Abstract
Pseudomonas aeruginosa is a gram-negative opportunistic human pathogen often infecting the lungs of individuals with the heritable disease cystic fibrosis and the peritoneum of individuals undergoing continuous ambulatory peritoneal dialysis. Often these infections are not caused by colonization with P. aeruginosa alone but instead by a consortium of pathogenic bacteria. Little is known about growth and persistence of P. aeruginosa in vivo, and less is known about the impact of coinfecting bacteria on P. aeruginosa pathogenesis and physiology. In this study, a rat dialysis membrane peritoneal model was used to evaluate the in vivo transcriptome of P. aeruginosa in monoculture and in coculture with Staphylococcus aureus. Monoculture results indicate that approximately 5% of all P. aeruginosa genes are differentially regulated during growth in vivo compared to in vitro controls. Included in this analysis are genes important for iron acquisition and growth in low-oxygen environments. The presence of S. aureus caused decreased transcription of P. aeruginosa iron-regulated genes during in vivo coculture, indicating that the presence of S. aureus increases usable iron for P. aeruginosa in this environment. We propose a model where P. aeruginosa lyses S. aureus and uses released iron for growth in low-iron environments.
Pseudomonas aeruginosa is a gram-negative opportunistic human pathogen commonly found in water and soil. P. aeruginosa causes a number of chronic and acute infections and is noted for its inherent resistance to many clinically relevant antibiotics. Two of the most common infections caused by P. aeruginosa are chronic colonization of the lungs of individuals with the genetic disease cystic fibrosis (CF) (19) and peritonitis in individuals undergoing continuous ambulatory peritoneal dialysis (CAPD) (25). The lungs of CF patients are commonly colonized before the age of 8, and most individuals maintain these infections throughout their lifetimes. High infection rates are also associated with CAPD, which is often used to treat end-stage renal disease.
P. aeruginosa physiology and gene expression during in vivo growth is largely unknown. Using in vivo expression technology (IVET), Wang et al. identified 19 P. aeruginosa genes inducible during growth in a neutropenic mouse (40). Although that study identified several new genes important for virulence in P. aeruginosa, it did not provide a comprehensive analysis of in vivo gene expression. Two recent studies using Pasteurella multocida and Borrelia burgdorferi have provided a more comprehensive view of in vivo bacterial gene expression by using DNA microarrays (3, 4, 33). These studies illustrate that a significant number of genes (2 to 8% of all the genes in the genomes) are differentially regulated in vivo, suggesting that the in vivo environment is distinct from normal in vitro culture conditions.
Although evaluation of the transcriptomes of in vivo-grown bacteria provides a snapshot of transcription under monoculture growth conditions, it is clear that many infections are not simply the result of colonization by one bacterium but rather the pathogenic contributions of several bacteria (10, 19, 20, 22, 46). Such is the case for P. aeruginosa infections, particularly in the CF lung, which often consist of a consortium of pathogenic bacteria, including Staphylococcus aureus and Streptococcus pneumoniae (19, 23). As with most bacteria, studies of P. aeruginosa pathogenesis have primarily focused on monoculture infections; consequently, little is known about interspecies interactions in polymicrobial infections. Although bacterium-bacterium interactions in vivo will be affected by numerous factors both spatial and temporal, it is possible that in some circumstances interspecies interactions may affect the course of disease.
To provide a more comprehensive analysis of P. aeruginosa gene expression in vivo, we used Affymetrix GeneChips to examine the transcriptome of P. aeruginosa growing as a monoculture and in coculture with S. aureus in the rat peritoneum. Our studies indicate that approximately 5% of all P. aeruginosa genes are differentially regulated during monoculture growth within the peritoneum compared to in vitro conditions. These results indicate that the peritoneum is a low-oxygen, iron-limited environment, and the presence of S. aureus increases usable iron for P. aeruginosa in vivo. We propose a model where P. aeruginosa lyses S. aureus during coculture and gains access to sequestered iron. Our data suggest that P. aeruginosa pathogenesis and physiology are influenced by the presence of S. aureus, thereby implicating the importance of studying interspecies interactions to understand P. aeruginosa pathogenesis.
MATERIALS AND METHODS
Bacterial strains, plasmids, and media.
P. aeruginosa strain UCBPP-PA14 (30) and S. aureus strain MN8 (35) were used in these studies. P. aeruginosa PA14-LM1 containing a Tn5 insertion in pqsA was obtained from a publicly available transposon database (http://pga.mgh.harvard.edu/cgi-bin/pa14/mutants/retrieve.cgi). Bacteria were grown in morpholinepropanesulfonic acid (MOPS) minimal medium containing 50 mM MOPS (pH 7.2), 93 mM NH4Cl, 43 mM NaCl, 2 mM KH2PO4, 1 mM MgSO4, 3.5 μM FeSO4, and 20 mM glucose or 20 mM sodium succinate as the sole source of carbon and energy. For in vitro growth of S. aureus, 0.5% yeast extract and 0.5% Casamino Acids were added to the MOPS medium. For low-iron medium, MOPS minimal medium without added FeSO4 and containing 20 mM sodium succinate was chelexed two times overnight at 4°C using Chelex 100 (10 g/liter; Sigma Chemical Co., St. Louis, Mo.). For differential isolation of P. aeruginosa and S. aureus in coculture, pseudomonas isolation agar and Baird Parker agar were used, respectively (Remel, Lenexa, Kans.). Brain heart infusion (BHI; Difco, Detroit, Mich.) agar was used for coculture on petri plates.
Dialysis membrane chambers.
Spectra/Por (Spectrum Medical Industries Inc., Los Angeles, Calif.) dialysis tubing with a molecular mass exclusion of molecules larger than 8,000 Da was rinsed in sterile water for 5 min and then boiled for 20 min in sterile water containing 1 mM EDTA. The bags were then rinsed with sterile water for 10 min, transferred to MOPS minimal medium to cool, and tied at one end using strict aseptic technique. Each bag was then filled with 10 ml of MOPS minimal medium containing either 2,000 P. aeruginosa cells/ml or a binary culture of P. aeruginosa and S. aureus each at 1,000 cells/ml. Stationary-phase cultures of MOPS medium-grown P. aeruginosa and S. aureus cells were the source of inocula for these experiments. After inoculation, the dialysis bags were tied at the other end, and excess tubing was removed from the ends of the dialysis bag. Four- to 6-week-old Sprague-Dawley rats (Sprague-Dawley Inc., Indianapolis, Ind.) were anesthetized by subcutaneous injection with a mixture of ketamine (50 mg/ml), xylazine (5 mg/ml), and acepromazine (1 mg/ml) at a dose of 1 ml/kg of body weight. The inoculated dialysis bags were then implanted into the rat's peritoneal cavity as outlined previously (1). At the desired time, the dialysis bags were explanted from the rats and rinsed with sterile MOPS minimal medium, and bacteria were removed using a sterile syringe.
Analysis of global gene expression using Affymetrix GeneChips.
Dialysis bags containing P. aeruginosa or P. aeruginosa-S. aureus cocultures were explanted from the rats at 18 h (optical density at 600 nm [OD600] = 0.3 to 0.4). In vivo-grown bacteria were removed from the dialysis bags by using a sterile syringe and mixed 1:1 with the RNA-stabilizing agent RNALater (Ambion, Austin, Tex.). To serve as an in vitro comparison, P. aeruginosa was grown aerobically in MOPS minimal medium containing either 20 mM glucose or 20 mM succinate to an OD600 of 0.4 and mixed 1:1 with RNALater. DNA-free RNA was isolated from in vitro- and in vivo-grown P. aeruginosa as outlined elsewhere, except for cocultures of P. aeruginosa and S. aureus, where the lysozyme-lysostaphin treatment was omitted (36). RNA integrity was monitored by gel electrophoresis of glyoxylated samples. Preparation of labeled cDNA and processing of the P. aeruginosa GeneChip arrays was performed as previously described (36). Washing, staining, and scanning of the GeneChips were performed by the University of Iowa DNA Core Facility using an Affymetrix fluidics station. GeneChips were performed in duplicate or triplicate for each culture condition. Data were analyzed using Microarray Suite software, and only genes exhibiting regulation levels of fivefold or greater are reported. Verification of GeneChip data was performed with semiquantitative reverse transcription-PCR (RT-PCR) using Superscript II (Invitrogen, Carlsbad, Calif.) as described previously (31). PA2426 (pvdS), PA2247 (bkdA1), and PA1717 (pscD) were confirmed using PA1802 (clpX) as the constitutively expressed control.
Lysis of S. aureus.
Lysis of S. aureus on petri plates was performed by thoroughly swabbing a BHI plate with an overnight culture of S. aureus diluted to an OD600 of 0.1. After drying, 5 μl of an overnight culture of P. aeruginosa was spotted onto the petri plate, dried, and incubated at 37°C for 24 h. Plates were imaged using an Alpha-Innotech documentation system.
