| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
BIOMERIT Research Centre, Department of Microbiology, National University of Ireland, Cork, Ireland
Correspondence
Fergal O'Gara
f.ogara{at}ucc.ie
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Genes with a statistically significant (P=0·05), minimum twofold change in expression in the rsmA mutant compared to P. aeruginosa PAO1 wild-type are shown in Supplementary Table S1, genes commonly altered in expression in the P. aeruginosa PAK vfr mutant and the P. aeruginosa PAO1 rsmA mutant are shown in Supplementary Table S2, and genes commonly altered in expression in the P. aeruginosa PAK retS mutant and the P. aeruginosa PAO1 rsmA mutant are shown in Supplementary Table S3, available with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Van Delden & Iglewski, 1998). The ability to sense aspects of the external environment and bring about timed, co-ordinate gene expression ensures that this nosocomial pathogen exploits its ecological adaptability to successfully live in the immunocompromised host. Virulence in P. aeruginosa is multifaceted, and involves the production of disease-causing secondary metabolites and virulence factors, nutrient scavenging, motility and biofilm formation. The GacA/RsmA/RsmB (RsmZ) signal transduction system is a conserved pathway involved in a variety of adaptive functions in P. aeruginosa. RsmA has numerous homologues in both Gram-negative and Gram-positive bacteria (Romeo, 1998); posttranscriptional regulatory proteins similar to RsmA have been described in Pseudomonas fluorescens CHA0 (RsmA), Escherichia coli (CsrA) and Erwinia carotovora (RsmA) (Blumer et al., 1999; Cui et al., 1995; Liu & Romeo, 1997). CsrA is a well-documented RNA binding protein of E. coli that mediates posttranscriptional control of genes involved in a number of physiological pathways. CsrA represses the genes involved in glycogen biosynthesis in E. coli by binding to the 5' end of the glgCAP mRNA transcript around the Shine–Dalgarno site; this prevents ribosome binding and translation and leads to rapid degradation of the mRNA (Baker et al., 2002). Positive posttranscriptional regulation has also been described in E. coli, in which CsrA binds to a yet-unidentified area in the 5' untranslated leader sequence of the flhDC master operon that is involved in flagellum biosynthesis, thus increasing the stability of the mRNA and facilitating its translation (Wei et al., 2001).
In P. aeruginosa PAO1, it has been shown that the response regulator GacA has an effect on RsmA levels, as a gacA mutant expresses 30 % less RsmA protein than wild-type in late-exponential phase (Pessi et al., 2001). GacA works in tandem with the transmembrane sensor kinase GacS. The GacS/GacA two-component system controls expression of virulence-related genes in response to unidentified external environmental signals. GacA also has a positive posttranscriptional effect on the hcnABC gene cluster (Pessi & Haas, 2001). Recently, both GacA and RsmA have been shown to modulate the levels of RsmB (RsmZ) by regulating its transcription. RsmB is the small, untranslated regulatory RNA molecule that works in tandem with RsmA (Burrowes et al., 2005; Heurlier et al., 2004). RsmB is proposed to bind and sequester RsmA, and hence limit its activity. Therefore, GacA also regulates the activity of the RsmA protein by controlling the levels of its partner molecule, RsmB. The posttranscriptional regulatory protein RsmA has been implicated in secondary metabolite production (Pessi et al., 2001).
Interestingly, in P. aeruginosa, RsmA and GacA are also involved in the regulation of quorum sensing (QS) signalling molecule synthesis. The QS system has key roles in activation and co-ordination of virulence gene expression in P. aeruginosa PAO1, in response to cell density and environmental signals. QS involves the accumulation of signal molecules, often N-acyl homoserine lactones (AHLs), to a threshold level to activate expression of certain genes (Fuqua & Greenberg, 1998). P. aeruginosa has two interdependent QS systems, the las and rhl systems. The LasI AHL synthase produces N-(3-oxododecanoyl)-L-homoserine lactone (3-oxo-c12-hsl) and the rhli ahl synthase produces n-butanoyl-L-homoserine lactone (C4-HSL). These AHLs, in association with their cognate regulators, LasR and RhlR, respectively, control the transcription of a number of virulence-related genes, such as alkaline protease, and the pyocyanin and hydrogen cyanide biosynthetic genes (Gambello et al., 1993; Reimmann et al., 1997). RsmA negatively regulates lasI, rhlI and the hydrogen cyanide biosynthetic genes (Pessi et al., 2001). In contrast, RsmA has a positive posttranscriptional effect on the expression of the genes encoding the secondary metabolites lipase and rhamnolipid (Heurlier et al., 2004). GacA is also a documented hierarchical member of the QS network and has a positive transcriptional effect on the las and rhl QS systems (Reimmann et al., 1997).
P. aeruginosa also produces a third intercellular signal molecule, 3,4-dihydroxy-2-heptylquinoline, known as the Pseudomonas quinolone signal (PQS). LasR has been shown to regulate PQS production, and RhlR is required for PQS activity (Pesci et al., 1999). PQS has been shown to induce expression of lasB, the gene encoding LasB elastase, and rhlI (McKnight et al., 2000; Pesci et al., 1999). PQS has also been implicated in the regulation of the secondary metabolites pyocyanin and hydrogen cyanide (Gallagher et al., 2002). Diggle et al. (2003) showed that mutants defective in PQS production or in PQS-mediated response produced substantially lower levels of rhl-dependent exoproducts, while C4-HSL levels were unaltered, suggesting that PQS also regulates rhl-dependent genes independently of C4-HSL. MvfR (PqsR) is a QS-associated regulator that controls the production of PQS, pyocyanin and elastases, and is required for full P. aeruginosa virulence (Deziel et al., 2005; Diggle et al., 2003; Gallagher et al., 2002; Rahme et al., 1997). A recent study has revealed that LasR and RhlR indirectly regulate PQS production via MvfR, and that MvfR binds directly to the pqsABCDE operon, positively regulating its transcription (Wade et al., 2005). So far, a direct impact of RsmA or GacA on PQS synthesis has not been reported.
