| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Max F. Perutz Laboratories, Department of Microbiology and Immunobiology, University of Vienna, Dr. Bohrgasse 9, A-1030 Vienna, Austria
2 Innsbruck Biocenter, Division of Genomics and RNomics, Medical University Innsbruck, Fritz-Pregl Str. 3, A-6020 Innsbruck, Austria
3 Bioinformatics Group, Department of Computer Science and Interdisciplinary Center for Bioinformatics, University of Leipzig, Härtelstraße 16-18, D-04107 Leipzig, Germany
Correspondence
Isabella Moll
Isabella.Moll{at}univie.ac.at
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Three supplementary figures showing the genetic organization and predicted secondary structures of PhrX and PhrY, the detection of transcription of the phrX and phrY loci by RT-PCR, and the transcription of loci 72/101 and 102/16, and two supplementary tables showing the DNA oligonucleotides used in this study, and the matching sRNAs/ncRNAs detected by RNomics, Livny et al. (2006)
and González et al. (2008), are available with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
; Sittka et al., 2007; Heurlier et al., 2004; Lenz et al., 2004; Wilderman et al., 2004; Zhang et al., 2003). Some of these sRNAs are involved in quorum sensing (QS) and bacterial pathogenesis, such as RsmY and RsmZ in PAO1 (Heurlier et al., 2004; Sonnleitner et al., 2006) and Qrr1-4 in V. cholerae (Lenz et al., 2004). Another group of sRNAs modulates the stress response by regulating the expression of the alternative sigma factor 
S (Repoila et al., 2003), while others are important for iron metabolism (Massé & Gottesman, 2002; Wilderman et al., 2004), catabolite regulation (Møller et al., 2002) and for remodelling of the outer membrane (Guillier et al., 2006). A large number of sRNAs has been shown to associate with Hfq and to require this protein for post-transcriptional regulation (Gottesman, 2005; Massé et al., 2003). Part of this requirement can be explained by Hfq-mediated stabilization of some of the studied sRNAs, e.g. DsrA, Spot42 RNA, RyhB and RsmY (Møller et al., 2002; Moll et al., 2003a; Sledjeski et al., 2001; Sonnleitner et al., 2006), as well as by the RNA chaperone function of Hfq (Moll et al., 2003b), which facilitates the interaction of sRNAs with their target mRNAs (Kawamoto et al., 2006; Zhang et al., 2002).
Hfq has been identified as a virulence factor in Yersinia enterocolitica, Brucella abortus, Legionella pneumophila, Salmonella typhimurium and V. cholerae (Sittka et al., 2007; Ding et al., 2004; McNealy et al., 2005; Nakao et al., 1995; Robertson & Roop, 1999). A P. aeruginosa PAO1hfq– mutant showed a reduced virulence in larvae of Galleria mellonella and in mice (Sonnleitner et al., 2003). As in E. coli (Sledjeski et al., 2001), Hfq is involved in regulation of rpoS expression in PAO1 (Sonnleitner et al., 2003). However, Hfq also exerts S-independent effects on catalase, pyocyanin and elastase production, and is required for PAO1 to swarm and twitch, which are important traits for colonization and biofilm development (Sonnleitner et al., 2003). A comparative transcriptome analysis of a PAO1 rpoS– and a rpoS–hfq– strain has indicated that Hfq affects approximately 5 % of the PAO1 transcripts (Sonnleitner et al., 2006). Among these transcripts, 72 are known to be regulated by QS, which is a cell density-dependent regulatory mechanism, impacting on virulence gene expression (Van Delden & Iglewski, 1998). Hfq has been suggested to affect QS and QS-controlled genes in at least two ways: by a non-specific positive regulation of the QS repressor QscR and of the pqsH gene, and by RsmY-mediated indirect positive regulation of the QS regulator RhlI (Sonnleitner et al., 2006).