P. aeruginosa growth yields.
For growth yield experiments using S. aureus as a source of iron, overnight bacterial cultures were centrifuged at 5,000 × g for 5 min and washed (three times) in chelexed MOPS minimal succinate medium. P. aeruginosa and S. aureus were then resuspended in chelexed MOPS minimal succinate medium (no added FeSO4) to an OD600 of 2.0 and 100, respectively. P. aeruginosa was then mixed 1:1 with S. aureus or chelexed MOPS minimal succinate medium (as a no-iron control). A sterile 25-mm polycarbonate membrane was placed onto a chelexed MOPS minimal succinate plate solidified with 1% agarose and allowed to dry. Five-microliter aliquots of these mixtures were placed onto the polycarbonate membrane, allowed to dry, and incubated at 37°C for 24 h (12). Membranes were resuspended in 1 ml MOPS minimal medium, and serial dilutions were performed. Viable bacteria were quantitated using pseudomonas isolation agar and Baird Parker agar plates. S. aureus does not grow on MOPS minimal succinate medium due to amino acid auxotrophy. To examine if P. aeruginosa could use lysed S. aureus as a source of iron, S. aureus (1 ml of OD600 = 100) was mechanically lysed by bead beating for 30 min in a Biospec Products minibead beater as outlined by the manufacturer (Biospec Products, Bartlesville, Okla.) with 0.1-mm beads. After filter sterilization, chelexed minimal succinate agarose plates containing mechanically lysed S. aureus (1/14 dilution) were used to grow P. aeruginosa as outlined above. To ensure the chelexed succinate plates were iron limited, P. aeruginosa was also grown with 10 μM FeSO4 added to the solidified medium.
RESULTS
Monitoring growth of P. aeruginosa in vivo.
The opportunistic human pathogen P. aeruginosa primarily causes infections in individuals with compromised immune systems. These infections often occur in the lungs of individuals with the heritable disease CF (19) and the peritoneum of individuals undergoing CAPD (25). Although a number of in vivo models exist to study P. aeruginosa lung and peritoneal pathogenesis (14, 29, 43-45), most of these models do not possess the versatility to perform genome-scale gene expression studies and study multispecies consortia. To begin to understand in vivo gene expression, we grew P. aeruginosa in a dialysis membrane chamber (DMC) implanted into the peritoneal cavity of a rat (1). In this in vivo model, P. aeruginosa undergoes a typical planktonic growth curve with a doubling time similar to that of in vitro glucose-grown bacteria (approximately 50 min in the DMC and 40 min in glucose minimal medium) (Fig. 1). Final bacterial densities achieved by P. aeruginosa in the DMC were greater than 1010 bacteria. Although the DMC model does not allow direct interactions with host cells, P. aeruginosa is growing on peritoneal contents; thus, we will refer to this model as an in vivo batch culture model.
FIG. 1.
Growth of P. aeruginosa in vitro in MOPS minimal medium with 20 mM glucose (▪) and in vivo in the DMC in the rat (▴). Representative growth curves are shown. Maximum doubling times were approximately 40 min in vitro and 50 min in vivo.
Identification and classification of P. aeruginosa genes differentially expressed in the peritoneum.
Although most laboratory experiments evaluating bacterial gene expression are conducted in vitro, in vivo growth conditions are difficult to mimic in the laboratory. To begin to understand the physiology of in vivo-grown P. aeruginosa, we performed transcriptome analyses of DMC-grown and in vitro-grown P. aeruginosa using Affymetrix GeneChips. As in vitro comparisons, P. aeruginosa was grown in MOPS minimal medium containing 20 mM glucose or 20 mM sodium succinate as the sole source of carbon and energy. To eliminate carbon source-specific genes from our analysis, only genes differentially expressed in vivo compared to glucose- and succinate-grown bacteria are reported.
Bacteria were harvested for GeneChip analysis at mid-logarithmic phase (OD600 = 0.3 to 0.5) under both in vitro and in vivo conditions, and gene expression profiles were compared using Affymetrix Microarray Suite software. On average, 71% of the gene-specific tiles present on the Affymetrix GeneChip hybridized at levels sufficient for statistical analysis; thus, we were able to compare expression profiles of approximately 4,000 genes. A total of 316 genes (approximately 5% of all P. aeruginosa genes) were differentially regulated at least fivefold during growth of P. aeruginosa in vivo compared to the in vitro controls. Of these genes, the majority were up-regulated (238 genes), while a smaller number were repressed (78 genes). As validation of our GeneChip results, we observed similar regulation patterns for PA2426 (pvdS), PA2247 (bkdA1), and PA1717 (pscD) using semiquantitaive RT-PCR (Fig. 2 and data not shown).
FIG. 2.
Lysis of S. aureus is necessary for iron acquisition by P. aeruginosa during coculture. (A) A BHI petri plate was swabbed with a confluent lawn of S. aureus, and 5 μl of an overnight culture of wild-type P. aeruginosa PA14 and P. aeruginosa PA14-LM1 containing a Tn5 insertion in pqsA were spotted on the plate and incubated at 37°C for 24 h. A zone of clearing (indicating lysis of S. aureus) is visible around the wild-type P. aeruginosa colony. Similar lysis results were also observed in test tube coculture experiments (data not shown). Although not shown here, approximately 20% of the time P. aeruginosa PA14-LM1 produces a small zone of lysis in this assay. (B) Semiquantitative RT-PCR analysis of the low-iron-inducible gene pvdS and the constitutively expressed clpX gene (42). P. aeruginosa was grown in vitro in glucose minimal medium containing excess FeSO4 (3.5 μM) or in vivo as a monoculture or in coculture with S. aureus. P. aeruginosa PA14-LM1, which is unable to lyse S. aureus but possesses no observed growth defects under iron-limited conditions, was also grown in vivo in coculture with S. aureus.
Genes in the sequenced P. aeruginosa genome are grouped into a number of classes based on functionality (www.pseudomonas.com). Classification of our in vivo-regulated genes revealed that most of these genes (33%) are of unknown function (Table 1). This is not surprising, given the observation that 45% of the 5,570 predicted P. aeruginosa genes have little homology to known proteins (37). The remaining genes are included in several classes, some of which possess distinct patterns of regulation. Most of the classes primarily include genes which are induced during in vivo growth, while all the genes in one group (bacteriophage/transposon) were repressed (Table 1). One of the most striking features of these data is that 29 transcriptional activators were induced during growth in vivo, indicating specific responses to the peritoneal environment. Only one gene involved in motility and attachment was differentially regulated during in vivo growth, indicating that these processes are quite similar with regard to transcription during in vitro and in vivo growth. Although no genes involved in lipopolysaccharide (LPS) biosynthesis were differentially expressed, it should be noted that most of the genes involved in O-antigen biosynthesis did not hybridize at levels sufficient for analysis. This is not surprising given the fact that we are not using the sequenced P. aeruginosa PAO1 strain, and O-antigen genes are highly divergent among P. aeruginosa strains (32).
TABLE 1.
Classes of in vivo-induced genes
| Functional classa | No. activatedb | No. repressedb |
|---|---|---|
| Adaptation, protection | 10 | 2 |
| Amino acid biosynthesis and metabolism | 17 | 6 |
| Biosynthesis of cofactors | 7 | 3 |
| Carbon compound catabolism | 3 | 2 |
| Cell wall-LPS | 1 | 0 |
| Central intermediary metabolism | 4 | 1 |
| Chaperones and heat shock protein | 2 | 0 |
| Energy metabolism | 25 | 6 |
| Membrane proteins | 2 | 0 |
| Motility and attachment | 1 | 0 |
| Nucleotide biosynthesis and metabolism | 1 | 0 |
| Protein-secretion-export apparatus | 4 | 0 |
| Putative enzymes | 15 | 2 |
| Related to phage, transposon, plasmid | 0 | 11 |
| Secreted factors | 2 | 0 |
| Transcriptional regulators | 29 | 2 |
| Translation, posttranslational modification, degradation | 1 | 0 |
| Transport of small molecules | 32 | 18 |
| Two-component regulatory systems | 3 | 0 |
| Hypothetical proteins | 65 | 19 |
| Unknown (conserved hypothetical) | 14 | 6 |
Functional classes are from the P. aeruginosa genome website (www.pseudomonas.com).
Genes that were activated or repressed in the DMC at least fivefold compared to cells grown in vitro in MOPS medium.
Correlation of GeneChip and IVET data.
Using IVET and a neutropenic mouse model, Wang et al. identified 19 genes induced during growth of P. aeruginosa in vivo (40). An examination of our in vivo-regulated genes indicated that of these 19 IVET genes, 6 did not hybridize at levels sufficient for GeneChip analysis, 10 were not differentially regulated, 2 were induced over fivefold in our DMC model (Table 2), and 1 was induced threefold in the DMC (PA4115). The two IVET genes induced over fivefold in our DMC model, np20 (PA5499) and fptA (PA4221), have been partially characterized in P. aeruginosa (2, 39, 40). The np20 gene was up-regulated approximately sixfold in the DMC and encodes a polypeptide with homology to a number of transcriptional regulators. The np20 locus is important for pathogenesis of P. aeruginosa (40). The fptA gene encodes a receptor for the siderophore pyochelin (discussed below) and is up-regulated approximately 81-fold in the DMC model (2).
