PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
 
PLoS One. 2011; 6(12): e29276.
Published online 2011 December 15. doi:  10.1371/journal.pone.0029276
PMCID: PMC3240657

Phenotypic and Genome-Wide Analysis of an Antibiotic-Resistant Small Colony Variant (SCV) of Pseudomonas aeruginosa

Jacques Ravel, Editor

Abstract

Background

Small colony variants (SCVs) are slow-growing bacteria, which often show increased resistance to antibiotics and cause latent or recurrent infections. It is therefore important to understand the mechanisms at the basis of this phenotypic switch.

Methodology/Principal Findings

One SCV (termed PAO-SCV) was isolated, showing high resistance to gentamicin and to the cephalosporine cefotaxime. PAO-SCV was prone to reversion as evidenced by emergence of large colonies with a frequency of 10−5 on media without antibiotics while it was stably maintained in presence of gentamicin. PAO-SCV showed a delayed growth, defective motility, and strongly reduced levels of the quorum sensing Pseudomonas quinolone signal (PQS). Whole genome expression analysis further suggested a multi-layered antibiotic resistance mechanism, including simultaneous over-expression of two drug efflux pumps (MexAB-OprM, MexXY-OprM), the LPS modification operon arnBCADTEF, and the PhoP-PhoQ two-component system. Conversely, the genes for the synthesis of PQS were strongly down-regulated in PAO-SCV. Finally, genomic analysis revealed the presence of mutations in phoP and phoQ genes as well as in the mexZ gene encoding a repressor of the mexXY and mexAB-oprM genes. Only one mutation occurred only in REV, at nucleotide 1020 of the tufA gene, a paralog of tufB, both encoding the elongation factor Tu, causing a change of the rarely used aspartic acid codon GAU to the more common GAC, possibly causing an increase of tufA mRNA translation. High expression of phoP and phoQ was confirmed for the SCV variant while the revertant showed expression levels reduced to wild-type levels.

Conclusions

By combining data coming from phenotypic, gene expression and proteome analysis, we could demonstrate that resistance to aminoglycosides in one SCV mutant is multifactorial including overexpression of efflux mechanisms, LPS modification and is accompanied by a drastic down-regulation of the Pseudomonas quinolone signal quorum sensing system.

Introduction

Pseudomonas aeruginosa is a ubiquitous Gram-negative bacterium found in diverse ecological habitats such as soils, marshes and coastal marine waters. As an opportunistic pathogen, P. aeruginosa is able to infect humans, animals and plants [1], [2], [3]. P. aeruginosa is a primary nosocomial diseases causative agent and represents the major cause of morbidity and mortality in patients with cystic fibrosis (CF). P. aeruginosa produces a large panel of secreted virulence factors like the phenazine pyocyanin, the siderophore pyoverdine, elastase, and toxins. It is also characterized by its high level of drug resistance involving the formation of antibiotic-resistant biofilms resulting from the emergence of phenotypic variants [2], [3]. During the course of infection, P. aeruginosa can efficiently adopt diverse strategies to evade antimicrobial stresses and the host immune system defenses, making it impossible to eradicate this bacterium permanently from CF lungs [2], [4]. Important phenotypic variations can occur during chronic colonization, such as conversion to mucoidy [5], the emergence of persister cells after antibiotics treatment [6], [7] or the occurrence of small colony variants with higher resistance to antibiotics [8], [9], [10], [11], [12]. Compared to wild-type P. aeruginosa, SCVs show increased antibiotic resistance, enhanced biofilm formation, reversion to wild-type-like morphotypes, reduced motility, and slow and auto-aggregative growth behavior [13], [14]. SCVs have been isolated from CF lungs or sputum [4], [8], [9], [12], laboratory-grown biofilms [11], [12], [14], in vitro selection upon antibiotic exposure [15], [16] or as a consequence of gene inactivation [17], [18]. Clinically, P. aeruginosa SCVs have already been proven to associate with chronic infections behaving as persisters in pathogenesis of CF patients and making it almost impossible for clinicians to eradicate the infections [8], [19], [20]. The intracellular second messenger cyclic-di-GMP (c-di-GMP) [21] has been recently shown to be involved in SCV phenotype switching in terms of biofilm formation, reduced motility, and exopolysaccharide (EPS) production [18], [22], [23], [24], [25], [26]. The “phenotypic variant regulator”, PvrR, containing a conserved EAL domain of phosphodiesterase (PDE) involved in the hydrolysis of c-di-GMP, has been identified to control the phenotypic switch from an antibiotic resistant and auto-aggregative rough SCV (RSCV) of P. aeruginosa strain PA14 to wild-type-like antibiotics susceptible revertants [15]. Another characteristic driven by the elevated level of c-di-GMP in SCVs is the contribution of two EPS-encoding loci in some P. aeruginosa strains (PA2231-PA2245 for psl and PA3058-PA3064 for pel) to auto-aggregation and hyper adherence phenotypes characterized by increased Congo Red dye binding [27], [28], [29]. Although antibiotics resistance of P. aeruginosa has been connected to biofilm formation and linked to phenotypic variation [15], the mechanisms underlying the extremely high antibiotic resistance of SCVs has not been reported extensively due to the unavailability, in some cases, of the WT counterpart for comparison.

In this study, we present the identification of a novel, reversion-prone, P. aeruginosa SCV with distinct features, including resistance to various antibiotics, defective motility, and absence of production of the quorum sensing PQS signal molecule. Using a combination of genomic, transcriptomic, proteomic and phenotypic approaches, we provide the first evidence of concerted mechanisms harnessed by this P. aeruginosa SCV leading to antibiotic resistance as well as down-regulation of acute virulence genes, probably involving the PhoP PhoQ two component system.

Results

Phenotypic characterization of a gentamicin-resistant P. aeruginosa PAO1-SCV and large colony pseudo-revertants

Following sub-culturing P. aeruginosa PAO1 (ATCC 15692) in the presence of high-concentration of gentamicin (200 µg ml−1, Gm), we isolated a Gm-resistant SCV designated PAO-SCV, which formed small (ca. 1/5 of the wild-type diameter), smooth colonies after three days of incubation at 37°C on LB agar plates (Figure 1A). PAO-SCV grown in liquid LB also showed a delayed entry in exponential phase compared to the wild-type (Figure 1B).

Figure 1
Growth phenotypes of PAO-SCV.

PAO-SCV showed high level of resistance towards gentamicin and cefotaxime (Table 1 and Figure 2). The persistence fraction of PAO-SCV after treatment with the fluoroquinolone antibiotic ofloxacin was approximately 2-fold higher compared to the PAO1 wild-type strain (Figure S1).

Figure 2
Sensitivity to antibiotics of wild-type, PAO-SCV, and PAO-REV.
Table 1
Minimal inhibitory concentrations (MICs) of different antibiotics against wild type (PAO), PAO-SCV and one selected pseudo-revertant (REV).

In the absence of Gm large colonies variants tended to appear, characterized by rough contours, at a frequency of 10−5 (Figure 1A and Figure 2) on agar plates. The frequency of reversion varied between 1.3×10−5 to 8.7×10−5 depending on the medium used (LB or CAA) or the incubation temperature (25°C or 37°C). Importantly, no large colonies appeared when the PAO-SCV was grown in the presence of Gm since the cells from large colonies regained Gm and cefotaxime sensitivity (Figure 2). Given its unstable character, PAO-SCV was kept on LB plates supplemented with Gm (200 µg ml−1) to avoid the emergence of pseudo-revertants. However, during experiments described below no antibiotic was added (unless mentioned in the text) in order to avoid Gm-induced changes independent of those caused by the SCV phenotype. At the end of experiments cell suspensions were diluted and the number of large colonies counted. When their number was less than 1/105 the experiment was considered to be valid.

