• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
Infect Immun. Nov 2007; 75(11): 5313–5324.
Published online Aug 27, 2007. doi:  10.1128/IAI.01807-06
PMCID: PMC2168270

In Vivo Evidence of Pseudomonas aeruginosa Nutrient Acquisition and Pathogenesis in the Lungs of Cystic Fibrosis Patients[down-pointing small open triangle]

Abstract

One of the hallmarks of Pseudomonas aeruginosa infection in cystic fibrosis (CF) patients is very-high-cell-density (HCD) replication in the lung, allowing this bacterium to induce virulence controlled by the quorum-sensing systems. However, the nutrient sources sustaining HCD replication in this chronic infection are largely unknown. Here, we performed microarray studies of P. aeruginosa directly isolated from the lungs of CF patients to demonstrate its metabolic capability and virulence in vivo. In vivo microarray data, confirmed by real-time reverse transcription-PCR, indicated that the P. aeruginosa population expressed several genes for virulence, drug resistance, and utilization of multiple nutrient sources (lung surfactant lipids and amino acids) contributing to HCD replication. The most abundant lung surfactant lipid molecule, phosphatidylcholine (PC), induces key genes of P. aeruginosa pertinent to PC degradation in vitro as well as in vivo within the lungs of CF patients. The results support recent research indicating that P. aeruginosa exists in the lungs of CF patients as a diverse population with full virulence potential. The data also indicate that there is deregulation of several pathways, suggesting that there is in vivo evolution by deregulation of a large portion of the transcriptome during chronic infection in CF patients. To our knowledge, this is the first in vivo transcriptome analysis of P. aeruginosa in a natural infection in CF patients, and the results indicate several important aspects of P. aeruginosa pathogenesis, drug resistance, nutrient utilization, and general metabolism within the lungs of CF patients.

Pseudomonas aeruginosa is the major cause of morbidity and mortality in lung diseases, including cystic fibrosis (CF) (6, 11, 32) and nosocomial pneumonia (3, 40). Over 93% of CF patients between 18 and 24 years old have been reported to have P. aeruginosa infections (11). In addition, nosocomial pneumonia is the second most common of all nosocomial infections, and P. aeruginosa was the most frequently isolated microbe involved from 1992 to 1997 (37). The pathogenesis of this organism has been intensively studied with respect to virulence and virulence expression (5, 26, 31, 36, 39), biofilm production (10, 43), and quorum sensing (15, 27, 28, 30). Several virulence factors that P. aeruginosa expresses (e.g., exotoxin A, exoenzyme S, cytotoxin, proteases, lipases, phospholipases, alginate, and hydrogen cyanide) all contribute to severe lung damage. High-cell-density (HCD) replication is necessary for many of these events to occur, and >109 bacteria/ml of sputum have been found in the lungs of CF patients (43, 46). The ability of P. aeruginosa to obtain nutrients in the lung for HCD replication and maintenance is the quintessential factor leading to quorum-sensing-induced virulence expression, which is a hallmark of chronic lung infections in CF patients. However, the nutrient requirements of P. aeruginosa in vivo are unknown. A recent in vitro study by Palmer et al. (25), in which P. aeruginosa was grown on minimal medium supplemented with lyophilized sputum, suggested that one of the nutrient sources in the lung is amino acids. Therefore, because the enigmatic nutrient source(s) in vivo has long been ignored and until now has been presumptive, an area of research that should be explored is the metabolic capability of this organism with respect to nutrient acquisition in the lungs of CF patients. Understanding the nutrient requirements that contribute to HCD in the lungs of CF patients may yield new insights into the pathogenesis and potential treatment of infections in CF patients caused by P. aeruginosa.

Recovery of an HCD P. aeruginosa population from the lungs of CF patients presents a unique opportunity to analyze the transcriptome of this organism in vivo. The primary focus of our approach was to reveal bacterial in vivo gene expression prior to collection of expectorate from a patient. Specifically, we examined bacteria immediately after they were collected from fresh sputum samples in order to decipher their metabolic activities in vivo. This was achieved by isolating bacterial mRNA immediately after a single expectoration from a 42-year-old CF patient (>109 bacteria/ml of sputum). The sample was used for microarray and real-time reverse transcription (RT)-PCR, and the results of gene expression profiling experiments were compared to the results for a clinical isolate pool from the same patient grown in vitro in 1× M9 medium containing citrate. This comparison, performed using Affymetrix P. aeruginosa GeneChips, revealed hundreds of genes that were induced ≥2-fold (P ≤ 0.05) in vivo compared to in vitro growth. Constitutively expressed genes were also observed by comparing the expression profile of the clinical isolate pool to that of PAO1, both grown in 1× M9 medium containing citrate. In this report, our data provide evidence showing that P. aeruginosa as a population exhibited a specific repertoire of highly expressed genes related to virulence, drug resistance, and nutrient utilization in vivo.

MATERIALS AND METHODS

Media and bacterial culturing conditions.

All in vitro studies were carried out using P. aeruginosa laboratory strain PAO1 and the clinical isolate pools. In this study, we utilized minimal medium with a defined carbon source (e.g., citrate) for comparisons to the lung environment rather than a rich medium, because the nutrient source metabolized by P. aeruginosa in the lung may also be present in rich medium (e.g., amino acid in LB medium) and therefore may abolish differential expression of genes essential for nutrient utilization in vivo. This medium was amenable to the identification of P. aeruginosa pathways induced in the lung that are not necessarily “on” when a single defined carbon source is metabolized. To obtain cultures for RNA isolation, starter cultures of PAO1 or the clinical isolate pools were grown in Pseudomonas isolation broth (Difco) overnight and inoculated (1:200 dilution) into 200 ml of 1× M9 minimal medium (38) with 0.4% (wt/vol) phosphatidylcholine (PC), 0.4% (wt/vol) C16:0, or 20 mM citrate and grown at 37°C with shaking at 225 rpm; in addition, all media contained 0.2% Brij 58 (Sigma, St. Louis, MO). When cells reached mid-log phase (optical density at 540 nm [OD540], ~0.6), cells were harvested for RNA isolation as described below.

Bacterial RNA isolation from sputum and bacterial culture.