TABLE 2.
P. aeruginosa genes differentially regulated during in vivo growth
| ORFa | Gene | Homology or functiona | Fold regulationb |
|---|---|---|---|
| 0149 | Probable sigma-70 factor | 7 | |
| 0171 | Hypothetical protein | −8 | |
| 0200 | Hypothetical protein | 7 | |
| 0280 | cysA | Sulfate transport protein | −11 |
| 0281 | cysW | Sulfate transport protein | −17 |
| 0282 | cysT | Sulfate transport protein | −11 |
| 0283 | sbp | Sulfate-binding protein precursor | −15 |
| 0284 | Hypothetical protein | −17 | |
| 0440* | Probable oxidoreductase | 7 | |
| 0466 | Hypothetical protein | −7 | |
| 0509* | nirN | Probablec-type cytochrome | 7 |
| 0510 | Probable uroporphyrin III c-methyltransferase | 13 | |
| 0511* | nirJ | Heme d1 biosynthesis protein | 7 |
| 0512 | Conserved hypothetical protein | 8 | |
| 0513 | Probable transcriptional regulator | 8 | |
| 0514* | nirL | Heme d1 biosynthesis protein | 11 |
| 0515 | Probable transcriptional regulator | 9 | |
| 0516* | nirF | Heme d1 biosynthesis protein | 8 |
| 0517* | nirC | Probablec-type cytochrome precursor | 11 |
| 0518* | nirM | Cytochromec551 precursor | 12 |
| 0519* | nirS | Nitrate reductase precursor | 15 |
| 0521 | Probable cytochrome c oxidase subunit | 6 | |
| 0523* | norC | Nitric oxide reductase subunit C | 32 |
| 0524* | norB | Nitric oxide reductase subunit B | 68 |
| 0525* | Probable denitrification protein | 15 | |
| 0587 | Conserved hypothetical protein | 5 | |
| 0613 | Hypothetical protein | −5 | |
| 0614 | Hypothetical protein | −5 | |
| 0617 | Probable bacteriophage protein | −10 | |
| 0621 | Conserved hypothetical protein | −5 | |
| 0622 | Probable bacteriophage protein | −8 | |
| 0627 | Conserved hypothetical protein | −5 | |
| 0629 | Conserved hypothetical protein | −7 | |
| 0631 | Hypothetical protein | −6 | |
| 0632 | Hypothetical protein | −7 | |
| 0633 | Hypothetical protein | −8 | |
| 0635 | Hypothetical protein | −11 | |
| 0636 | Hypothetical protein | −8 | |
| 0639 | Conserved hypothetical protein | −9 | |
| 0641 | Probable bacteriophage protein | −19 | |
| 0674 | Hypothetical protein | 25 | |
| 0675 | Probable sigma-70 factor | 5 | |
| 0713 | Hypothetical protein | 26 | |
| 0714 | Hypothetical protein | 24 | |
| 0781 | Hypothetical protein | 18 | |
| 0782† | putA | Proline dehydrogenase | 8 |
| 0865† | hpd | 4-Hydroxyphenylpyruvate dioxygenase | 5 |
| 0872† | phhA | Phenylalanine-4-hydroxylase | 31 |
| 0910 | Hypothetical protein | −18 | |
| 0911 | Hypothetical protein | −6 | |
| 0921 | Hypothetical protein | −6 | |
| 0984 | Colicin immunity protein | 7 | |
| 1123 | Hypothetical protein | 5 | |
| 1178 | oprH | Outer membrane protein H1 precursor | −17 |
| 1179 | phoP | Two-component response regulator | −5 |
| 1195 | Hypothetical protein | 6 | |
| 1196 | Probable transcriptional regulator | 5 | |
| 1382 | Probable type II secretion system protein | 12 | |
| 1414 | Hypothetical protein | 5 | |
| 1516 | Hypothetical protein | 5 | |
| 1537 | Probable short-chain dehydrogenase | 6 | |
| 1540 | Conserved hypothetical protein | 7 | |
| 1541 | Probable drug efflux transporter | 8 | |
| 1552 | Probable cytochrome c | −12 | |
| 1553 | Probable cytochrome c oxidase subunit | −13 | |
| 1554 | Probable cytochrome oxidase subunit | −5 | |
| 1673 | Hypothetical protein | 5 | |
| 1742 | Probable amidotransferase | 7 | |
| 1746 | Hypothetical protein | 9 | |
| 1837 | Hypothetical protein | −13 | |
| 1838 | cysI | Sulfite reductase | −10 |
| 1871 | lasA | LasA protease precursor | 6 |
| 1887 | Hypothetical protein | 11 | |
| 1888 | Hypothetical protein | 10 | |
| 1892 | Hypothetical protein | −9 | |
| 1894 | Hypothetical protein | −9 | |
| 1895 | Hypothetical protein | −10 | |
| 1896 | Hypothetical protein | −9 | |
| 1897 | Hypothetical protein | −35 | |
| 1911 | Probable transmembrane sensor | 6 | |
| 1912 | Probable sigma-70 factor | 7 | |
| 1922 | Probable TonB-dependent receptor | 11 | |
| 1924 | Hypothetical protein | 9 | |
| 1925 | Hypothetical protein | 8 | |
| 1999† | Probable CoA transferase, subunit A | 35 | |
| 2000† | Probable CoA transferase, subunit B | 44 | |
| 2001 | atoB | Acetyl-CoA acetyltransferase | 13 |
| 2003 | bdhA | 3-Hydroxybutyrate dehydrogenase | 6 |
| 2004 | Conserved hypothetical protein | 8 | |
| 2016 | Probable transcriptional regulator | 5 | |
| 2114 | Probable MFS transporter | −7 | |
| 2202 | Probable amino acid permease | −7 | |
| 2204 | Probable binding protein component of ABC transporter | −10 | |
| 2223 | Hypothetical protein | 6 | |
| 2247† | bkdA1 | 2-Oxoisovalerate dehydrogenase (alpha subunit) | 38 |
| 2248† | BkdA2 | 2-Oxoisovalerate dehydrogenase (beta subunit) | 14 |
| 2249† | bkdB | Branched-chain alpha-keto acid dehydrogenase (lipoamide component) | 15 |
| 2250† | lpdV | Lipoamide dehydrogenase (−Val) | 9 |
| 2466 | Probable TonB-dependent receptor | 5 | |
| 2534 | Probable transcriptional regulator | 5 | |
| 2567 | Hypothetical protein | 5 | |
| 2648 | nuoM | NADH dehydrogenase I chain | −5 |
| 2686 | pfeR | Two-component response regulator | 6 |
| 2687 | pfeS | Two-component sensor | 5 |
| 2753 | Hypothetical protein | 7 | |
| 2807 | Hypothetical protein | −19 | |
| 2825 | Probable transcriptional regulator | 5 | |
| 2826 | Probable glutathione peroxidase | 5 | |
| 2929 | Hypothetical protein | 5 | |
| 2931 | Probable transcriptional regulator | 19 | |
| 2932 | morB | Morphinone reductase | 71 |
| 2933 | Probable MFS transporter | 31 | |
| 2934 | Probable hydrolase | 16 | |
| 2945 | Conserved hypothetical protein | −6 | |
| 3038 | Probable porin | −6 | |
| 3118† | leuB | 3-Isopropylmalate dehydrogenase | −5 |
| 3120† | leuD | 3-Isopropylmalate dehydratase small subunit | −6 |
| 3121† | leuC | 3-Isopropylmalate dehydratase large subunit | −6 |
| 3126 | ibpA | Heat shock protein | 7 |
| 3188 | Probable permease of ABC sugar transporter | −382 | |
| 3189 | Probable permease of ABC sugar transporter | −79 | |
| 3195 | gapA | Glyceraldehyde-3-phosphate dehydrogenase | −6 |
| 3249 | Probable transcriptional regulator | 5 | |
| 3309 | Conserved hypothetical protein | 6 | |
| 3313 | Hypothetical protein | −5 | |
| 3391* | nosR | Regulatory protein | 54 |
| 3392* | nosZ | Nitrous oxide reductase precursor | 25 |
| 3393* | nosD | NosD protein | 11 |
| 3394* | nosF | NosF protein | 16 |
| 3395* | nosY | NosY protein | 8 |
| 3396* | nosL | NosL protein | 8 |
| 3404 | Probable secretion protein | 7 | |
| 3405 | hasE | Metalloprotease secretion protein | 32 |
| 3406 | hasD | Transport protein | 16 |
| 3415 | Probable dihydrolipoamide acetyltransferase | 9 | |