PQS production is strongly decreased in PAO-SCV

We observed that the small colony variant showed reduced production of some known quorum sensing-dependent virulence factors (pyocyanin, pyoverdine, elastase, and a total absence of motility [Figure S2]). Likewise, the PAO-SCV showed strongly reduced virulence using both plants (Belgian endive) and Drosophila as hosts (Figure S3). This prompted us to look at the production of quorum sensing signal molecules themselves, including N-3-(oxododecanoyl)-L-homoserine lactone (3-oxo-C12-HSL) for the LasR–LasI system and N-butyryl-L-homoserine lactone (C4-HSL) for the RhlR–RhlI system [30], [31]. Finally, we also checked the production of 4-quinolones such as 2-heptyl-4-quinolone (HHQ) and 2-heptyl-3-hydroxy-4-quinolone (PQS) [32]. The levels of 3-oxo-C12-HSL and C4-HSL in the cell culture supernatants were similar for the wild-type, PAO-SCV and the pseudo-revertant (results not shown). However, in PAO-SCV a strong decrease in the production of both HHQ and PQS was observed as compared to that of wild-type while the wild type level was restored in the pseudo-revertant (Figure 3).

Figure 3
Production of signal molecules.

Comparison of proteome profiles of PAO-SCV and wild-type P. aeruginosa

Because profound phenotypic changes were detected in PAO-SCV, we decided to compare the proteomes of PAO-SCV and wild-type cells (Figure 4). After protein identification with MALDI-TOF MS analysis, we found at least 24 differentially expressed proteins, whereby 16 proteins were less abundant and 8 more abundant in PAO-SCV (Table 2). The proteins showing differential abundance are involved in amino acid biosynthesis and metabolism, motility, transport of small molecules and transcriptional regulation. According to this analysis, the two-component response regulator PhoP is one of the most prominently induced proteins in PAO-SCV. Another finding is the over-expression of the major outer membrane protein OprF in PAO-SCV, which is the P. aeruginosa major non-specific porin allowing diffusion of various solutes, such as nitrates or nitrites under anaerobic conditions or small oligosaccharides with a molecular weight up to 1519 Da [33], [34]. We also found decreased expression of the anaerobiosis-induced outer membrane porin OprE, which, similarly to OprD, was predicted to be involved in outer membrane permeability of the β-lactam antibiotic imipenem and basic amino acids [35], [36]. The SCV proteome showed a drastic decrease in the production of the translation elongation factor Tu, TufB (PA4277) and a less marked decrease in elongation factor Ts (Tsf, PA3655).

Figure 4
Comparison of proteomes of wild-type and PAO-SCV.
Table 2
Identification of differentially produced proteins in P. aeruginosa PAO-SCV by MALDI-TOF MS peptide mass mapping (PMP).

Genome-wide transcriptional profile of PAO-SCV and PAO1

Since some of the differentially produced proteins could already give clue to the changes occurring in the SCV mutant, we decided to further investigate which global changes in gene expression could account for this phenotypic variation. The gene transcription profiles of PAO-SCV and WT strains were compared in early and late stationary-phase of growth, corresponding to incubation times of 20 and 40 h respectively, using P. aeruginosa Affymetrix GeneChips. The results are presented using Venn diagrams and pie charts for simplicity, facilitating the understanding and interpretation of the overall genome transcriptional profile [37]. The tables showing the complete lists of differentially expressed genes are shown as supplementary material (Tables S1, S2, S3, S4 and S5). As shown in Figure 5 and in supplementary Tables S1, S2, S3, S4 and S5), during stationary phase, a total of 642 genes representing approximately 12% of the entire genome displayed a differential expression pattern in PAO-SCV compared to that of wild type PAO1 (P value <0.05, Student's t-test). Among these 642 genes, 466 were up-regulated (≈ 73% of differentially regulated genes, from 2- to 26-fold, see Table S3) and 176 were down-regulated (≈ 27% of differentially regulated genes, from 2- to 16-fold, see Table S4). Interestingly, remarkable differences were observed for up-regulated genes (Figure 5B), among which 356 genes were found to be highly expressed during late stationary phase while only 164 genes were up-regulated during early stationary phase as compared to the wild-type. Genes involved in amino acid biosynthesis and metabolism showed an increased transcription level in both early and late stationary phase of growth of PAO-SCV (Table S2). Genes involved in antibiotic resistance and genes coding for membrane proteins were highly expressed in the SCV mutant in early stationary phase. Conversely, some genes involved in the production of secreted factors and those related to phage, transposon and plasmids were expressed at a lower level in PAO-SCV compared to the wild-type.

Figure 5
Transcriptome analysis.

Figure 6A shows that some of the genes known to be involved in antibiotic resistance are up-regulated in the PAO-SCV. These could be classified into four different functional groups, linked to four distinct resistance mechanisms (See lists of selected genes in Table S3). Among these are efflux pump systems genes known to contribute to resistance to aminoglycosides, including mexAB-oprM and mexXY and their respective mexR and mexZ regulatory genes [38]. The observed higher expression of these efflux pumps is in agreement with the results showing a higher resistance to all aminoglycosides and to the cephalosporin antibiotic cefotaxime (Figure 2 and Table 1). We also found that the cat gene encoding the chloramphenicol acetyl transferase is up-regulated in PAO-SCV, both in exponential and stationary phase. Interestingly, expression of another resistance-nodulation-cell division (RND) efflux pump, MexGHI-OpmD, is reduced in PAO-SCV in late stationary phase. This efflux system has been shown to be important for PQS-mediated signaling, pyocyanin production, and is thought to be a general phenazine transporter, including pyocyanin [39], [40], [41]. Again, this observation is in line with the reduced production of pyocyanin by PAO-SCV and the quasi-absence of HHQ and PQS in culture supernatants (Figure 3). Among PAO-SCV up-regulated genes are those involved in LPS modification, including migA (PA0705) encoding a glycosyl transferase, and the gene cluster PA3552-PA3559 (arnBCADTEF-PA3559, Figure 6B), which are homologues of the pmrHFIJLKM genes of Salmonella enterica involved in lipid A modification [42], [43], [44]. Interestingly, phoP-phoQ, together with the upstream porin protein gene oprH was markedly up-regulated throughout the stationary phase in PAO-SCV, forming the third functional group, and explaining the overexpression of migA and arnBCADTEF-PA3559 (Figure 6C). As already mentioned, higher levels of the transcriptional regulator PhoP were also detected by 2D-PAGE analysis. The PhoP-PhoQ system is known to be involved in aminoglycoside resistance in P. aeruginosa [45].

Figure 6
Analysis of differentially expressed genes involved in transport, efflux, and membrane modification.

A fourth functional group of genes markedly up-regulated in PAO-SCV included those encoding membrane proteins, transcriptional regulators and transporters of small molecules (Figure 6D). More specifically, several genes encoding outer membrane proteins are up-regulated in PAO-SCV: the previously mentioned oprH, oprD, PA1198 (encoding a lipoprotein), oprQ, opdQ, opdP, and the lipoprotein gene omlA. OprQ, and OpdP belong to the OprD family and have been proposed to contribute to the transport of arginine [46]. In this context, it is interesting to note that the genes PA5152 (ABC transporter, ATP binding component), and, to a large extent, PA5153 (periplasmic binding protein), probably involved in the transport of arginine, are also up-regulated.

The transcriptome analysis not only provided insights into the PAO-SCV mechanisms involved in aminoglycoside-resistance, but also explained some of the prominent phenotypic changes. As shown in Figure 7A, transcript levels of the pqsABCDE genes as well as for the two neighboring anthranilate synthase genes phnA and phnB were strongly reduced in PAO-SCV, in line with the results presented in Figure 3 showing a strong decrease in HHQ and PQS production. As a result of the down-regulation of PQS genes (pqsA-E, pqsH, phnAB), genes such as lasA (coding for elastase), phzC2-G2, phzB1, phzS (for pyocyanin biosynthesis), hcnC (for HCN production) and rhlA (for rhamnolipids synthesis) were also down-regulated. Lower rhamnolipid production could also partly explain the observed decreased swarming motility and the absence of channels in PAO-SCV biofilms [47], [48]. In agreement with the absence of changes in AHLs production, the transcription of lasI and rhlI coding for the 3-oxo-C12-HSL and C4-HSL synthases was unchanged.