Recent studies have shed more light on the regulation of rsmA. Deziel et al. (2005) reported that in P. aeruginosa PA14, rsmA is posttranscriptionally regulated by MvfR. Another regulatory factor known to influence the levels of free, active RsmA is the recently identified hybrid sensor kinase/response regulator RetS (RtsM) (Goodman et al., 2004; Laskowski et al., 2004). Transcriptome profiling of the P. aeruginosa PAK retS mutant revealed a decrease in the expression of genes involved in type III secretion and virulence and an increase in the biosynthetic genes (the psl and pel operons) involved in the formation of the mannose- and glucose-rich exopolysaccharide component of biofilms (Goodman et al., 2004). Interestingly, these phenotypes were suppressed by transposon insertions in gacS, gacA and rsmB. The authors propose that, under conditions of acute infection, environmental signals favour activation of RetS, resulting in repression of the GacS/GacA/RsmB pathway. A mutation in retS results in de-repression of the levels of GacS, GacA and RsmB, and hence a decrease in the level of active RsmA. A retS mutation is, therefore, thought to cause alterations in gene expression via decreased levels of free, active RsmA (Goodman et al., 2004).
Although RsmA may be an element in signal transduction pathways with key roles in virulence, the full regulatory impact of RsmA on cellular activities in P. aeruginosa PAO1 is not understood. Here we have used transcriptome profiling to examine the global effects of an rsmA mutation on gene expression in P. aeruginosa PAO1. Thus, we have identified a set of genes whose expression is most probably indirectly affected by the absence of RsmA. Validation of the transcriptome data using lacZ transcriptional fusions confirmed that, amongst other targets, RsmA is involved in the modulation of pvdS, vfr and pilM transcription. Analysis of these data identified a number of candidate genes, including some encoding transcriptional regulators, that may be novel targets for direct RsmA action.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. P. aeruginosa PAO1 strains and E. coli strains were routinely cultured at 37 °C in Luria–Bertani (LB) broth or on LB plates. Antibiotics were used at the following concentrations: for P. aeruginosa, 70 µg tetracycline ml–1, 500 µg chloramphenicol ml–1, 200 µg carbenicillin ml–1; for E. coli, 25 µg tetracycline ml–1, 30 µg chloramphenicol ml–1.
|
). Small-scale plasmid DNA isolation was performed using the Qiagen Plasmid Mini kit according to the manufacturers' specifications. Restriction fragments were purified from agarose gels using the Qiagen Gel Extraction kit. PCR products were purified using the High Pure PCR Product Purification kit (Boehringer). Plasmids were introduced into Pseudomonas strains via electroporation.
Microarray sample preparation and analysis.
Duplicate cultures of P. aeruginosa PAO1 wild-type and the rsmA mutant were grown in LB broth to an OD600 of 0·8, and 4 ml of each culture was treated with RNAprotect bacterial reagent (Qiagen), according to the manufacturers' instructions. Cells were harvested by centrifugation and stored at –80 °C. Total RNA was extracted using the RNeasy Midi RNA Isolation kit (Qiagen) according to the manufacturers' instructions. The RNA was treated with RQ1 RNase-free DNase (Promega) for 1 h at 37 °C and purified using the Clean-Up protocol of the RNeasy Midi RNA Isolation kit according to the manufacturers' instructions. The RNA was precipitated with ethanol and the RNA pellet was resuspended in sterile distilled H2O to a final volume of 20 µl; the integrity of the RNA was checked on an agarose gel and quantified using a spectrophotometer. cDNA was synthesized according to the Affymetrix Expression Analysis Protocol guide; spike RNA transcripts from Bacillus subtilis genes were included in the reaction. The cDNA was treated with 2 µg DNase-free RNase (Roche) in a final volume of 100 µl for 1 h at 37 °C, purified using the QIAquick PCR Purification kit (Qiagen) and eluted in 45 µl sterile distilled H2O. The cDNA was fragmented using RQ1 RNase-free DNase for 1 h at 37 °C, and biotin-labelled using the BioArray Terminal Labelling kit with biotin–ddUTP (Enzo Life Sciences). The cDNA was snap-frozen on dry ice and analysed at the MRC Geneservice (Cambridge, England) on Affymetrix GeneChip P. aeruginosa Genome arrays. A Student's t test was carried out on the data generated from the microarray analysis using GeneDirector software (BioDiscovery) to give a list of genes with altered expression greater than twofold and a P value 
0·05.
Semi-quantitative RT-PCR analysis.
To confirm the microarray data, semi-quantitative RT-PCR analysis was carried out. RNA was isolated from cultures of PAO1 wild-type and the rsmA mutant grown to an OD600 of 0·8 in LB broth, and converted to cDNA using the procedure already described. The primers used to amplify the genes gloA2, pilM, pvdS, oprN, pilU and proC are listed in Table 1
. PCRs were carried out using PCR Master Mix (Promega) and 50 ng cDNA template for 20–25 cycles. The constitutively expressed control gene proC (Savli et al., 2003) was used as the internal control to verify the absence of significant variation in cDNA levels in all samples.
Construction and analysis of transcriptional fusions.
The promoters and start sites of pvdS, pilM and vfr were amplified from P. aeruginosa PAO1 DNA using primers listed in Table 1. The primers (pvdSFZ, pvdSRZ, pilMFZ, pilMRZ, vfrFZ, vfrRZ) include the recognition sequences (Table 1, italics) for the restriction enzymes Xba1 and Kpn1. The resultant PCR products were digested using Xba1 or Kpn1, PCR-purified, and cloned into the reporter fusion pMP220 (Spaink et al., 1987), upstream of the promoterless lacZ gene. 
-Galactosidase activity was measured by standard methods (Sambrook et al., 1989). Due to the clumping phenotype of the rsmA mutant, total protein was extracted from 2 ml of cells at each time-point analysed, and this number was substituted for OD600 to calculate -galactosidase activity.
Western blot analysis.