| 3416 | Probable pyruvate dehydrogenase E1 component, beta chain | 36 | |
| 3417 | Probable pyruvate dehydrogenase E1 component, alpha subunit | 7 | |
| 3418† | ldh | Leucine dehydrogenase | 8 |
| 3441 | Probable molybdopterin-binding protein | −12 | |
| 3442 | Probable ATP-binding component of ABC transporter | −21 | |
| 3443 | Probable permease of ABC transporter | −43 | |
| 3444 | Conserved hypothetical protein | −161 | |
| 3445 | Conserved hypothetical protein | −62 | |
| 3446 | Conserved hypothetical protein | −19 | |
| 3450 | Probable antioxidant protein | −8 | |
| 3516 | Probable lyase | −5 | |
| 3581 | glpF | Glycerol uptake facilitator protein | 9 |
| 3582 | glpK | Glycerol kinase | 6 |
| 3584 | glpD | Glycerol-3-phosphate dehydrogenase | 44 |
| 3598 | Conserved hypothetical protein | 7 | |
| 3600 | Conserved hypothetical protein | 139 | |
| 3601 | Conserved hypothetical protein | 83 | |
| 3719 | Hypothetical protein | 21 | |
| 3720 | Hypothetical protein | 30 | |
| 3721 | Probable transcriptional regulator | 5 | |
| 3837 | Probable permease of ABC transporter | −5 | |
| 3862 | Hypothetical protein | 6 | |
| 3870 | moaA1 | Molybdopterin biosynthetic protein | 7 |
| 3871 | Probable peptidyl-prolyl cis-trans isomerase | 7 | |
| 3872* | narI | Respiratory nitrate reductase gamma chain | 9 |
| 3873* | narJ | Respiratory nitrate reductase delta chain | 11 |
| 3874* | narH | Respiratory nitrate reductase beta chain | 24 |
| 3875* | narG | Respiratory nitrate reductase alpha chain | 19 |
| 3876* | narK2 | Nitrite extrusion protein 2 | 10 |
| 3877* | narK1 | Nitrite extrusion protein 1 | 6 |
| 3914 | moeA1 | Molybdenum cofactor biosynthetic protein | 7 |
| 3915 | moaB1 | Molybdopterin biosynthetic protein | 5 |
| 3922 | Conserved hypothetical protein | −6 | |
| 3931 | Conserved hypothetical protein | −16 | |
| 3933 | Probable choline transporter | 10 | |
| 3935 | tauD | Taurine dioxygenase | −16 |
| 3936 | Probable permease of ABC taurine transporter | −35 | |
| 3937 | Probable ATP-binding component of ABC taurine transporter | −25 | |
| 3938 | Probable periplasmic taurine-binding protein precursor | −21 | |
| 4063 | Hypothetical protein | 15 | |
| 4064 | Probable ATP-binding component of ABC transporter | 7 | |
| 4065 | Hypothetical protein | 5 | |
| 4156 | Probable TonB-dependent receptor | 10 | |
| 4160 | fepD | Ferric enterobactin transport protein | 5 |
| 4167 | Probable oxidoreductase | 13 | |
| 4168 | Probable TonB-dependent receptor | 17 | |
| 4181 | Hypothetical protein | 16 | |
| 4182 | Hypothetical protein | 10 | |
| 4197 | Probable two-component sensor | 10 | |
| 4288 | Probable transcriptional regulator | 8 | |
| 4333 | Probable fumarase | −5 | |
| 4364 | Hypothetical protein | 113 | |
| 4365 | Probable transporter | 54 | |
| 4442† | cysN | ATP sulfurylase GTP-binding subunit, APS kinase | −22 |
| 4443† | cysD | ATP sulfurylase small subunit | −10 |
| 4577 | Hypothetical protein | 5 | |
| 4588† | gdhA | Glutamate dehydrogenase | −6 |
| 4610 | Hypothetical protein | 9 | |
| 4621 | Probable oxidoreductase | 5 | |
| 4657 | Hypothetical protein | 5 | |
| 4738 | Conserved hypothetical protein | −6 | |
| 4739 | Conserved hypothetical protein | −9 | |
| 4834 | Hypothetical protein | 18 | |
| 4835 | Hypothetical protein | 16 | |
| 4836 | Hypothetical protein | 32 | |
| 4837 | Probable outer membrane protein | 38 | |
| 4838 | Hypothetical protein | 7 | |
| 5024 | Conserved hypothetical protein | −13 | |
| 5027 | Hypothetical protein | 5 | |
| 5088 | Hypothetical protein | 7 | |
| 5100† | hutU | Urocanase | 7 |
| 5102 | Hypothetical protein | −8 | |
| 5106 | Conserved hypothetical protein | 6 | |
| 5170† | arcD | Arginine-ornithine antiporter | 11 |
| 5171† | arcA | Arginine deiminase | 9 |
| 5172† | arcB | Ornithine carbamoyltransferase, catabolic | 10 |
| 5173† | arcC | Carbamate kinase | 6 |
| 5302† | dadX | Catabolic alanine racemase | 6 |
| 5303 | Conserved hypothetical protein | 9 | |
| 5304† | dadA | d-Amino acid dehydrogenase, small subunit | 9 |
| 5351 | Rubredoxin | −5 | |
| 5372† | betA | Choline dehydrogenase | 12 |
| 5373† | betB | Betaine aldehyde dehydrogenase | 12 |
| 5374 | betI | Transcriptional regulator | 11 |
| 5427 | adhA | Alcohol dehydrogenase | 6 |
| 5446 | Hypothetical protein | 13 | |
| 5475 | Hypothetical protein | 5 | |
| 5481 | Hypothetical protein | −7 | |
| 5482 | Hypothetical protein | −8 | |
| 5499‡ | np20 | Transcriptional regulator | 6 |
| 5500 | znuC | Zinc transport protein | 6 |
| 5532 | Hypothetical protein | 8 | |
| 5534 | Hypothetical protein | 13 | |
| 5535 | Conserved hypothetical protein | 15 | |
| 5536 | Conserved hypothetical protein | 52 | |
| 5538 | amiA | N-Acetylmuramoyl-l-alanine amidase | 28 |
| 5539 | Hypothetical protein | 8 | |
| 5540 | Hypothetical protein | 17 | |
| 5541 | Probable dihydroorotase | 11 |
From the P. aeruginosa genome website, www.pseudomonas.com. Genes involved in anaerobic metabolism (*), transport and metabolism of amino acids (†), or identified previously using IVET (‡) are listed in bold.
Fold change in P. aeruginosa mRNA level when grown in vivo in the DMC compared to that in bacteria grown in glucose minimal medium in vitro. Positive numbers represent induction, and negative numbers indicate repression in the DMC. All genes listed were also >5- fold regulated when DMC-grown and succinate-grown P. aeruginosa cultures were compared.
Iron acquisition in vivo.
Most bacteria require iron for growth, and acquisition of iron is a significant challenge in most in vivo environments where little free iron exists. P. aeruginosa and many other bacteria produce high-affinity iron chelators called siderophores to acquire iron in vivo (7). These extracellular chelators are able to scavenge iron from in vivo sources and deliver it to the bacterium. P. aeruginosa produces two well-studied siderophores, pyoverdine and pyochelin, and the genes important for their syntheses are induced in low-iron environments (7). Ochsner et al. recently reported a transcriptome analysis of P. aeruginosa grown in vitro under low- and high-iron conditions (28). Of the 113 genes which were induced at least fivefold in this study, under low-iron conditions, 82 were also induced at least fivefold in our DMC model (Table 3). Included within these in vivo-induced genes were genes involved in synthesis and binding of pyoverdine and pyochelin, as well as multiple genes important for heme uptake (Table 3). Heme may serve as a source of iron in vivo, and the hasAp gene involved in heme utilization was the most highly induced in vivo gene (over 2,000-fold induction).
TABLE 3.