Figure 7
Analysis of down-regulated genes in PAO-SCV.

Some genes involved in energy generation via respiration were also differentially regulated since we observed a higher expression in PAO-SCV of cytochromes genes such as PA0105-0108 (encoding cytochrome c oxidase subunits I–III), PA1175-1177 (napDFE encoding the components of nitrate reduction), and a marked up-regulation of PA1983 (exaB, encoding a cytochrome c550). Other genes involved in energy generation were found down-regulated in PAO-SCV such as PA4133 (ccoN).

Another interesting finding is the differential expression of genes associated with biofilm formation. We found that in PAO-SCV, flagellar synthesis genes expression was reduced compared to wild-type PAO1 which was also confirmed by proteomic analysis (see Figure 5 and Table 2). This result could explain the total absence of motility of PAO-SCV (Figure S2). The third interesting functional group is formed by phage-related genes including phage those involved in Pf1 phage production and the PA0616-PA0647 cluster, the expression of which was greatly reduced in PAO-SCV compared to wild-type (Figure 7C).

Validation of microarray results via Quantitative RT PCR

Quantitative real time PCR was used to measure the level of transcripts of the phoP and phoQ genes in wild type, SCV, and one pseudo-revertant. As shown in Figure 8, the level of phoP and phoQ transcription was increased in the SCV while the levels were similar for wild-type and the revertant large colony variant.

Figure 8
Analysis of phoP and phoQ gene expression.

Whole genome analysis of P. aeruginosa SCV

The genome of the originally selected pseudo-revertant (REV) of PAO-SCV was fully sequenced using the Illumina Genome Analyzer. The choice to re-sequence the revertant only was justified by the fact that it should contain all mutations present in three strains (PAO1, PAO-SCV, revertant), and indels and single nucleotide polymorphisms (SNPs) could be easily checked for their presence in the genomes of PAO-SCV and its clonal wild-type using a combination of PCR amplification and Sanger sequencing. A limited list of sequence variations in relation to the PAO1 sequence was found (Table 3), most of which were already detected when we re-sequenced these regions in our own PAO1 lab strain and in the PAO1 strain of Chronic Pseudomonas Infection Group in Helmholtz Infection Research Centre [49]. In order to exclude the possibility that genetic variations have been overlooked due to the limitations of reference based read mapping, the reads were additionally assembled in a de novo approach using Velvet and the resulting contigs were aligned with the PAO1 genome to find genetic variations. Finally, eight changes remained that were unique to REV after elimination of the mutations also found in PAO1 wild-type (Table 3). For seven of these regions, PCR amplification and re-sequencing via the Sanger method revealed that these differences in sequence were present in both REV and SCV. In only one instance, a mutation was found that was unique to REV, causing a base change at position 1020 of the tufA gene (PA4277), a paralog of tufB (PA4265). The two genes are nearly identical in sequence, except at this position where tufB has a GAC codon for aspartate while tufa has the rarely used codon GAT at the same place. In REV, a mutation caused a silent mutation in tufA, changing the GAT codon to GAC, like in the tufB gene. However, no additional variations were identified and all contigs aligned to at least one genomic region of PAO1 excluding the possibility of an unrecognized insertion of exogenous genetic elements (e.g., phages or transposons). The detection of genetic variations usually assumes a clonal population and SNPs and indels of low abundance are mostly interpreted as sequencing errors and excluded from the analysis. Nevertheless, the absence of REV specific mutations (except in one case) may be due to several mutations that independently cause the phenotypic conversion and are present in different clones coexisting in a mixed population. The original data were therefore re-analyzed to identify SNPs and indels that were present in at least 10% of the reads. However, this analysis also identified only those variations that were already detected in the initial approach.

Table 3
List of genetic variations in P. aeruginosa SCV revertant.

Interestingly, changes were identified in phoPQ and mexZ, in line with the results of the transcriptional analysis. Specifically, the phoP gene contains a SNP which confers a histidine (H) to arginine (R) change while, remarkably, phoQ harbors an in-frame 39-bp deletion in its coding sequence, deleting a 13 amino acids sequence RLLRSEHKQRERY between residues 226 and 239. In PAO-SCV the mexZ gene was inactivated by the introduction of a stop codon, which could explain the over-expression of the MexXY pump involved in aminoglycosides and fluoroquinolone resistance.

Discussion

The SCV phenotype observed in this study is reminiscent of the observations made by Tarighi et al. who found that knocking out the the ppgL gene (PA4204) of P. aeruginosa caused a SCV phenotype with the apparition of large colony variants [17]. In this particular case the SCV phenotype was thought to be due to the accumulation of a toxic intermediate: gluconolactone [17]. This observation suggests that SCV phenotypes can be the results of exposures to different stresses. In P. aeruginosa, several mechanisms of aminoglycoside resistance have been described: resistance through efflux systems, by alteration of porins or outer membrane properties (including LPS modification), resistance through chromosomal mutations of regulatory genes, and resistance through enzymatic drug modification, including both intrinsic and acquired resistance [50], [51], [52], [53], [54], [55]. P. aeruginosa can use these mechanisms in combination, to reach high-level of resistance to certain antibiotics, which is precisely what we observed in this study since we found an overexpression of two efflux systems (MexXY-OprM, MexAB-OprM, increased expression of the arnBCADTEF-PA3559 LPS modification genes, of the porins OprH, OprF, and decreased expression of OprE. The tripartite efflux pump MexXY-OprM is known to be a major contributor to aminoglycoside resistance in P. aeruginosa [56], [57], [58], [59] as well as the efflux of the drug tigecycline [60]. In addition, MexAB-OprM was also proven to confer resistance to β-lactams, fluoroquinolones [51] and to contribute to aminoglycosides resistance [61]. Both efflux pumps share the same efflux porin, OprM [62]. In P. aeruginosa, the PhoP-PhoQ two-component regulatory system is known to be induced upon Mg2+ starvation to up-regulate the production of the outer-membrane protein OprH and to increase the resistance to the polycationic antibiotic polymyxin B [63]. In addition, PhoP-PhoQ is also involved in resistance to antimicrobial cationic peptides and aminoglycoside antibiotics [45], in good agreement with the higher resistance of PAO-SCV to this class of antibiotics. The arnBCADTEF-PA3559 genes could also be involved in conferring a higher resistance to aminoglycosides. Intriguingly, PA3559 encodes a UDP-glucose dehydrogenase that is induced by low concentrations of Mg2+ [43] and its expression depends on the PmrA-PmrB two-component regulatory system, which is itself regulated by the PhoP-PhoQ two-component system [64], [65], [66], [67]. The general porin OprF, which is overexpressed in the SCV, has been recently shown to participate in resistance mechanisms to a broad spectrum of antibiotics such as β-lactams, cephalosporins, and fluoroquinolones [68], supporting the role of OprF as an intrinsic antibiotic resistance contributor as well as partially explaining the cefotaxime resistance revealed by phenotypic assays. Interestingly, we observed a down-regulation of the cytochrome c ccoN gene (PA4133), which was shown by others to be involved in small colony variant formation leading to higher resistance to aminoglycosides [69]. In PAO-SCV we observed a down-regulation of phage genes compared to the wild-type. The expression of phage genes has been shown to reciprocally associate with biofilm formation and antibiotic resistance as evidenced by Whiteley and colleagues [70]. These authors found that phages genes were up-regulated in mature biofilms as compared to planktonic cultures while they were shown to be down-regulated in biofilms exposed to the aminoglycoside tobramycin. A puzzling observation is the strong down-regulation in PAO-SCV of PQS biosynthesis genes and of the PQS-related efflux pump MexGHI-OpmD [39], which is confirmed by the near absence of PQS production presented in Figure 3. It has been suggested that high production of PQS, which results in an autolysis phenotype, could be explained by the induction of prophages [22], which fits with the results presented here.

