medium as well as in subinhibitory concentrations of gentamicin (Fig. 1a⇓).

    Figure image not available in archive
    Fig. 1.

    (a) Biofilm formation by P. aeruginosa PAO1 gentamicin-induced lux fusions. Bacteria were grown for 24 h at 37 °C in 96-well polystyrene microtitre plates containing LB (black bars) or LB supplemented with 0.5 μg gentamicin ml−1 (white bars). I, Biofilm-repressed phenotype under both growth conditions; II, gentamicin-mediated biofilm repression. Surface-associated biofilm formation was analysed by crystal violet staining of the adherent biofilm followed by the extraction of the crystal violet with ethanol and measurement of A595. Means±sd of three independent experiments are plotted. Changes were statistically significant (P<0.05) compared to the parental strain PAO1. (b) Growth of P. aeruginosa PAO1 (⧫, □) in comparison to lon : : lux (▴, ▵) in LB (⧫, ▴) and LB supplemented with 0.5 μg gentamicin ml−1 (□, ▵). Growth was measured every 20 min during incubation at 37 °C using a TECAN Spectrafluor Plus.

    When grown in LB medium, seven lux fusions from our mutant pool showed impaired biofilm formation compared to the parent PAO1 strain (Fig. 1a⇑, black bars). Five of these lux fusions mapped within genes that are involved in flagella synthesis, two were undesignated genes (although PA3352 is homologous to E. coli flgN, a putative chaperone involved in flagella synthesis), while the most substantially affected mutant was PA1803, which is annotated as the Lon protease in P. aeruginosa. All mutants belonging to our gentamicin-induced mutant pool were also tested for biofilm formation when grown in the presence of sub-MIC gentamicin. Besides the lux fusions that already showed an impaired biofilm formation phenotype during growth in LB we found 11 additional lux fusions that were only impaired in the gentamicin induction of biofilm formation (Fig. 1a⇑, white bars). Again, lon : : lux showed the weakest biofilm formation phenotype among the tested mutants. Interestingly there was no obvious biofilm defect when grown in LB alone in the pilY1 mutant, which encodes a tip-associated (adhesin-like) component of type IV pili, despite the known role of pili in biofilm formation.

    To independently confirm the crippled biofilm formation nature of lon : : lux fusions, we tested the biofilm formation phenotype of two further lon : : lux fusions from our PAO1mini-Tn5-luxCDABE transposon mutant library (; 31_C9 and 24_A2). The same, biofilm-impaired phenotype was observed (data not shown). In addition, complementation of lon : : lux with the entire ORF of PA1803 (designated clon : : lux) restored the biofilm formation phenotype to wild-type levels (data not shown). To rule out the possibility of impaired growth of lon : : lux resulting in the observed phenotype, we monitored the growth of lon : : lux and its parental strain PAO1 under the growth conditions used for the biofilm studies above. No obvious growth defect of lon : : lux was detected (Fig. 1b⇑).

    The Lon protease of P. aeruginosa is involved in swimming, swarming and twitching motility

    Important factors for biofilm formation by P. aeruginosa are its ability to swim and twitch (Ramsey & Whiteley, 2004). Loss of surface motility influences phage uptake via type IV pili, conferring phage sensitivity to P. aeruginosa. To determine whether lon : : lux was impaired in type IV pili formation, a phage overlay assay was performed to compare phage sensitivity of lon : : lux with its parental strain PAO1 as well as with the pilus-deficient strain pilR : : lux (strain 80_E5 from our mutant library). Pilus-specific bacteriophage PO4 formed plaques on strain lon : : lux to the same extent as on PAO1, but not on the negative control strain pilR : : lux, indicating the normal presence of pili in lon : : lux (data not shown). Since crippled pili may retain enough function to induce phage uptake but not enough to drive coordinated cell motility, we tested lon : : lux for twitching and swarming as well as for the flagellum-based movement of swimming. The lon : : lux mutant showed dramatically reduced swarming motility, a moderate change in twitching motility and a small impairment in swimming motility compared to the parent strain PAO1, while this phenotype was restored in the PA1803 complemented clon : : lux construct (Fig. 2⇓).

