| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Microbiology, College of Natural Sciences, University of Hawaii at Manoa, Honolulu, HI 96822, USA
Correspondence
Tung T. Hoang
tongh{at}hawaii.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
psrA mutant indicated its importance in regulating β-oxidative enzymes. These microarray data were confirmed by real-time RT-PCR analyses of the fadB5 and lipA (encoding a lipase) genes. Induction of the fadBA5 operon was demonstrated to respond to novel LCFA signals, and this induction required the presence of PsrA, suggesting that LCFAs bind to PsrA to derepress fadBA5. Electrophoretic mobility shift assays indicate specific binding of PsrA to the fadBA5 promoter region. This binding is disrupted by specific LCFAs (C18 : 19, C16 : 0, C14 : 0 and, to a lesser extent, C12 : 0), but not by other medium- or short-chain fatty acids or the first intermediate of β-oxidation, acyl-CoA. It is shown here that PsrA is a fadBA5 regulator that binds and responds to LCFA signals in P. aeruginosa.
These authors contributed equally to this work.
A supplementary table listing sequenced chromosomal transposition sites in strain PAO1-PfadBA5–lacZ and a supplementary figure showing the confirmation of three regulatory mutants affecting the expression of fadBA5 are available with the online version of this paper.
The array data discussed in this paper have been deposited in the NCBI Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO series accession number GSE8083.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
; Bowton, 1999; Fleiszig et al., 1995; Greenberger, 1997; Lode et al., 2000; Pruitt et al., 1998; Reyes & Lerner, 1983; Richards et al., 1999; Schaberg et al., 1991). P. aeruginosa lung infections transpire in nosocomial pneumonia and in CF patients. Nosocomial pneumonia is the second most common of all nosocomial infections, and P. aeruginosa has been the most frequently isolated microbe responsible (Pennington, 1994; Richards et al., 1999). In addition, over 93 % of CF patients between the ages of 18 and 24 have been infected with P. aeruginosa (Doring, 1997).
Fatty acid degradation (Fad) pathways may play critical roles in P. aeruginosa pathogenesis. It has recently been demonstrated that the Fad pathways of P. aeruginosa may have important implications in nutrient acquisition and pathogenesis within the CF lung (Son et al., 2007), through the degradation of an essential lung surfactant component phosphatidylcholine (PC). The degradation of one PC component, long-chain fatty acid (LCFA), potentially occurs through three different β-oxidation pathways, involving three fadBA operons (fadAB1, PA1737 and PA1736; fadAB4, PA4786 and PA4785; and fadBA5, PA3014 and PA3013) (Son et al., 2007, 2008; unpublished data). Thus, understanding the genetic regulation of Fad could have direct relevance to the physiology and pathogenesis of this micro-organism within the context of CF infections. The aerobic Fad pathway in Escherichia coli is well established, and several of the genes are negatively regulated by FadR in the absence of external fatty acids (FAs) 
C12 (Cronan & Satyanarayana, 1998; DiRusso et al., 1992), including fadA, fadB, fadD, fadE and fadL (Campbell & Cronan, 2002). The fad genes are derepressed by acyl-CoA (>C10) binding to FadR, and this complex no longer activates fab genes (DiRusso et al., 1992; Henry & Cronan, 1992). However, the regulation of Fad in P. aeruginosa has not been characterized, and deciphering the regulation mechanism of Fad is important to further understand the physiology and pathogenesis of P. aeruginosa.
In this study, we focused on identifying the regulation mechanism of the P. aeruginosa fadBA5 operon, which is involved in Fad. We performed transposon mutagenesis to identify a regulator of the fadBA5 operon. We showed that this regulator, PsrA (PA3006), represses the fadBA5 operon. This repression is relieved by LCFA signals, as shown by gene fusion data, and this was consistent with our in vitro DNA binding results. We performed microarray experiments to further identify other genes within the regulon controlled by this regulatory protein. The microarray data were confirmed by real-time RT-PCR for fadB5 and lipA.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
) was routinely used as a strain for cloning and was cultured in Luria–Bertani (LB) medium (Difco). P. aeruginosa strain PAO1 (Holloway et al., 1994; Stover et al., 2000) and derivatives were cultured on Pseudomonas Isolation Agar (PIA; Difco) or LB medium. All chemicals were purchased from Sigma. FA (butyric acid, C4 : 0; N-caproic acid, C6 : 0; caprylic acid, C8 : 0; capric acid, C10 : 0; lauric acid, C12 : 0; myristic acid, C14 : 0; palmitic acid, C16 : 0 and oleic acid, C18 : 19) stock solutions at 3 % (w/v) were made with equimolar KOH and 1 % Brij-58 [poly(oxyethylene) cetyl ether]. Growth curves (Fig. 1) were performed in 1x M9 (containing 0.5 mM MgCl2 and 0.02 mM CaCl2)+1 % Brij-58 supplemented with 20 mM glucose or 0.2 % of different FAs as sole carbon source. Growth curves for β-galactosidase activity (Fig. 3) were performed in 1x M9+1 % Brij-58+40 mM glucose with or without 0.1 % of the particular FA. Complementation of the psrA mutation (Fig. 5b) was performed in LB+1 % Brij-58 with or without 0.2 % C18 : 19. Unless indicated otherwise, the cultures were grown at 37 °C with a shaking speed of 250 r.p.m.
|
|
|
. Oligonucleotide primers (Table 1) were synthesized by Integrated DNA Technologies (IDT). Southern hybridization analysis was performed as described previously (Hoang et al., 1998). P. aeruginosa competent cells were prepared as described elsewhere (Choi et al., 2006). Generally, we performed the various PCRs by initial denaturation for 1 min at 94 °C and 34 cycles of 45 s at 94 °C, 30 s at 58 °C, and 1 min kb–1 at 72 °C, and a final step of 10 min at 72 °C was included. Pfu was purchased from Stratagene.
|
fadBA5 mutant strain and complementation study.fadBA5 mutant strain, PAO1-fadBA5, has been described previously (Son et al., 2007). Complementation of the PAO1-fadBA5 mutant was done as follows. The fadBA5 operon, including its native promoter region, was amplified from the PAO1 chromosome by PCR using oligos #287 and #273. The PCR product was digested with EcoRI and cloned into mini-CTX2 (Hoang et al., 2000) digested with SmaI+EcoRI, yielding mini-CTX2-fadBA5. Next, mini-CTX2-fadBA5 was transformed into the PAO1-fadBA5 mutant to integrate into its chromosome and complement the fadBA5 mutation.