Collectively, our whole genome expression analysis of PAO-SCV versus wild-type PAO1 allowed us to get a good correlation between phenotypic traits (antibiotic resistance, PQS and virulence factors production), proteomic and gene expression data. Intriguingly, we found only one mutation to be present in the pseudo-revertant only. This mutation could cause a higher translation rate of the tufA messenger RNA since it changed a rarely used GAU codon for aspartate to the more frequently used GAC. It is also interesting to notice that in the SCV proteome there was a strong decrease of the TufB protein (Figure 4). In Salmonella enterica, the translation factor Ef-Tu is also encoded by two paralogous genes, tufA and tufB, and it has been shown that mutations in tufB, which impair growth, can be rescued by compensatory mutations in the RNAse E gene or in tufA [71], situation similar to what we observed here. However, we have to be cautious before concluding that the tufA mutation is the only cause of the reversion associated with a higher growth rate. Our results do not explain either why the level of the TufB protein is strongly decreased in the SCV since no changes were found at the transcriptional level. We should not exclude either the possibility of a phenotypic switch without mutation, like the recently described bi-stable phenotypic switch due to the LysR regulator BexR [72]. Phenotypic switches have been described in Pseudomonas brassicacearum leading to two types of colonies with different abilities to colonize plant roots [73]. In rhizosphere pseudomonads phase variation is commonly observed and in one case the involvement of GacS/GacA two-component system as well as MutS and RpoS in phase variation has been described [74], [75], [76]. However, we could not detect any nucleotide change in gacA/gacS or in mutS and rpoS, which suggests that the switch has another, unidentified, origin and excluding the possibility that the SCV strain has a hypermutator phenotype.

In conclusion, our study demonstrates that the SCV phenotype could arise due to the accumulation of several mutations and that the resistance to aminoglycosides expressed by the SCV clone is multifactorial.

Methods

Bacterial strains and culture conditions

P. aeruginosa PAO1 Strain (ATTC 15692) and its gentamicin-resistant mutant PAO-SCV were used in this study. P. aeruginosa strains were grown at 37°C in Luria-Bertani (LB) broth or on LB agar plates, iron poor casamino acids (CAA) medium (Difco Laboratories) or Pseudomonas agar medium (Difco Laboratories). The antibiotics gentamicin (Gm) at 200 µg ml−1 and spectinomycin (Sp) at 50 µg ml−1 were used when necessary. Growth rate of three replicates for each strain was monitored spectrophotometrically (Bioscreen C, Thermo Labsystems).

Motility assay

Swarming, swimming and twitching motility were determined as previously described [77]. To investigate swarming motility, 4 µl of overnight cultures of P. aeruginosa grown in LB (1×109 cells) were placed in the center of 0.4% agar LB or CAA plates while swimming motility was evaluated using 0.3% agar LB or CAA plates. For twitching motility, LB or CAA plates containing 1.5% agar were inoculated with a toothpick by stabbing the plates. The plates were then incubated at 37°C. In the case of twitching, after incubation the LB- or CAA-agar media was removed from the plates and plates were stained with 1% crystal violet (Merck) in 33% acetic acid for a minimum of 20 min. Spreading of bacteria from the inoculation point was measured and pictures were taken. Three independent experiments were performed.

Detection and analysis of signal molecules

Rapid detection of N-acyl homoserine lactones (AHLs) in filter sterilized (0.2 µm pore-size filters) culture supernatants was done using AHL reporter plate bioassays by either E. coli JM109 carrying the plasmid pSB401 for the detection of N-(butanoyl)-L-homoserine lactone (C4-HSL) [78] or E. coli MH155 [79] for the detection of N-(3-oxododecanoyl)-L-homoserine lactone (3-oxo-C12-HSL). For accurate AHL quantification, 100 ml of acidified filter-sterilized culture supernatants were extracted with equal volumes of dichloromethane. The organic phase was removed and dried by evaporation in vacuum. Extracts were re-dissolved in 1 ml of 50% acetonitrile. Thin-layer chromatography (TLC) plates Silica gel 60 F254 (Merck) and RP-18 F245 (Merck) were used for detection of 3-oxo-C12-HSL and C4-HSL, respectively. Twenty µl of extracted AHLs were fractionated on TLC plates. After development in a solvent mixture of methanol/water (60[ratio]40, vol/vol), the plates were dried and overlaid with 50 ml of soft top LB-agar mixed with 1 ml of overnight culture of an E. coli MH155 strain harboring the reporter plasmid pUCP22NotI-PlasB::gfp(ASV)Plac ::lasR (to detect 3-oxo-C12-HSL) or 1 ml of overnight culture of E. coli JM109 pSB401 (to detect C4 and C6-HSL). After 18 h incubation at 37°C the production of 3-oxo-C12-HSL was detected under UV by visualization of green fluorescent spots and production of C4 and C6 was visible by the production of light.

The alkyl-hydroxy quinolones PQS and HHQ were extracted from 10 ml of early stationary phase filtered supernatants by adding equal volumes of acidified ethyl acetate. The organic phase was dried and the residue re-suspended in 50 ml methanol. Ten µl samples of this extract were spotted onto normal phase silica 60 F254 (Merck) TLC plates, pre-treated by soaking in 5% K2HPO4 for 30 min and activated at 100°C for 1 h. Extracts were separated using a dichloromethane:methanol (95[ratio]5, vol/vol) solvent system until the solvent front reached the top of the plate. PQS was visualised under UV light and specific detection was done using soft top LB-agar including 1 ml of overnight culture of a P. aeruginosa lecA::lux ΔpqsA strain as bioreporter [80]. Bioluminescence was detected and quantified with a Bio Imaging System (Syngene). Amounts of C4-HSL, C6-HSL, 3-oxo-C12-HSL and PQS were determined by measuring the diameter of the spots.

Persistence assay

The persistence assay was performed essentially as described previously [6]. Shortly, cultures were grown overnight at 37°C in 100 ml LB medium in Erlenmeyer flasks. One mL of a stationary phase culture was treated with 10 µL of ofloxacin at a final concentration of 5 µg mL−1; a control treatment was performed with sterile water. Both treatments were performed at 37°C, shaking at 200 rpm, during five hours, after which the number of colony forming units were determined by plate counts. The persister fraction is defined as the number of surviving cells after treatment with ofloxacin, divided by the number of cells after the control treatment. The relative persister fraction for each strain is the persister fraction of the strain divided by that of the wild-type. The mean relative persister fraction is calculated as the inverse logarithm of the mean of the logarithmic values of these relative persister fractions of separate experiments. The mean relative persister fractions are displayed with the bars representing the 25th and 75th percentiles as shown in Figure S1. Each experiment was repeated at least three times.