    Figure image not available in archive
    Fig. 2.

    Swarming (I), swimming (II) and twitching (III) motility of P. aeruginosa PAO1, lon : : lux and clon : : lux. For swarming, bacteria were spot-inoculated onto BM2 swarm plates containing 0.5 % agar. Swimming motility was evaluated on BM2 plates containing 0.3 % agar. Colony diameters of swarming and swimming colonies were measured after 20 h incubation at 37 °C. Twitching was measured after 24–48 h incubation at 37 °C on LB broth plates containing 1 % agar. Means±sd of four independent experiments are plotted.

    DISCUSSION

    In this report we show that the transcription of 33 genes in P. aeruginosa is activated when cells are cultured in the presence of sub-MIC aminoglycosides on LB agar plates. Most of these lux fusions map within genes that are not apparently involved in aminoglycoside resistance mechanisms but rather in host adaptation processes. This is supported by our finding that the majority of the gentamicin-induced lux fusions listed in Table 1⇑ showed the same MIC values for gentamicin as did the parental strain PAO1 (Table 1⇑). Since aminoglycosides induce biofilm formation in P. aeruginosa (Hoffman et al., 2005), we tested our gentamicin-induced lux fusions for possible involvement in this phenotype. In addition to mutants that were deficient in functional flagella synthesis, we found lon mutants to be totally deficient in biofilm formation (Fig. 1a⇑).

    PA1803 encodes a protein that is 84 % similar to Lon protease (). Lon belongs to a class of ATP-dependent proteases, conserved in bacteria and even in bovine mitochondria (recently reviewed by Tsilibaris et al., 2006). It is known to be involved in the degradation of abnormal proteins, e.g. during heat shock, and in protein quality control (Tsilibaris et al., 2006) but also influences certain short-lived regulatory proteins. It is well known that such proteases are essential for virulence gene regulation (Robertson et al., 2000; Summers et al., 2000).

    Biofilm formation by opportunistic pathogens is devastating because the bacteria in these structured communities can withstand host immune responses, and are resistant to the highest deliverable doses of antibiotics (Stewart & Costerton, 2001). One mechanism that explains this tolerance is that bacteria in a biofilm experience nutrient limitation that leads to a localized slow growth (Brown et al., 1988; Xu et al., 2000). Many antibiotics are known to be less effective against non-growing cells than against rapidly growing cells. During amino acid starvation, protein degradation by ATP-dependent proteases is stimulated in order to increase the amino acid pool that can be used for the synthesis of specific enzymes required for adaptation to this condition (Miller, 1996). Upon nutritional downshift and at the onset of starvation, E. coli accumulates poly-P, a linear polymer of orthophosphate residues. Poly-P binds to Lon and redirects its activity towards free ribosomal proteins (Kuroda et al., 2001). Thus, in the centre of a biofilm, P. aeruginosa cells could recover from amino acid starvation by stimulating Lon-mediated turnover of free ribosomal proteins.

    Recent evidence indicates that P. aeruginosa residing as biofilms in airway mucus of cystic fibrosis patients is undergoing anaerobic metabolism (Hassett et al., 2002). In E. coli, the SoxRS regulon is induced under aerobic growth conditions. Once the inducing oxygen stress has subsided, the Lon protease was found to be responsible for degrading the de novo-synthesized regulator SoxS (Griffith et al., 2004). We thus speculate that the Lon protease in P. aeruginosa might have a function in the early onset of biofilm formation.