Harvested cell cultures described above and 30 ml of a fresh sputum sample (single expectoration with >109 CFU/ml sputum) collected from a 42-year-old CF patient were kept on ice and processed immediately as follows. The approach and method used for sputum collection were approved by the Institutional Review Board Committee on Human Studies at the University of Hawaii at Manoa, and full informed consent was obtained from the patient. The cell pellets from cultures and the fresh sputum sample were treated on ice for 20 min with an equal volume of Sputolysin reagent (Calbiochem, La Jolla, CA) plus Sigma DNase I (10 U/ml of cell suspension) and proteinase K (600 μg/ml of cell suspension) with intermittent mixing to remove extracellular chromosomal DNA and proteins. Samples were then centrifuged at 8,000 × g, and the cell pellets were washed twice with ice-cold sterile double-distilled water (DDW) to lyse eukaryotic cells and remove soluble cell debris in the sputum. For the sputum sample, an aliquot of the washed cells was serially diluted and plated on no-salt Luria-Bertani medium (Teknova, Hollister, CA) and Pseudomonas isolation agar (Difco) to obtain a clinical isolate pool and for determining bacterial counts. The use of a clinical isolate pool, as opposed to a single clinical isolate, eliminated the possibility of any bias towards a particular isolate, and the pool was more representative of the P. aeruginosa population in vivo. Total RNA was isolated from P. aeruginosa using the established protocol of Stephen Lory (http://cfgenomics.unc.edu/protocols_rna_prep.htm), with a few minor modifications. Briefly, cells were harvested at 4°C and then resuspended and sonicated in Trizol reagent (Invitrogen, Carlsbad, CA) and treated with chloroform. An additional phenol-chloroform extraction was performed, and total nucleic acids were precipitated with ethanol. DNA was digested with DNase I (Promega, Madison, WI), and the RNA was phenol-chloroform extracted, ethanol precipitated, and resuspended in diethyl pyrocarbonate (DEPC)-treated DDW. tRNA was removed using QIAgen RNeasy purification columns as recommended by the manufacturer (QIAGEN, Valencia, CA), with an additional on-column DNase I digestion. The final RNA concentrations and purities were determined with a Beckman DU7500 spectrophotometer (A260/A280, between 1.8 and 2.0). The RNA yield for the sputum sample was ~55 μg and was sufficient for performing up to three GeneChip analyses and several real-time PCR runs. In our experience, the volume of sputum was not as critical as the P. aeruginosa cell counts in the sputum for successful microarray and real-time RT-PCR experiments. In particular, we ensured that processed sputum had an HCD (≥109 CFU/ml of sputum).

cDNA synthesis, labeling, and hybridization and microarray data analyses.

cDNA was synthesized, fragmented, labeled, and processed as recommended by Affymetrix (Affymetrix, Santa Clara, CA). Hybridization to P. aeruginosa GeneChips was performed at the Greenwood Biotechnology Facility, University of Hawaii. Raw data were obtained using the Affymetrix GeneChip Operating System 1.4 software (Affymetrix, Santa Clara, CA). The tab-delimited files were then imported into the GeneSpring 7.0 software (Agilent Technologies, Redwood City, CA) for further analysis. Analysis was done by conducting pairwise comparisons between duplicate or triplicate GeneChips for two conditions at a time (i.e., PAO1 grown in PC versus PAO1 grown in citrate, PAO1 grown in C16:0 versus PAO1 grown in citrate, in vivo sputum versus in vitro-grown clinical pool, in vitro-grown clinical pool versus PAO1 grown in citrate). The list of genes was subjected to statistical analysis using analysis of variance (ANOVA) and including the two-tailed Student t test, and only significant expression data (P ≤ 0.05) were kept. The gene list was further analyzed based on fold changes, where only genes showing a change of twofold or greater were kept, and all hypothetical genes were removed from the final list. Fold change values were averaged for nine independent pairwise comparisons (three GeneChips per in vitro condition) with P values of ≤0.05. However, for in vivo conditions, fold change values were averaged for six independent pairwise comparisons (two GeneChips for in vivo conditions and three GeneChips for in vitro conditions) with P values of ≤0.05. Average expression levels of a few selected genes (glp genes, mexY, and plcH) that did not pass the statistical stringency test (P ≤ 0.05) were analyzed individually by determining the relative expression level for each pairwise comparison, and the average expression levels were calculated (see Fig. Fig.22).

FIG. 2.
Average fold changes for pairwise comparisons of glp genes, plcH, and mexY, which were not detected due to the statistical stringency (P ≤ 0.05) imposed on the GeneChip data for glp genes (Table (Table4)4) and plcH and mexY (see Table ...

Gene description and pathway designation.

Gene descriptions and pathway designations were assigned according to the Kyoto Encyclopedia of Genes and Genomes (http://www.genome.ad.jp/kegg/) (24).

cDNA synthesis for real-time RT-PCR.

Three micrograms of the same purified mRNA for each condition used for microarrays was used for cDNA synthesis. An iScript cDNA synthesis kit was used as recommended by the manufacturer (Bio-Rad, Hercules, CA). A “no reverse-transcriptase” control was included for each sample during cDNA synthesis to ensure that there was no DNA carryover contamination from the RNA isolation. Final volumes were brought to 1,400 μl with DDW (with no DEPC), and 10 μl was used for each real-time PCR as described below.

Bacterial mRNA processing of a sample from a second CF patient for real-time RT-PCR.

A sputum sample was obtained from a young adult (an 18-year-old patient). The total volume of sputum was approximately 2 ml, and the P. aeruginosa plate count was 2.3 × 108 CFU/ml. Total RNA was initially isolated as described above, but the subsequent cDNA synthesis procedure was slightly modified as follows. The total RNA on a QIAGEN RNeasy column was eluted in 30 μl of DEPC-treated DDW and protected with 40 U of rRNasin (Promega, Madison, WI). A 2.75-μg aliquot of RNA was treated with 2 U of RQ1 RNase-free DNase I (Promega) for a second time in 2 μl of iScript cDNA synthesis buffer (Bio-Rad) for 10 min at 37°C. The DNase I was inactivated at 70°C for 10 min, and the sample was chilled on ice. Then 14 μl of iScript cDNA synthesis buffer, 4 μl of iScript reverse transcriptase (Bio-Rad), and 23 μl of DEPC-treated DDW were added to the same tube. cDNA was synthesized according to the recommendations of Bio-Rad. The volume of the final product was adjusted to 1,200 μl with DDW (with no DEPC) rather than 1,400 μl because there was less input RNA, and 10 μl was used as a template for each quantitative real-time TaqMan PCR as described below.

Primers and TaqMan probe design.

Primers and probes (see Table S3 in the supplemental material) for each gene in the real-time PCRs were designed using Integrated DNA Technologies Primer Quest software (http://www.idtdna.com). Briefly, the amplicon sizes ranged from 67 to 86 bp and the primer melting temperatures were designed for 62°C, with a melting temperature difference of less than 4°C for each primer pair. The probe melting temperatures were designed to be 5 to 10°C higher than the melting temperatures for the corresponding primer pairs. Primer and probe sequences were also subjected to BLAST analysis against the P. aeruginosa PAO1 genome to eliminate the possibility of nonspecific binding. It was particularly ensured that the primers had no significant complementarity at the 3′ end and the probe had no significant complementarity at the 5′ end to other nonspecific locations on the PAO1 genome.

Real-time PCR.

Supermixtures for all reactions were made and aliquoted into subsupermixtures for each gene assayed. Essentially, each real-time PCR mixture (final volume, 25 μl) contained 10 μl of cDNA, 12.5 μl of iQ Supermix (Bio-Rad), 120 nM of each forward and reverse primer, and 12 nM probe. Real-time PCR was performed with an iCycler iQ (Bio-Rad) using the following protocol: denaturation at 95°C for 10 min and then 55 cycles of amplification and quantification at 95°C for 20 s and at 65°C for 45 s. To control for variations between runs, all three housekeeping genes and the various target genes for each individual condition were amplified at the same time on a 96-well plate.

Real-time PCR data analyses.