Proteome analysis by two-dimensional (2D) gel electrophoresis

P. aeruginosa cells were harvested in early stationary phase by centrifugation (4,000 g, 10 min, 4°C) and washed three times with Tris-HCl buffer (pH 8.0). To prepare extracts of cellular proteins, bacterial cells were washed twice in PBS buffer (pH 8.0) and re-suspended in a solution containing 40 ml of 2.5 mM Tris-HCl (pH 8.0), one tablet of protease inhibitor (Sigma), 80 µl of 0.5 M Na2EDTA, and 400 µl of DNase at 10 mg ml−1. After lysis of the cells by sonication with a Branson Sonifier 250, each suspension was centrifuged (2,500 g, 15 min, 4°C) to remove the cell debris and unbroken cells. The supernatant was then subjected to a second centrifugation (50,000 g, 40 min, 4°C) to remove the insoluble components, and the protein concentration in the resulting supernatant was determined by a Bradford Protein assay (Bio-Rad). Deoxycholic acid (sodium salt) was added to a final concentration of 0.2 mg ml−1. After 30 min of incubation on ice, the proteins were precipitated by addition of 6% (wt/vol) trichloroacetic acid and incubated at 4°C for 2 hr. After centrifugation (18,000 g, 30 min, 4°C) the precipitated proteins were re-suspended in distilled water, and eight volumes of cold acetone (−20°C) were added. After incubation at −20°C for 2 h, the mixture was centrifuged (3,500 g, 20 min, 4°C), and the pellet was allowed to dry for 5 min before it was dissolved in an appropriate amount of solubilisation buffer. After centrifugation (50,000 g, 40 min, 4°C) to remove the insoluble components, the protein concentration of the remaining supernatant was determined. Protein extracts were either used immediately for 2-D gel electrophoresis or stored at −80°C. Isoelectric focusing was performed with the IPGphor system and Immobiline DryStrip gel strips (GE Healthcare). Equal quantities of solubilised proteins from the different P. aeruginosa strains were diluted to obtain a final volume of 360 µl with solubilisation solution and applied to the Immobiline gel strips by in-gel rehydration. Linear immobilized pH gradients (pH 4 to 7) were used. Thirty to 50 µg of protein was applied for analytical gels (silver staining), and 200 to 500 µg of protein was loaded for Coomassie staining. After rehydration under silicone oil for 10 hr, the proteins were focused for a total of 120 kV/h at 20°C. The proteins were reduced by equilibration of the strips in equilibration solution (6 M urea, 30% glycerol, 2% [wt/vol] sodium dodecyl sulfate, and 1% [wt/vol] dithiothreitol in 0.05 M Tris-HCl (pH 8.8) for 15 min and then carbamidomethylated in the same solution containing 260 mM iodoacetamide for 15 min. The strips were transferred to 12% acrylamide gradient gels and electrophoresis was performed overnight at 125 V at 10°C. Gels were stained with Coomassie brilliant blue solution and spots of interest were then further analyzed through peptide mass fingerprinting according to the described protocol [81]. Peptides examined on a MALDI-TOF mass spectrometer (Bruker) and analyzed by MASCOT (Matrix Science) were used to identify proteins from peptide identifications using the NCBInr database.

Microarray and quantitative real time PCR analysis

Cultures were grown in triplicate until early (24 h) and late (48 h) stationary phase, respectively, allowing three biological replicates per condition. Total RNA was obtained from the cultures of early and late stationary phase by first treating the cells with RNAprotect Bacteria Reagent (Qiagen) as recommended by the manufacturer. Cells were then lysed and total RNA was extracted using the RNeasy Midi Kit (Qiagen), on-column DNase digestion was performed using the RNase-free DNase Set (Qiagen) according to the manufactures instructions. RNA integrity was assessed using the 2100 bioanalyzer (Agilent Technologies Inc.). cDNA synthesis, fragmentation and labeling were performed according to the supplier protocol for the P. aeruginosa Genechip genome array (Affymetrix) at 50°C. Washing and staining of the arrays was performed according to the manufacturer's instructions using a fluidics station 400 (Affymetrix). Slides were scanned using the 2500A GeneArray Scanner (Agilent Technologies Inc.) and Affymetrix MAS 5.0. Data analysis was performed using GeneSpring GX (Agilent Technologies Inc.) in which the scaled data was further normalized by per Chip and per Gene median normalizations. Filtering of genes was performed to find genes that had changed in expression by a magnitude of 2-fold (P value<0.05, Student's t-test).

Bacterial cells were harvested in stationary phase, bacterial RNA was extracted by using RNeasy Midi Kit (QIAGEN). The purity and concentration of the RNA was determined by spectrophotometry (NanoDrop, Thermo Scientific). First-strand cDNA was reverse transcribed from one microgram of total RNA by using First-strand cDNA Synthesis Kit (Amersham Biosciences, GE Healthcare). qRT-PCR was performed in a Bio-Rad (Hercules, CA, USA) iCycler with Bio-Rad iQ SYBR Green Supermix. For all primer sets, the following cycling parameters were used: 94°C for 3 min followed by 40 cycles of 94°C for 30 s, 55°C for 45 s and 72°C for 30 s, followed by 72°C for 7 min. The outer membrane lipoprotein oprI gene was used to normalize gene expression [82],[82],[82],[81]. Amplification products were electrophoresed on 0.8% agarose gels. For statistical analysis of relative gene expression, the 2[big up triangle, open][big up triangle, open]CT method was used [83]. All experiments were carried out in triplicate.

Whole genome sequencing

Genomic DNA was isolated from thawed pellets using the DNeasy Blood & Tissue Kit (Qiagen) according to the manufacturer's instructions. DNA samples were further prepared and sequenced using paired end sequencing in 76 cycles on a Genome Analyzer II–X (Illumina). Libraries of 300 bp were prepared according the manufacturer's instructions “Preparing Samples for Paired-End-Sequencing”. Cluster generation was performed using the Illumina cluster station, sequencing for read 1 and read 2 on the Genome Analyzer followed a standard protocol. The fluorescent images were processed to sequences using the Genome Analyzer Pipeline Analysis software 1.3.2 (Illumina).

Sequencing the DNA library resulted in a total number of 12,302,376 read pairs of 2×76 nt length. Reads were mapped to the reference genome of strain PAO1, which was obtained from the Pseudomonas genome database [84]. The 3′-ends of reads were trimmed using Perl script Trim.pl (http://bioinformatics.ucdavis.edu/index.php/Trim.pl) with the adaptive window option and a quality threshold of 10 to remove sequences of low read quality. Reads pairs containing one or two reads that were trimmed to a length less than 20 nt were discarded leaving 11,991,338 read pairs (97.5%). The free license version of Novoalign (www.novocraft.com) was used for mapping, because this software includes a gapped alignment algorithm, which improves the detection of indels [85]. 11,824,266 read pairs (96.1% of the original reads) were mapped to unique locations resulting in a median read depth (genome coverage) of 237. Single nucleotide polymorphisms (SNPs) and indels were detected using the MAQ software [86] using its built-in functions “cns2snp” and “indelpe”, respectively. Initially, only SNPs with a minimal consensus quality of 30 and indels that were supported by at least 50% of the reads overlapping the indel position were considered for further analysis as potential true positives. All positively filtered SNPs and indels were checked by visual inspection for correct base/indel calling. Genomic regions showing a read depth of less than 30 were also checked by visual inspection for the occurrence of larger indels that cannot be detected by the combination of Novoalign and MAQ alone. Furthermore, the confirmation of positives by Sanger sequencing has been performed and aligned with the sequences extracted from www.pseudomonas.com.

Additionally, the quality-trimmed read pairs were also assembled de novo using Velvet [87] yielding 1181 contigs with an n50 of 18.4 kbp and maximum length of 83.0 kbp. SNPs and indels were detected by aligning the contigs to the PAO1 reference genome using mummer3 [88].

Virulence assays

P. aeruginosa strains were grown overnight in LB medium to reach stationary phase at 37°C. After centrifugation, bacterial cells were washed with 10 mM MgSO4 and diluted to 108 CFU/ml and then 10 µl of this suspension were injected with a syringe into the main vein of Belgian chicory (Cichorium intybus) leaves. The leaves were placed on dishes containing a Whatman filter impregnated with sterilized water. The plates were kept in an incubator at 37°C and watery rot symptoms monitored daily for 72 h.