Growth curve studies were performed on three strains (PAO1, PAO1-fadBA5 and PAO1-fadBA5/miniCTX2-fadBA5). These strains were grown overnight at 37 °C in PIA medium. Overnight cultures were washed twice with one volume of 1x M9 buffer and resuspended in an equal volume of the same buffer. Resuspended cultures were then diluted 100-fold into fresh 1x M9+20 mM glucose or 0.2 % of a FA (C4 : 0, C6 : 0, C8 : 0, C10 : 0, C12 : 0, C14 : 0, C16 : 0 or C18 : 19). Growth was then initiated. At each time point, aliquots of each culture were diluted fourfold in 4 % Brij-58 and OD540 measurements were taken.
Promoter mapping.
We mapped transcript start sites for fadB5 (PA3014), according to a method for mapping eukaryotic transcript start sites described elsewhere (Tabansky & Nurminsky, 2003). Cells, grown in 1x M9+palmitate, were harvested for promoter mapping in exponential phase (OD540 
0.6), early stationary phase (10 h after exponential phase), and late stationary phase (18 h after early stationary phase). All reagents required for transcriptional start-site mapping were supplied in the SMART RACE cDNA Amplification kit (Clontech, BD Biosciences). Briefly, total mRNA was isolated with an on-column DNase I treatment, using the Qiagen RNeasy mini kit. A 1 µg sample of total mRNA was assembled with a SMART primer and the first gene-specific primer (primers #244 and #279, Table 1
); each primer was at 2 µM final concentration. cDNA synthesis was carried out with the reagents supplied and as recommended by the manufacturer. The diluted cDNA was used as a template in a PCR with a second nested gene-specific primer (Table 1, primer #280) and the SMART RACE primer. PCR products were purified from gel slices and sequenced with a third nested gene-specific primer (Table 1, primer #281) at our local core facility.
PfadBA5–lacZ fusion construction and β-galactosidase assay.
The PfadBA5–lacZ transcriptional fusion was constructed by amplifying the fadBA5 promoter region, encompassing nucleotides 148–266 upstream of the start codon. PCR was performed using primers #287 and #289 with PAO1 chromosomal DNA. The resulting 120 bp fragment was cut with HindIII and cloned into HindIII- and SmaI-digested pTZ120 (a derivative of pTZ110) (Schweizer & Chuanshuen, 2001) to yield pTZ120-PfabAB. For integration into the P. aeruginosa genome, the PfadBA5–lacZ fusion was removed by AflIII digestion, blunt-ended, and then digested with HindIII. This fragment was then subcloned between the HindIII and SmaI sites of mini-CTX2 to yield mini-CTX2-PfadBA5–lacZ. Chromosomal integration of this mini-CTX2–lacZ fusion vector in P. aeruginosa, excision of unwanted plasmid sequences and verification of insertion at the chromosomal attB locus were performed as described previously (Hoang et al., 2000).
β-Galactosidase activities were measured for the integrated PfadBA5–lacZ fusion under various growth conditions. Cells harbouring the fusions were grown overnight in PIA medium. Overnight cultures were washed twice with one volume of 1x M9 buffer and resuspended in an equal volume of the same buffer. Resuspended cultures were then diluted 100-fold into fresh 1x M9+40 mM glucose±0.1 % FA (C4 : 0, C6 : 0, C8 : 0, C10 : 0, C12 : 0, C14 : 0, C16 : 0 or C18 : 19). Growth curves were determined for each culture by diluting them fourfold in 4 % Brij-58, after which OD540 was measured. All growth curves had identical growth rates (data not shown). Cell cultures (1 ml) were taken at mid-exponential phase (OD540 1.4) during the growth curve experiments, and β-galactosidase assays were performed from the same cultures. β-Galactosidase assays were done in triplicate and measured in Miller units (Miller, 1992; mean±SEM).
Identification and characterization of transposon mutants affecting fadBA5 expression.
The fusion strain PAO1-PfadBA5–lacZ was subjected to transposon mutagenesis using the mariner transposon vector pBT20. The transposon in pBT20 was conjugally transferred by biparental mating into PAO1-PfadBA5–lacZ, following a protocol described elsewhere (Kulasekara et al., 2005). Mating mixtures were scraped and resuspended in 3 ml LB. Then, 300 µl of this suspension was plated onto PIA+gentamicin (Gm) at 150 µg ml–1+X-Gal (40 µg ml–1) and incubated at 37 °C. Colonies that appeared more blue were selected. Transposon insertion sites were determined through sequencing the flanking region of the transposon by a semi-random PCR method, as described elsewhere (Levano-Garcia et al., 2005), using random primer #524 and transposon-specific primer #463, followed by the nested primers #525 and #526 (Table 1). β-Galactosidase activities were measured and compared with those of the parental strain (PAO1-PfadBA5–lacZ) for the three mutants identified as regulatory mutants (Supplementary Fig. S1), where all were grown in LB.
Transduction and complementation of the PAO1-PfadBA5–lacZ/psrA : : Tn-Gmr mutant.