Total protein was extracted from strains of P. aeruginosa PAO1, grown in LB broth to an OD600 of 0·8 and quantified using the Bio-Rad Protein Assay Dye Reagent. Proteins and All Blue Precision Plus Protein standard (Bio-Rad) were electrophoresed on 5 % polyacrylamide SDS stacking gels and 15 % polyacrylamide SDS resolving gels at 130 V using the Mini-PROTEAN II cell (Bio-Rad). Proteins were either stained with Coomassie brilliant blue (BDH Chemicals) or transferred to a nitrocellulose membrane using a Mini Trans-Blot cell (Bio-Rad). The membrane was blocked with 5 % non-fat dried milk, 0·1 % (v/v) Tween 20 in Tris-buffered saline (TBS) for 1 h, washed three times in TBS/Tween and incubated with anti-PvdS monoclonal antibodies (received from Professor M. Vasil, University of Colorado) for 2 h at room temperature. The membrane was washed three times in TBS/Tween and incubated with anti-mouse secondary antibodies (Dakocytomation) for 1 h. Detection was performed using an ECL+ kit (Amersham Pharmacia Biotech) and chemiluminescent film (Kodak).
Pyocyanin quantification.
Production of the phenazine pyocyanin by P. aeruginosa PAO1 wild-type and the rsmA mutant was measured as described elsewhere (Essar et al., 1990).
Motility assay.
Swimming was analysed on tryptone swim plates (1 %, w/v, tryptone, 0·5 %, w/v, NaCl, 0·3 %, w/v, agar), which were inoculated using a sterile toothpick with bacteria from colonies grown overnight on LB agar at 37 °C. The plates were incubated at 37 °C, and motility was assessed qualitatively by examining the circular turbid zone formed by the bacterial cells migrating away from the point of inoculation.
Adhesion assay.
Adhesion on polyvinyl chloride (PVC) was tested using non-coated 24-well tissue-culture flat-well plates (Sarstedt). LB (1 ml) was added to each well and inoculated to a starting OD600 of 0·05. The plates were incubated at 37 °C without agitation for 12 h. The wells were rinsed with water and the bacterial film was stained with 1 % crystal violet. Six replicates were performed per sample. The wells were then washed with 1 ml 95 % ethanol and the OD570 of a 1 : 100 dilution in sterile distilled H2O was obtained.
PQS extraction and analysis.
Cultures of P. aeruginosa PAO1 wild-type, PAZH13 (rsmA mutant), PAO1 pCU705 (rsmA overexpressing strain), the P. aeruginosa pqsR mutant (Gallagher et al., 2002) and the P. aeruginosa phnA mutant (Jacobs et al., 2003) were grown to early stationary phase in LB broth. Culture (3 ml) was extracted with 3 ml acidified ethyl acetate, the solvent removed by rotary evaporation and the extract resuspended in 1 : 1 acidified ethyl acetate/acetonitrile. Silica gel 60 F254 aluminium TLC plates (Merck) were activated by soaking for 15 min in 5 % KH2PO4, wrapping in tin foil and heating to 100 °C for 1 h. Samples, PQS standard (Diggle et al., 2003) and anthranilate standard (Sigma) were separated on the TLC plates using 95 : 5 dichloromethane/methanol in a solvent chamber. PQS appeared as a blue/purple spot under UV light.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Therefore, in order to study the impact of the loss of rsmA on gene expression, duplicate transcriptome analyses were carried out on P. aeruginosa PAO1 wild-type and PAZH13, the rsmA mutant, grown to late-exponential phase in LB broth at 37 °C. Under these conditions, the growth rate and final OD600 of PAO1 wild-type and the rsmA mutant were identical (data not shown). The transcriptome analysis was carried out using four Affymetrix GeneChip P. aeruginosa Genome arrays (see Methods). A Student's t test was carried out on the comparative data generated from the microarrays using GeneDirector software (BioDiscovery) to give a list of genes with a statistically significant (P0·05), minimum twofold, change in expression (Table 2 and Supplementary Table S1).
|
), which encodes a pyrroline-5-carboxylate reductase, and whose expression was not altered in the rsmA mutant, was used as an internal control in these reactions. The RT-PCR analyses correlated with the microarray alterations in gene expression: pilM and pilU showed reduced expression, and gloA2, pvdS and oprN showed increased expression in the rsmA mutant compared to wild-type (Fig. 1).
|
. A large number of altered genes (n=222 or 44 %) were categorized as Class 4 or of unknown function according to the Pseudomonas website classification system (http://www.pseudomonas.com; Supplementary Table S1). Where possible, genes of altered expression were classified into functional groups (Table 2). The expression of RNA complementary to 29 intergenic regions was also significantly altered by inactivation of rsmA. This may indicate yet-unidentified ORFs or the presence of untranslated regulatory RNA molecules. Indeed, RsmA has been shown previously to positively regulate the transcription of the gene encoding its cognate regulatory untranslated RNA, RsmB (Burrowes et al., 2005; Heurlier et al., 2004). However, the rpoS/fdxA intergenic region, in which rsmB is located, is not represented on the Affymetrix P. aeruginosa PAO1 GeneChip.
Functional categories of genes with increased expression in the rsmA mutant compared to PAO1 wild-type
Iron acquisition.
The expression of genes involved in iron acquisition was increased in the rsmA mutant compared to PAO1 wild-type. The expression of a number of pvd and pch genes, which direct the biosynthesis of the siderophores pyoverdine and pyochelin, respectively, was increased in the rsmA mutant compared to wild-type. In addition, expression of fpvA, the ferric-pyoverdine receptor, and fptA, the ferric-pyochelin receptor, was increased in the rsmA mutant compared to PAO1 wild-type. Interestingly, the microarray analysis revealed that expression of the gene encoding the iron-regulated sigma factor PvdS was increased 11-fold in the rsmA mutant compared to wild-type. The fact that PvdS levels were increased in the rsmA mutant may explain the increase in expression of the pvd genes (Beare et al., 2003; Visca et al., 2002) but not the pch operon. Transcription of pchABCDE and fptA is activated by the AraC-like regulator PchR in the presence of pyochelin and repressed in the absence of pyochelin (Heinrichs & Poole, 1996). Although expression of the pch genes was increased in the rsmA mutant, the levels of pchR transcript were not altered. Therefore, the increase in pch transcript levels may reflect a change in mRNA stability rather than in transcription rate, or the implementation of an alternate regulator. To confirm that the observed increase in pvdS levels in the rsmA mutant compared to PAO1 wild-type was due to a change in transcription rate, a pvdS–lacZ transcriptional fusion (pCUB25) was constructed in the fusion vector pMP220. Transcription of pvdS was increased in the rsmA mutant compared to PAO1 wild-type at all time-points analysed (Fig. 2a). To ascertain if this increase in transcription corresponded to an increase in PvdS protein levels, Western blot analysis was carried out on total protein isolated from PAO1 wild-type and from the rsmA mutant grown to an OD600 of 0·8 in LB broth. The 21·2 kDa PvdS protein was detected with an anti-PvdS antibody via chemiluminescence, and PvdS levels were found to be higher in the rsmA mutant compared to PAO1 wild-type, albeit by a moderate increase (Fig. 3). Band intensities were quantified using GeneTools software (version 3.05.03, Syngene). The mean band intensity in the rsmA mutant sample, as a percentage of mean band intensity of the wild-type sample, was 155±20 % (n=3).