For analysis we used the method of Peirson et al. (29) as previously reported, which gives more accurate quantitative real-time PCR data. Because normalization by geometric averaging using multiple housekeeping genes has been shown to yield more accurate fold changes than normalization using a single housekeeping gene (47), we opted to use three housekeeping genes for our normalization analyses. Real-time PCR was conducted for each of the 11 target genes and the three housekeeping genes in eight replicates. Real-time PCR fold change values were averaged for eight replicates for each gene and were determined by comparison with the geometric mean of three housekeeping genes. A list of housekeeping genes was determined based on in vivo and in vitro microarray data from over 40 GeneChips for 16 different growth conditions in various growth phases, from mid-log phase to stationary phase (data not shown). These housekeeping genes were analyzed using GeneSpring 7.0 and were found to be expressed consistently across all conditions (fold change, <2; P ≤ 0.05). Three housekeeping genes (PA1769, PA1795, and PA1805) which showed the least amount of variability and were expressed consistently across all conditions (as determined using Spearman's correlation coefficient calculations) were selected. Real-time PCR fold changes were calculated using the amplification plot method and the available macro for data analyses of real-time PCR (DART-PCR) (29). Our requirement for real-time PCR is that the efficiencies of each gene (including housekeeping genes) are within 5% from one condition to the next, but similar efficiencies are not necessarily required for the different genes. Accordingly, the average efficiencies of each gene in this study were very similar for the conditions compared (<4.8% efficiency differences), allowing accurate analysis. We recommend not using the standard curve method for clinical samples, since it requires at least three serial dilution points in order to generate a reasonable standard curve. Due to the initial limited amount of input RNA isolated directly from sputum and the possibility of rare transcripts and low transcript abundance within bacterial cells, we were unable to use the dilution-standard curve method to obtain consistent standard curves and efficiencies (<5% difference) for different conditions for many genes. Using the amplification plot analysis method solves this problem, and this method has been demonstrated (29) to yield very good results comparable to those of standard curve analysis and the absolute quantification method, and it is much more accurate than the ΔΔCT and the efficiency correction method.

Assay for constitutive alginate production.

Starter cultures of PAO1 and the clinical isolate pool from the 42-year-old CF patient were grown overnight in Pseudomonas isolation broth and then subcultured (1:200 dilution) into no-salt Luria-Bertani medium (Teknova) and incubated at 250 rpm and 37°C. Cell density (OD540) was measured at regular intervals to monitor cell growth and the growth phases, and aliquots were taken for a simultaneous alginate assay. The amount of alginate produced for each strain was normalized to the corresponding OD540. Cells were harvested at early, mid-, and late-log and early stationary phases, and the alginate assays with carbazole reagent were performed as previously described (13, 16). Extracellular alginate production was quantified in triplicate (± standard error of the mean) at various growth phases. The amount of alginate produced by the clinical isolate pool at early stationary phase was defined as 100%, and all other alginate quantities were normalized to this value.

Microarray data accession number.

Our microarray data have been deposited in the NCBI Gene Expression Omnibus (GEO) repository under accession number GSE7704.

RESULTS AND DISCUSSION

Virulence expression and deregulation of P. aeruginosa genes in vivo.

Affymetrix P. aeruginosa GeneChips were used for monitoring the profile of gene expression in P. aeruginosa, based on hybridization with mRNA directly isolated from the sputum of a chronically infected patient. In vivo induced genes were determined by comparing the gene expression of bacteria obtained directly from the sputum sample to that of the isolated bacterial pool from the same sputum sample grown in vitro. The large data set collected in the whole-genome expression profile experiments for P. aeruginosa from a 42-year-old CF patient with chronic lung infection is summarized in Table Table11 (see Table S1 in the supplemental material). There was a limited set of induced genes that were involved in virulence; this set included three genes for alginate synthesis, five genes for iron acquisition, eight genes for flagellar assembly and function, and four genes for neutralizing oxidative stress (see Table S1 in the supplemental material).

TABLE 1.
Summary of genes induced in vivoa

The lack of induction of more virulence genes in this initial comparison, especially biofilm synthesis genes (e.g., alg and muc genes) that create an obvious mucoid phenotype, was quite puzzling (Table (Table1).1). Mucoid (biofilm) P. aeruginosa is by far the predominant organism in the lungs of most CF patients and is the cause of high morbidity and mortality associated with CF (17). Interestingly, several genes expected to be upregulated (e.g., more alg genes for alginate biosynthesis) were not present in the initial list (Table (Table1;1; see Table S1 in the supplemental material), although we clearly isolated organisms with the mucoidy phenotype among the clinical isolates. Perhaps, in our attempt to detect in vivo induced genes by comparing microarray data for in vivo organisms to data for the in vitro-grown isolate pool (Table (Table1;1; see Table S1 in the supplemental material), we may have missed many constitutively expressed genes that have lost regulatory mechanisms (e.g., alg genes). Therefore, we hypothesized that many genes were constitutively expressed or deregulated, and there were no differences in the expression of many virulence genes between the in vivo condition and the isolated bacterial pool grown in vitro. The loss of these regulatory mechanisms likely represents the many years of evolution of P. aeruginosa during the chronic infection within the host.

This observation prompted our efforts to analyze the GeneChip expression profiles of the in vitro-grown clinical pool to those of prototype strain PAO1, both grown in 1× M9 medium containing citrate. Our results revealed hundreds of genes constitutively expressed compared to the expression in PAO1, where the expression of these genes in PAO1 is more controlled (Table (Table2;2; see Table S2 in the supplemental material). As suspected and summarized in Table Table22 (see Table S2 in the supplemental material), the clinical isolate pool obtained from the 42-year-old patient had higher constitutive expression levels of several classical virulence factors (muc and alg genes for biofilm synthesis, lipase, phospholipase, and several different proteases, rhl for rhamnolipid hemolysin, hcn for hydrogen cyanide production, and pch for pyochelin). The significance of biofilms in the lungs of CF patients cannot be emphasized enough, since the levels of expression of many alg genes were several-hundred-fold; the results for one of these genes (algK) were confirmed by real-time RT-PCR (Table (Table3).3). This constitutive expression of biofilm synthesis genes in the CF patient isolate pool relative to the basal biofilm production in PAO1 translated into greater alginate production at all growth phases (Fig. (Fig.1).1). The lack of virulence gene (plcH, encoding phospholipase C) expression, despite the presence of plcR expression in vivo, was due to larger variations in the microarray data (P > 0.05) for this gene, since PlcR is the accessory protein that is required for PlcH secretion (see Table S2 in the supplemental material) (9). Although not detected by microarrays at high stringency (P ≤ 0.05), plcH was expressed in vivo, and this was confirmed by real-time RT-PCR (Table (Table33 and Fig. Fig.2F).2F). There were nonidentical fold changes in some genes that may be organized in an operon (e.g., bet genes and PA5372 to PA5375; and alg genes and PA3540 to PA3551) (see Tables S1 and S2 in the supplemental material), which are normal and typical of many microarray data sets described previously (41, 48, 50). However, the general trends for genes expressed in the same direction (positive increase) were observed for genes arranged in operons. In a few cases (e.g., mexX, arcC, and arcD) (see below), gene expression was observed both in vitro and in vivo (see Tables S1 and S2 in the supplemental material). We reasoned that these genes were expressed in a gradient fashion, with the expression highest in vivo, lower in the in vitro clinical isolate pool, and lowest or repressed in PAO1. Therefore, Tables Tables11 and and22 represent a summary of in vivo expressed genes for in vivo survival and virulence; some genes are regulated (Table (Table1;1; see Table S1 in the supplemental material), while others have evolved to be constitutive (Table (Table2;2; see Table S2 in the supplemental material).