To date, the functions of only a small number of sRNAs are known in PAO1. The sRNAs RsmY (Valverde et al., 2003) and RsmZ (Heurlier et al., 2004) act by sequestration of the regulatory protein RsmA (Pessi et al., 2001). As RsmA acts as a translational repressor of several virulence genes (Pessi et al., 2001), RsmY and RsmZ RNA stimulate indirectly the synthesis of these virulence factors. The sRNAs PrrF1 and PrrF2 (Wilderman et al., 2004) are orthologues of the E. coli sRNA RyhB, and are involved in regulation of iron acquisition and storage functions.
While this work was in progress, 25 small PAO1 RNAs of unknown function have been computationally predicted based on sequence conservation, and their expression has been experimentally verified (Livny et al., 2006; González et al., 2008). In contrast to those studies, here we have used (i) a shotgun-cloning approach (RNomics) (Vogel et al., 2003), and (ii) bioinformatics tools based on the evolutionary conservation of RNA structure rather than on sequence conservation to reveal candidate sRNAs in PAO1. In summary, RNomics revealed three novel sRNA candidates, PhrD, PhrX and PhrY, the transcriptional start sites of which were analysed. Another sRNA, PhrS, which has likewise been identified in recent screens (Livny et al., 2006; González et al., 2008), was studied in more detail. The bioinformatic tool RNAz (Washietl et al., 2005a) identified seven novel sRNA and/or ncRNA loci, of which two were shown to be transcribed. A compilation of matching PAO1 sRNAs/ncRNAs detected by RNomics, RNAz, Livny et al. (2006) and González et al. (2008) is provided.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
. The bacterial cultures were grown at 37 °C in Luria–Bertani (LB) medium (Miller, 1972) supplemented with appropriate antibiotics. Antibiotics were added to final concentrations of 200 µg spectinomycin ml–1, 200 µg streptomycin ml–1, 50 µg gentamicin ml–1, 125 µg tetracycline ml–1, 100 µg ampicillin ml–1 and 100 µg rifampicin ml–1.
|
50 and 500 nt were excised from the gel, eluted and ethanol-precipitated. Ten micrograms of gel-purified RNA was added to 1 nmol purified PAO1 Hfq. After 10 min on ice, 10 µl polyclonal rabbit anti-Hfq IgG was added and incubation was extended for 45 min on ice. Then, 20 µl Dynabeads Protein G (Dynal Biotech), washed once with 0.1 M sodium acetate and twice with VD buffer containing 2 mM magnesium acetate, was added. After 40 min at room temperature the mixture was subjected to a magnetic device (Dynal MPC) to capture Hfq-bound RNAs. To remove unbound components, the beads were washed four times with VD buffer containing 2 mM magnesium acetate and 480 mM NH4Cl. The Hfq-bound RNAs were eluted from the beads by phenol/chloroform extraction, and subsequently precipitated with ethanol. For the cDNA library, the RNAs were polyC-tailed by poly(A) polymerase (Invitrogen), the C-tailed RNAs were ligated to a 5' oligonucleotide linker, and the RNAs were then converted into cDNA by RT-PCR, as described by Hüttenhofer & Vogel (2006), employing primers complementary to the 5' linker and the poly(C) tail, followed by cloning into the pGEM-T vector (Promega). The cDNA clones were sequenced (Lung et al., 2006), and the cDNA sequences were compared with one another using the Lasergene Seqman II program package to identify identical sequences (DNASTAR). Following a BLASTN search against the PAO1 genome (http://www.pseudomonas.com), all sequences corresponding to intergenic regions or to the opposite strand of protein-coding regions were considered to encode potential novel sRNAs/ncRNAs.
Construction of plasmids and RNA preparation.
For in vitro transcription of PhrD and PhrS, the plasmids pUC19-T7phrD and pUC19-T7phrS (Table 1
) were constructed by inserting the PCR products generated with primer pair Z32 and Y32 (phrD; Supplementary Table S1), and primer pair J33 and I33 (phrS; Supplementary Table S1), respectively, into the XbaI–PstI site of plasmid pUC19. The forward primers Z32 and J33 contained a T7
10 promoter sequence.