|
|
; West et al., 1994). Although PvdS levels were increased in the rsmA mutant, toxA showed a 2·8-fold decrease in expression in the rsmA mutant. This may be due to the drop in vfr levels in the rsmA mutant (see below), suggesting that under the conditions of this experiment, the loss of the positive influence of Vfr on toxA transcription is more potent than the increase in PvdS levels on toxA transcription.
The expression of genes involved in resistance to oxidative stress, namely sodM and fumC (Hassett et al., 1997) was also increased. This is consistent with an increase in iron uptake, which generates reactive hydroxyl radicals via the Fenton reaction.
Antibiotic resistance.
Loss of rsmA resulted in increased expression of the genes encoding the MexEF-OprN pump. In P. aeruginosa, the MexEF-OprN pump removes numerous antibiotics and PQS from the cell (Poole, 2001). Several proteins have been implicated in the positive regulation of expression of this Mex system, including MexT, an activator of MexEF-OprN (Maseda et al., 2000). In addition, spontaneous nfxC mutants show an increase in expression of the MexEF-OprN pump (Kohler et al., 2001). Examination of MexT protein levels in the rsmA mutant showed that they were not altered (data not shown). However, levels of mexS mRNA, which encodes a transcriptional repressor of MexEF-OprN, (Sobel et al., 2005) were increased fourfold in the rsmA mutant compared to PAO1. These results reflect the complexity of Mex pump regulation.
Functional categories of genes with decreased expression in the rsmA mutant compared to PAO1 wild-type
PQS biosynthesis.
PQS is a third signal molecule in addition to C4-HSL and 3-oxo-C12-HSL produced by P. aeruginosa, and PQS production occurs at the onset of stationary phase (Diggle et al., 2003; Pesci et al., 1999). Microarray analysis revealed that phnA expression was decreased by 2·6-fold in the rsmA mutant compared to PAO1 wild-type. The phnAB operon encodes an anthranilate synthase, which converts chorismate to anthranilate, a precursor of PQS and pyocyanin. PhnA is essential for PQS production in P. aeruginosa, as a phnA mutant does not make PQS (Gallagher et al., 2002). However, levels of the pqsABCDE transcripts, which encode the PQS biosynthetic enzymes, did not appear to be altered in the comparative microarray analyses. To validate a transcript recorded as ‘no change’ on the microarray, confirmatory analysis was performed with a pqsA–lacZ transcriptional fusion (pLP0996) (Fig. 2b). As expected, there was no significant difference in the transcription of the pqsABCDE operon in PAO1 wild-type and the rsmA mutant. To examine the impact of the decrease in expression of phnA on PQS levels in the rsmA mutant, PQS extractions were carried out on supernatants of PAO1, the rsmA mutant, PAO1 pCU705 (rsmA++), MP551 (PAO1 pqsR) and PAO1 phnA grown to stationary phase (OD600 of 3·0) in LB broth. Fig. 4 shows that the rsmA mutant produced less PQS than PAO1 wild-type in stationary phase and that PAO1 overexpressing rsmA produced more PQS than PAO1 wild-type. This was accompanied by a parallel reduction in the intensity of the upper band, namely compound B; this compound has not yet been identified, but its levels have been found to parallel PQS levels in certain P. aeruginosa strains, including PAO1 (Collier et al., 2002). The large spot present in the TLC analysis of the rsmA, pqsR and phnA mutants is a precursor of PQS, namely anthranilate (Baysse et al., 2005). As expected, neither PQS nor compound B was present in the pqsR mutant, as this gene product is essential for their synthesis. In addition, it was interesting to note that the PQS profile of the rsmA mutant was similar to that of the phnA mutant, supporting the transcriptome finding that expression of phnA was reduced in the rsmA mutant.
|
); levels of pyocyanin in stationary-phase cultures of PAO1 and the rsmA mutant were examined. This revealed that loss of RsmA substantially reduced the production of pyocyanin when the bacteria were cultured in LB; PAO1 wild-type produced 4·2 µg ml–1 (±0·1, n=3) of pyocyanin, while the rsmA mutant only produced 0·3 µg ml–1 (±0·05, n=3) of pyocyanin in stationary phase (OD600 of 4). However, under the same experimental conditions in glycerol/alanine medium, which was used as a control, the rsmA mutant showed an increase in production of pyocyanin compared to wild-type: PAO1 wild-type produced 0·23 µg ml–1 (±0·03, n=3) of pyocyanin, while the rsmA mutant produced 3·27 µg ml–1 (±0·01, n=3) of pyocyanin, which is in accordance with the previous observation by Pessi et al. (2001). The result obtained in LB is in agreement with the observed accumulation of the metabolic precursor anthranilate and the drop in levels of PQS in the rsmA mutant. Indeed, PQS has been shown to be required for phenazine gene expression in PAO1 (Diggle et al., 2003). Interestingly, expression of phzA1, the first gene in one of the pyocyanin biosynthetic operons (Mavrodi et al., 2001), was decreased by approximately threefold in the rsmA mutant compared to PAO1 wild-type. The other genes in the operon were recorded as ‘absent’ for the rsmA mutant, which is most likely due to mRNA or cDNA degradation during sample preparation for microarray analysis.
Motility.