    To maximize movement and colonization in a viscous environment and on solid surfaces, numerous bacteria differentiate into swarmer cells, which may be a prerequisite to biofilm formation. Swarmer cell differentiation in peritrichous bacteria involves upregulation of the number of flagella otherwise used to swim. In polarly flagellated bacteria, swarmer cell differentiation involves the induction of an alternate flagella system. Interestingly, Lon appears to be involved in the control of both types of swarmer cell differentiation (Tsilibaris et al., 2006). We assessed the influence of lon : : lux on motility using soft BM2 agar for swimming and swarming as well as 1 % LB agar plates for twitching motility assays. The lon : : lux mutant showed impaired motility compared to PAO1. The defect in biofilm formation of lon mutants could thus be in part due to impaired motility, although the defect in biofilm formation seems to correlate more with the drastic defect on swarming than the rather more modest change in twitching (∼60 %) and swimming (∼30 %) motility. A swimming-impaired phenotype of a lon mutant has also been observed for Agrobacterium tumefaciens (Su et al., 2006).

    Emphasizing the importance of motility in the presence of gentamicin, our screen for gentamicin-induced gene expression also revealed seven lux fusions where the transposon insertion site was located within motility genes such as fliF, which was shown to play a role in bacterial adherence to mucin (Arora et al., 1996), and PA4554, a gonococcal PilC homologue involved in adhesion (Alm et al., 1996). In this context it is interesting that we observed aminoglycoside regulation of transcription of the lux fusion mapping immediately upstream of amrZ. AmrZ is required for proper surface localization of type IV pili, and it was previously shown that an amrZ mutant is twitching negative (Baynham et al., 2006). Independent of its role in twitching, amrZ is also essential for alginate production (Baynham et al., 1999). Cystic fibrosis patients are highly susceptible to chronic pulmonary disease caused by mucoid P. aeruginosa strains that overproduce the exopolysaccharide alginate (Silo-Suh et al., 2005). Interestingly we also found two PA3182 : : lux fusions to be up-regulated under the tested aminoglycoside induction conditions. PA3182 (pgl) is included in the same operon as zwf, which encodes glucose-6-phosphate dehydrogenase. A mutation in zwf leads to a 90 % reduction in alginate production in the mucoid cystic fibrosis isolate P. aeruginosa FRD1 (Silo-Suh et al., 2005). In addition both, pgl and zwf are more highly expressed in the Liverpool epidemic strain LES400 as well as in the strain FRD1 cystic fibrosis isolate (Silo-Suh et al., 2005), compared to strain PAO1 (Salunkhe et al., 2005; Silo-Suh et al., 2005).

    Our screen for gentamicin-induced genes revealed three encoding transport systems. The MexXY proteins form a functional tripartite efflux machinery with outer-membrane component OprM (Mine et al., 1999; Aires et al., 1999). In addition to its ability to accommodate a wide range of antibiotics the MexXY system has the distinctive property of being able to export aminoglycosides (Masuda et al., 2000a, b; Aires et al., 1999) and was shown to be necessary for adaptive resistance of P. aeruginosa to aminoglycosides (Hocquet et al., 2003). The PA4226 to PA4220 operon, which encodes the receptor and biosynthesis enzymes for the siderophore pyochelin, was also aminoglycoside regulated. Interestingly, this system is known to be involved in P. aeruginosa pathogenesis in that pyochelin stimulates bacterial growth during murine infections (Cox, 1982), reverses iron deprivation caused by human serum and transferrin (Ankenbauer et al., 1985) and efficiently removes iron from transferrin (Sriyosachati & Cox, 1986) while mutants with defects in pyochelin-mediated Fe(III) transport are less virulent than wild-type strains of P. aeruginosa (Sokol, 1987). Pyochelin is induced by the respiratory mucus derived P. aeruginosa strains from cystic fibrosis patients (Palmer et al., 2005; Wang et al., 1996). We also found a putative ABC transporter (PA0603–PA0606) with an unknown substrate to be inducible by gentamicin and tobramicin.