Expression of P. aeruginosa iron-regulated genes during monoculture and coculture growth in vivo
| ORFa | Gene | Homology or functiona | Fold regulationb
|
|
|---|---|---|---|---|
| In vivo vs in vitro | Coculture vs monoculture | |||
| 0471 | Probable transmembrane sensor | 7 | −17 | |
| 0472 | Probable sigma-70 factor | 8 | −34 | |
| 0500 | bioB | Biotin synthase | −9 | NC |
| 0672 | Hypothetical protein | 44 | −44 | |
| 0707 | toxR | Transcriptional regulator | 6 | −4 |
| 1245 | Hypothetical protein | 5 | −3 | |
| 1300 | Probable sigma-70 factor | 35 | −37 | |
| 1301 | Probable transmembrane sensor | 13 | −12 | |
| 2033 | Hypothetical protein | 20 | −23 | |
| 2034 | Hypothetical protein | 20 | −16 | |
| 2384 | Hypothetical protein | 41 | −46 | |
| 2385 | Probable acylase | 34 | −30 | |
| 2386 | pvdA | l-Ornithine N5-oxygenase | 85 | −74 |
| 2389 | Conserved hypothetical protein | 9 | −9 | |
| 2390 | Probable ATP-binding/permease fusion ABC transporter | 7 | −8 | |
| 2391 | Probable outer membrane protein | 5 | −7 | |
| 2392 | Hypothetical protein | 18 | −12 | |
| 2393 | Probable dipeptidase precursor | 53 | −127 | |
| 2394 | Probable aminotransferase | 57 | −60 | |
| 2395 | Hypothetical protein | 13 | −38 | |
| 2396 | Hypothetical protein | 13 | −7 | |
| 2397 | pvdE | Pyoverdine biosynthesis protein | 44 | −25 |
| 2398 | fpvA | Ferripyoverdine receptor | 30 | −86 |
| 2399 | pvdD | Pyoverdine synthetase | 40 | −23 |
| 2400 | Probable nonribosomal peptide synthetase | 94 | −28 | |
| 2401 | Probable nonribosomal peptide synthetase | 94 | −40 | |
| 2402 | Probable nonribosomal peptide synthetase | 46 | −28 | |
| 2403 | Hypothetical protein | 58 | −22 | |
| 2404 | Hypothetical protein | 43 | −39 | |
| 2405 | Hypothetical protein | 43 | −37 | |
| 2406 | Hypothetical protein | 24 | −21 | |
| 2407 | Probable adhesion protein | 14 | −18 | |
| 2408 | Probable ATP-binding component of ABC transporter | 38 | −11 | |
| 2411 | Probable thioesterase | 84 | −111 | |
| 2412 | Conserved hypothetical protein | 197 | −203 | |
| 2413 | Probable class III aminotransferase | 51 | −136 | |
| 2424 | Probable nonribosomal peptide synthetase | 112 | −53 | |
| 2425 | Probable thioesterase | 44 | −51 | |
| 2426 | pvdS | Sigma factor | 66 | −117 |
| 2427 | Hypothetical protein | 14 | −8 | |
| 2451 | Hypothetical protein | 9 | −9 | |
| 2452 | Hypothetical protein | 73 | −110 | |
| 2467 | Probable transmembrane sensor | 5 | −8 | |
| 2468 | Probable sigma-70 factor | 6 | −13 | |
| 3407 | hasAp | Heme acquisition protein | 2,180 | −724 |
| 3408 | hasR | Heme acquisition protein | 74 | −74 |
| 3409 | Probable transmembrane sensor | 9 | −7 | |
| 3410 | Probable sigma-70 factor | 14 | −18 | |
| 3530 | Conserved hypothetical protein | 9 | −39 | |
| 3899 | Probable sigma-70 factor | 17 | −26 | |
| 3900 | Probable transmembrane sensor | 7 | −11 | |
| 3901 | fecA | Fe(III) dicitrate transport protein | 9 | −10 |
| 4158 | fepC | Ferric enterobactin transport protein | 13 | −13 |
| 4218 | Probable transporter | 40 | −22 | |
| 4219 | Hypothetical protein | 109 | −66 | |
| 4220 | Hypothetical protein | 288 | −48 | |
| 4221 | fptA | Fe(III) pyochelin receptor precursor | 81 | −92 |
| 4222 | Probable ATP-binding component of ABC transporter | 34 | −13 | |
| 4223 | Probable ATP-binding component of ABC transporter | 26 | −15 | |
| 4224 | Hypothetical protein | 58 | −34 | |
| 4225 | pchF | Pyochelin synthetase | 52 | −65 |
| 4226 | pchE | Dihydroaeruginoic acid synthetase | 53 | −46 |
| 4227 | pchR | Transcriptional regulator | 21 | −76 |
| 4228 | pchD | Pyochelin biosynthesis protein | 74 | −66 |
| 4229 | pchC | Pyochelin biosynthesis protein | 99 | −163 |
| 4230 | pchB | Salicylate biosynthesis protein | 126 | −118 |
| 4231 | pchA | Salicylate biosynthesis isochorismate synthetase | 122 | −89 |
| 4359 | Conserved hypothetical protein | 5 | −10 | |
| 4467 | Hypothetical protein | 21 | −10 | |
| 4468 | sodM | Superoxide dismutase | 40 | −48> |
| 4469 | Hypothetical protein | 48 | −53 | |
| 4470 | fumC1 | Fumarate hydratase | 60 | −106 |
| 4471 | Hypothetical protein | 100 | −131 | |
| 4570 | Hypothetical protein | 37 | −25 | |
| 4708 | Hypothetical protein | 9 | −6 | |
| 4709 | Probable hemin degrading factor | 9 | −8 | |
| 4710 | Probable outer membrane hemin receptor | 19 | −30 | |
| 4895 | Probable transmembrane sensor | 12 | −13 | |
| 4896 | Probable sigma-70 factor | 14 | −25 | |
| 4973 | thiC | Thiamin biosynthesis protein | −14 | NC |
| 5312 | Probable aldehyde dehydrogenase | 6 | NC | |
| 5313 | Probable pyridoxal-dependent aminotransferase | 7 | NC | |
From the P. aeruginosa genome website, www.pseudomonas.com.
Regulation of P. aeruginosa iron-regulated genes (28) as determined by Affymetrix GeneChip analysis. Two conditions were compared: monoculture growth in vivo versus monoculture in vitro growth in glucose minimal medium with added FeSO4 (positive numbers represent induction in vivo); and in vivo coculture growth versus in vivo monoculture growth (positive numbers represent induction during coculture growth). NC, no change.
Oxygen levels in the peritoneum.
P. aeruginosa grows anaerobically by using nitrate and its reduced derivatives as terminal electron acceptors. If nitrate is not present, arginine may be metabolized fermentatively by P. aeruginosa (38). The genes encoding proteins critical for anaerobic nitrate and arginine metabolism are positively regulated under low-oxygen conditions (34). An examination of our in vivo GeneChip data indicates that several P. aeruginosa operons involved in utilization of nitrate, nitrite, nitric oxide, nitrous oxide, and arginine were highly up-regulated during in vivo growth compared to aerobic glucose-grown cells (Table 2). Thus, we hypothesize that P. aeruginosa would possess the enzymatic capabilities of anaerobically grown bacteria, including the ability to reduce nitrate via the nitrate reductase enzyme. An evaluation of the nitrate reductase activity of in vivo-grown P. aeruginosa revealed that in vivo-grown P. aeruginosa was capable of reducing exogenously added nitrate to levels below detection (data not shown).
P. aeruginosa physiology and metabolism in vivo.
To cause more than a transient infection, P. aeruginosa must survive and grow within the host environment. However, the growth environment within the host is not well understood, other than the observation that P. aeruginosa mutants unable to synthesize purine precursors grow poorly in mouse models (40). Thus, it is critical to understand the growth environment of P. aeruginosa in vivo. Our in vitro conditions required P. aeruginosa to metabolize glucose or succinate as the sole carbon source and synthesize all essential metabolic precursors de novo. Under these in vitro conditions as our GeneChip control, our data indicate that genes encoding proteins involved in transport and metabolism of amino acids, particularly aromatic (PA0865 and PA0872) and branched-chain (PA2247 to PA2250) amino acids, are highly induced during growth in the DMC (Table 2). These data suggest that P. aeruginosa is using amino acids as a carbon source in the peritoneum.
Growth of P. aeruginosa and S. aureus cocultures in vitro and in vivo.
It is clear that a number of bacterial infections are not simply the result of colonization by one microorganism, but of the pathogenic contributions of several organisms (13, 46). Such is the case with chronic lung infections and peritoneal infections in patients undergoing dialysis, where multispecies consortia, often including P. aeruginosa and S. aureus (10, 19, 20, 22, 46), are associated with disease. Although these bacteria may reside in the same pathogenic environment, essentially nothing is known about if or how they interact in vivo. To begin to understand the interactions between P. aeruginosa and S. aureus in vivo, these bacteria were grown in monoculture and in coculture in the DMC model. When grown in monoculture, S. aureus possesses a faster doubling time than P. aeruginosa in the DMC (30 min for S. aureus versus 50 min for P. aeruginosa). When P. aeruginosa and S. aureus are started at identical densities and grown in the DMC model, P. aeruginosa grows similar to that under monoculture conditions, reaching densities of 5 × 108 bacteria/ml after 15 to 18 h. S. aureus reaches slightly lower growth yields at this time during coculture in the DMC (4 × 109 bacteria/ml) compared to monoculture growth (1010 bacteria/ml).
Gene expression of P. aeruginosa cocultured in vivo with S. aureus.
To determine the impact of S. aureus on P. aeruginosa physiology during growth in the peritoneum, we isolated RNA from P. aeruginosa grown in vivo with S. aureus. As mentioned above, P. aeruginosa reaches similar densities after 15 to 18 h in the DMC when growing as a monoculture or in coculture with S. aureus; thus, we sampled bacteria for RNA isolation at these time points under both conditions. To enrich for P. aeruginosa RNA during isolation, S. aureus lysis was prevented by omitting the lysozyme-lysostaphin treatment. If any contaminating S. aureus RNA were present, it would likely not cross-hybridize because of the significant differences in the GC content of these bacteria and the built-in mismatch controls on the Affymetrix GeneChips.