The chromosomal disrupted psrA : : Tn-Gmr mutation was transduced back into the original parental PAO1-PfadBA5–lacZ fusion strain following a protocol described elsewhere (Choi et al., 2006). Transposon mutants harbouring the PfadB5–lacZ reporter gene fusions and four independent isolates of the retransduced PAO1-psrA : : Tn-Gmr mutants were grown overnight in PIA+Gm (150 µg ml–1), along with PAO1-PfadBA5–lacZ in PIA. Overnight cultures were washed twice and resuspended in an equal volume of LB medium. Resuspended cultures were then diluted 100-fold into fresh LB. Growth curves were performed, and β-galactosidase assays at mid-exponential phase were carried out for each strain.
For the complementation study, the coding region of psrA was amplified from PAO1 chromosomal DNA using oligonucleotides #691 and #692. The NdeI/BamHI-digested PCR product was cloned into pUCP-Nde (Cronin & McIntire, 1999) that had been digested with the same enzymes. Competent cells of the PAO1-PfadBA5–lacZ/psrA : : Tn-Gmr mutant were prepared and electroporated with 50 ng plasmid DNA of pUCP-Nde-psrA or pUCP-Nde, as described elsewhere (Choi et al., 2006), and selected on PIA+carbenicillin (Cb;500 µg ml–1). Transposon mutants with the PfadBA5–lacZ reporter gene fusion, harbouring pUCP-Nde-psrA or pUCP-Nde, were grown overnight in PIA+Cb500, washed twice with LB, and diluted 100-fold into fresh LB+Cb500+1 % Brij-58 with or without 0.2 % C18 : 19. Growth curves were performed and β-galactosidase assays at mid-exponential phase were carried out for each strain.
Purification of His6–PsrA.
His6–PsrA was expressed and purified using the E. coli T7 RNA polymerase expression system. The expression vector pET15b-psrA was constructed by cloning the psrA-encoding region from pUCP-Nde-psrA into pET15b as an NdeI–BamHI fragment, and then transformed into the BL21-Codonplus(DE3)-RIPL strain (Stratagene). The transformed BL21-Codonplus(DE3)-RIPL strain containing pET15b-psrA was grown and PsrA was expressed as described previously (Hoang et al., 2002). After expression, cells were harvested and protein purification was performed as described previously (Hoang et al., 2002). Protein concentration was determined by Bradford Assay (Bio-Rad), using BSA as a standard, prior to electrophoretic mobility shift assay (EMSA) studies.
EMSA of PsrA binding to PfadBA5.
Purified His6–PsrA was incubated with the PfadBA5 promoter DNA fragment (amplified with oligonucleotides #287 and #289), and a gel mobility shift was performed to demonstrate specific DNA binding. Essentially, 100 ng purified PfadBA5 DNA fragment was incubated with 1 µg of purified His6–PsrA in 1x binding buffer [20 mM HEPES, pH 7.6, 1 mM EDTA, 10 mM (NH4)2SO4, 1 mM DTT, 1 % (w/v) Tween 20, 150 mM KCl] with 1 µg poly(dI-C) and 0.1 µg poly-L-lysine for 30 min at room temperature. For all the FAs used in EMSA (C18 : 19, C16 : 0, C14 : 0, C12 : 0, C10 : 0, C8 : 0, C6 : 0 and C4 : 0), 1 mM stocks were made by dissolving the FA in 0.2 % Brij-58 with equimolar KOH. Oleoyl-CoA was diluted to 1 mM with 1x binding buffer+0.2 % Brij-58. Master-mix containing binding buffer, poly(dI-C), poly-L-lysine, and the DNA fragment of PfadBA5 was made for each gel, then FAs (2.5–50 µM) or oleoyl-CoA (50 µM) were added to each aliquot to the appropriate concentration. Purified His6–PsrA was always the last component to be added to the binding reactions. Each binding reaction was loaded on an 8 % native polyacrylamide gel (pre-run at 90 V for 1 h) and ran for 120 min at 90 V. Gels were stained with SYBR Green as recommended by the manufacturer of EMSA kit E33075 (Invitrogen/Molecular Probes).
RNA isolation, cDNA synthesis and labelling, and Affymetrix GeneChip processing.
Starter cultures of PAO1 or the PAO1-psrA : : Tn-Gmr mutant derivative were grown in LB for 14 h at 37 °C with aeration, and a 1 : 100 dilution of the starter cultures was inoculated into 20 ml LB in a 125 ml flask. Cultures were then incubated at 37 °C with shaking and grown up to mid-exponential phase (OD540 0.6). Cells were then harvested and P. aeruginosa mRNA isolated. cDNA syntheses and labelling, as well as GeneChip processing and microarray analysis, were performed as described previously (Son et al., 2007). Analysis was done by conducting pair-wise comparisons between two replicates of PAO1 and three replicates of PAO1-psrA : : Tn-Gmr grown in LB. Genes that exhibited a twofold or greater increase in expression in PAO1-psrA : : Tn-Gmr relative to PAO1 form one list (PsrA-repressed genes; Table 2), while genes that showed a twofold or greater increase in expression in PAO1 compared with PAO1-psrA : : Tn-Gmr form another (PsrA-activated genes; Table 3).
|
|
Real-time TaqMan RT-PCR and data analysis.