    The Pseudomonas quinolone signal (PQS) was shown to be a quorum-sensing regulatory network in P. aeruginosa (Pesci et al., 1999). PQS is required for phenazine production in P. aeruginosa (Mavrodi et al., 2001; McKnight et al., 2000). PA0998 (pqsC), which was shown in our screen to be up-regulated in the presence of aminoglycosides, and PA0999 are required for PQS production (Gallagher et al., 2002). Emphasizing the role of pqsC in infection, the Liverpool epidemic strain LES431 expresses pqsC to a higher extent than PAO1 (Salunkhe et al., 2005) and it was shown that the production of PQS is increased and dysregulated in P. aeruginosa isolates from infant patients with cystic fibrosis (Guina et al., 2003).

    This study has strengthened our understanding of the potential diversity of impacts of subinhibitory concentrations of aminoglycosides on microbial persistence in P. aeruginosa. Further studies are under way in our laboratory to study the important function of Lon in the pathogenesis of P. aeruginosa, as well as the complex gene regulation that is triggered in this pathogen by sensing aminoglycosides in the environment.

    Acknowledgments

    This work was supported by grants from the Canadian Cystic Fibrosis Foundation and the Canadian Institutes of Health Research. J. O. was a recipient of postdoctoral fellowships from the Alexander von Humboldt Foundation and Canadian Cystic Fibrosis Foundation. R. E. W. H. holds a Canada Research Chair. The authors thank John Hale for critically reading the manuscript and Agnes Che Yan Yuen for excellent technical assistance.