We compared the transcriptomes of P. aeruginosa grown in vivo in monoculture to those of P. aeruginosa grown in coculture with S. aureus in vivo. A total of 178 genes were differentially expressed at least fivefold during in vivo growth with S. aureus, with the majority of these genes (131) repressed in the coculture situation. One of the most striking features of these data is the fact that over 95% (78 of 82) of the genes repressed during in vivo growth with S. aureus are regulated by iron availability (Table 3). This includes genes involved in pyoverdine and pyochelin biosynthesis and suggests that P. aeruginosa growing in coculture with S. aureus perceives its environment as high in iron, in contrast to the monoculture in vivo situation, where the perceived environment is low in iron (Table 3). In fact, no differences in iron-regulated genes were observed when the transcriptomes of high-iron, in vitro-grown P. aeruginosa were compared to those of P. aeruginosa grown in vivo with S. aureus (data not shown). This suggests that the presence of S. aureus must locally concentrate useful iron for P. aeruginosa or increase the concentration of free iron in the peritoneum. The former is likely, since the molecular weight exclusion of the dialysis membranes prevents diffusion of most S. aureus toxins and most in vivo iron is chelated by transferrin or lactoferrin.
Lysis of S. aureus is required for iron acquisition.
Given the observation that P. aeruginosa lyses several bacteria, including S. aureus (11, 17, 26), we hypothesized that the increased iron levels perceived by P. aeruginosa during in vivo coculture may be partially explained by S. aureus lysis and subsequent release of intracellular iron. To test this hypothesis, we cocultured S. aureus in vivo with P. aeruginosa PA14 and P. aeruginosa PA14-LM1, which contains a Tn5 insertion in pqsA and exhibits reduced lysis of S. aureus (Fig. 2A). The pqsA gene is involved in biosynthesis of several quinolone molecules in P. aeruginosa important for cell-cell signaling and lysis of S. aureus but possesses no known involvement in regulation of low-iron-inducible genes (data not shown). We used the low-iron-inducible gene pvdS as a marker gene for iron limitation and evaluated transcript levels of pvdS in these cocultures using RT-PCR. As expected from our GeneChip data, the wild-type PA14 perceives the in vivo environment as high in iron when cocultured with S. aureus (low levels of pvdS transcript), while PA14-LM1 perceives its environment as low in iron (high levels of pvdS mRNA) in coculture (Fig. 2B).
P. aeruginosa can use S. aureus as an iron source.
These data suggest that P. aeruginosa can use S. aureus as an iron source. If this were true, P. aeruginosa should grow to higher densities in iron-limited media when grown in coculture with S. aureus or when grown in the presence of lysed S. aureus cells. To test this hypothesis, we grew P. aeruginosa PA14 in iron-replete medium in vitro as a monoculture, as a monoculture in the presence of mechanically lysed S. aureus, and in coculture with viable S. aureus. Final P. aeruginosa growth yields showed that the wild-type strain grew to higher densities when grown in coculture with S. aureus or in the presence of S. aureus lysate (Fig. 3), indicating that viable and lysed S. aureus can be used as a source of iron by wild-type P. aeruginosa. To assess if lysis is required for acquisition of iron from viable S. aureus cells, we grew P. aeruginosa PA14-LM1 in the presence and absence of viable and lysed S. aureus (Fig. 3). Growth of PA14-LM1 in the presence of lysed S. aureus greatly increased the growth yields over monoculture growth, indicating that this mutant can use lysed S. aureus as an iron source. However, significantly increased growth yields were not observed for PA14-LM1 when cocultured with viable S. aureus, suggesting that the inability of PA14-LM1 to effectively lyse S. aureus impairs its growth under low-iron coculture conditions.
FIG. 3.
Growth yields of wild-type P. aeruginosa and P. aeruginosa PA14-LM1 grown in iron-limited MOPS minimal medium (see Materials and Methods) as a monoculture, with 10 μM FeSO4, in coculture with S. aureus, and with the addition of mechanically lysed S. aureus. Data are expressed as (the ratio of P. aeruginosa cell yield in the presence of the addition)/(monoculture cell yields with no addition) (numbers >1 indicate increased cell yield during growth with the indicated addition). For the two no-addition controls, the ratio was calculated by pair-wise comparisons of four individual replicates. S. aureus does not grow in this minimal medium due to amino acid auxotrophy. Less than 103 S. aureus cells were present in PA14-S. aureus coculture colonies at the time of sampling, while >109 remained in the PA14-LM1-S. aureus cocultures. Error bars represent the standard deviation for three to four experimental repeats.
DISCUSSION
The goal of this study was to examine gene expression of P. aeruginosa during in vivo growth in order to better understand the physiology of this bacterium during infection. We used a rat DMC model in conjunction with DNA microarrays to identify P. aeruginosa genes differentially regulated during growth in vivo as a monoculture and in coculture with S. aureus. This model involves growing P. aeruginosa in a dialysis bag implanted in the peritoneum of the rat. The DMC model does not allow direct interactions with host molecules larger than 8 kDa, including host cells; thus, genes differentially regulated by such interactions will not be detected. However, this model makes microarray and proteomic approaches viable and is easily manipulated to observe interactions between bacteria in an in vivo growth environment. P. aeruginosa grows well in the DMC, with a doubling time of less than 1 h and reaching final bacterial densities of over 1010 bacteria, illustrating that the peritoneum is a high-nutrient growth environment.
To identify in vivo-specific genes, we compared the transcriptome of in vivo-grown P. aeruginosa to bacteria grown in vitro in MOPS minimal medium with glucose or succinate as the sole source of carbon and energy. These carbon sources were chosen for comparison since most P. aeruginosa metabolic studies have focused on growth using these substrates. By comparing our in vivo transcriptome results to two in vitro growth conditions, we were able to decrease the number of in vivo-regulated genes by 30 to 40% compared to single in vitro growth conditions. We believe this analysis effectively eliminated many carbon source-specific genes from our analysis. It should be noted that both of our in vitro growth conditions require P. aeruginosa to synthesize all anabolic precursors de novo, thus allowing us to make predictions about the nutritional environment of the peritoneum. Our data indicate that amino acids are likely a substrate for growth in the peritoneum, since genes involved in catabolism and transport of branched-chain and aromatic amino acids are induced during growth in the DMC (Table 2). Although further studies are necessary to determine the specific growth substrates, we feel the DMC model provides an easily manipulatable preliminary model for these studies. Metabolic genes identified as important for growth in the DMC can then be tested in more relevant animal systems.
Our transcriptome analysis of in vivo monocultures did not show activation of most of the P. aeruginosa genes identified by Wang et al. as in vivo-induced by IVET (Tables 2 and 3). There are a number of reasons to account for this difference, the most obvious being the differences in the mouse models used. Wang et al. used a neutropenic mouse model and harvested bacteria from the mouse liver, whereas our peritoneal model is a much simpler in vivo batch growth model. Our study also used glucose- or succinate-grown P. aeruginosa as an in vitro control, whereas Wang et al. used a common complex medium for IVET library construction. Regardless of the differences in the studies, it is an important observation that three genes were common between the studies. Of these three genes, much is known about the FptA pyochelin receptor, which is important for iron acquisition in low-iron environments. Much less is known about np20, other than it is induced in vivo and is important for production of the extracellular virulence factor pyocyanin, and nothing is known about the third gene (PA4115), which was only induced threefold during growth in the peritoneum. Our data in combination with the previous IVET analysis indicate that these three genes are members of a core set of genes inducible during in vivo growth in at least two disparate animal models. The np20 gene is the most intriguing, given its homology with a number of transcriptional regulators. Further study of these genes should provide clues to the in vivo environment and potentially identify processes important for in vivo growth and pathogenesis.
Our transcriptome analysis of in vivo P. aeruginosa monocultures revealed that the peritoneal environment is low in iron (Table 3). These data are not surprising and coincide with a number of studies evaluating bacterial growth in vivo (21, 24, 40). It is interesting that the in vitro response of P. aeruginosa to low iron reported by Ochsner et al. (28) is remarkably similar to our in vivo results, indicating that in vitro studies evaluating the response to low iron are applicable to understanding in vivo growth and pathogenesis. It should be noted that although our data indicate that the peritoneal environment is low in free iron, we do not believe that iron is severely limiting growth, since P. aeruginosa grows near its maximum doubling rate in the rat (50-min doubling time). Thus, it appears that although low in free iron, as expected P. aeruginosa is well adapted to growth in the peritoneum.
Our transcriptome data also indicate that the peritoneum is low in oxygen, since many genes known to be regulated by anaerobiosis were highly induced during growth in vivo (Table 2). Based on the fast growth rate of P. aeruginosa in the DMC, it is likely that sufficient levels of oxygen and/or nitrate are available in the peritoneum for growth, since the maximum doubling time anaerobically in vitro using arginine is approximately 5 h (data not shown). Nitrate is present within the peritoneum, as we detected low micromolar levels of nitrate from an uninoculated DMC chamber (data not shown). Anaerobic growth in the peritoneum is relevant to pathogenesis, given the recent evidence that the ability of P. aeruginosa to form antibiotic-resistant biofilms is enhanced in low-oxygen environments (18). Many P. aeruginosa infections, including those in the CF lung and in the peritoneum, are a consequence of biofilm formation (8, 9, 18). The observation that P. aeruginosa growing in the peritoneum perceives its environment as low in oxygen enhances the importance of this model to pathogenic studies, including biofilm studies. Although in this study we focused on planktonic bacteria, we have preliminary evidence that P. aeruginosa readily forms antibiotic-resistant biofilms on catheter tubing placed in the DMC (data not shown).