The plasmids pME9651 and pME9651-1, harbouring the phrS-ORF'–'lacZ and the phrS-ORFAUG
CUG'–'lacZ fusion genes, respectively, were constructed as follows. The promoter region of phrS and the first 47 bp of phrS were amplified (Fig. 1) using primer C2_phrSlacZfw together with primer D2_phrSlacZrev or primer O5_phrSstop (Supplementary Table S1) and genomic DNA of PAO1 as template. The PCR fragments were digested with EcoRI and PstI and ligated into the corresponding sites of pME6013 (Table 1), resulting in pME9651 and pME9651-1, respectively.
|
) were constructed. The respective DNA sequences of phrD and phrS, including their authentic promoters, were amplified by PCR using the primer pairs T33 and Y32 for phrD and H33 and I33 for phrS (Supplementary Table S1), containing a BamHI or PstI restriction site (highlighted in bold type in Supplementary Table S1). The PCR fragments were then cloned into the corresponding sites of plasmid pME4510 (Table 1), resulting in pMEphrD and pMEphrS, respectively.
Northern blotting.
Total RNA of PAO1 and PAO1hfq– was purified using the hot phenol method (Lin-Chao & Bremer, 1986). The abundance of the PhrD and PhrS RNAs was determined by Northern blotting using 15 µg total RNA. The signals were normalized to the hybridization signal obtained for ribosomal 5S rRNA, as described by Sonnleitner et al. (2006). The RNA samples were denatured for 5 min at 65 °C in loading buffer containing 50 % formamide, separated on 8 % polyacrylamide/8 M urea gels, and then transferred to nylon membranes by electroblotting. The RNAs were cross-linked to the membrane by exposure to UV light. The membranes were hybridized with gene-specific 32P-end-labelled oligonucleotides (PhrD, P31; PhrS, O30; 5S rRNA, I26; see Supplementary Table S1), and the hybridization signals were visualized using a PhosphorImager (Molecular Dynamics).
Determination of the half-lives of PhrD and PhrS in PAO1 and PAO1hfq–.
PAO1 and PAOhfq– were grown in LB medium to OD600 2.0, then rifampicin (final concentration 100 µg ml–1) was added to both strains and 10 ml aliquots were withdrawn at 0, 5, 15, 30 and 45 min thereafter for isolation of total RNA (see above). Total RNA (6 µg) of each sample was loaded on an 8 % polyacrylamide/8 M urea gel and blotted to a Hybond-N membrane (Amersham). PhrD and PhrS were visualized with DIG-labelled double-stranded probes. Chromosomal DNA of PAO1 was used as a template for the DIG-labelling procedure, as described by the manufacturer (DIG DNA Labeling Mix, Roche) with the primer pairs PhrDfw and PhrDrev for PhrD, PhrSfw and PhrSrev for PhrS, and 5S-rRNA-1 and 5S-rRNA-2 for 5S rRNA (Supplementary Table S1), which served as a loading control.
Primer extension analysis.
For determination of the 5' ends of PhrD and PhrS (Fig. 2), and PhrX and PhrY (Supplementary Fig. S1), total RNA was purified using the hot phenol method (Lin-Chao & Bremer, 1986) after growth of PAO1 in LB medium to OD600 2.0. Primer extension was performed with AMV reverse transcriptase (Promega) using 2 µg total RNA and the 5' end-labelled oligonucleotides P31 for PhrD, O30 for PhrS, I37 and B32 for PhrX, and K37 and R30 for PhrY (Supplementary Table S1). The plasmids pMEphrD and pMEphrS were used as templates in DNA sequencing reactions.
|
.
RT-PCR.