    References

    1. Aires, J. R., Kohler, T., Nikaido, H. & Plesiat, P. (1999). Involvement of an active efflux system in the natural resistance of Pseudomonas aeruginosa to aminoglycosides. Antimicrob Agents Chemother 43, 2624–2628.
    2. Alm, R. A., Hallinan, J. P., Watson, A. A. & Mattick, J. S. (1996). Fimbrial biogenesis genes of Pseudomonas aeruginosa: pilW and pilX increase the similarity of type 4 fimbriae to the GSP protein-secretion systems and pilY1 encodes a gonococcal PilC homologue. Mol Microbiol 22, 161–173.
    3. Ankenbauer, R., Sriyosachati, S. & Cox, C. D. (1985). Effects of siderophores on the growth of Pseudomonas aeruginosa in human serum and transferrin. Infect Immun 49, 132–140.
    4. Arora, S. K., Ritchings, B. W., Almira, E. C., Lory, S. & Ramphal, R. (1996). Cloning and characterization of Pseudomonas aeruginosa fliF, necessary for flagellar assembly and bacterial adherence to mucin. Infect Immun 64, 2130–2136.
    5. Baynham, P. J., Brown, A. L., Hall, L. L. & Wozniak, D. J. (1999). Pseudomonas aeruginosa AlgZ, a ribbon-helix-helix DNA-binding protein, is essential for alginate synthesis and algD transcriptional activation. Mol Microbiol 33, 1069–1080.
    6. Baynham, P. J., Ramsey, D. M., Gvozdyev, B. V., Cordonnier, E. M. & Wozniak, D. J. (2006). The Pseudomonas aeruginosa ribbon-helix-helix DNA-binding protein AlgZ (AmrZ) controls twitching motility and biogenesis of type IV pili. J Bacteriol 188, 132–140.
    7. Brazas, M. D. & Hancock, R. E. W. (2005). Using microarray gene signatures to elucidate mechanisms of antibiotic action and resistance. Drug Discov Today 10, 1245–1252.
    8. Brown, M. R., Allison, D. G. & Gilbert, P. (1988). Resistance of bacterial biofilms to antibiotics: a growth-rate related effect? J Antimicrob Chemother 22, 777–780.
    9. Cox, C. D. (1982). Effect of pyochelin on the virulence of Pseudomonas aeruginosa. Infect Immun 36, 17–23.
    10. Doring, G., Conway, S. P., Heijerman, H. G., Hodson, M. E., Hoiby, N., Smyth, A. & Touw, D. J. (2000). Antibiotic therapy against Pseudomonas aeruginosa in cystic fibrosis: a European consensus. Eur Respir J 16, 749–767.
    11. Friedman, L. & Kolter, R. (2004). Genes involved in matrix formation in Pseudomonas aeruginosa PA14 biofilms. Mol Microbiol 51, 675–690.
    12. Gallagher, L. A., McKnight, S. L., Kuznetsova, M. S., Pesci, E. C. & Manoil, C. (2002). Functions required for extracellular quinolone signaling by Pseudomonas aeruginosa. J Bacteriol 184, 6472–6480.
    13. Goh, E. B., Yim, G., Tsui, W., McClure, J., Surette, M. G. & Davies, J. (2002). Transcriptional modulation of bacterial gene expression by subinhibitory concentrations of antibiotics. Proc Natl Acad Sci U S A 99, 17025–17030.
    14. Griffith, K. L., Shah, I. M. & Wolf, R. E., Jr (2004). Proteolytic degradation of Escherichia coli transcription activators SoxS and MarA as the mechanism for reversing the induction of the superoxide (SoxRS) and multiple antibiotic resistance (Mar) regulons. Mol Microbiol 51, 1801–1816.
    15. Guina, T., Purvine, S. O., Yi, E. C., Eng, J., Goodlett, D. R., Aebersold, R. & Miller, S. I. (2003). Quantitative proteomic analysis indicates increased synthesis of a quinolone by Pseudomonas aeruginosa isolates from cystic fibrosis airways. Proc Natl Acad Sci U S A 100, 2771–2776.
    16. Hancock, R. E. W. (1997). The bacterial outer membrane as a drug barrier. Trends Microbiol 5, 37–42.
    17. Hancock, R. E. W. & Speert, D. P. (2000). Antibiotic resistance in Pseudomonas aeruginosa: mechanisms and impact on treatment. Drug Resist Updat 3, 247–255.
    18. Hassett, D. J., Cuppoletti, J., Trapnell, B., Lymar, S. V., Rowe, J. J., Yoon, S. S., Hilliard, G. M., Parvatiyar, K., Kamani, M. C. & other authors (2002). Anaerobic metabolism and quorum sensing by Pseudomonas aeruginosa biofilms in chronically infected cystic fibrosis airways: rethinking antibiotic treatment strategies and drug targets. Adv Drug Deliv Rev 54, 1425–1443.