Virulence was also tested on third instar larvae of the fruitfly (Drosophila melanogaster). Larvae were pricked with needles dipped in a concentrated suspension of P. aeruginosa strains grown in LB medium with an absorbance of 1.0 at 600 nm. After pricking, the larvae were placed in plates containing a piece of paper impregnated with 10% sugar water. For each condition 20 plates with 10 larvae were used. During 8 h, the number of dead larvae was counted every 30 min. A final count was held 24 h after initial pricking. Data were statistically analyzed by means of repeated measures ANOVA.

GEO Accession Number: GSE34141

Supporting Information

Figure S1

Mean relative persister fraction of wild-type PAO1 and PAO-SCV after exposure to ofloxacin. See text for details.

(TIF)

Figure S2

Production of virulence factors pyocyanin, pyoverdine, and elastase (A); motility of wild type PAO1 and PAO-SCV on swimming (top), swarming (middle), and twitching (bottom) plates (B); atomic force microscopy images of wild-type and SCV showing the loss of flagella (C).

(TIF)

Figure S3

Virulence of wild-type, PAO-SCV and revertant (A) in plants (Cychorium intybus) and (B) of wild-type and PAO-SCV in Drosophila melanogaster larvae.

(TIF)

Table S1

Number of differentially expressed genes between P. aeruginosa PAO-SCV and its clonal wild-type PAO1 during stationary phase.

(DOC)

Table S2

Functional classification of differentially expressed genes between P. aeruginosa PAO-SCV and its clonal wild-type PAO1 during stationary phase.

(DOC)

Table S3

Up-regulated genes in P. aeruginosa PAO-SCV compared to its clonal wild-type PAO1 during stationary phase.

(DOC)

Table S4

Down-regulated genes in P. aeruginosa PAO-SCV compared to its clonal wild-type PAO1 during stationary phase.

(DOC)

Table S5

Differential expression of selected genes in P. aeruginosa PAO-SCV compared to its clonal wild-type PAO1 during stationary phase.

(DOC)

Acknowledgments

We wish to thank Dr. Jean Paul Pirnay, Daniel De Vos, and Florence Bilocq from the Military hospital in Brussels for the VITEK analysis. We would like to thank De Meuter Sonja and Van Hemelrijck Herman for technical assistance.

Footnotes

Competing Interests: The authors have declared that no competing interests exist.