Total RNA of PAO1 was purified using the RNA/DNA Maxi kit for low-molecular-weight RNAs (Qiagen). Purified RNA (10 µg) was treated with 20 U RNase-free DNase I (Roche). For cDNA synthesis, 20 pmol of primers I37 (PhrX), K37 (PhrY), T34 (RsmY), E47 (RNA 72/101) and G47 (RNA 102/16) (Supplementary Table S1) were annealed to 2 µg RNA for 2 min at 80 °C. Upon cooling on ice, RNase-free AMV Reverse Transcriptase buffer (Promega) and dNTPs (10 mM) in a total volume of 20 µl were added. Then, 30 U AMV reverse transcriptase (Promega) was added, and the reaction was allowed to proceed for 1 h at 42 °C. Aliquots of 0.5 µl of these cDNA reactions were used as templates in 25 µl PCR amplification reactions using primers H37 and I37 for PhrX, J37 and K37 for PhrY, G40 and T34 for RsmY, D47 and E47 for RNA 72/101, and F47 and G47 for RNA 102/16 (1 µM final concentrations; Supplementary Table S1) and GoTaq Green Master Mix (Promega). The PCR fragments generated were analysed on 6 % polyacrylamide gels stained with ethidium bromide. Chromosomal DNA of PAO1 was used as a positive control and RT-PCR performed in the absence of reverse transcriptase was used as a negative control.
Gel mobility shift assays.
The PhrD and PhrS RNAs were transcribed in vitro using T7 polymerase (Fermentas) and the PstI-linearized plasmids pUC-T7phrD and pUC-T7phrS, as described by Sonnleitner et al. (2006). The in vitro-transcribed RNAs were 5' end-labelled with [
-32P]ATP (Amersham) and purified on 6 % polyacrylamide/8 M urea gels. Labelled RNA (0.05 pmol) was incubated with increasing amounts of purified Hfq hexamer protein (Hfq6), as described by Sonnleitner et al. (2006). Non-labelled PhrD, PhrS and RsmY RNA, respectively, were used as specific competitors, and E. coli bulk tRNA as a non-specific competitor. Immediately before loading, the samples were mixed with 4 µl loading dye (25 % glycerol, 0.2 mg xylencyanol l–1 and bromphenol blue), and loaded on a native 4 % polyacrylamide gel. Electrophoresis was performed in Tris-acetate/EDTA buffer at 160 V. The radioactively labelled bands were visualized with a PhosphorImager (Molecular Dynamics).
2D gel analysis.
To detect PhrS-mediated regulatory effects, differences in the proteome profile were analysed, upon plasmid-directed overexpression of PhrS RNA. Total cellular extracts of strain PAO1 harbouring plasmid pME4510 or pMEphrS were extracted and compared by 2D gel electrophoresis. The experiment was performed twice to ensure reproducibility. The strains were grown in LB medium to OD600 1.5, and equal amounts of cells were dissolved in lysis buffer [8 M urea, 4 % (w/v) CHAPS, 40 mM Tris base]. The cells were disrupted by repeated freezing in liquid N2 and thawing at 37 °C, followed by incubation at 37 °C for 1 h to achieve complete lysis. For the first dimension, the Immobiline Dry Strip pH 3–10 (18 cm; Amersham Pharmacia Biotech) was used with the following IEF program: 12 h rehydration at 40 V, 0.5 h at 300 V, 0.5 h at 1000 V, 0.5 h at 2000 V, 0.5 h at 3000 V, 9 h at 7000 V at 18 °C (IPGphor isoelectric focusing system). Resolution in the second dimension was performed on 12 % SDS-polyacryamide gels for 15 min at 15 mA, and then for 4 h at 20 mA. Buffers and conditions were used according to the manufacturer's instructions. The gels were silver-stained as described elsewhere (Shevchenko et al., 1996). Selected protein spots were excised from the gel and the protein identities were assessed by MS.
RNAz screen.