RsmA positively regulates fimbrial biosynthetic genes in P. aeruginosa PAO1. Transcriptome analysis revealed that expression of genes involved in the formation of type IV pili, fimbriae and chemotaxis was decreased in the rsmA mutant compared to PAO1 wild-type (Table 2). Expression of the pilMNOPQ operon and pilU, which encode proteins of unknown function that are thought to be involved in the biosynthesis of pili (Mattick et al., 1996), was decreased in the rsmA mutant compared to wild-type. Expression of a third pil gene, pilC, was also decreased in the rsmA mutant compared to PAO1. To examine the altered transcription of the pilMNOPQ operon, a pilM–lacZ transcriptional fusion was constructed in pMP220 (pCUB24) and analysed over time in PAO1 and the rsmA mutant (Fig. 2c). This confirmed that transcription of the pilMNOPQ operon was decreased in the mutant throughout all phases of growth. Depending on the carbon source used, type IV pili are not always necessary for the initial attachment of P. aeruginosa PAO1, but they are always necessary for normal biofilm development, irrespective of growth medium (Klausen et al., 2003). Under the conditions used for this transcriptome study (i.e. LB broth), the P. aeruginosa rsmA mutant (OD570 0·20±0·06, n=5) was significantly impaired for biofilm formation on PVC compared to wild-type (OD570 0·34±0·05, n=5) (Fig. 5a).
|
). The cup genes are thought to be involved in the assembly of novel P. aeruginosa fimbrial structures that are different from type IV pili; the cupA gene cluster is required for cell adhesion and initiation of biofilm formation (Vallet et al., 2001). In the rsmA mutant, expression of cupA2, which encodes a fimbrial chaperone protein, was down ninefold and expression of cupC3, which encodes an usher protein, was down approximately threefold compared to PAO1. Interestingly, mutations in cupA2 have been shown to cause cell clumping in P. aeruginosa PAK (Vallet et al., 2001), so it is possible that the clumping phenotype observed for the rsmA mutant (Heurlier et al., 2004) is caused by the drop in cupA2 levels. The expression of a number of genes involved in chemotaxis was also decreased in the rsmA mutant compared to PAO1. PA0173, which encodes a putative response regulator for chemotaxis, was decreased tenfold, and expression of PA1251, which encodes a probable chemotaxis transducer, was down eightfold in the rsmA mutant compared to wild-type. The decreased expression of genes involved in chemotaxis may explain the drop in swimming motility in the rsmA mutant compared to PAO1 (Fig. 5b). The motA and motB genes encode proteins that make up the stator or stationary component of the flagellar motor (Toutain et al., 2005). Expression of motA and motB was down twofold in the rsmA mutant compared to PAO1 (Table 2). From these data, it is evident that RsmA is involved in a complex regulatory cascade that controls the expression of genes involved in motility in P. aeruginosa.
Vfr transcription.
Microarray analysis revealed that expression of vfr was decreased twofold in the rsmA mutant compared to PAO1 wild-type. Vfr is a homologue of the E. coli cAMP receptor protein Crp. Crp is activated by binding cAMP and then binds to a specific sequence in the promoter of Crp-controlled genes. In P. aeruginosa, Vfr has been implicated in the regulation of twitching motility, QS, pyocyanin production, repression of flagellar biosynthesis and the expression of numerous proteins secreted by the type II pathway, including exotoxin A (Albus et al., 1997; Beatson et al., 2002; Dasgupta et al., 2002; West et al., 1994). As mentioned previously, toxA expression was decreased approximately 2·8-fold in the rsmA mutant compared to PAO1. The expression of regA, lasR and fleQ, other known Vfr targets, was not altered in the rsmA mutant. However, the transcription of regA, lasR and fleQ is also controlled by numerous other regulators (Dasgupta & Ramphal, 2001; Dasgupta et al., 2003; Hunt et al., 2002; Juhas et al., 2005), which could compensate for the drop in vfr levels. To confirm the novel impact of RsmA on Vfr levels, a vfr–lacZ transcriptional fusion (pCUB26) was constructed and analysed in PAO1 and the rsmA mutant (Fig. 2d). This analysis showed that vfr expression peaked after 10 h of growth, corresponding to late-exponential/early stationary phase, in both the wild-type and the rsmA mutant. Expression of vfr was decreased in the rsmA mutant compared to PAO1 throughout the growth cycle until late stationary phase, in which vfr transcription was the same in both backgrounds.
Iron storage.
Finally, the expression of bfrB, the gene encoding the bacterioferritin iron-storage protein, was down-regulated in the rsmA mutant compared to PAO1, which is in agreement with the increase in expression of genes involved in siderophore-mediated iron uptake (Table 2). Therefore, although grown in iron-rich LB medium, the rsmA mutant behaved partly as though it was cultured in iron-deplete conditions.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A notable feature of this transcriptome analysis was the far-reaching consequences of an rsmA mutation, emphasizing the role of RsmA as a global regulator. RsmA is known to indirectly control the transcription of numerous genes through regulation of the QS network (Pessi et al., 2001). Hence, some of the observed alterations in the transcriptome profile of the rsmA mutant may be due to the direct effect of the rsmA mutation on the QS network, since RsmA negatively regulates lasI and rhlI (Pessi et al., 2001). For example, the increase in 3-oxo-C12-HSL in the rsmA mutant (Pessi et al., 2001) may contribute to the increase in expression of the pyoverdine biosynthetic genes (Stintzi et al., 1998; Whiteley et al., 1999) and sodM (Hassett et al., 1999), both regulated by the QS system. However, although inactivation of rsmA affects AHL synthesis, this does not appear to have a dramatic effect on expression of all QS-regulated genes. Moreover, alteration in AHL levels in the rsmA mutant cannot explain all of the changes in gene expression brought about by the loss of RsmA, so further explanations must be considered. An important feature of this study, which cannot be overlooked, is that microarray experiments measure mRNA steady-state levels, which are influenced by transcription rate and mRNA stability and decay. The only known method of RsmA-mediated regulation of gene expression occurs through alteration of mRNA stability and subsequent translation. However, the known influence of RsmA on the stability of certain mRNAs may not be sufficient to explain the extent of changes observed in the comparative transcriptome profile. Therefore, given that several phenotypes were greatly affected by inactivation of RsmA, we believe that RsmA posttranscriptionally regulates one or more transcriptional regulators, which in turn affect the transcription of a number of genes.