Primers, probes and real-time RT-PCR were exactly the same as described previously (Son et al., 2007). RNAs used were from the same preparation as described above for microarray experiments. Because normalization by geometric averaging using multiple housekeeping genes has been shown to yield more accurate fold-changes than using a single housekeeping gene (Vandesompele et al., 2002), we opted to use three housekeeping genes for our normalizations. Three housekeeping genes described previously (PA1769, PA1795 and PA1805; Son et al., 2007) were selected and utilized to normalize fold-changes for lipA and fadB5 in the PAO1-psrA : : Tn-Gmr mutant compared with PAO1.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
), and the organism is capable of degrading short-, medium- and long-chain fatty acids (SCFAs, MCFAs and LCFAs) as sole carbon sources. Among FadB and FadA proteins in P. aeruginosa, FadB5 (PA3014) and FadA5 (PA3013) have the highest homology to their E. coli counterparts (Campbell & Cronan, 2002; Yang et al., 1991), with similarities of 72 % (54 % identity) and 76 % (61 % identity), respectively. Initial characterization demonstrated that the fadBA5 operon is involved in LCFA utilization (Son et al., 2007). In our previous study, fadBA5 was not fully characterized with respect to the utilization of FA of different chain lengths as sole carbon sources. In this study, the PAO1-fadBA5 mutant was further characterized. The PAO1-fadBA5 mutant had no growth defect when grown on glucose SCFA or MCFA (C4 : 0 to C10 : 0) as sole carbon source, but it showed defects in growth using LCFAs as sole carbon source (C12 : 0, C14 : 0, C16 : 0 and C18 : 19, Fig. 1). The slower growth and lower cell density of the mutant compared with the wild-type suggest a defect in the ability to degrade these LCFAs and the accumulation of growth-inhibitory intermediates. When the fadBA5 operon was cloned into mini-CTX2 and integrated as a single copy into the PAO1-fadBA5 mutant background, it completely complemented this mutation (Fig. 1), indicating that there were no other spurious mutations on the chromosome of the mutant strain other than fadBA5. There was an incomplete lack of growth using LCFAs as a carbon source for this mutant, because P. aeruginosa has two other fadAB operons [fadAB1 (PA1737 and PA1736; Son et al., 2008) and fadAB4 (PA4786 and PA4785; unpublished data)], which have overlapping functions with fadBA5 for MCFAs and LCFAs. Although the functions of fadAB4 have yet to be fully characterized, we have recently shown that the fadAB1 operon is induced by MCFAs and to a lesser extent by LCFAs (Son et al., 2008), perhaps for the degradation of these carbon sources, suggesting that FadBA1 has overlapping functions with FadBA5 in the metabolism of different FA chain lengths. We have shown here that fadBA5 is important for growth on LCFAs in vitro. In vivo during lung infection, fadBA5 was significantly induced, suggesting that LCFAs are also available in vivo as a source of nutrients (Son et al., 2007). However, the regulation of this operon is undefined and poorly understood in P. aeruginosa.
Regulatory region of fadBA5
Based on an established non-radioactive method for mapping transcriptional start sites (Tabansky & Nurminsky, 2003), we mapped the promoter of the fadBA5 operon. Our data shown in Fig. 1 indicate induction of fadBA5 by palmitate (C16 : 0). Therefore, to map the fadBA5 transcript start site, we utilized RNA isolated from PAO1 grown in 1x M9 minimal media containing palmitic acid as the sole carbon source to induce the expression of these fad genes. We mapped the transcript start site for fadBA5 to 185 bp upstream of the start codon of fadB5 (Fig. 2a, c). We utilized free software (http://nostradamus.cs.rhul.ac.uk/leo/sak_demo/) for the prediction of prokaryotic 
70 promoters using sequence alignment kernel methods (Gordon et al., 2003), which predicted a perfect match to our fadBA5 transcriptional start-site mapping data. This gives further confirmation of the promoter mapping method. There is a pseudo-palindromic sequence, with one base mismatch, at the –10 and –35 region of this mapped 70 promoter (blue arrows in Fig. 2a). Since the mapping method is based on RT-PCR and is semiquantitative, the band in Fig. 2(b) (lane 1) indicated the expression of fadBA5 in the exponential phase during active degradation of palmitate as the sole carbon source, and the expression of fadBA5 was not detected in early or late stationary phase (Fig. 2b).
|
) and we had mapped the promoter of fadBA5 (Fig. 2), we next wanted to determine which FA chain lengths induce fadBA5. A PfadBA5–lacZ fusion was constructed in mini-CTX2 and introduced as a stable single copy into the P. aeruginosa chromosome to study the regulation of fadBA5 by LCFAs. We analysed the expression of the PfadBA5–lacZ fusion in strain PAO1 grown in 1x M9 glucose media±FA of different chain lengths. As shown in Fig. 3, fadBA5 was tightly repressed when grown with glucose as the sole carbon source at mid-exponential phase. This operon was also repressed in the presence of FA from C4 : 0 to C10 : 0 at mid-exponential phase. However, at mid-exponential phase, fadBA5 was induced by LCFAs (C14 : 0, C16 : 0 and C18 : 19) and to a lesser extent by C12 : 0 (Fig. 3). We had determined that at 40 mM glucose in all media with and without FA, the growth rates for the fusion strain were identical throughout the exponential phase for all cultures (data not shown). Therefore, the induction of the fadBA5 operon by LCFAs at mid-exponential phase was not due to differences in growth rates. This induction of fadBA5 by LCFAs (Fig. 3) also correlates well with its metabolic function and with its growth on similar FA chain lengths (Fig. 1).