The tool NcDNAlign (Rose et al., 2008) was used to construct multiple sequence alignments based on the known genome sequences of P. aeruginosa PAO1 (NC 002516), P. aeruginosa UCBPP-PA14 (NC 008463), Pseudomonas entomophila (NC 008027), Pseudomonas fluorescens Pf-5 (NC 004129), P. fluorescens PfO-1 (NC 007492), Pseudomonas putida KT2440 (NC 002947), Pseudomonas syringae pv. phaseolicola 1448A (NC 005773), P. syringae (NC 007005), P. syringae pv. tomato str. DC3000 (NC 004578) and P. fluorescens SBW25 (ftp://ftp.sanger.ac.uk/pub/pathogens/pf/PF.dbs). Briefly, the program is based on pairwise BLASTN (Altschul et al., 1990) (e-value <10–3) comparisons, which are combined to multiple alignments. For comparison, MultiZ (Blanchette et al., 2004) was additionally used to produce an alternative set of genome-wide alignments. Both alignments were used as input for the RNAz pipeline (Washietl et al., 2005a, b). The RNAz approach searches for signatures of conservation of RNA secondary structure in a multiple sequence alignment and uses a support vector machine (SVM) to distinguish conserved RNA structure elements from genomic background. In order to assess the reliability of the predictions, the procedure was repeated two times with two different sets of input alignments. In each case the procedure outlined in the current version of the RNAz manual was used (Washietl, 2006). As the SVM of RNAz cannot handle more than six aligned sequences, from alignments exceeding six, the pipeline selects a subset of six sequences that have approximately equal pairwise sequence similarities. Input alignments are cut into windows of length 120 with 40 nt overlap between adjacent slices, which are scored individually. Finally, the predictions are combined to contiguous structured loci at the reference genome, in this case PAO1. As RNAz returns a classification confidence, we report the results for two cutoff values: p >0.5 and a high confidence set with p >0.9. The predicted structured RNAs were compared using BLASTN to the available public databases, the sequences reported by Livny et al. (2006) and González et al. (2008), and the Hfq-binding RNAs listed in Table 2. To identify known PAO1 ncRNAs among the predictions, rnazAnnotate.pl and a minimum overlap of 0.7 were used.
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
; Hüttenhofer & Vogel, 2006; Vogel et al., 2003). In an initial attempt, we shotgun-cloned size-fractionated total RNA (50–300 nt). However, as revealed by subsequent sequencing, this approach yielded mainly RNA fragments originating from rRNA and tRNA (not shown). In a second experiment, a cell extract was used for co-immunoprecipitation of Hfq-binding RNAs with PAO1-Hfq specific antibodies upon removal of ribosomes by centrifugation. Although this modification of the protocol significantly decreased the number of rRNA clones, the majority of the cloned RNAs originated from larger mRNA fragments, in accordance with the capacity of Hfq to bind to mRNAs (Vecerek et al., 2003). Therefore, total RNA obtained after removal of ribosomes was first size-fractionated (50–300 nt), followed by co-immunoprecipitation of Hfq-binding RNAs. The procedure was carried out with cells grown to early stationary phase or with cells exposed to human serum (see Methods) to mimic systemic PAO1 infection (Van Delden, 2004). The protocol used here deviates from the co-immunoprecipitation of sRNA with Hfq antibodies from cell lysates (Zhang et al., 2003), and is biased against sRNAs that eventually associate with ribosomes (Worhunsky et al., 2003).
A total of 400 cDNA sequences were analysed by first grouping identical cDNA clones, followed by their location by bioinformatics on the PAO1 genome. Although the cell extracts were centrifuged at 150 000 g to remove ribosomes, 15 % of the sequenced clones from stationary-phase or serum-treated cells represented rRNA fragments. Of the clones, 40 and 20 % were derived from coding regions of PAO1 (presumably representing mRNA degradation intermediates) and from the opposite strand of protein-coding regions, respectively. Eleven candidate sRNA-encoding genes (Table 2) were predicted to localize to intergenic regions based on the presence of putative rho-independent terminators.