RsmA is involved in the expression of pvdS and siderophore biosynthetic genes. A recently identified link between the QS network and iron acquisition in P. aeruginosa is VqsR (Juhas et al., 2004). VqsR is a regulator of the QS network in P. aeruginosa TB, and has been shown to regulate genes involved in iron acquisition. vqsR mutants have reduced C4-HSL and 3-oxo-C12-HSL production, reduced expression of both the phnAB and pqsABCDE genes, and reduced expression of pvdS, along with reduced transcription of other genes in the pyoverdine and pyochelin biosynthetic loci (Juhas et al., 2004). As both RsmA and VqsR influence AHL production, and the expression of phnA and the genes involved in iron acquisition, it is possible to consider a potential link between these two global regulators, which has not been previously recognized. Clearly more work needs to be carried out to unravel the involvement of RsmA in the complex, highly regulated iron acquisition and storage system in P. aeruginosa.
During late-exponential and stationary phase, cultures of the rsmA mutant produced less PQS than PAO1 wild-type. The QS network has been implicated in the regulation of PQS production: the las system positively controls the transcription of pqsA, while the rhl system negatively controls the transcription of pqsA (McGrath et al., 2004), via the impact of both systems on pqsR expression (Wade et al., 2005). However, as neither pqsR nor pqsA transcription was altered in the rsmA mutant, it is unlikely that the increased AHL levels (Pessi et al., 2001) influence the decrease in PQS production in the rsmA mutant. Therefore, another explanation for this drop in PQS levels was sought. In P. aeruginosa, there are two anthranilate synthase systems: PhnA and PhnB (anthranilate synthase II) and TrpE and TrpG (anthranilate synthetase I) (Essar et al., 1990). Both systems have different functions, in that anthranilate synthetase I converts chorismate to the anthranilate used in tryptophan biosynthesis, while anthranilate synthase II converts chorismate to the anthranilate used in phenazine biosynthesis (Essar et al., 1990). It has been suggested recently that anthranilate produced by anthranilate synthetase I (the Trp pathway) is not available for PQS biosynthesis (Aendekerk et al., 2005). As trpG and trpE expression was not altered in the rsmA mutant compared to PAO1, the anthranilate present in the rsmA mutant extract is likely to be an accumulation of anthranilate formed via the anthranilate synthetase I system. As phnA transcription was decreased by approximately threefold in the rsmA mutant compared to PAO1 wild-type, while pqsABCDE levels were not altered, it is possible that the anthranilate required for PQS formation is not available in the rsmA mutant, hence the reduction in PQS levels. Interestingly, loss of RsmA brought about a dramatic drop in levels of the phenazine pyocyanin. This finding is consistent with the decrease in phnA expression in the rsmA mutant compared to PAO1 wild-type, since a phnA mutant has been shown elsewhere to have reduced pyocyanin levels (Essar et al., 1990). In addition, the decrease in expression of phzA1 in the rsmA mutant may contribute to the decrease in pyocyanin production. Other studies have shown that loss of RsmA results in an increase in pyocyanin production compared to P. aeruginosa PAO1 wild-type, when the bacteria are cultured in glycerol/alanine broth (Pessi et al., 2001). However, our data show that this is not the case when the bacteria are cultured in LB broth. It is possible that the different carbon sources used in the two experiments alter pyocyanin production, via the activation of the PhnAB versus the TrpEG pathway, which would explain the discrepancy in the results.
We have shown that RsmA affects expression of vfr, encoding a homologue of the E. coli cAMP receptor protein. Comparison of the transcriptome profiles of the P. aeruginosa PAK vfr mutant (Wolfgang et al., 2003) and the P. aeruginosa PAO1 rsmA mutant showed that 19 genes and one intergenic region were commonly altered in expression (Supplementary Table S2). The influence of RsmA on the expression of this subset of genes may therefore be through regulation of vfr, particularly with respect to the pil genes involved in the biosynthesis of pili. Vfr-null mutants have been shown to exhibit severely reduced twitching motility with barely detectable levels of type IV pili, as the mutant is defective in type IV pilus assembly (Beatson et al., 2002). Beatson et al. (2002) propose that it is most likely that Vfr acts as a transcriptional regulator of a gene(s) encoding functional elements of type IV pili or signal transduction pathways that control twitching motility. If this is indeed the case, the drop in vfr levels in the rsmA mutant could explain the reduced expression of genes involved in pilus biosynthesis. Another candidate target of direct RsmA control is the regulatory protein MvaT. Recently, MvaT has been shown to be a negative regulator of the transcription of cupA genes and, to a lesser extent, of cupB and cupC expression (Vallet et al., 2004). Expression of cupA2 and cupC3 was decreased in the rsmA mutant compared to PAO1 wild-type. As a regulator of the cup genes, it is therefore possible that RsmA has posttranscriptional control over mvaT. Since mvaT transcript levels were not altered in the rsmA mutant, as revealed by the transcriptome assay, it is likely that RsmA affects the translation of mvaT rather than the stability of its mRNA.
Since the retS mutant is thought to alter gene expression in part by controlling the level of free, active RsmA (Goodman et al., 2004), we assessed the similarities between the transcriptome profiles of the P. aeruginosa PAK retS mutant and the P. aeruginosa PAO1 rsmA mutant, and genes of altered expression in both backgrounds were compared. This revealed that 53 altered genes were common to both mutants (Supplementary Table S3), hence strengthening the proposal put forward by Goodman et al. (2004) that RsmA is part of the RetS regulon.
Recent studies conducted by Heeb et al. (2005) have shown that in P. fluorescens CHA0, RsmA posttranscriptionally regulates rpoS. The stationary-phase sigma factor RpoS has been implicated in the regulation of numerous functions in P. aeruginosa, including QS-regulated processes (Schuster et al., 2004). To assess the posttranscriptional impact of the loss of RsmA on rpoS, an rpoS–lacZ translational fusion was analysed in P. aeruginosa PAO1 wild-type and the rsmA mutant over time. However, there was no difference in the translation of rpoS between the two strains (data not shown). Therefore, in contrast to P. fluorescens CHA0, P. aeruginosa PAO1 RsmA does not posttranscriptionally regulate rpoS, under the experimental conditions outlined.