Funding: This work was supported by two research grants from the FWO Belgium (Fonds voor Wetenschappelijk Onderzoek Vlaanderen) and a grant from BBSRC UK (BBF0143921). QW has a CSC-VUB scholarship and got a FEMS student fellowship to perform experiments in the lab of Dr. Susanne Haüssler. BB and AB are research assistants and PV is a postdoctoral researcher of the FWO-Vlaanderen. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1. Stover CK, Pham XQ, Erwin AL, Mizoguchi SD, Warrener P, et al. Complete genome sequence of Pseudomonas aeruginosa PAO1, an opportunistic pathogen. Nature. 2000;406:959–964. [PubMed]
2. Lyczak JB, Cannon CL, Pier GB. Lung infections associated with cystic fibrosis. Clin Microbiol Rev. 2002;15:194–222. [PMC free article] [PubMed]
3. Lyczak JB, Cannon CL, Pier GB. Establishment of Pseudomonas aeruginosa infection: lessons from a versatile opportunist. Microbes Infect. 2000;2:1051–1060. [PubMed]
4. Boles BR, Thoendel M, Singh PK. Self-generated diversity produces “insurance effects” in biofilm communities. Proc Natl Acad Sci U S A. 2004;101:16630–16635. [PubMed]
5. Govan JR, Deretic V. Microbial pathogenesis in cystic fibrosis: mucoid Pseudomonas aeruginosa and Burkholderia cepacia. Microbiol Rev. 1996;60:539–574. [PMC free article] [PubMed]
6. De Groote VN, Verstraeten N, Fauvart M, Kint CI, Verbeeck AM, et al. Novel persistence genes in Pseudomonas aeruginosa identified by high-throughput screening. FEMS Microbiol Lett. 2009;297:73–79. [PubMed]
7. Moker N, Dean CR, Tao J. Pseudomonas aeruginosa increases formation of multidrug-tolerant persister cells in response to quorum-sensing signaling molecules. J Bacteriol. 2010;192:1946–1955. [PMC free article] [PubMed]
8. Haussler S, Tummler B, Weissbrodt H, Rohde M, Steinmetz I. Small-colony variants of Pseudomonas aeruginosa in cystic fibrosis. Clin Infect Dis. 1999;29:621–625. [PubMed]
9. Haussler S, Ziegler I, Lottel A, von Gotz F, Rohde M, et al. Highly adherent small-colony variants of Pseudomonas aeruginosa in cystic fibrosis lung infection. J Med Microbiol. 2003;52:295–301. [PubMed]
10. von Gotz F, Haussler S, Jordan D, Saravanamuthu SS, Wehmhoner D, et al. Expression analysis of a highly adherent and cytotoxic small colony variant of Pseudomonas aeruginosa isolated from a lung of a patient with cystic fibrosis. J Bacteriol. 2004;186:3837–3847. [PMC free article] [PubMed]
11. Kirisits MJ, Prost L, Starkey M, Parsek MR. Characterization of colony morphology variants isolated from Pseudomonas aeruginosa biofilms. Appl Environ Microbiol. 2005;71:4809–4821. [PMC free article] [PubMed]
12. Starkey M, Hickman JH, Ma L, Zhang N, De Long S, et al. Pseudomonas aeruginosa rugose small-colony variants have adaptations that likely promote persistence in the cystic fibrosis lung. J Bacteriol. 2009;191:3492–3503. [PMC free article] [PubMed]
13. Proctor RA, von Eiff C, Kahl BC, Becker K, McNamara P, et al. Small colony variants: a pathogenic form of bacteria that facilitates persistent and recurrent infections. Nat Rev Microbiol. 2006;4:295–305. [PubMed]
14. Deziel E, Comeau Y, Villemur R. Initiation of biofilm formation by Pseudomonas aeruginosa 57RP correlates with emergence of hyperpiliated and highly adherent phenotypic variants deficient in swimming, swarming, and twitching motilities. J Bacteriol. 2001;183:1195–1204. [PMC free article] [PubMed]
15. Drenkard E, Ausubel FM. Pseudomonas biofilm formation and antibiotic resistance are linked to phenotypic variation. Nature. 2002;416:740–743. [PubMed]
16. Nelson LK, Stanton MM, Elphinstone RE, Helwerda J, Turner RJ, et al. Phenotypic diversification in vivo.: Pseudomonas aeruginosa gacS- strains generate small colony variants in vivo that are distinct from in vitro variants. Microbiology. 2010. [PubMed]
17. Tarighi S, Wei Q, Camara M, Williams P, Fletcher MP, et al. The PA4204 gene encodes a periplasmic gluconolactonase (PpgL) which is important for fitness of Pseudomonas aeruginosa. Microbiology. 2008;154:2979–2990. [PubMed]
18. Malone JG, Jaeger T, Spangler C, Ritz D, Spang A, et al. YfiBNR mediates cyclic di-GMP dependent small colony variant formation and persistence in Pseudomonas aeruginosa. PLoS Pathog. 2010;6:e1000804. [PMC free article] [PubMed]
19. Schneider M, Muhlemann K, Droz S, Couzinet S, Casaulta C, et al. Clinical characteristics associated with isolation of small-colony variants of Staphylococcus aureus and Pseudomonas aeruginosa from respiratory secretions of patients with cystic fibrosis. J Clin Microbiol. 2008;46:1832–1834. [PMC free article] [PubMed]
20. Mulcahy LR, Burns JL, Lory S, Lewis K. Emergence of Pseudomonas aeruginosa strains producing high levels of persister cells in patients with cystic fibrosis. J Bacteriol. 2010. [PMC free article] [PubMed]
21. Hengge R. Principles of c-di-GMP signalling in bacteria. Nat Rev Microbiol. 2009;7:263–273. [PubMed]
22. D'Argenio DA, Calfee MW, Rainey PB, Pesci EC. Autolysis and autoaggregation in Pseudomonas aeruginosa colony morphology mutants. J Bacteriol. 2002;184:6481–6489. [PMC free article] [PubMed]
23. Hickman JW, Tifrea DF, Harwood CS. A chemosensory system that regulates biofilm formation through modulation of cyclic diguanylate levels. Proc Natl Acad Sci U S A. 2005;102:14422–14427. [PubMed]
24. Kuchma SL, Brothers KM, Merritt JH, Liberati NT, Ausubel FM, et al. BifA, a cyclic-Di-GMP phosphodiesterase, inversely regulates biofilm formation and swarming motility by Pseudomonas aeruginosa PA14. J Bacteriol. 2007;189:8165–8178. [PMC free article] [PubMed]
25. Kulasekara BR, Kulasekara HD, Wolfgang MC, Stevens L, Frank DW, et al. Acquisition and evolution of the exoU locus in Pseudomonas aeruginosa. J Bacteriol. 2006;188:4037–4050. [PMC free article] [PubMed]
26. Meissner A, Wild V, Simm R, Rohde M, Erck C, et al. Pseudomonas aeruginosa cupA-encoded fimbriae expression is regulated by a GGDEF and EAL domain-dependent modulation of the intracellular level of cyclic diguanylate. Environ Microbiol. 2007;9:2475–2485. [PubMed]
27. Friedman L, Kolter R. Two genetic loci produce distinct carbohydrate-rich structural components of the Pseudomonas aeruginosa biofilm matrix. J Bacteriol. 2004;186:4457–4465. [PMC free article] [PubMed]
28. Jackson KD, Starkey M, Kremer S, Parsek MR, Wozniak DJ. Identification of psl, a locus encoding a potential exopolysaccharide that is essential for Pseudomonas aeruginosa PAO1 biofilm formation. J Bacteriol. 2004;186:4466–4475. [PMC free article] [PubMed]
29. Matsukawa M, Greenberg EP. Putative exopolysaccharide synthesis genes influence Pseudomonas aeruginosa biofilm development. J Bacteriol. 2004;186:4449–4456. [PMC free article] [PubMed]
30. Williams P. Quorum sensing. Int J Med Microbiol. 2006;296:57–59. [PubMed]
31. Venturi V. Regulation of quorum sensing in Pseudomonas. FEMS Microbiol Rev. 2006;30:274–291. [PubMed]
32. Diggle SP, Cornelis P, Williams P, Camara M. 4-quinolone signalling in Pseudomonas aeruginosa: old molecules, new perspectives. Int J Med Microbiol. 2006;296:83–91. [PubMed]
33. Nestorovich EM, Sugawara E, Nikaido H, Bezrukov SM. Pseudomonas aeruginosa porin OprF: properties of the channel. J Biol Chem. 2006;281:16230–16237. [PMC free article] [PubMed]
34. Yoon SS, Hennigan RF, Hilliard GM, Ochsner UA, Parvatiyar K, et al. Pseudomonas aeruginosa anaerobic respiration in biofilms: relationships to cystic fibrosis pathogenesis. Dev Cell. 2002;3:593–603. [PubMed]
35. Yamano Y, Nishikawa T, Komatsu Y. Cloning and nucleotide sequence of anaerobically induced porin protein E1 (OprE) of Pseudomonas aeruginosa PAO1. Mol Microbiol. 1993;8:993–1004. [PubMed]
36. Yamano Y, Nishikawa T, Komatsu Y. Involvement of the RpoN protein in the transcription of the oprE gene in Pseudomonas aeruginosa. FEMS Microbiol Lett. 1998;162:31–37. [PubMed]
37. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci U S A. 1998;95:14863–14868. [PubMed]
38. Llanes C, Hocquet D, Vogne C, Benali-Baitich D, Neuwirth C, et al. Clinical strains of Pseudomonas aeruginosa overproducing MexAB-OprM and MexXY efflux pumps simultaneously. Antimicrob Agents Chemother. 2004;48:1797–1802. [PMC free article] [PubMed]
39. Aendekerk S, Diggle SP, Song Z, Hoiby N, Cornelis P, et al. The MexGHI-OpmD multidrug efflux pump controls growth, antibiotic susceptibility and virulence in Pseudomonas aeruginosa via 4-quinolone-dependent cell-to-cell communication. Microbiology. 2005;151:1113–1125. [PubMed]
40. Aendekerk S, Ghysels B, Cornelis P, Baysse C. Characterization of a new efflux pump, MexGHI-OpmD, from Pseudomonas aeruginosa that confers resistance to vanadium. Microbiology. 2002;148:2371–2381. [PubMed]
41. Dietrich LE, Price-Whelan A, Petersen A, Whiteley M, Newman DK. The phenazine pyocyanin is a terminal signalling factor in the quorum sensing network of Pseudomonas aeruginosa. Mol Microbiol. 2006;61:1308–1321. [PubMed]