FIG. 1.
Constitutive expression of P. aeruginosa biofilm/alginate biosynthetic genes results in high production of alginate at all growth phases by the clinical isolate pool from the CF patient relative to the production by strain PAO1. Although the growth of ...
TABLE 2.
Summary of constitutively expressed genesa
TABLE 3.
Verification of microarray data by real-time RT-PCR for genes required for in vivo PC degradation, drug resistance (mexY), and biofilm biosynthesis (algK) by P. aeruginosa

The expression of many genes would have been overlooked if we had not considered the potential constitutive expression and evolution of the organism, by deregulation of the transcriptome, within the lungs of CF patients with chronic infections. A recent study by Smith et al. (45), who compared the genomes of two isolates in the same CF patient 7.5 years apart, identified several differences between the two isolates, which suggested that there was in vivo evolution of P. aeruginosa during chronic lung infection in a CF patient. Further detection of the same mutations in several other isolates of the population in one CF patient over time indicated that there was significant evolution and population diversity of P. aeruginosa within the lungs of this CF patient (45). Our data complement and support this important genomic finding (45) with evidence for deregulation of the transcriptome in the bacterial population. Smith et al. (45) also showed that isolates from CF patients have mutations in genes important for O-antigen biosynthesis, type III secretion, twitching motility, exotoxin A regulation, multidrug efflux, osmotic balance, phenazine biosynthesis, quorum sensing, and iron acquisition. Most importantly, they found that in many, but not all, of the 29 CF patients examined, P. aeruginosa isolates showed mutations in lasR (one of the major virulence and biofilm activators at HCD) to abolish virulence in order to evade immune detection, which was paradoxical in view of the current belief that extracellular virulence factors and biofilm synthesis genes are important during chronic infection in patients with CF. However, immune evasion seems paradoxical, considering the chronic inflammation in the lungs of CF patients during chronic infection (32). Our transcriptome data indicated that many virulence and biofilm synthesis genes, controlled by LasR, were expressed in the bacterial population (see Tables S1 and S2 in the supplemental material), which may also seem paradoxical in view of the previous study (45). However, this finding does not contradict the findings of Smith et al. (45), and it agrees with current dogma concerning infection in CF patients, considering population diversity and the diverse microenvironment of each individual bacterium in the population. Therefore, looking at gene expression of individual isolates from CF patients, although important, may not reveal the complete picture of the whole cooperative population. In support of the concept of a cooperative P. aeruginosa population during infection of CF patients, it has also been shown that different P. aeruginosa isolates in the same CF patient could produce completely different extracellular protein profiles, whereas the cellular protein profiles are quite conserved (51). This difference in the extracellular protein profiles of different isolates in the same CF patient suggests that there is cooperation between members of the bacterial population, where each member may have specific roles that serve the whole population. For example, secreted virulence elastases (LasA and LasB) may benefit both producers and nonproducers. In addition, the presence of a heterogeneous environment in the lungs of CF patients through space and time has recently been suggested by Nguyen and Singh (21), which may contribute to population expression diversity. As complex as the population diversity is, we agree that the diverse microenvironments in the lungs of CF patients, within which distinct isolates occur and adapt, signal individual bacteria to behave distinctly and may even select for certain mutations. For example, for osmotic regulation, twitching, or type III secretion, bacteria found on the surface of the biofilm and in contact with epithelial cells may express the genes encoding these properties or even deregulate these genes differently than bacteria growing and surviving inside the biofilm.

In vivo expression of drug resistance genes.

In terms of mechanisms of resistance to antibacterial agents, P. aeruginosa is second to none, having several such mechanisms in its huge arsenal, including biofilm production, target modification and overexpression, drug modification, and more than 12 multidrug efflux pumps (1, 18, 22, 33). All possible mechanisms of drug resistance were observed in our data (see Tables S1 and S2 in the supplemental material), with the exception of target modification since our approach cannot detect such changes. At the time of sputum collection, the patient was being treated with tobramycin (an aminoglycoside) by inhalation, tobramycin and cefepime (a fourth-generation cephalosporin) intravenously, and Bactrim (trimethoprim-sulfamethoxazole) orally. Certainly, the highly expressed biofilm biosynthesis genes could present biofilm as the first barrier for drug resistance. The target of tobramycin is at the level of the ribosome to inhibit protein translation, and we observed significant overexpression of 38 different ribosomal proteins (target overexpression) (see Table S1 in the supplemental material). In addition to the induction of two aminoglycoside drug modification proteins, PA1409 (aphA) and PA4119 (aph), we also observed induction of genes for the aminoglycoside efflux pump, MexXY-OprM (19) (Table (Table33 and Fig. Fig.2G;2G; see Table S1 in the supplemental material). Mutations in mexZ, a negative regulator of mexX and mexY, have been found in the majority of isolates from several CF patients (45). It has been previously demonstrated that MexXY-OprM also effluxes trimethoprim (7). Therefore, expression of this pump may also provide resistance against administered trimethoprim. Increased expression of five folate biosynthetic genes was also observed (cofactor [see Table S1 in the supplemental material]), likely due to the inhibition of folate synthesis by the trimethoprim-sulfamethoxazole, and one of the proteins overexpressed was dihydropteroate synthase (FolP, PA4750), which is the target of sulfamethoxazole (target overexpression). For potential resistance to the β-lactam cefepime, overexpression of β-lactamases was observed in vivo (ampC, 87.9-fold [see Table S1 in the supplemental material]; PA5514, 13.9-fold [see Table S2 in the supplemental material]). A previous study showed that antibiotic resistance mechanisms arise from hypermutable mutS variants (44), and isolates from CF patients do harbor mutS mutations (45). Some of the antibiotic resistance mechanisms discussed here in the population may have arisen from such mutS variants. Regardless, our important findings suggest that P. aeruginosa utilizes several mechanisms in vivo to deal with antibacterial agents in a natural infection, by inducing many genes to counteract the negative effects of the administered drugs (see Tables S1 and S2 in the supplemental material).

General metabolism.