Interestingly, the gene with the highest fold increase in expression (1173-fold) (Table 2) in the rsmA mutant compared to PAO1 wild-type was gloA2, which encodes an enzyme involved in the breakdown of the toxic electrophile methylglyoxal, a by-product of glycolysis. This suggests that, like CsrA in E. coli (Romeo, 1998), RsmA posttranscriptionally regulates genes involved in carbon metabolism in P. aeruginosa.
In conclusion, this study has revealed the impact of RsmA on gene expression and the regulation of key cellular processes in P. aeruginosa. In addition, evidence is provided suggesting that RsmA may function as a downstream effector of hierarchical regulatory proteins, such as RetS and MvfR.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Albus, A. M., Pesci, E. C., Runyen-Janecky, L. J., West, S. E. & Iglewski, B. H. (1997). Vfr controls quorum sensing in Pseudomonas aeruginosa. J Bacteriol 179, 3928–3935.
Baker, C. S., Morozov, I., Suzuki, K., Romeo, T. & Babitzke, P. (2002). CsrA regulates glycogen biosynthesis by preventing translation of glgC in Escherichia coli. Mol Microbiol 44, 1599–1610.[CrossRef][Medline]
Baysse, C., Cullinane, M., Dénervaud, V., Burrowes, E., Dow, J. M., Morrissey, J. P., Tam, L., Trevors, J. T. & O'Gara, F. (2005). Modulation of quorum sensing in Pseudomonas aeruginosa through alteration of membrane properties. Microbiology 151, 2529–2542.
Beare, P. A., For, R. J., Martin, L. W. & Lamont, I. L. (2003). Siderophore-mediated cell signalling in Pseudomonas aeruginosa: divergent pathways regulate virulence factor production and siderophore receptor synthesis. Mol Microbiol 47, 195–207.[CrossRef][Medline]
Beatson, S. A., Whitchurch, C. B., Sargent, J. L., Levesque, R. C. & Mattick, J. S. (2002). Differential regulation of twitching motility and elastase production by Vfr in Pseudomonas aeruginosa. J Bacteriol 184, 3605–3613.
Bertani, I., Sevo, M., Kojic, M. & Venturi, V. (2003). Role of GacA, LasI, RhlI, Ppk, PsrA, Vfr and ClpXP in the regulation of the stationary-phase sigma factor rpoS/RpoS in Pseudomonas. Arch Microbiol 180, 264–271.[CrossRef][Medline]
Blumer, C., Heeb, S., Pessi, G. & Haas, D. (1999). Global GacA-steered control of cyanide and exoprotease production in Pseudomonas fluorescens involves specific ribosome binding sites. Proc Natl Acad Sci U S A 96, 14073–14078.
Burrowes, E., Abbas, A., O'Neill, A., Adams, C. & O'Gara, F. (2005). Characterisation of the regulatory RNA RsmB from Pseudomonas aeruginosa PAO1. Res Microbiol 156, 7–16.[Medline]
Collier, D. N., Anderson, L., McKnight, S. L., Noah, T. L., Knowles, M., Boucher, R., Schwab, U., Gilligan, P. & Pesci, E. C. (2002). A bacterial cell-to-cell signal in the lungs of cystic fibrosis patients. FEMS Microbiol Lett 215, 41–46.[CrossRef][Medline]
Cui, Y., Chatterjee, A., Liu, Y., Dumenyo, C. K. & Chatterjee, A. K. (1995). Identification of a global repressor gene, rsmA, of Erwinia carotovora subsp. carotovora that controls extracellular enzymes, N-(3-oxohexanoyl)-L-homoserine lactone, and pathogenicity in soft-rotting Erwinia spp. J Bacteriol 177, 5108–5115.
Dasgupta, N. & Ramphal, R. (2001). Interaction of the antiactivator FleN with the transcriptional activator FleQ regulates flagellar number in Pseudomonas aeruginosa. J Bacteriol 183, 6636–6644.
Dasgupta, N., Ferrell, E. P., Kanack, K. J., West, S. E. & Ramphal, R. (2002). fleQ, the gene encoding the major flagellar regulator of Pseudomonas aeruginosa, is 
70 dependent and is downregulated by Vfr, a homolog of Escherichia coli cyclic AMP receptor protein. J Bacteriol 184, 5240–5250.
Dasgupta, N., Wolfgang, M. C., Goodman, A. L., Arora, S. K., Jyot, J., Lory, S. & Ramphal, R. (2003). A four-tiered transcriptional regulatory circuit controls flagellar biogenesis in Pseudomonas aeruginosa. Mol Microbiol 50, 809–824.[CrossRef][Medline]
Deziel, E., Gopalan, S., Tampakaki, A. P., Lepine, F., Padfield, K. E., Saucier, M., Xiao, G. & Rahme, L. G. (2005). The contribution of MvfR to Pseudomonas aeruginosa pathogenesis and quorum sensing circuitry regulation: multiple quorum sensing-regulated genes are modulated without affecting lasRI, rhlRI or the production of N-acyl-L-homoserine lactones. Mol Microbiol 55, 998–1014.[CrossRef][Medline]
Diggle, S. P., Winzer, K., Chhabra, S. R., Worrall, K. E., Camara, M. & Williams, P. (2003). The Pseudomonas aeruginosa quinolone signal molecule overcomes the cell density-dependency of the quorum sensing hierarchy, regulates rhl-dependent genes at the onset of stationary phase and can be produced in the absence of LasR. Mol Microbiol 50, 29–43.[CrossRef][Medline]
Essar, D. W., Eberly, L., Hadero, A. & Crawford, I. P. (1990). Identification and characterization of genes for a second anthranilate synthase in Pseudomonas aeruginosa: interchangeability of the two anthranilate synthases and evolutionary implications. J Bacteriol 172, 884–900.
Fuqua, C. & Greenberg, E. P. (1998). Self-perception in bacteria: quorum sensing with acylated homoserine lactones. Curr Opin Microbiol 1, 183–189.[CrossRef][Medline]
Gallagher, L. A., McKnight, S. L., Kuznetsova, M. S., Pesci, E. C. & Manoil, C. (2002). Functions required for extracellular quinolone signalling by Pseudomonas aeruginosa. J Bacteriol 184, 6472–6480.