Identification of PsrA (PA3006) as the fadBA5 regulator
To initiate studies on the fadBA5 regulation mechanism, we searched for a regulator of fadBA5. We utilized a transposon mutagenesis approach similar to that described elsewhere, using the gentamicin-resistant (Gmr) transposon vector pBT20 (Kulasekara et al., 2005) to randomly mutagenize the chromosome of strain PAO1 carrying a PfadBA5–lacZ fusion. Since it is our hypothesis that there is a repressor that regulates fadBA5, we looked for Gmr-transposon mutants of PAO1-PfadBA5–lacZ that were bluer on X-Gal-containing media, signifying transposon inactivation of a potential negative regulator. We sequenced 67 bluer transposon mutants as described elsewhere (Levano-Garcia et al., 2005) (Supplementary Table S1) to identify the inactivation of any potential regulator, and this is summarized in Fig. 4. It is interesting to see that a large number of bluer colonies affecting fadBA5 expression were associated with LPS biosynthesis (wpb genes), suggesting a possible connection between Fad and LPS biosynthesis. However, further research is required to determine the connection between Fad and LPS biosynthesis. Of interest were three probable transcriptional regulators (PA2601, PA3006 or PsrA, and PA3508; Supplementary Table S1). More thorough measurements of β-galactosidase activities of these three regulatory mutants showed increased β-galactosidase activities (> twofold) for the PA2601 : : Tn-Gmr and psrA : : Tn-Gmr mutants, which significantly affected the expression of fadBA5 during exponential phase (Supplementary Fig. S1). Southern hybridization analysis indicated that there were two transposons in strain PAO1-PfadBA5-lacZ/PA2601 : : Tn-Gmr and a single transposon in the chromosome of strain PAO1-PfadBA5-lacZ/psrA : : Tn-Gmr (Supplementary Fig. S1). We separated and sequenced the other mutation in the strain carrying the PA2601 : : Tn-Gmr mutation, and this indicated that there was another transposon mutation in the LPS biosynthetic gene, wbpM. Separating these two mutations by retransduction of chromosomal fragments (Choi et al., 2006) into the PAO1-PfadBA5–lacZ fusion strain indicated that PA2601 does not affect the expression of fadBA5 (data not shown), and that the increased PfadBA5 promoter activity was due to the wbpM mutation. In addition, purified His-tagged PA2601 did not bind to the fadBA5 promoter region (data not shown), showing further that PA2601 was not involved in the regulation of fadBA5.
|
and Supplementary Fig. S1). To rule out potential fadBA5 promoter mutation(s) causing constitutive PfadBA5–lacZ expression in the psrA : : Tn-Gmr mutant, we transduced this mutation back into the parental PAO1-PfadBA5–lacZ fusion strain, as described elsewhere (Choi et al., 2006). Four independent transductants were again measured for the expression of the PfadBA5–lacZ fusion, and all four mutants showed constitutive expression of fadBA5 (Fig. 5a), indicating that the derepression of fadBA5 is due to a single mutation in psrA. We further cloned and complemented the psrA : : Tn-Gmr mutation, and the clones complemented the psrA mutation to repress PfadBA5–lacZ back to wild-type levels (Fig. 5b). In addition, upon adding oleate (C18 : 19) to the LB media, these complemented clones derepressed fadBA5 similarly to wild-type. The vector-alone control showed no repression of fadBA5 (Fig. 5b), and no further increase in fadBA5 expression was observed upon addition of the FA oleate to the vector-alone control strain. Together, these data for the PAO1-psrA : : Tn-Gmr mutant indicate a requirement for LCFAs to induce the fadBA5 operon in LB media. The growth phase-independent constitutive expression of fadBA5 in the PAO1-PfadBA5–lacZ/psrA : : Tn-Gmr strain suggests a direct regulatory role of PsrA (PA3006) on fadBA5 (Supplementary Fig. S1a). We demonstrate below that PsrA binds directly to the fadBA5 promoter region.
PsrA binds to the promoter region of fadBA5 and is inhibited by LCFAs
To demonstrate direct binding of purified His6–PsrA to the promoter region of fadBA5 (PfadBA5), we performed in vitro EMSA. The data showed that PsrA binds specifically to PfadBA5 (Fig. 6a, lanes 3 and 4) and that His6–PsrA does not bind to other tested promoters or DNA (e.g. the 400 bp aacC1 gentamicin-resistance promoter; data not shown). Since PsrA regulates fadBA5 and fadBA5 is involved in growth on LCFAs, we wanted to see if free LCFAs or long-chain acyl-CoAs could inhibit the binding of PsrA to PfadBA5 by acting as an inducer. The LCFA oleate could inhibit PsrA binding. However, oleoyl-CoA, which acts as an inducer of E. coli fad genes by binding to FadR (DiRusso et al., 1992), could not inhibit PsrA binding to PfadBA5 (Fig. 6a, b). We next tested the range of FA chain lengths that PsrA could recognize and thereby inhibit the shifting of PfadBA5. Only LCFAs (C12 : 0, C14 : 0, C16 : 0 and C18 : 19), but not MCFAs or SCFAs, could visibly inhibit the binding of PsrA to PfadBA5 in vitro (Fig. 6c). The inhibition of PsrA binding to PfadBA5 DNA was in a gradient fashion from C12 : 0 to C14 : 0 and from C16 : 0 to C18 : 19 (Fig. 6c), which is consistent with the induction of PfadBA5–lacZ by LCFAs (Fig. 3) and the growth characteristics of the fadBA5 mutant (Fig. 1). Together, these data, as well as the growth curves and gene fusion experiments (Figs 1 and 3), consistently indicate that LCFAs are the inducer of fadBA5, and bind to PsrA to derepress the fadBA5 operon in P. aeruginosa.
|
). Since we utilized SYBR Green from an established EMSA kit (E33075, Invitrogen/Molecular Probes) as an alternative to radioactivity for this DNA binding study, a larger amount of DNA (100 ng; 1.28 pmol) was used to visualize the PfadBA5 DNA band rather than that typically used for radioactivity-based EMSA (25 attomol DNA) (DiRusso et al., 1992). Therefore, the protein (1 µg His6–PsrA) to DNA mole ratio was approximately 30 : 1 in our study. This ratio is significantly lower than that in the earlier radioactivity-based study with the E. coli FadR (DiRusso et al., 1992), where the protein : DNA mole ratio used was 1000 : 1. Along with the inability of His6–PsrA to bind other control DNA (data not shown), this appropriate protein : DNA ratio further supports the hypothesis that the His6–PsrA binding to PfadBA5 DNA observed in our study is specific. Given the high concentrations of DNA and His6–PsrA used, the 50 µM FA concentration used in this study is reasonable, yielding an inducer (FA) to protein mole ratio of 27 : 1. This value is comparable to that in the earlier E. coli FadR study (DiRusso et al., 1992), which utilized an inducer (acyl-CoA) to protein mole ratio of 50 : 1. Our data strongly indicate that the binding of His6–PsrA to PfadBA5 DNA is specific and is inhibited by LCFAs. The data support the suggestion that the novel signals for PsrA to regulate fadBA5 are LCFAs.