Various central pathways expressed by P. aeruginosa (Tables (Tables11 and and2)2) indicate active metabolism in vivo. Gene expression in vivo for central pathways, including those of saccharide metabolism, the Krebs and glyoxylate cycles, the pentose phosphate pathway, and oxidative phosphorylation, along with vitamin/cofactor biosyntheses and nutrient transport, indicates that P. aeruginosa is very metabolically active in vivo. The carbon sources utilized by P. aeruginosa seem to be both lipids of lung surfactant and amino acids, as discussed below. All these activities may contribute to replication and division in vivo, as evident from the expression of genes for cell wall metabolism and division. The data also indicate that P. aeruginosa actively senses and interacts with the in vivo environment by expressing many genes involved in two-component systems, secretion, and transport. In addition, various processes are controlled by induction of a large number of transcriptional regulators and sigma factors. Finally, in an animal model, the expression of the napEFDABC and arcDABC gene clusters has been monitored to indicate anaerobic respiration by reflecting an increase in nitrate and arginine turnover (4). Our data demonstrate that there is significant expression of one of these two gene clusters, arcDABC (13.1-, 28.0-, 8.8-, and 4.2-fold changes, respectively) (see Table S1 in the supplemental material). Together with expression of genes involved in oxidative respiration, this suggests that anaerobic respiration also occurs in the population, which may result from the diverse microenvironment within the lung or biofilm (21). As alluded to by Nguyen and Singh (21), we believe that in a sputum biofilm in a CF patient there may be subpopulations of P. aeruginosa undergoing oxidative respiration on the biofilm surface, while other subpopulations are anaerobically respiring within the biofilm or oxygen-deficient areas, as indicated by the expression of both aerobic and anaerobic respiration genes in the population.

Active amino acid degradation contributing to HCD replication in vivo.

Our data indicate that P. aeruginosa metabolizes amino acids as a nutrient source in vivo. Previous studies utilizing sterilized mucopurulent respiratory liquid or sputum from CF patients to grow P. aeruginosa in vitro have discovered several sputum-induced genes (25, 53). One of these in vitro studies (25), in which P. aeruginosa was grown on a pool of several lyophilized sputum samples, showed that there was induction of a subset of 20 genes for amino acid degradation, suggesting that P. aeruginosa metabolizes amino acids as nutrient sources in vivo. However, this hypothesis requires in vivo confirmation. Our in vivo data (see Tables S1 and S2 in the supplemental material), obtained by isolating mRNA directly from a single expectoration source of sputum, indicate that there is induction of a much larger set of genes for amino acid transport and degradation (36 genes induced and 26 genes constitutively expressed). This set represents a total of 60 different genes (since arcC and arcD are present in both Table S1 and Table S2 in the supplemental material) that are involved in amino acid degradation, which suggests that amino acid degradation in vivo could be much more important than previously thought (25). Since proteases were expressed in vivo (LasA, LasB, and others) (see Table S2 in the supplemental material) to cleave proteins into peptides and amino acids, it was significant to find that many amino acid and peptide transporters were also expressed in our data. However, these data indicate that P. aeruginosa may utilize more than amino acids and that lung surfactant lipids may also serve as a great nutrient source in vivo.

Degradation of lung surfactant lipids by P. aeruginosa.

Lung surfactant components, especially lipids, are essential for proper lung functions, and the lack of these components can lead to respiratory distress syndromes in adults and prematurely born infants. Pulmonary surfactant consists of ~10% surfactant proteins (SP-A, SP-B, SP-C, and SP-D) and ~90% lipids (PC, phosphatidylglycerol, phosphatidylethanolamine, phosphatidylinositol, phosphatidylserine, and small amounts of other lipids and free fatty acids), and PC accounts for ~80% of the lipids (2, 14). Thus, the most plentiful sources of potential bacterial nutrients in lung surfactant are lipids, especially PC.

We initially performed GeneChip studies to determine which genes were induced by P. aeruginosa strain PAO1 during growth on PC in vitro as the sole carbon source (Table (Table4).4). By identifying the genes involved in PC degradation in vitro, we could relate and search for the expression of the same set of genes in vivo to determine if P. aeruginosa metabolizes PC in vivo. Through the action of lipases (Fig. (Fig.3),3), the richest nutrient obtained from the lung surfactant PC molecule in vivo is fatty acids, mostly palmitic acid (C16:0) (50 to 60%) but also approximately 10 to 20% each C14:0, C16:1, C18:1, and C18:2 (34). Accordingly, the PC utilized in our experiments (Sigma, St. Louis, MO) contained mostly long-chain fatty acids (LCFA) (C16:0, 33%; C18:1, 30%; C18:2, 14%; C18:0, 14%). These LCFA can be metabolized via β-oxidation in the fatty acid degradation (Fad) pathway (Fig. (Fig.3C).3C). As observed for the relevant genes in Table Table4,4, several key genes signal the degradation of PC and LCFA in vitro. Specifically, lipases (LipA and LipC) and phospholipases (PlcH and the accessory protein PlcR) were expressed to break down PC into smaller constituents: fatty acid, glycerol, and phosphorylcholine. These constituents could be further metabolized in vitro, via the expression of Bet enzymes (choline head group metabolism) (49), Glp enzymes (glycerol metabolism) (42), and Fad enzymes (fatty acid degradation) (8). Fold changes for the glycerol degradation genes in Table Table44 were determined with less stringency, because of the larger variations (P > 0.05) between the pairwise comparisons in our microarray data for these genes, and all showed positive induction (Fig. (Fig.2).2). Expression of the fadBA5 operon was observed only with C16:0 and not with PC relative to citrate (Table (Table4),4), although these genes were involved in PC degradation in vitro (Fig. (Fig.4)4) and were expressed in vivo (Table (Table3).3). FadB5 has 73% homology to the well-characterized Escherichia coli FadB protein (35). To demonstrate that FabBA5 is involved in PC degradation, we created a ΔfadBA5 deletion mutant. This mutant showed a decrease in the ability to grow on LCFA and PC (Fig. (Fig.4).4). The ΔfadBA5 mutant could still grow on LCFA, because there are at least two other fadBA operons in P. aeruginosa (unpublished data). Of course, there was less of a defect for this mutant to grow on PC than for it to grow on palmitate, since the PC molecule still contains two other nutrient sources (i.e., glycerol and choline) (Fig. (Fig.33).

FIG. 3.
(A) Lung surfactant, made of 90% lipids and 10% proteins, coats the trachea, bronchioles, and alveoli of the lung. (B) Of the 90% lipids, 80% consists of PC, which can be cleaved by P. aeruginosa lipases and phospholipase ...
FIG. 4.
fadBA5 operon is involved in PC degradation. (A) P. aeruginosa fusion strain PAO1-PfadBA5-lacZ was grown in LB, PC, citrate, and palmitate (C16:0), and the β-galactosidase (β-Gal) assay was performed (20). The results indicate the reason ...
TABLE 4.
In vitro microarray analysis of relevant genes expressed twofold or more in P. aeruginosa strain PAO1grown on 1× M9 medium containing 0.4% PC or 0.4% palmitate (C16:0) compared to the expression with 20 mM citrate

In vivo (see Tables S1 and S2 in the supplemental material), we observed that the majority of the same genes were induced for PC degradation. Our microarray data were confirmed, both in vitro and in vivo, by real-time RT-PCR for several of these genes (Tables (Tables33 and and4).4). The fadD2, betA, glpK, glpD, lipA, and plcH genes for PC degradation were also expressed in vitro when P. aeruginosa was grown on PC as the sole carbon source (Table (Table4).4). FadD1 and FadD2 have very high homology (72%) to E. coli FadD and are also involved in β-oxidation (unpublished data), and all three FadDs (FadD1, FadD2, and FadD4) have a very conserved ATP/AMP binding signature motif compared to the E. coli signature motif (52). The in vivo gene expression data (Table (Table3;3; see Tables S1 and S2 in the supplemental material) strongly suggest that PC is metabolized in vivo. P. aeruginosa can metabolize PC and all three components of PC (fatty acid, glycerol, and choline) (Fig. (Fig.3),3), at least in vitro (data not shown). The individual constituents are further imported and degraded, via high levels of expression of many fad, glp, and bet genes (Fig. (Fig.3).3). With the exception of one lipase gene, previous in vitro growth of P. aeruginosa in pooled and lyophilized sputum did not reveal any other genes induced for PC degradation (25). We reasoned that the previous study was performed under in vitro conditions and could have missed less soluble nutrients (such as PC and other lipids) in the lung surfactant that coats the surface of the lung and are continuously being replenished due to its essentiality in vivo.