Gambello, M. J., Kaye, S. & Iglewski, B. H. (1993). LasR of Pseudomonas aeruginosa is a transcriptional activator of the alkaline protease gene (apr) and an enhancer of exotoxin A expression. Infect Immun 61, 1180–1184.
Goodman, A. L., Kulasekara, B., Rietsch, A., Boyd, D., Smith, R. S. & Lory, S. (2004). A signalling network reciprocally regulates genes associated with acute infection and chronic persistence in Pseudomonas aeruginosa. Dev Cell 7, 745–754.[CrossRef][Medline]
Hassett, D. J., Howell, M. L., Ochsner, U. A., Vasil, M. L., Johnson, Z. & Dean, G. E. (1997). An operon containing fumC and sodA encoding fumarase C and manganese superoxide dismutase is controlled by the ferric uptake regulator in Pseudomonas aeruginosa: fur mutants produce elevated alginate levels. J Bacteriol 179, 1452–1459.
Hassett, D. J., Ma, J. F., Elkins, J. G. & 10 other authors (1999). Quorum sensing in Pseudomonas aeruginosa controls expression of catalase and superoxide dismutase genes and mediates biofilm susceptibility to hydrogen peroxide. Mol Microbiol 34, 1082–1093.[CrossRef][Medline]
Heeb, S., Valverde, C., Gigot-Bonnefoy, C. & Haas, D. (2005). Role of the stress sigma factor RpoS in GacA/RsmA-controlled secondary metabolism and resistance to oxidative stress in Pseudomonas fluorescens CHA0. FEMS Microbiol Lett 243, 251–258.[CrossRef][Medline]
Heinrichs, D. E. & Poole, K. (1996). PchR, a regulator of ferripyochelin receptor gene (fptA) expression in Pseudomonas aeruginosa, functions both as an activator and as a repressor. J Bacteriol 178, 2586–2592.
Heurlier, K., Williams, F., Heeb, S., Dormond, C., Pessi, G., Singer, D., Camara, M., Williams, P. & Haas, D. (2004). Positive control of swarming, rhamnolipid synthesis, and lipase production by the posttranscriptional RsmA/RsmZ system in Pseudomonas aeruginosa PAO1. J Bacteriol 186, 2936–2945.
Hunt, T. A., Peng, W. T., Loubens, I. & Storey, D. G. (2002). The Pseudomonas aeruginosa alternative sigma factor PvdS controls exotoxin A expression and is expressed in lung infections associated with cystic fibrosis. Microbiology 148, 3183–3193.
Jacobs, M. A., Alwood, A., Thaipisuttikul, I. & 12 other authors (2003). Comprehensive transposon mutant library of Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 100, 14339–14344.
Juhas, M., Wiehlmann, L., Huber, B. & 8 other authors (2004). Global regulation of quorum sensing and virulence by VqsR in Pseudomonas aeruginosa. Microbiology 150, 831–841.
Juhas, M., Eberl, L. & Tummler, B. (2005). Quorum sensing: the power of cooperation in the world of Pseudomonas. Environ Microbiol 7, 459–471.[CrossRef][Medline]
Klausen, M., Heydorn, A., Ragas, P., Lambertsen, L., Aaes-Jorgensen, A., Molin, S. & Tolker-Nielsen, T. (2003). Biofilm formation by Pseudomonas aeruginosa wild type, flagella and type IV pili mutants. Mol Microbiol 48, 1511–1524.[CrossRef][Medline]
Kohler, T., van Delden, C., Curty, L. K., Hamzehpour, M. M. & Pechere, J. C. (2001). Overexpression of the MexEF-OprN multidrug efflux system affects cell-to-cell signalling in Pseudomonas aeruginosa. J Bacteriol 183, 5213–5222.
Lamont, I. L., Beare, P. A., Ochsner, U., Vasil, A. I. & Vasil, M. L. (2002). Siderophore-mediated signalling regulates virulence factor production in Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 99, 7072–7077.
Laskowski, M. A., Osborn, E. & Kazmierczak, B. I. (2004). A novel sensor kinase-response regulator hybrid regulates type III secretion and is required for virulence in Pseudomonas aeruginosa. Mol Microbiol 54, 1090–1103.[CrossRef][Medline]
Liu, M. Y. & Romeo, T. (1997). The global regulator CsrA of Escherichia coli is a specific mRNA-binding protein. J Bacteriol 179, 4639–4642.
Maseda, H., Saito, K., Nakajima, A. & Nakae, T. (2000). Variation of the mexT gene, a regulator of the MexEF-oprN efflux pump expression in wild-type strains of Pseudomonas aeruginosa. FEMS Microbiol Lett 192, 107–112.[CrossRef][Medline]
Mattick, J. S., Whitchurch, C. B. & Alm, R. A. (1996). The molecular genetics of type-4 fimbriae in Pseudomonas aeruginosa – a review. Gene 179, 147–155.[CrossRef][Medline]
Mavrodi, D. V., Bonsall, R. F., Delaney, S. M., Soule, M. J., Phillips, G. & Thomashow, L. S. (2001). Functional analysis of genes for biosynthesis of pyocyanin and phenazine-1-carboxamide from Pseudomonas aeruginosa PAO1. J Bacteriol 183, 6454–6465.
McGrath, S., Wade, D. S. & Pesci, E. C. (2004). Dueling quorum sensing systems in Pseudomonas aeruginosa control the production of the Pseudomonas quinolone signal (PQS). FEMS Microbiol Lett 230, 27–34.[CrossRef][Medline]
McKnight, S. L., Iglewski, B. H. & Pesci, E. C. (2000). The Pseudomonas quinolone signal regulates rhl quorum sensing in Pseudomonas aeruginosa. J Bacteriol 182, 2702–2708.
Pesci, E. C., Milbank, J. B., Pearson, J. P., McKnight, S., Kende, A. S., Greenberg, E. P. & Iglewski, B. H. (1999). Quinolone signalling in the cell-to-cell communication system of Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 96, 11229–11234.
Pessi, G. & Haas, D. (2001). Dual control of hydrogen cyanide biosynthesis by the global activator GacA in Pseudomonas aeruginosa PAO1. FEMS Microbiol Lett 200, 73–78.[CrossRef][Medline]