PsrA affects gene expression globally
It has been suggested elsewhere that PsrA, a member of the TetR family of repressors, could affect global gene expression (Kojic et al., 2005; Shen et al., 2006), including activation of rpoS (Kojic et al., 2002) and the type III secretion system exsCEBA operon (Shen et al., 2006). A proteomic study has shown that PsrA influences the expression of a few proteins, including FadBA5, but the signal (inducer) and mechanism of regulation by PsrA have not been demonstrated (Kojic et al., 2005). The full array of genes influenced by PsrA has not been shown. Hence, we wanted to utilize a genomic approach to determine the extent of PsrA-regulated genes, using Affymetrix GeneChips to show differential expression between wild-type PAO1 and the PAO1-psrA : : Tn-Gmr mutant. We grew both strains, with identical growth rates in LB (data not shown), and harvested both strains at the same cell density at mid-exponential phase to determine the difference in gene expression. Table 2 lists genes negatively affected by PsrA. First, PsrA regulates itself, which has been demonstrated elsewhere (Kojic et al., 2002), and our genomic data indicate a difference of 6.2-fold at mid-exponential phase. Recently, the expression of psrA and fadBA5 in vivo in the CF lung have also been observed (Son et al., 2007), suggesting that the bacteria sense free FA in the lung for PsrA to derepress itself and fadBA5. Second, several genes involved in FA metabolism were also negatively regulated by PsrA (Table 2), including fadBA5 (PA3014 and PA3013) and another possible fad gene, fadE (PA0506). Electron transport genes were significantly induced (PA2951–2953), along with cytochrome P450 (PA3331) (Table 2). This group of genes is significant for Fad, because the result of β-oxidation is the production of reducing power in the form of FADH2 and NADH. Hence, Fad is closely coupled to electron transport. It has been suggested that the E. coli FadE contains sequences that bind and shuffle its electrons (FADH2) to electron transport protein(s) (Campbell & Cronan, 2002). Cytochrome P450 has been demonstrated to be involved in the oxidation of LCFAs (Salaun & Helvig, 1995), and it exists within the gene cluster for FA catabolism of Pseudonocardia autotrophica that is also regulated by LCFAs (Chen et al., 2005). It is interesting that we saw a phospholipase D (PA3487) and a lipase (lipA, PA2862) also negatively regulated by PsrA. Of particular significance is the lipase, which could potentially liberate LCFAs from phospholipids (e.g. lung surfactant phosphatidylcholine) (Son et al., 2007). These LCFAs would then feed directly into β-oxidation through fadBA5. The microarray data were confirmed and were in close agreement with real-time RT-PCR data for both fadB5 and lipA (Table 2 in parentheses), supporting the role of PsrA as a repressor of genes in the Fad pathway. The PsrA binding or recognition consensus sequence has been demonstrated to be G/CAAACN2–4GTTTG/C (Kojic et al., 2005; Shen et al., 2006). Out of hundreds of genes regulated by PsrA in the microarray data (Tables 2 and 3), we focused on genes that potentially contribute to Fad. We found and aligned the consensus sequences upstream of these genes involved in Fad (Fig. 7), and this suggested a direct regulation mechanism of these genes by PsrA. It is interesting that there are two potential PsrA binding consensus sequences in the fadBA5 promoter region (Figs 2a and 7). The location of these binding sites relative to the –10 and –35 sequences confirms the role of PsrA as a repressor for the fadBA5 operon (Fig. 2a). Together with the microarray data and the existence of PsrA binding consensus sequences upstream, several potential fad genes seem to be directly repressed by PsrA, suggesting that one of its major physiological roles is as a regulator of FA metabolism.
|
shows the list of genes potentially activated by PsrA, directly or indirectly. The majority of the genes regulated by PsrA are unknown hypothetical proteins, the future characterization of which will be interesting. One fadE homologue, PA2815, seems to be positively regulated by PsrA. Several genes involved in the production of type IV pili were expressed, and seem to be activated by PsrA. It is curious that the stationary-phase sigma factors rpoS and exsC of the type III secretion system, shown elsewhere to be positively regulated by PsrA (Kojic et al., 2002; Shen et al., 2006), were not present in Table 3. However, our GeneChip data are only representative of a single point during exponential growth, and the regulation of these two genes could occur later in the growth phases or be less than the twofold cut-off for our growth conditions.
Conclusions
Our laboratory has been interested in Fad of P. aeruginosa with respect to the lung environment (Son et al., 2007). We have previously demonstrated the expression of genes in vivo in the lungs of CF patients, including fadBA5, which is required for PC and FA degradation in vitro (Son et al., 2007). Cleavage of PC by P. aeruginosa phospholipases and subsequently by lipases (Martinez et al., 1999; Rosenau & Jaeger, 2000) will free the readily available LCFAs to the bacteria. These FAs are made mostly of C16 : 0 (50–60 %) and 10–20 % of each of C14 : 0, C16 : 1, C18 : 1 and C18 : 2 (Postle et al., 1999). These LCFAs would serve as inducers of Fad enzymes to metabolize LCFAs via β-oxidation. Indeed, the expression of psrA and fadBA5 has also been observed in vivo in the CF lung (Son et al., 2007), suggesting that the bacteria sense free FAs in the lung for PsrA to derepress itself and fadBA5 to metabolize LCFAs.