Expression of PC degradation genes in a second CF patient as determined by real-time RT-PCR.

To further demonstrate the importance of potential PC degradation in vivo through gene expression, we analyzed the expression of lipA, plcH, fadB5, fadD1, fadD2, fadD4, betA, glpD, and glpK from P. aeruginosa isolated directly from an 18-year-old CF patient by real-time RT-PCR. However, not armed with microarray data for this patient, we wanted to perform more thorough and careful analyses to clearly show gene expression as it relates to PC degradation. Table Table55 indicates that genes required for PC degradation in vitro were also expressed in vivo, compared to both the same clinical isolate pool or PAO1 grown in vitro in 1× M9 medium with 20 mM citrate. As Table Table55 shows, using gene expression in PAO1 as a reference is essential to get a baseline for quantifying the relative expression of many genes. In this second CF patient, as in the first patient described above, P. aeruginosa clearly expressed genes essential for the degradation of PC. It is interesting that glp genes, fadD2, and fadD4 in the first patient were deregulated (Table (Table3).3). Likewise, these same genes were also constitutively expressed in the second patient (Table (Table5).5). These gene expression data further support our overall hypothesis that P. aeruginosa may utilize PC as one of the major nutrient sources in vivo.

TABLE 5.
Expression of genes involved in PC degradation in a second CF patient by real-time RT-PCR

Future analyses of the available microarray data.

The microarray data have been deposited in the NCBI Gene Expression Omnibus repository for future analyses. Low signals in microarray data have previously been observed for some genes, for example, the exotoxin A-encoding gene (23). Due to the inherent variability in microarray data that could lead to missing some expressed genes shown in this study (Fig. (Fig.2),2), we suspect that much more could be gleaned from our microarray data by investigators interested in certain aspects of P. aeruginosa beyond those that are discussed here. For example, we did not observe expression of the virulence factor PA1249 for alkaline protease (aprA), but the accessory proteins encoded in the same operon (PA1248 [aprF] and PA1250 [aprI]) were constitutively expressed (see Table S2 in the supplemental material). Perhaps aprA is expressed, and this may be a situation similar to the situation for plcH and plcR discussed above. Since performing pairwise comparisons for all genes in the P. aeruginosa genome is an enormous undertaking, we selected a few genes to examine in this study (Fig. (Fig.2).2). On average, these genes were induced in pairwise comparisons (Fig. (Fig.2).2). Indeed, these genes were induced when we performed real-time RT-PCR (e.g., glp genes in Table Table44 and mexY and plcH in Table Table3).3). Therefore, we hope that the deposited microarray data may be of some use in future analyses performed by other investigators who may use lower stringency (higher P values) or pairwise comparisons and further testing.

In summary, our study revealed several important aspects of P. aeruginosa pathogenesis, drug resistance, and metabolism during infection in CF patients in relation to population diversity. P. aeruginosa has a huge arsenal of metabolic capabilities, which allows it to exploit many possible nutrients in the lung environment. This is the first study that defines the metabolic capability of P. aeruginosa in vivo and the host pulmonary nutrient factors that may contribute to bacterial replication. Our objective in this study was to identify the expression of metabolic pathways which may contribute to nutrient acquisition, leading to replication and maintenance of P. aeruginosa in the lung at an HCD. Our data suggested that P. aeruginosa degrades amino acids as a nutrient source in the lung. In addition, P. aeruginosa also induces genes in vivo to potentially metabolize lung surfactant lipids, adapting to the many possible nutrient sources.

The gene expression of P. aeruginosa in vivo may be influenced by years of evolution during chronic lung infection and by the microenvironment of each bacterium, which leads to a diverse infectious population. Our study revealed a different level of evolution (i.e., deregulation) for several genes in the transcriptome rather than the deletion or acquisition of genes previously demonstrated in the genome (12, 54). We suggest that the population diversity and microenvironment might influence an individual bacterium's behavior, perhaps contributing to a cooperative and diverse infectious population. The shift in thinking of the bacteria in a chronic infection in a CF patient as a diverse infectious population and not as individual isolates may yield novel insight into future treatment. For example, the existence of hypermutable mutS variants coupled to the bacterial population diversity found in the lungs of CF patients with chronic lung infections as opposed to acute infections has important ramifications for future testing for effective antibiotic combinations and concentrations. Thus, future tests of antibiotics on planktonic cells or biofilm would be performed more effectively with a clinical population or a pool of isolates than with a single isolate, in order to obtain the true resistance potential of the infectious P. aeruginosa population. Therefore, both the diversity of the population and the microenvironment may contribute to the full potential of P. aeruginosa within the lungs of CF patients, such that each member of the population plays a distinct role in order to contribute to the pathogenesis of the infectious population. However, this view awaits novel technology to dissect the metabolism and virulence of individual bacteria in the population relative to the microenvironment, which should answer some questions concerning the extent of bacterial cooperation.

The results of this study should serve as an initial model for understanding (i) how P. aeruginosa reacts to antibacterial treatment in vivo, (ii) nutrient acquisition for extracellular pathogens of the lung, and (iii) the detrimental effects of nutrient metabolism in the lung. Finally, this study opens opportunities for further and more thorough analyses of these important pathogenesis aspects with clinical samples from other CF patients in order to determine the transcriptome conservation and heterogeneity of P. aeruginosa populations in vivo.

Supplementary Material

[Supplemental material]

Acknowledgments

This publication was made possible by pilot funding from grant P20RR018727 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH). Graduate salary support for D.T.N. was provided through a Hawaii Community Foundation grant awarded to T.T.H.

The contents of this paper are solely the responsibility of the authors and do not necessarily represent the official views of NCRR or NIH.

We are very grateful to E. P Greenberg and M. J. Schurr for their microarray data, which allowed us to search for appropriate housekeeping genes with our data. We thank Joseph Lam and Chad Walton for critical reading of the manuscript.

Notes

Editor: V. J. DiRita

Footnotes

[down-pointing small open triangle]Published ahead of print on 27 August 2007.

Supplemental material for this article may be found at http://iai.asm.org/.