The most interesting observation from this study is that wild-type P. aeruginosa perceives its environment as high in iron when grown with S. aureus in the DMC. Our data suggest that the change in iron perception by P. aeruginosa during coculture with S. aureus is dependent on lysis of S. aureus, since P. aeruginosa PA14-LM1, which is unaffected in iron acquisition or growth in vivo but cannot effectively lyse S. aureus (Fig. 2A), perceives its environment as low in iron during coculture growth with S. aureus (Fig. 2B). This would not be surprising, since P. aeruginosa produces several extracellular antimicrobial molecules (many controlled by the pqsA-E operon) known to be important for lysis of S. aureus (11, 17, 26, 27). The DMC coculture experiments were also supported by in vitro data, which showed that growth yields of P. aeruginosa were increased in iron-limited media when grown in coculture with S. aureus (Fig. 3). This again was dependent on lysis of S. aureus, since growth yields of PA14-LM1 were not increased in coculture. This deficiency in growth by PA14-LM1 cannot be explained by competition with S. aureus for iron, since S. aureus is auxotrophic and will not grow in the minimal succinate medium used in these experiments. We believe these data together suggest that P. aeruginosa lyses S. aureus and can use it as an iron source.
S. aureus is capable of undergoing autolysis during in vitro growth (5, 6, 15, 16). Our data do not distinguish whether P. aeruginosa is causing direct lysis of S. aureus or inducing autolysis; however, our results with P. aeruginosa PA14-LM1 indicate that P. aeruginosa-independent autolysis by S. aureus is most likely not the primary mechanism of iron acquisition during coculture. We are currently dissecting the specific mechanism(s) of S. aureus lysis, and our preliminary work and the work of others indicate that it is most likely multifactorial (11, 26). Lysis of S. aureus by P. aeruginosa is not unique to the laboratory strains used in this study. Our laboratory P. aeruginosa strain exhibited visible lysis of seven S. aureus strains tested, and 24 of 29 (83%) P. aeruginosa CF lung isolates tested produced visible lysis of our laboratory S. aureus strain (data not shown).
Although lysis appears to be important for iron acquisition during coculture in vitro and in vivo, the source of the iron is unknown. Our experiments indicate that P. aeruginosa can use mechanically lysed S. aureus as a source of iron, and it is likely that iron-containing proteins released from the lysed S. aureus cells serve as the source of iron. The DMC dialysis membrane will prevent diffusion of proteins larger than 8,000 Da, which may act to concentrate iron-containing proteins; thus, we cannot be certain whether iron acquisition through S. aureus lysis is important for P. aeruginosa in vivo. However, it is clear that the physiology of P. aeruginosa with regard to iron availability is dramatically altered during coculture growth with S. aureus in the DMC. Iron acquisition through siderophore biosynthesis and response is an energy-intensive process; thus, increasing local iron levels through lysis of S. aureus during growth in low-iron environments, such as those encountered in vivo, might be beneficial. We do not believe our data are limited to P. aeruginosa-S. aureus interactions. P. aeruginosa also lyses several other gram-positive pathogens, including S. pneumoniae and Bacillus anthracis (data not shown), which it may also encounter during infection.
Although we have shown that P. aeruginosa may obtain iron through lysis of S. aureus, our results do not preclude other mechanisms of iron acquisition during coculture. P. aeruginosa may also obtain iron by binding and uptake of siderophores produced by other bacteria. A recent study of cocultures of P. aeruginosa and Burkholderia cepacia indicated that P. aeruginosa specifically responds to the presence of B. cepacia siderophores during coculture (41). It is likely that the mechanistic details of iron acquisition will vary depending on the constituents of the coculture, but our results indicate that antagonistic interactions between species may help define community structure. Spatial and temporal factors will also affect these interactions, and it is likely that P. aeruginosa iron acquisition through S. aureus lysis will only be important when these bacteria are in very close proximity, such as in multispecies biofilms.
Our transcriptome analysis provides a more comprehensive view of in vivo gene expression of P. aeruginosa. Our data suggest that the peritoneum is a low-oxygen, iron-limited environment with sufficient nutrients to support large numbers of P. aeruginosa. Our studies verified three previously described in vivo-induced P. aeruginosa genes and identified a large number of other genes differentially expressed during growth in the peritoneum. Our data also provide a genomic analysis of a common pathogenic coculture and indicate that the physiology of P. aeruginosa in vivo is drastically altered in regards to iron availability during coculture with S. aureus. These results may have significant ramifications for in vivo growth of P. aeruginosa in polymicrobial communities and underscore the need for analysis of multispecies infections. Although we focused on iron acquisition in this study, it is plausible that other nutrients or cofactors may be obtained through competitive interactions. It is likely that the benefits of such interactions are not confined to pathogenesis but instead reflect processes important for survival and growth in many polymicrobial environments.
Acknowledgments
We thank the MGH-Parabiosys: NHLBI Program for Genomic Applications, Massachusetts General Hospital and Harvard Medical School, Boston, Mass. (http://pga.mgh.harvard.edu/cgi-bin/pa14/mutants/retrieve.cgi) for the pqsA mutant and S. Jin for IVET sequences.
This work was supported by a grant from the National Institutes of Health (1P20RR15564-01 to D.R.A. and M.W.).
REFERENCES
- 1.Akins, D. R., K. W. Bourell, M. J. Caimano, M. V. Norgard, and J. D. Radolf. 1998. A new animal model for studying Lyme disease spirochetes in a mammalian host-adapted state. J. Clin. Investig. 101:2240-2250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ankenbauer, R. G., and H. N. Quan. 1994. FptA, the Fe(III)-pyochelin receptor of Pseudomonas aeruginosa: a phenolate siderophore receptor homologous to hydroxamate siderophore receptors. J. Bacteriol. 176:307-319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Boyce, J. D., I. Wilkie, M. Harper, M. L. Paustian, V. Kapur, and B. Adler. 2002. Genomic scale analysis of Pasteurella multocida gene expression during growth within the natural chicken host. Infect. Immun. 70:6871-6879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Brooks, C. S., P. S. Hefty, S. E. Jolliff, and D. R. Akins. 2003. Global analysis of Borrelia burgdorferi genes regulated by mammalian host-specific signals. Infect. Immun. 71:3371-3383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brunskill, E. W., and K. W. Bayles. 1996. Identification and molecular characterization of a putative regulatory locus that affects autolysis in Staphylococcus aureus. J. Bacteriol. 178:611-618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Brunskill, E. W., B. L. de Jonge, and K. W. Bayles. 1997. The Staphylococcus aureus scdA gene: a novel locus that affects cell division and morphogenesis. Microbiol. 143:2877-2882. [DOI] [PubMed] [Google Scholar]
- 7.Budzikiewicz, H. 2001. Siderophores of the human pathogenic fluorescent pseudomonads. Curr. Top. Med. Chem. 1:1-6. [DOI] [PubMed] [Google Scholar]
- 8.Costerton, J. W., Z. Lewandowski, D. E. Caldwell, D. R. Korber, and H. M. Lappin-Scott. 1995. Microbial biofilms. Annu. Rev. Microbiol. 49:711-745. [DOI] [PubMed] [Google Scholar]
- 9.Costerton, J. W., P. S. Stewart, and E. P. Greenberg. 1999. Bacterial biofilms: a common cause of persistent infections. Science 284:1318-1322. [DOI] [PubMed] [Google Scholar]
- 10.Cystic Fibrosis Foundation. 1999. Cystic fibrosis patient registry annual data report. Cystic Fibrosis Foundation, Bethesda, Md.
- 11.Deziel, E., F. Lepine, S. Milot, J. He, M. N. Mindrinos, R. G. Tompkins, and L. G. Rahme. 2004. Analysis of Pseudomonas aeruginosa 4-hydroxy-2-alkylquinolines (HAQs) reveals a role for 4-hydroxy-2-heptylquinoline in cell-to-cell communication. Proc. Natl. Acad. Sci. USA 101:1339-1344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Dodds, M. G., K. J. Grobe, and P. S. Stewart. 2000. Modeling biofilm antimicrobial resistance. Biotechnol. Bioeng. 68:456-465. [PubMed] [Google Scholar]
- 13.Domann, E., G. Hong, C. Imirzalioglu, S. Turschner, J. Kuhle, C. Watzel, T. Hain, H. Hossain, and T. Chakraborty. 2003. Culture-independent identification of pathogenic bacteria and polymicrobial infections in the genitourinary tract of renal transplant recipients. J. Clin. Microbiol. 41:5500-5510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Findon, G., and T. Miller. 1995. Bacterial peritonitis in continuous ambulatory peritoneal dialysis: effect of dialysis on host defense mechanisms. Am. J. Kidney Dis. 26:765-773. [DOI] [PubMed] [Google Scholar]
- 15.Fujimoto, D. F., and K. W. Bayles. 1998. Opposing roles of the Staphylococcus aureus virulence regulators, Agr and Sar, in Triton X-100- and penicillin-induced autolysis. J. Bacteriol. 180:3724-3726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Groicher, K. H., B. A. Firek, D. F. Fujimoto, and K. W. Bayles. 2000. The Staphylococcus aureus lrgAB operon modulates murein hydrolase activity and penicillin tolerance. J. Bacteriol. 182:1794-1801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Haba, E., A. Pinazo, O. Jauregui, M. J. Espuny, M. R. Infante, and A. Manresa. 2003. Physicochemical characterization and antimicrobial properties of rhamnolipids produced by Pseudomonas aeruginosa 47T2 NCBIM 40044. Biotechnol. Bioeng. 81:316-322. [DOI] [PubMed] [Google Scholar]
- 18.Hassett, D. J. , J. Cuppoletti, B. Trapnell, S. V. Lymar, J. J. Rowe, S. S. Yoon, G. M. Hilliard, K. Parvatiyar, M. C. Kamani, D. J. Wozniak, S. H. Hwang, T. R. McDermott, and U. A. Ochsner. 2002. Anaerobic metabolism and quorum sensing by Pseudomonas aeruginosa biofilms in chronically infected cystic fibrosis airways: rethinking antibiotic treatment strategies and drug targets. Adv. Drug Deliv. Rev. 54:1425-1443. [DOI] [PubMed] [Google Scholar]
- 19.Hoiby, N. 1998. Pseudomonas in cystic fibrosis: past, present, and future. Cystic Fibrosis Trust, Kent, United Kingdom.