Of the 11 candidate sRNAs, two ncRNAs, RsmY and tmRNA (Table 2), have been previously described (Sonnleitner et al., 2006; Williams & Bartel, 1996). When PAO1 was exposed to human serum, the RsmY RNA gene was repeatedly present in the cDNA library (nine cDNA clones were found), which verified the RsmY interaction with Hfq (Sonnleitner et al., 2006) and could point to its involvement in the regulation of virulence factors (Kay et al., 2006). The tmRNA, also known as ssrA RNA or 10Sa RNA, functions as both tRNA and mRNA (Komine et al., 1994). A fragment of the PAO1 tmRNA (Table 2) co-immunoprecipitated with Hfq. However, as the E. coli tmRNA does not bind to Hfq (Wassarman et al., 2001) and the PAO1 tmRNA has previously been annotated, we did not further study whether full-length PAO1 tmRNA binds to Hfq. In addition, one Hfq-binding RNA fragment was identified as part of the leader RNA of the amidase operon (Table 2; Drew, 1984; Wilson & Drew, 1995), and another Hfq-binding RNA termed PhrW (Table 2) corresponded to P28/2510, which was recently detected by Livny et al. (2006) and González et al. (2008). PhrW/P28/2510 shows significant homology with RnpB, which in E. coli is part of RNase P (Livny et al., 2006). Because of their known or inferred function, the last two RNAs were likewise not further studied.
Next, specific probes were designed for the detection of the remaining seven sRNA candidates by Northern hybridization performed with total RNA from PAO1 cultures grown to early stationary phase and after exposure to human serum. Only two Pseudomonas Hfq-binding RNAs, termed PhrD and PhrS (Fig. 1a, b), isolated from cells exposed to serum and from stationary-phase cells, respectively, were detected (Fig. 2a, b, Table 2). PhrS matched with the recently described, but not further characterized, sRNA P20/1887 (Livny et al., 2006; González et al., 2008). From the determination of their major 5' ends by primer extension (Fig. 2c, d), and from the position of the putative transcriptional terminator, the sizes of these sRNAs were estimated to be 72 nt (PhrD) and 212 nt (PhrS), which was in agreement with the size of the corresponding signals detected on Northern blots (Fig. 2a, b). A putative –10 box (TATGAT) and a putative –35 box (TTGCAT) were found upstream of the predicted transcription start site of phrD. A putative 70 promoter with a –10 box (TAATCT) and a –35 box (TTGTGC) was likewise detected upstream of the transcriptional start site of phrS (Fig. 2e). However, with only 13 bp, the spacing between the –35 and the –10 region would deviate from that of canonical 70 promoters. Nonetheless, as shown below (Fig. 5), this putative promoter was able to direct transcription of the plasmid-borne phrS gene. A BLASTN search of PhrD with all bacterial sequences in the NCBI homepage (http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi) revealed that PhrD is only present in PAO1, whereas homologues of PhrS are found in different isolates of P. aeruginosa (PACS2, C3719, 2192, UCBPP-PA14 and PA7), although not in other Gram-negative bacteria.
|
) and between argA, encoding an N-acetylglutamate synthase, and PA5205, encoding a conserved hypothetical membrane protein (Supplementary Fig. S1b, Table 2), respectively.
PhrS contains an ORF
Inspection of the phrS gene revealed an ORF with the capacity to encode a 37 aa peptide (Fig. 1b). To date, only a few regulatory RNAs have been shown to have protein-coding capacity. For instance, the RNAIII gene of Staph. aureus contains an ORF hld, which encodes delta-haemolysin (Janzon & Arvidson, 1990), and the E. coli SrgS sRNA, besides being a riboregulator, encodes the 43 aa SgrT polypeptide with a function in glucose uptake (Wadler & Vanderpool, 2007). To test whether the internal ORF of phrS (phrS-ORF) is expressed, a translational fusion between the phrS-ORF and the lacZ reporter gene was engineered. In plasmid pME9651 transcription of the phrS-ORF–lacZ gene is driven by the authentic phrS promoter. The control plasmid pME9651-1, which harbours the same phrS-ORF–lacZ gene, but wherein the start codon of phrS-ORF is changed to a CUG (phrS-ORFAUGCUG–lacZ), did not direct synthesis of the fusion protein. In contrast, plasmid pME9651 directed synthesis of the PhrSLacZ protein (Fig. 1b, inset), indicating that the internal ORF of phrS is indeed translated. Thus, PhrS appears to be another candidate for a bifunctional sRNA acting as a riboregulator (see below) and as mRNA. The function of the encoded peptide remains to be elucidated.