We initiated this study to decipher the regulatory mechanism of the P. aeruginosa fadBA5 operon. FadBA5 was shown to be important for the utilization of LCFAs as sole carbon sources (Fig. 1). We mapped the transcript start site and identified the promoter region of fadBA5 (Fig. 2). The promoter region of fadBA5 was utilized to construct a PfadBA5–lacZ chromosomal fusion, allowing us to demonstrate the regulation of fadBA5 by LCFAs in P. aeruginosa (Fig. 3). The PAO1-PfadBA5–lacZ fusion strain was used in transposon mutagenesis to identify a regulator for fadBA5, which is PsrA (PA3006). The DNA recognition sequence for PsrA binding is well established (Kojic et al., 2005; Shen et al., 2006). Downstream of the fadBA5 promoter region, there were two putative PsrA binding sites (Figs 2a and 7). Our EMSA confirmed this, and indicated that PsrA binds to PfadBA5 in vitro and that this binding is inhibited by LCFAs and not by long-chain acyl-CoAs (Fig. 6). FadR, which regulates fad genes in E. coli, has been shown to bind to acyl-CoAs of C12 to derepress Fad (DiRusso et al., 1992; Henry & Cronan, 1992). This represents an interesting and novel difference between P. aeruginosa and E. coli LCFA degradation. Whether free FAs get into the cytoplasm to bind PsrA, or whether PsrA interacts closely with the membrane or other membrane proteins to obtain free FAs from the membrane, is a current enigma. Thus, how PsrA gains access to free FAs will be an interesting subject for future investigations. We further showed that PsrA is a global regulator of several genes, both positively and negatively. Among the genes repressed by PsrA were the ones leading to LCFA degradation, including fadBA5 and lipA, and this was confirmed by real-time RT-PCR. Finally, although PsrA seems to regulate some virulence genes (lipA and type IV pili genes), we focused on fadBA5 and propose here that PsrA is a β-oxidation regulator responding to LCFAs. It is exciting to anticipate the full physiological role of PsrA, once the functions of the many hypothetical proteins that it regulates are determined.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: P. Cornelis
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Bowton, D. L. (1999). Nosocomial pneumonia in the ICU – year 2000 and beyond. Chest 115, (3 Suppl.), 28S–33S.[CrossRef][Medline]
Campbell, J. W. & Cronan, J. E., Jr (2002). The enigmatic Escherichia coli fadE gene is yafH. J Bacteriol 184, 3759–3764.
Chen, C.-H., Cheng, J.-C., Cho, Y.-C. & Hsu, W.-H. (2005). A gene cluster for the fatty acid catabolism from Pseudonocardia autotrophica BCRC12444. Biochem Biophys Res Commun 329, 863–868.[CrossRef][Medline]
Choi, K.-H., Kumar, A. & Schweizer, H. P. (2006). A 10-min method for preparation of highly electrocompetent Pseudomonas aeruginosa cells: application for DNA fragment transfer between chromosomes and plasmid transformation. J Microbiol Methods 64, 391–397.[CrossRef][Medline]
Cronan, J. E., Jr & Satyanarayana, S. (1998). FadR, transcriptional co-ordination of metabolic expediency. Mol Microbiol 29, 937–943.[CrossRef][Medline]
Cronin, C. N. & McIntire, W. S. (1999). pUCP-Nco and pUCP-Nde: Escherichia–Pseudomonas shuttle vectors for recombinant protein expression in Pseudomonas. Anal Biochem 272, 112–115.[CrossRef][Medline]
DiRusso, C. C., Heimert, T. L. & Metzger, A. K. (1992). Characterization of FadR, a global transcription regulator of fatty acid metabolism in Escherichia coli. J Biol Chem 267, 8685–8691.
Doring, G. (1997). Cystic fibrosis respiratory infections: interactions between bacteria and host defense. Monaldi Arch Chest Dis 52, 363–366.[Medline]
Fleiszig, S. M. J., Zaidl, T. S. & Pier, G. B. (1995). Pseudomonas aeruginosa invasion of and multiplication within corneal epithelial cells in vitro. Infect Immun 63, 4072–4077.[Abstract]
Gordon, L., Chervonenkis, A. Y., Gammerman, A. J., Shahmuradov, I. A. & Solovyev, V. V. (2003). Sequence alignment kernel for recognition of promoter regions. Bioinformatics 19, 1964–1971.
Greenberger, P. A. (1997). Immunologic aspects of lung diseases and cystic fibrosis. JAMA 278, 1924–1930.
Henry, M. F. & Cronan, J. E. (1992). A new mechanism of transcriptional regulation: release of an activator triggered by small molecule binding. Cell 70, 671–679.[CrossRef][Medline]
Hoang, T. T., Karkhoff-Schweizer, R. R., Kutchma, A. J. & Schweizer, H. P. (1998). A broad-host-range Flp-FRT recombination system for site-specific excision of chromosomally-located DNA sequences: application for isolation of unmarked Pseudomonas aeruginosa mutants. Gene 212, 77–86.[CrossRef][Medline]
Hoang, T. T., Kutchma, A. J., Becher, A. & Schweizer, H. P. (2000). Integration-proficient plasmids for Pseudomonas aeruginosa: site-specific integration and use for engineering of reporter and expression strains. Plasmid 43, 59–72.[CrossRef][Medline]
Hoang, T. T., Sullivan, S. A., Cusick, K. C. & Schweizer, P. H. (2002). β-Ketoacyl carrier protein reductase (FabG) activity of the fatty acid biosynthetic pathway is a determining factor of 3-oxo-homoserine lactone acyl chain lengths. Microbiology 148, 3849–3856.
Holloway, B. W., Roemling, U. & Tuemmler, B. (1994). Genomic mapping of Pseudomonas aeruginosa PAO. Microbiology 140, 2907–2929.
Kojic, M., Aguilar, C. & Venturi, V. (2002). TetR family member PsrA directly binds the Pseudomonas rpoS and psrA promoters. J Bacteriol 184, 2324–2330.