- 20.Holley, J. L., J. Bernardini, and B. Piraino. 1992. Polymicrobial peritonitis in patients on continuous peritoneal dialysis. Am. J. Kidney Dis. 19:162-166. [DOI] [PubMed] [Google Scholar]
- 21.Janakiraman, A., and J. M. Slauch. 2000. The putative iron transport system SitABCD encoded on SPI1 is required for full virulence of Salmonella typhimurium. Mol. Microbiol. 35:1146-1155. [DOI] [PubMed] [Google Scholar]
- 22.Kiernan, L., F. O. Finkelstein, A. S. Kliger, N. Gorban-Brennan, P. Juergensen, A. Mooraki, and E. Brown. 1995. Outcome of polymicrobial peritonitis in continuous ambulatory peritoneal dialysis patients. Am. J. Kidney Dis. 25:461-464. [DOI] [PubMed] [Google Scholar]
- 23.Kolak, M., F. Karpati, H. J. Monstein, and J. Jonasson. 2003. Molecular typing of the bacterial flora in sputum of cystic fibrosis patients. Int. J. Med. Microbiol. 293:309-317. [DOI] [PubMed] [Google Scholar]
- 24.Kosorok, M. R., L. Zeng, S. E. West, M. J. Rock, M. L. Splaingard, A. Laxova, C. G. Green, J. Collins, and P. M. Farrell. 2001. Acceleration of lung disease in children with cystic fibrosis after Pseudomonas aeruginosa acquisition. Pediatr. Pulmonol. 32:277-287. [DOI] [PubMed] [Google Scholar]
- 25.Lo, C. Y., W. L. Chu, K. M. Wan, S. Y. Ng, W. L. Lee, M. F. Chu, S. W. Cheng, and W. K. Lo. 1998. Pseudomonas exit-site infections in CAPD patients: evolution and outcome of treatment. Perit. Dial. Int. 18:637-640. [PubMed] [Google Scholar]
- 26.Machan, Z. A., G. W. Taylor, T. L. Pitt, P. J. Cole, and R. Wilson. 1992. 2-Heptyl-4-hydroxyquinoline N-oxide, an antistaphylococcal agent produced by Pseudomonas aeruginosa. J. Antimicrob. Chemother. 30:615-623. [DOI] [PubMed] [Google Scholar]
- 27.Morse, S. A., B. V. Jones, and P. G. Lysko. 1980. Pyocin inhibition of Neisseria gonorrhoeae: mechanism of action. Antimicrob. Agents Chemother. 18:416-423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.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 pyoverdine biosynthesis genes. Mol. Microbiol. 45:1277-1287. [DOI] [PubMed] [Google Scholar]
- 29.Pedersen, S. S., G. H. Shand, B. L. Hansen, and G. N. Hansen. 1990. Induction of experimental chronic Pseudomonas aeruginosa lung infection with P. aeruginosa entrapped in alginate microspheres. APMIS 98:203-211. [PubMed] [Google Scholar]
- 30.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. [DOI] [PubMed] [Google Scholar]
- 31.Ramsey, M. M., and M. Whiteley. 2004. Pseudomonas aeruginosa attachment and biofilm development in dynamic environments. Mol. Microbiol. 53:1075-1088. [DOI] [PubMed] [Google Scholar]
- 32.Raymond, C. K., E. H. Sims, A. Kas, D. H. Spencer, T. V. Kutyavin, R. G. Ivey, Y. Zhou, R. Kaul, J. B. Clendenning, and M. V. Olson. 2002. Genetic variation at the O-antigen biosynthetic locus in Pseudomonas aeruginosa. J. Bacteriol. 184:3614-3622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Revel, A. T., A. M. Talaat, and M. V. Norgard. 2002. DNA microarray analysis of differential gene expression in Borrelia burgdorferi, the Lyme disease spirochete. Proc. Natl. Acad. Sci. USA 99:1562-1567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sawers, R. G. 1991. Identification and molecular characterization of a transcriptional regulator from Pseudomonas aeruginosa PAO1 exhibiting structural and functional similarity to the FNR protein of Escherichia coli. Mol. Microbiol. 5:1469-1481. [DOI] [PubMed] [Google Scholar]
- 35.Schlievert, P. M., and D. A. Blomster. 1983. Production of staphylococcal pyrogenic exotoxin type C: influence of physical and chemical factors. J. Infect. Dis. 147:236-242. [DOI] [PubMed] [Google Scholar]
- 36.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Stover, C. K., X. Q. Pham, A. L. Erwin, S. D. Mizoguchi, P. Warrener, M. J. Hickey, F. S. Brinkman, W. O. Hufnagle, D. J. Kowalik, M. Lagrou, R. L. Garber, L. Goltry, E. Tolentino, S. Westbrock-Wadman, Y. Yuan, L. L. Brody, S. N. Coulter, K. R. Folger, A. Kas, K. Larbig, R. Lim, K. Smith, D. Spencer, G. K. Wong, Z. Wu, and I. T. Paulsen. 2000. Complete genome sequence of Pseudomonas aeruginosa PA01, an opportunistic pathogen. Nature 406:959-964. [DOI] [PubMed] [Google Scholar]
- 38.Vander Wauven, C., A. Pierard, M. Kley-Raymann, and D. Haas. 1984. Pseudomonas aeruginosa mutants affected in anaerobic growth on arginine: evidence for a four-gene cluster encoding the arginine deiminase pathway. J. Bacteriol. 160:928-934. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wang, J. , S. Lory, R. Ramphal, and S. Jin. 1996. Isolation and characterization of Pseudomonas aeruginosa genes inducible by respiratory mucus derived from cystic fibrosis patients. Mol. Microbiol. 22:1005-1012. [DOI] [PubMed] [Google Scholar]
- 40.Wang, J. , A. Mushegian, S. Lory, and S. Jin. 1996. Large-scale isolation of candidate virulence genes of Pseudomonas aeruginosa by in vivo selection. Proc. Natl. Acad. Sci. USA 93:10434-10439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Weaver, V. B., and R. Kolter. 2004. Burkholderia spp. alter Pseudomonas aeruginosa physiology through iron sequestration. J. Bacteriol. 186:2376-2384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Yanagihara, K., K. Tomono, Y. Imamura, Y. Kaneko, M. Kuroki, T. Sawai, Y. Miyazaki, Y. Hirakata, H. Mukae, J. Kadota, and S. Kohno. 2002. Effect of clarithromycin on chronic respiratory infection caused by Pseudomonas aeruginosa with biofilm formation in an experimental murine model. J. Antimicrob. Chemother. 49:867-870. [DOI] [PubMed] [Google Scholar]
- 44.Yanagihara, K., K. Tomono, T. Sawai, Y. Hirakata, J. Kadota, H. Koga, T. Tashiro, and S. Kohno. 1997. Effect of clarithromycin on lymphocytes in chronic respiratory Pseudomonas aeruginosa infection. Am. J. Respir. Crit. Care Med. 155:337-342. [DOI] [PubMed] [Google Scholar]
- 45.Yanagihara, K., K. Tomono, T. Sawai, M. Kuroki, Y. Kaneko, H. Ohno, Y. Higashiyama, Y. Miyazaki, Y. Hirakata, S. Maesaki, J. Kadota, T. Tashiro, and S. Kohno. 2000. Combination therapy for chronic Pseudomonas aeruginosa respiratory infection associated with biofilm formation. J. Antimicrob. Chemother. 46:69-72. [DOI] [PubMed] [Google Scholar]
- 46.Yinnon, A. M., D. Gabay, D. Raveh, Y. Schlesinger, I. Slotki, D. Attias, and B. Rudensky. 1999. Comparison of peritoneal fluid culture results from adults and children undergoing CAPD. Perit. Dial. Int. 19:51-55. [PubMed] [Google Scholar]