42. Trent MS, Ribeiro AA, Lin S, Cotter RJ, Raetz CR. An inner membrane enzyme in Salmonella and Escherichia coli that transfers 4-amino-4-deoxy-L-arabinose to lipid A: induction on polymyxin-resistant mutants and role of a novel lipid-linked donor. J Biol Chem. 2001;276:43122–43131. [PubMed]
43. Hung RJ, Chien HS, Lin RZ, Lin CT, Vatsyayan J, et al. Comparative analysis of two UDP-glucose dehydrogenases in Pseudomonas aeruginosa PAO1. J Biol Chem. 2007;282:17738–11748. [PubMed]
44. Poon KK, Westman EL, Vinogradov E, Jin S, Lam JS. Functional characterization of MigA and WapR: putative rhamnosyltransferases involved in outer core oligosaccharide biosynthesis of Pseudomonas aeruginosa. J Bacteriol. 2008;190:1857–1865. [PMC free article] [PubMed]
45. Macfarlane EL, Kwasnicka A, Hancock RE. Role of Pseudomonas aeruginosa PhoP-phoQ in resistance to antimicrobial cationic peptides and aminoglycosides. Microbiology. 2000;146:2543–2554. [PubMed]
46. Tamber S, Ochs MM, Hancock RE. Role of the novel OprD family of porins in nutrient uptake in Pseudomonas aeruginosa. J Bacteriol. 2006;188:45–54. [PMC free article] [PubMed]
47. Davey ME, Caiazza NC, O'Toole GA. Rhamnolipid surfactant production affects biofilm architecture in Pseudomonas aeruginosa PAO1. J Bacteriol. 2003;185:1027–1036. [PMC free article] [PubMed]
48. Caiazza NC, Shanks RM, O'Toole GA. Rhamnolipids modulate swarming motility patterns of Pseudomonas aeruginosa. J Bacteriol. 2005;187:7351–7361. [PMC free article] [PubMed]
49. Klockgether J, Munder A, Neugebauer J, Davenport CF, Stanke F, et al. Genome diversity of Pseudomonas aeruginosa PAO1 laboratory strains. J Bacteriol. 2010;192:1113–1121. [PMC free article] [PubMed]
50. Poole K. Outer membranes and efflux: the path to multidrug resistance in Gram-negative bacteria. Curr Pharm Biotechnol. 2002;3:77–98. [PubMed]
51. Alekshun MN, Levy SB. Molecular mechanisms of antibacterial multidrug resistance. Cell. 2007;128:1037–1050. [PubMed]
52. Gooderham WJ, Hancock RE. Regulation of virulence and antibiotic resistance by two-component regulatory systems in Pseudomonas aeruginosa. FEMS Microbiol Rev. 2009;33:279–294. [PubMed]
53. Mingeot-Leclercq MP, Glupczynski Y, Tulkens PM. Aminoglycosides: activity and resistance. Antimicrob Agents Chemother. 1999;43:727–737. [PMC free article] [PubMed]
54. Poole K. Efflux-mediated antimicrobial resistance. J Antimicrob Chemother. 2005;56:20–51. [PubMed]
55. Poole K. Aminoglycoside resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother. 2005;49:479–487. [PMC free article] [PubMed]
56. Aires JR, Kohler T, Nikaido H, Plesiat P. Involvement of an active efflux system in the natural resistance of Pseudomonas aeruginosa to aminoglycosides. Antimicrob Agents Chemother. 1999;43:2624–2628. [PMC free article] [PubMed]
57. Hocquet D, Vogne C, El Garch F, Vejux A, Gotoh N, et al. MexXY-OprM efflux pump is necessary for a adaptive resistance of Pseudomonas aeruginosa to aminoglycosides. Antimicrob Agents Chemother. 2003;47:1371–1375. [PMC free article] [PubMed]
58. Sobel ML, McKay GA, Poole K. Contribution of the MexXY multidrug transporter to aminoglycoside resistance in Pseudomonas aeruginosa clinical isolates. Antimicrob Agents Chemother. 2003;47:3202–3207. [PMC free article] [PubMed]
59. Vogne C, Aires JR, Bailly C, Hocquet D, Plesiat P. 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. 2004;48:1676–1680. [PMC free article] [PubMed]
60. Dean CR, Visalli MA, Projan SJ, Sum PE, Bradford PA. Efflux-mediated resistance to tigecycline (GAR-936) in Pseudomonas aeruginosa PAO1. Antimicrob Agents Chemother. 2003;47:972–978. [PMC free article] [PubMed]
61. 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. [PMC free article] [PubMed]
62. Masuda N, Sakagawa E, Ohya S, Gotoh N, Tsujimoto H, et al. Contribution of the MexX-MexY-oprM efflux system to intrinsic resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother. 2000;44:2242–2246. [PMC free article] [PubMed]
63. Macfarlane EL, Kwasnicka A, Ochs MM, Hancock RE. PhoP-PhoQ homologues in Pseudomonas aeruginosa regulate expression of the outer-membrane protein OprH and polymyxin B resistance. Mol Microbiol. 1999;34:305–316. [PubMed]
64. Kwon DH, Lu CD. Polyamines induce resistance to cationic peptide, aminoglycoside, and quinolone antibiotics in Pseudomonas aeruginosa PAO1. Antimicrob Agents Chemother. 2006;50:1615–1622. [PMC free article] [PubMed]
65. McPhee JB, Lewenza S, Hancock RE. 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. 2003;50:205–217. [PubMed]
66. Gunn JS, Miller SI. PhoP-PhoQ activates transcription of pmrAB, encoding a two-component regulatory system involved in Salmonella typhimurium antimicrobial peptide resistance. J Bacteriol. 1996;178:6857–6864. [PMC free article] [PubMed]
67. Gooderham WJ, Gellatly SL, Sanschagrin F, McPhee JB, Bains M, et al. The sensor kinase PhoQ mediates virulence in Pseudomonas aeruginosa. Microbiology. 2009;155:699–711. [PubMed]
68. Dotsch A, Becker T, Pommerenke C, Magnowska Z, Jansch L, et al. Genomewide identification of genetic determinants of antimicrobial drug resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother. 2009;53:2522–2531. [PMC free article] [PubMed]
69. Schurek KN, Marr AK, Taylor PK, Wiegand I, Semenec L, et al. Novel genetic determinants of low-level aminoglycoside resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother. 2008;52:4213–4219. [PMC free article] [PubMed]
70. Whiteley M, Bangera MG, Bumgarner RE, Parsek MR, Teitzel GM, et al. Gene expression in Pseudomonas aeruginosa biofilms. Nature. 2001;413:860–864. [PubMed]
71. Hammarlof DL, Hughes D. Mutants of the RNA-processing enzyme RNase E reverse the extreme slow-growth phenotype caused by a mutant translation factor EF-Tu. Mol Microbiol. 2008;70:1194–1209. [PubMed]
72. Turner KH, Vallet-Gely I, Dove SL. Epigenetic control of virulence gene expression in Pseudomonas aeruginosa by a LysR-type transcription regulator. PLoS Genet. 2009;5:e1000779. [PMC free article] [PubMed]
73. Achouak W, Conrod S, Cohen V, Heulin T. Phenotypic variation of Pseudomonas brassicacearum as a plant root-colonization strategy. Mol Plant Microbe Interact. 2004;17:872–879. [PubMed]
74. van den Broek D, Chin AWTF, Bloemberg GV, Lugtenberg BJ. Role of RpoS and MutS in phase variation of Pseudomonas sp. PCL1171. Microbiology. 2005;151:1403–1408. [PubMed]
75. van den Broek D, Chin AWTF, Bloemberg GV, Lugtenberg BJ. Molecular nature of spontaneous modifications in gacS which cause colony phase variation in Pseudomonas sp. strain PCL1171. J Bacteriol. 2005;187:593–600. [PMC free article] [PubMed]
76. Sanchez-Contreras M, Martin M, Villacieros M, O'Gara F, Bonilla I, et al. Phenotypic selection and phase variation occur during alfalfa root colonization by Pseudomonas fluorescens F113. J Bacteriol. 2002;184:1587–1596. [PMC free article] [PubMed]
77. Kinscherf TG, Willis DK. Swarming by Pseudomonas syringae B728a requires gacS (lemA) and gacA but not the acyl-homoserine lactone biosynthetic gene ahlI. J Bacteriol. 1999;181:4133–4136. [PMC free article] [PubMed]
78. Swift S, Karlyshev AV, Fish L, Durant EL, Winson MK, et al. Quorum sensing in Aeromonas hydrophila and Aeromonas salmonicida: identification of the LuxRI homologs AhyRI and AsaRI and their cognate N-acylhomoserine lactone signal molecules. J Bacteriol. 1997;179:5271–5281. [PMC free article] [PubMed]
79. Hentzer M, Riedel K, Rasmussen TB, Heydorn A, Andersen JB, et al. Inhibition of quorum sensing in Pseudomonas aeruginosa biofilm bacteria by a halogenated furanone compound. Microbiology. 2002;148:87–102. [PubMed]
80. Fletcher MP, Diggle SP, Camara M, Williams P. Biosensor-based assays for PQS, HHQ and related 2-alkyl-4-quinolone quorum sensing signal molecules. Nat Protoc. 2007;2:1254–1262. [PubMed]
81. Shevchenko A, Tomas H, Havlis J, Olsen JV, Mann M. In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nat Protoc. 2006;1:2856–2860. [PubMed]
82. Cornelis P, Bouia A, Belarbi A, Guyonvarch A, Kammerer B, et al. Cloning and analysis of the gene for the major outer membrane lipoprotein from Pseudomonas aeruginosa. Mol Microbiol. 1989;3:421–428. [PubMed]
83. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25:402–408. [PubMed]
84. Winsor GL, Van Rossum T, Lo R, Khaira B, Whiteside MD, et al. Pseudomonas Genome Database: facilitating user-friendly, comprehensive comparisons of microbial genomes. Nucleic Acids Res. 2009;37:D483–488. [PMC free article] [PubMed]
85. Krawitz P, Rodelsperger C, Jager M, Jostins L, Bauer S, et al. Microindel detection in short-read sequence data. Bioinformatics. 2010;26:722–729. [PubMed]
86. Li H, Ruan J, Durbin R. Mapping short DNA sequencing reads and calling variants using mapping quality scores. Genome Res. 2008;18:1851–1858. [PubMed]
87. Zerbino DR, Birney E. Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res. 2008;18:821–829. [PubMed]
88. Kurtz S, Phillippy A, Delcher AL, Smoot M, Shumway M, et al. Versatile and open software for comparing large genomes. Genome Biol. 2004;5:R12. [PMC free article] [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science