Binding of PhrD and PhrS to Hfq
As PhrD and PhrS were abundant RNA species (Fig. 2), we focused in further experiments on these sRNAs. First, we verified that they bind to Hfq by performing band-shift assays with purified PAO1 Hfq protein. Hfq6 was added in increasing molar ratios to 5 nM of the respective 5' end-labelled RNA. When Hfq6 was added to PhrD in a molar ratio of 1 : 1, two band shifts were observed (Fig. 3a, lane 2; B1 and B2). However, the amount of PhrD present in shift B2 was only marginal and did not increase with higher concentrations of Hfq6. The apparent Kd value was 5.5±3.7 nM when 50 % of PhrD RNA was present in complex B1. The competition experiment suggested that PhrD binds specifically to Hfq. Unlabelled PhrD RNA competed with the Hfq–PhrD complex (Fig. 3a, lane7) when added in twofold molar excess over Hfq, whereas the non-specific competitor E. coli tRNA did not (Fig. 3a, lanes 8–9).
|
, lanes 2–5; B1 and B2). The second shift could indicate a 2 : 1 stoichiometry of Hfq6 to PhrS, which is consistent with the observation that the second shift is more prominent at a higher molar excess of Hfq6 (Fig. 3b, lane 5). The apparent Kd values were 13.9±3.6 nM for the first shift and 32.2±6.8 nM for the supershift. Binding of Hfq to PhrS RNA was verified as described above with unlabelled PhrS RNA (Fig. 3b, lanes 6 and 7), with the Hfq-binding PAO1 RsmY RNA (Fig. 3b, lanes 8 and 9) (Sonnleitner et al., 2006), and the non-specific competitor E. coli tRNA (Fig. 3b, lanes 10 and 11).
Abundance of PhrD and PhrS in the presence and absence of Hfq
To test whether Hfq affects the abundance of PhrD and PhrS, their steady-state levels were assessed by Northern blotting using total RNA isolated from PAO1, from PAO1hfq– grown in LB medium to OD600 2 (Fig. 4), or after growth in LB to OD600 1.0, followed by exposure to non-inactivated human serum in PBS buffer (Serum; Fig. 4). When compared with PAO1, the steady-state level of PhrD was reduced 50 % in the hfq– strain under both conditions (Fig. 4a). PhrS was predominantly synthesized in early stationary phase, and was absent upon serum exposure (Fig. 4b). When compared to PAO1, the steady-state level of PhrS was 50 % reduced in the PAO1hfq– mutant (Fig. 4b). These experiments indicated that Hfq affects the abundance of PhrD and PhrS, which could result from direct or indirect effects of Hfq on transcription of the two sRNAs, or from Hfq-mediated protection of these sRNAs from degradation (Moll et al., 2003a; Sorger-Domenigg et al., 2007). To distinguish between these possibilities, we next determined the half-lives of both sRNAs. The half-lives of PhrD and PhrS were comparable in PAO1 and in the PAO1hfq– mutant strain (Fig. 4c). Hence, Hfq appears to affect expression of these sRNAs rather than stabilizing them.
|
, lanes 3 and 4), whereas the abundance of PhrS was decreased at an OD600 of 2.5 (Fig. 5a, lanes 5 and 6).