    19. Hocquet, D., Vogne, C., El Garch, F., Vejux, A., Gotoh, N., Lee, A., Lomovskaya, O. & Plesiat, P. (2003). MexXY-OprM efflux pump is necessary for adaptive resistance of Pseudomonas aeruginosa to aminoglycosides. Antimicrob Agents Chemother 47, 1371–1375.
    20. Hoffman, L. R., D'Argenio, D. A., MacCoss, M. J., Zhang, Z., Jones, R. A. & Miller, S. I. (2005). Aminoglycoside antibiotics induce bacterial biofilm formation. Nature 436, 1171–1175.
    21. Jo, J. T., Brinkman, F. S. & Hancock, R. E. W. (2003). Aminoglycoside efflux in Pseudomonas aeruginosa: involvement of novel outer membrane proteins. Antimicrob Agents Chemother 47, 1101–1111.
    22. Karlowsky, J. A., Zelenitsky, S. A. & Zhanel, G. G. (1997). Aminoglycoside adaptive resistance. Pharmacotherapy 17, 549–555.
    23. Kohler, T., Curty, L. K., Barja, F., van Delden, C. & Pechere, J. C. (2000). Swarming of Pseudomonas aeruginosa is dependent on cell-to-cell signaling and requires flagella and pili. J Bacteriol 182, 5990–5996.
    24. Kovach, M. E., Elzer, P. H., Hill, D. S., Robertson, G. T., Farris, M. A., Roop, R. M., 2nd & Peterson, K. M. (1995). Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene 166, 175–176.
    25. Kuroda, A., Nomura, K., Ohtomo, R., Kato, J., Ikeda, T., Takiguchi, N., Ohtake, H. & Kornberg, A. (2001). Role of inorganic polyphosphate in promoting ribosomal protein degradation by the Lon protease in E. coli. Science 293, 705–708.
    26. Lewenza, S., Falsafi, R. K., Winsor, G., Gooderham, W. J., McPhee, J. B., Brinkman, F. S. & Hancock, R. E. W. (2005). Construction of a mini-Tn5-luxCDABE mutant library in Pseudomonas aeruginosa PAO1: a tool for identifying differentially regulated genes. Genome Res 15, 583–589.
    27. Lin, J. T., Connelly, M. B., Amolo, C., Otani, S. & Yaver, D. S. (2005). Global transcriptional response of Bacillus subtilis to treatment with subinhibitory concentrations of antibiotica that inhibit protein synthesis. Antimicrob Agents Chemother 49, 1915–1926.
    28. Macfarlane, E. L., Kwasnicka, A. & Hancock, R. E. W. (2000). Role of Pseudomonas aeruginosa PhoP-PhoQ in resistance to antimicrobial cationic peptides and aminoglycosides. Microbiology 146, 2543–2554.
    29. Mah, T. F., Pitts, B., Pellock, B., Walker, G. C., Stewart, P. S. & O'Toole, G. A. (2003). A genetic basis for Pseudomonas aeruginosa biofilm antibiotic resistance. Nature 426, 306–310.
    30. Masuda, N., Sakagawa, E., Ohya, S., Gotoh, N., Tsujimoto, H. & Nishino, T. (2000a). Contribution of the MexX-MexY-OprM efflux system to intrinsic resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother 44, 2242–2246.
    31. Masuda, N., Sakagawa, E., Ohya, S., Gotoh, N., Tsujimoto, H. & Nishino, T. (2000b). Substrate specificities of MexAB-OprM, MexCD-OprJ, and MexXY-OprM efflux pumps in Pseudomonas aeruginosa. Antimicrob Agents Chemother 44, 3322–3327.
    32. Mavrodi, D. V., Bonsall, R. F., Delaney, S. M., Soule, M. J., Phillips, G. & Thomashow, L. S. (2001). Functional analysis of genes for biosynthesis of pyocyanin and phenazine-1-carboxamide from Pseudomonas aeruginosa PAO1. J Bacteriol 183, 6454–6465.
    33. McKnight, S. L., Iglewski, B. H. & Pesci, E. C. (2000). The Pseudomonas quinolone signal regulates rhl quorum sensing in Pseudomonas aeruginosa. J Bacteriol 182, 2702–2708.
    34. McPhee, J. B., Lewenza, S. & Hancock, R. E. W. (2003). Cationic antimicrobial peptides activate a two-component regulatory system, PmrA-PmrB, that regulates resistance to polymyxin B and cationic antimicrobial peptides in Pseudomonas aeruginosa. Mol Microbiol 50, 205–217.
    35. Miller, C. G. (1996). Protein degradation and proteolytic modification. In Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn, vol. 1, pp. 938–954. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.