REFERENCES

1. Akasaka, T., M. Tanaka, A. Yamaguchi, and K. Sato. 2001. Type II topoisomerase mutations in fluoroquinolone-resistant clinical strains of Pseudomonas aeruginosa isolated in 1998 and 1999: role of target enzyme in mechanism of fluoroquinolone resistance. Antimicrob. Agents Chemother. 45:2263-2268. [PMC free article] [PubMed]
2. Bernhard, W., J.-Y. Wang, T. Tschernig, B. Tummler, H. J. Hedrich, and H. von der Hardt. 1997. Lung surfactant in a cystic fibrosis animal model: increased alveolar phospholipid pool size without altered composition and surface tension function in cftrm1HGU/m1HGU mice. Thorax 52:723-730. [PMC free article] [PubMed]
3. Bowton, D. L. 1999. Nosocomial pneumonia in the ICU—year 2000 and beyond. Chest 115:28S-33S. [PubMed]
4. Bragonzi, A., D. Worlitzsch, G. B. Pier, P. Timpert, M. Ulrich, M. Hentzer, J. B. Andersen, M. Givskov, and M. Conese. 2005. Nonmucoid Pseudomonas aeruginosa expresses alginate in the lungs of patients with cystic fibrosis and in a mouse model. J. Infect. Dis. 192:410-419. [PMC free article] [PubMed]
5. Brint, J. M., and D. E. Ohman. 1995. Synthesis of multiple exoproducts in Pseudomonas aeruginosa is under the control of RhlR-RhlI, another set of regulators in strain PAO1 with homology to the autoinducer-responsive LuxR-LuxI family. J. Bacteriol. 177:7155-7163. [PMC free article] [PubMed]
6. Burns, J. L., J. Emerson, J. R. Tapp, D. L. Yim, J. Krzewinski, L. Louden, B. W. Ramsey, and C. R. Clausen. 1998. Microbiology of sputum from patients at cystic fibrosis centers in the United States. Clin. Infect. Dis. 27:158-163. [PubMed]
7. Chuanchuen, R., K. Beinlich, T. T. Hoang, A. Becher, R. R. Karkhoff-Schweizer, and H. Schweizer. 2001. Cross-resistance between triclosan and antibiotics in Pseudomonas aeruginosa is mediated by multidrug efflux pumps: exposure of a susceptible mutant strain to triclosan selects nfxB mutants overexpressing MexCD-OprJ. Antimicrob. Agents Chemother. 45:428-432. [PMC free article] [PubMed]
8. Clark, D. P., and J. E. Cronan, Jr. 1996. Two-carbon compounds and fatty acids as carbon sources, p. 343-358. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. American Society for Microbiology, Washington, DC.
9. Cota-Gomez, A., A. I. Vasil, J. Kadurugamuwa, T. J. Beveridge, H. P. Schweizer, and M. L. Vasil. 1997. PlcR1 and PlcR2 are putative calcium binding proteins required for secretion of the hemolytic phospholipase C of Pseudomonas aeruginosa. Infect. Immun. 65:2904-2913. [PMC free article] [PubMed]
10. Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, and E. P. Greenberg. 1998. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280:295-298. [PubMed]
11. Doring, G. 1997. Cystic fibrosis respiratory infections: interactions between bacteria and host defense. Monaldi Arch. Chest Dis. 52:363-366. [PubMed]
12. Ernst, R. K., D. A. D'Argenio, J. K. Ichikawa, M. G. Bangera, S. Selgrade, J. L. Burns, P. Hiatt, K. McCoy, M. Brittnacher, A. Kas, D. H. Spencer, M. V. Olson, B. W. Ramsey, S. Lory, and S. I. Miller. 2003. Genome mosaicism is conserved but not unique in Pseudomonas aeruginosa isolates from the airways of young children with cystic fibrosis. Environ. Microbiol. 5:1341-1349. [PubMed]
13. Franklin, M. J., and D. E. Ohman. 1993. Identification of algF in the alginate biosynthetic gene cluster of Pseudomonas aeruginosa which is required for alginate acetylation. J. Bacteriol. 175:5057-5065. [PMC free article] [PubMed]
14. Hite, R. D. 2002. Surfactant deficiency in adults. Clin. Pulm. Med. 9:39-45.
15. Holden, M. T. G., S. R. Chhabra, R. Nys, P. Stead, N. J. Bainton, P. J. Hill, M. Manefield, N. Kumar, M. Labatte, D. England, S. Rice, M. Givskov, G. P. C. Salmond, G. S. A. B. Stewart, B. W. Bycroft, S. Kjelleberg, and P. Williams. 1999. Quorum-sensing cross talk: isolation and chemical characterization of cyclic dipeptides from Pseudomonas aeruginosa and other Gram-negative bacteria. Mol. Microbiol. 33:1254-1266. [PubMed]
16. Knutson, C. A., and A. Jeanes. 1968. A new modification of the carbazole analysis: application to heteropolysaccharides. Anal. Biochem. 24:470-481. [PubMed]
17. Lam, J., R. Chan, and J. W. Costerton. 1980. Production of mucoid microcolonies by Pseudomonas aeruginosa within infected lungs in cystic fibrosis. Infect. Immun. 28:546-556. [PMC free article] [PubMed]
18. Mah, T.-F. C., and G. A. O'Toole. 2001. Mechanisms of biofilm resistance to antimicrobial agents. Trends Microbiol. 9:34-39. [PubMed]
19. Masuda, N., E. Sakagawa, S. Ohya, N. Gotoh, H. Tsujimoto, and T. Nishino. 2000. Contribution of the MexX-MexY-OprM efflux system to intrinsic resistance in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 44:2242-2246. [PMC free article] [PubMed]
20. Miller, J. H. 1992. A short course in bacterial genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
21. Nguyen, D., and P. K. Singh. 2006. Evolving stealth: Genetic adaptation of Pseudomonas aeruginosa during cystic fibrosis infection. Proc. Natl. Acad. Sci. USA 103(22):8305-8306. [PMC free article] [PubMed]
22. Nichols, W. W., S. M. Dorrington, M. P. Slack, and H. L. Walmsley. 1988. Inhibition of tobramycin diffusion by binding to alginate. Antimicrob. Agents Chemother. 32:518-523. [PMC free article] [PubMed]
23. Ochsner, U. A., P. J. Wilderman, A. I. Vasil, and M. L. Vasil. 2002. GeneChip expression analysis of the iron starvation response in Pseudomonas aeruginosa: identification of novel pyroverdine biosynthesis genes. Mol. Microbiol. 45:1277-1287. [PubMed]
24. Ogata, H., S. Goto, W. Fujibuchi, and M. Kanehisa. 1998. Computation with the KEGG pathway database. BioSystems 47:119-128. [PubMed]
25. Palmer, K. L., L. M. Mashburn, P. K. Singh, and M. Whiteley. 2005. Cystic fibrosis sputum supports growth and cues key aspects of Pseudomonas aeruginosa physiology. J. Bacteriol. 187:5267-5277. [PMC free article] [PubMed]
26. Passador, L., J. M. Cook, M. J. Gambello, L. Rust, and B. H. Iglewski. 1993. Expression of Pseudomonas aeruginosa virulence genes requires cell-to-cell communication. Science 260:1127-1130. [PubMed]