Kojic, M., Jovcic, B., Vindigni, A., Odreman, F. & Venturi, V. (2005). Novel target genes of PsrA transcriptional regulator of Pseudomonas aeruginosa. FEMS Microbiol Lett 246, 175–181.[CrossRef][Medline]
Kulasekara, H. D., Ventre, I., Kulasekara, B. R., Lazdunski, A., Filloux, A. & Lory, S. (2005). A novel two-component system controls the expression of Pseudomonas aeruginosa fimbrial cup genes. Mol Microbiol 55, 368–380.[CrossRef][Medline]
Levano-Garcia, J., Verjovski-Almeida, S. & da Silva, A. C. R. (2005). Mapping transposon insertion sites by touchdown PCR and hybrid degenerate primers. Biotechniques 38, 225–229.[Medline]
Lode, H., Raffenberg, M., Erbes, R., Geerdes-Fenge, H. & Mauch, H. (2000). Nosocomial pneumonia: epidemiology, pathogenesis, diagnosis, treatment and prevention. Curr Opin Infect Dis 13, 377–384.[Medline]
Martinez, A., Ostrovsky, P. & Nunn, D. N. (1999). LipC, a second lipase of Pseudomonas aeruginosa, is LipB and Xcp dependent and is transcriptionally regulated by pilus biogenesis components. Mol Microbiol 34, 317–326.[CrossRef][Medline]
Miller, J. H. (1992). A Short Course in Bacterial Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Pennington, J. E. (1994).. Pseudomonas aeruginosa pneumonia and other respiratory tract infections. In Pseudomonas aeruginosa: Infections and Treatment, pp. 159–181. Edited by A. L. Baltch & R. P. Smith. New York: Marcel Dekker.
Postle, A. D., Mander, A., Reid, K. B. M., Wang, J.-Y., Wright, S. M., Moustaki, M. & Warner, J. O. (1999). Deficient hydrophilic lung surfactant proteins A and D with normal surfactant phospholipid molecular species in cystic fibrosis. Am J Respir Cell Mol Biol 20, 90–98.
Pruitt, B. A., Jr, McManus, A. T., Kim, S. H. & Goodwin, C. W. (1998). Burn wound infections: current status. World J Surg 22, 135–145.[CrossRef][Medline]
Reyes, M. P. & Lerner, A. M. (1983). Current problems in the treatment of infective endocarditis due to P. aeruginosa. Rev Infect Dis 5, 314–321.[Medline]
Richards, M. J., Edwards, J. R., Culver, D. H. & Gaynes, R. P. (1999). Nosocomial infections in medical intensive care units in the United States. Crit Care Med 27, 887–892.[CrossRef][Medline]
Rosenau, F. & Jaeger, K.-E. (2000). Bacterial lipases from Pseudomonas: regulation of gene expression and mechanisms of secretion. Biochimie 82, 1023–1032.[Medline]
Salaun, J. P. & Helvig, C. (1995). Cytochrome P450 dependent oxidation of fatty acids. Drug Metabol Drug Interact 12, 261–283.[Medline]
Sambrook, J. & Russell, D. W. (2001). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Schaberg, D. R., Culver, D. H. & Gaynes, R. P. (1991). Major trends in the microbial etiology of nosocomial infection. Am J Med 91 (suppl. 3B), 72S–75S.[Medline]
Schweizer, H. P. (1994). A method for construction of bacterial hosts for lac-based cloning and expression vectors: 
-complementation and regulated expression. Biotechniques 17, 452–456.[Medline]
Schweizer, H. P. & Chuanshuen, R. (2001). A small broad-host-range lacZ operon fusion vector with low background activity. Biotechniques 31, 1258–1262.[Medline]
Shen, D. K., Filopon, D., Kuhn, L., Polack, B. & Toussaint, B. (2006). PsrA is a positive transcriptional regulator of the type III secretion system in Pseudomonas aeruginosa. Infect Immun 74, 1121–1129.
Son, M. S., Matthews, W. J., Jr, Kang, Y., Nguyen, D. T. & Hoang, T. T. (2007). In vivo evidence of Pseudomonas aeruginosa nutrient acquisition and pathogenesis in the lungs of cystic fibrosis patients. Infect Immun 75, 5313–5324.
Son, M. S., Nguyen, D. T., Kang, Y. & Hoang, T. T. (2008). Engineering of FRT–lacZ fusion constructs: induction of the Pseudomonas aeruginosa fadBA1 operon by medium and long chain-length fatty acids. Plasmid 59, 111–118.[CrossRef][Medline]
Stover, C. K., Pham, X. Q., Erwin, A. L., Mizoguchi, S. D., Warrener, P., Hickey, M. J., Brinkman, F. S., Hufnagle, W. O. & other authors (2000). Complete genome sequence of Pseudomonas aeruginosa PAO1, an opportunistic pathogen. Nature 406, 959–964.[CrossRef][Medline]
Tabansky, I. & Nurminsky, D. I. (2003). Mapping of transcription start sites by direct sequencing of SMARTTM RACE products. Biotechniques 34, 482–486.[Medline]
Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A. & Speleman, F. (2002). Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3, research0034[Medline]
Yang, X. Y., Schulz, H., Elzinga, M. & Yang, S. Y. (1991). Nucleotide sequence of the promoter and fadB gene of the fadBA operon and primary structure of the multifunctional fatty acid oxidation protein from Escherichia coli. Biochemistry 30, 6788–6795.[CrossRef][Medline]
Received 2 March 2008;
accepted 26 March 2008.
This article has been cited by other articles:
|
|
 |
|
  Y. Kang, M. H. Norris, A. R. Barrett, B. A. Wilcox, and T. T. Hoang Engineering of Tellurite-Resistant Genetic Tools for Single-Copy Chromosomal Analysis of Burkholderia spp. and Characterization of the Burkholderia thailandensis betBA Operon Appl. Envir. Microbiol., June 15, 2009; 75(12): 4015 - 4027. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Kazakov, D. A. Rodionov, E. Alm, A. P. Arkin, I. Dubchak, and M. S. Gelfand Comparative Genomics of Regulation of Fatty Acid and Branched-Chain Amino Acid Utilization in Proteobacteria J. Bacteriol., January 1, 2009; 191(1): 52 - 64. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