    36. Mine, T., Morita, Y., Kataoka, A., Mizushima, T. & Tsuchiya, T. (1999). Expression in Escherichia coli of a new multidrug efflux pump, MexXY, from Pseudomonas aeruginosa. Antimicrob Agents Chemother 43, 415–417.
    37. Palmer, K. L., Mashburn, L. M., Singh, P. K. & Whiteley, M. (2005). Cystic fibrosis sputum supports growth and cues key aspects of Pseudomonas aeruginosa physiology. J Bacteriol 187, 5267–5277.
    38. Pesci, E. C., Milbank, J. B., Pearson, J. P., McKnight, S., Kende, A. S., Greenberg, E. P. & Iglewski, B. H. (1999). Quinolone signaling in the cell-to-cell communication system of Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 96, 11229–11234.
    39. Poole, K. (2005a). Aminoglycoside resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother 49, 479–487.
    40. Poole, K. (2005b). Efflux-mediated antimicrobial resistance. J Antimicrob Chemother 56, 20–51.
    41. Ramsey, M. M. & Whiteley, M. (2004). Pseudomonas aeruginosa attachment and biofilm development in dynamic environments. Mol Microbiol 53, 1075–1087.
    42. Robertson, G. T., Kovach, M. E., Allen, C. A., Ficht, T. A. & Roop, R. M., 2nd (2000). The Brucella abortus Lon functions as a generalized stress response protease and is required for wild-type virulence in BALB/c mice. Mol Microbiol 35, 577–588.
    43. Salunkhe, P., Smart, C. H., Morgan, J. A., Panagea, S., Walshaw, M. J., Hart, C. A., Geffers, R., Tummler, B. & Winstanley, C. (2005). A cystic fibrosis epidemic strain of Pseudomonas aeruginosa displays enhanced virulence and antimicrobial resistance. J Bacteriol 187, 4908–4920.
    44. Silo-Suh, L., Suh, S. J., Phibbs, P. V. & Ohman, D. E. (2005). Adaptations of Pseudomonas aeruginosa to the cystic fibrosis lung environment can include deregulation of zwf, encoding glucose-6-phosphate dehydrogenase. J Bacteriol 187, 7561–7568.
    45. Sokol, P. A. (1987). Surface expression of ferripyochelin-binding protein is required for virulence of Pseudomonas aeruginosa. Infect Immun 55, 2021–2025.
    46. Sriyosachati, S. & Cox, C. D. (1986). Siderophore-mediated iron acquisition from transferrin by Pseudomonas aeruginosa. Infect Immun 52, 885–891.
    47. Stewart, P. S. & Costerton, J. W. (2001). Antibiotic resistance of bacteria in biofilms. Lancet 358, 135–138.
    48. Su, S., Stephens, B. B., Alexandre, G. & Farrand, S. K. (2006). Lon protease of the α-proteobacterium Agrobacterium tumefaciens is required for normal growth, cellular morphology and full virulence. Microbiology 152, 1197–1207.
    49. Summers, M. L., Botero, L. M., Busse, S. C. & McDermott, T. R. (2000). The Sinorhizobium meliloti Lon protease is involved in regulating exopolysaccharide synthesis and is required for nodulation of alfalfa. J Bacteriol 182, 2551–2558.
    50. Tsilibaris, V., Maenhaut-Michel, G. & Van Melderen, L. (2006). Biological roles of the Lon ATP-dependent protease. Res Microbiol 157, 701–713.
    51. Vogne, C., Aires, J. R., Bailly, C., Hocquet, D. & Plesiat, P. (2004). Role of the multidrug efflux system MexXY in the emergence of moderate resistance to aminoglycosides among Pseudomonas aeruginosa isolates from patients with cystic fibrosis. Antimicrob Agents Chemother 48, 1676–1680.
    52. Wang, J., Lory, S., Ramphal, R. & Jin, S. (1996). Isolation and characterization of Pseudomonas aeruginosa genes inducible by respiratory mucus derived from cystic fibrosis patients. Mol Microbiol 22, 1005–1012.
    53. Whiteley, M., Bangera, M. G., Bumgarner, R. E., Parsek, M. R., Teitzel, G. M., Lory, S. & Greenberg, E. P. (2001). Gene expression in Pseudomonas aeruginosa biofilms. Nature 413, 860–864.
    54. Xu, K. D., McFeters, G. A. & Stewart, P. S. (2000). Biofilm resistance to antimicrobial agents. Microbiology 146, 547–549.