27. Pearson, J., K. Gray, L. Passador, K. Tucker, A. Eberhard, B. Iglewski, and E. Greenberg. 1994. Structure of the autoinducer required for expression of Pseudomonas aeruginosa virulence genes. Proc. Natl. Acad. Sci. USA 91:197-201. [PMC free article] [PubMed]
28. Pearson, J., L. Passador, B. Iglewski, and E. Greenberg. 1995. A second N-acylhomoserine lactone signal produced by Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 92:1490-1494. [PMC free article] [PubMed]
29. Peirson, S. N., J. N. Butler, and R. G. Foster. 2003. Experimental validation of novel and conventional approaches to quantitative real-time PCR data analysis. Nucleic Acids Res. 31:e73. [PMC free article] [PubMed]
30. Pesci, E. C., and B. H. Iglewski. 1997. The chain of command in Pseudomonas quorum sensing. Trends Microbiol. 5:132-134. [PubMed]
31. Pesci, E. C., J. P. Pearson, P. C. Seed, and B. H. Iglewski. 1997. Regulation of las and rhl quorum sensing in Pseudomonas aeruginosa. J. Bacteriol. 179:3127-3132. [PMC free article] [PubMed]
32. Pier, G. B. 1998. Pseudomonas aeruginosa: a key problem in cystic fibrosis. ASM News 64:339-347.
33. Poole, K. 2001. Multidrug efflux pumps and antimicrobial resistance in Pseudomonas aeruginosa and related organisms. J. Mol. Microbiol. Biotechnol. 3:255-264. [PubMed]
34. Postle, A. D., A. Mander, K. B. M. Reid, J.-Y. Wang, S. M. Wright, M. Moustaki, and J. O. Warner. 1999. Deficient hydrophilic lung surfactant proteins A and D with normal surfactant phospholipid molecular species in cystic fibrosis. Am. J. Respir. Cell Mol. Biol. 20:90-98. [PubMed]
35. Pramanik, A., S. Pawar, E. Antonian, and H. Schulz. 1979. Five different enzymatic acitivities are associated with the multienzyme complex of fatty acid oxidation in Escherichia coli. J. Bacteriol. 137:469-473. [PMC free article] [PubMed]
36. Rahme, L. G., E. J. Stevens, S. F. Wolfort, J. Shao, R. G. Tompkins, and F. M. Ausubel. 1995. Common virulence factors for bacterial pathogenicity in plants and animals. Science 268:1899-1902. [PubMed]
37. Richards, M. J., J. R. Edwards, D. H. Culver, and R. P. Gaynes. 1999. Nosocomial infections in medical intensive care units in the United States. Crit. Care Med. 27:887-892. [PubMed]
38. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
39. Sato, H., J. B. Feix, C. J. Hillard, and D. W. Frank. 2005. Characterization of phospholipase activity of the Pseudomonas aeruginosa type III cytotoxin, ExoU. J. Bacteriol. 187:1192-1195. [PMC free article] [PubMed]
40. Schaberg, D. R., D. H. Culver, and R. P. Gaynes. 1991. Major trends in the microbial etiology of nosocomial infection. Am. J. Med. 91(Suppl. 3B):72S-75S. [PubMed]
41. Schuster, M., C. P. Lostroh, T. Ogi, and E. P. Greenberg. 2003. Identification, timing, and signal specificity of Pseudomonas aeruginosa quorum-controlled genes: a transcriptome analysis. J. Bacteriol. 185:2066-2079. [PMC free article] [PubMed]
42. Schweizer, H. P., C. Po, and R. Jump. 1997. Structure and gene-polypeptide relationships of the region encoding glycerol diffusion facilitator (glpF) and glycerol kinase (glpK) of Pseudomonas aeruginosa. Microbiology 143:1287-1297. [PubMed]
43. Singh, P. K., A. L. Schaefer, M. R. Parsek, T. O. Moninger, M. J. Welsh, and E. P. Greenberg. 2000. Quorum-sensing signals indicate that cystic fibrosis lungs are infected with bacterial biofilms. Nature 407:762-764. [PubMed]
44. Smania, A. M., I. Segura, R. J. Pezza, C. Becerra, I. Albesa, and C. E. Argarana. 2004. Emergence of phenotypic variants upon mismatch repair disruption in Pseudomonas aeruginosa. Microbiology 150:1327-1338. [PubMed]
45. Smith, E. E., D. G. Buckley, Z. Wu, C. Saenphimmachak, L. Hoffman, D. D'Argenio, S. Miller, B. W. Ramsey, D. P. Speert, S. M. Moskowitz, J. L. Burns, R. Kaul, and M. V. Olson. 2006. Genetic adaptation by Pseudomonas aeruginosa to the airways of cystic fibrosis patients. Proc. Natl. Acad. Sci. USA 103:8487-8492. [PMC free article] [PubMed]
46. Storey, D. G., E. E. Ujack, and H. R. Rabin. 1992. Population transcript accumulation of Pseudomonas aeruginosa exotoxin A and elastase in sputa from patients with cystic fibrosis. Infect. Immun. 60:4687-4694. [PMC free article] [PubMed]
47. Vandesompele, J., D. P. K., F. Pattyn, B. Poppe, N. Van Roy, A. De Paepe, and F. Speleman. 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 3:1-12. [PMC free article] [PubMed]
48. Vasil, M. L. 2003. DNA microarray in analysis of quorum sensing: strengths and limitations. J. Bacteriol. 185:2061-2065. [PMC free article] [PubMed]
49. Velasco-Garcia, R., C. Mujica-Jimenez, G. Mendoza-Hernandez, and R. A. Munoz-Clares. 1999. Rapid purification and properties of betaine aldehyde dehydrogenase from Pseudomonas aeruginosa. J. Bacteriol. 181:1292-1300. [PMC free article] [PubMed]
50. Wagner, V. E., D. Bushnell, L. Passador, A. I. Brooks, and B. H. Iglewski. 2003. Microarray analysis of Pseudomonas aeruginosa quorum-sensing regulons: effect of growth phase and environment. J. Bacteriol. 185:2080-2095. [PMC free article] [PubMed]
51. Wehmhoner, D., S. Haussler, B. Tummler, L. Jansch, F. Bredenbruch, J. Wehland, and I. Steinmetz. 2003. Inter- and intraclonal diverstiy of the Pseudomonas aeruginosa proteome manifests within the secretome. J. Bacteriol. 185:5807-5814. [PMC free article] [PubMed]
52. Weimar, J. D., C. C. DiRusso, R. Delio, and P. N. Black. 2002. Functional role of fatty acid acyl-coenzyme A synthetase in the transmembrane movement and activation of exogenous long-chain fatty acids. J. Biol. Chem. 277:29369-29376. [PubMed]
53. Wolfgang, M. C., J. Jyot, A. L. Goodman, R. Ramphal, and S. Lory. 2004. Pseudomonas aeruginosa regulates flagellin expression as part of a global response to airway fluid from cystic fibrosis patients. Proc. Natl. Acad. Sci. USA 101:6664-6668. [PMC free article] [PubMed]
54. Wolfgang, M. C., B. R. Kulasekara, X. Liang, D. Boyd, K. Wu, Q. Yang, C. G. Miyada, and S. Lory. 2003. Conservation of genome content and virulence determinants among clinical and environmental isolates of Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 100:8484-8489. [PMC free article] [PubMed]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...