Proteome analyses upon overproduction of PhrS
Under the premise that riboregulation by an sRNA can result in a decreased or increased synthesis of the protein encoded by the target mRNA, plasmid-directed overexpression of phrS followed by proteomics was used as a means to identify putative PhrS targets. When compared to chromosome-directed background synthesis in strain PAO1(pME4510), the levels of PhrS were 12-fold increased in strain PAO1(pMEphrS) when the cells entered early stationary phase (Fig. 5a, lanes 3 and 4). Therefore, samples for 2D gel analysis were withdrawn at OD600 1.5. The 2D protein pattern of PAO1(pME4510) was compared to that of PAO1(pMEphrS) (Fig. 5b), and three distinct protein spots were selected for identification by MS. The three possible targets of PhrS included the heat-shock chaperonin GroEL, the outer membrane porin OprD and the putative periplasmic binding protein PA5153, all of which were upregulated in the presence of increased levels of PhrS (Fig. 5b). GroEL is among the most highly conserved proteins in nature (Segal & Ron, 1996), and functions together with GroES to maintain protein integrity, which enables cells to survive a variety of environmental stresses (Hendrick & Hartl, 1993). A strong antibody response to GroEL has been found in cystic fibrosis patients with chronic pulmonary infection caused by P. aeruginosa (Ulanova et al., 1997), which suggests a role for PhrS under these conditions. The outer membrane porin OprD has been shown to facilitate diffusion of basic amino acids as well as of small peptides, and also serves as a channel for the β-lactam antibiotic imipenem (Trias & Nikaido 1990), whereby the loss of OprD can lead to imipenem resistance (Lynch et al., 1987). The proteome analysis upon overproduction of PhrS suggests that the sRNA could function as a riboregulator. Studies are under way to test whether PhrS regulates the identified protein genes in a direct or indirect manner.
PAO1 sRNAs predicted by RNAz
In addition to the RNomic approach, we made use of recently developed bioinformatics tools to search for sRNAs/ncRNAs in PAO1. The majority of the computational approaches to detect small ncRNAs in bacterial genomes, such as sRNAPredict (Livny et al., 2006), search for putative genes without a recognizable ORF. In contrast, RNAz (Washietl et al., 2005a) used here extracts information based on the evolutionary conservation of RNA structure from a multiple sequence alignment based on the notion that structured RNAs fold into more stable secondary structures than the genomic background sequence of the same composition. RNAz also evaluates the pattern of substitutions in a multiple sequence alignment of related species. Substitutions that are consistent with preserving a base pair (e.g. GCGU) or that are compensatory (e.g. GCUA) provide direct evidence for the conservation of secondary structure. RNAz uses an SVM approach to decide based on this input information whether a multiple sequence alignment contains a conserved RNA structure. As RNAz relies on both sequence and structure conservation across genomes, it cannot detect unstructured anti-sense regulators or species-specific sRNAs. Therefore, none of the candidate Hfq-binding sRNAs encoded in the opposite strand of protein-coding regions detected by RNomics was predicted by RNAz (not shown).
Using NcDNAlign alignments as input we found 115 structured candidate loci, of which 101 were previously known and 14 are novel sRNA/ncRNA predictions. Based on the less restrictive MultiZ alignments, 221 candidates were predicted, of which 85 correspond to known sRNAs/ncRNAs. Details of the predicted loci are provided at http://www.bioinf.uni-leipzig.de/Publications/SUPPLEMENTS/07-023/. The results of the RNAz screen for both MultiZ- and NcDNAlign-generated alignments are shown in Table 3. The number of RNAz hits that matched with sRNAs /ncRNAs predicted or experimentally tested by Livny et al. (2006) and González et al. (2008) are indicated in Table 3. Information on these matches can be found at http://www.bioinf.uni-leipzig.de/Publications/SUPPLEMENTS/07-023/. MultiZ-based alignments result in significantly more RNAz hits, both at high and low confidence. NcDNAlign, however, identifies more annotatable hits than MultiZ at the cost of only a few novel sRNAs. Only 57 hits are shared between MultiZ- and NcDNAlign-based screens. Of these hits, only seven are novel, and all of them are high-confidence RNAz predictions (see Table 4). In an effort to demonstrate expression of these seven candidate sRNAs/ncRNAs, RT-PCR was exploited. As shown in Supplementary Fig. S3, transcription of the sRNAs/ncRNAs encoded by the loci 72/101 and 102/16 (Table 4) predicted by NcDNAlign and MultiZ, respectively, could be verified.
|
|
) and PhrH (input NcDNAlign, locus 20; input MultiZ, locus 77), which represents tmRNA, are highly conserved among pseudomonads (see http://www.bioinf.uni-leipzig.de/Publications/SUPPLEMENTS/07-023/). PhrS, which corresponds to P20 in Livny et al. (2006) and 1887 in González et al. (2008), was also predicted by RNAz using MultiZ alignments as input (locus 114; see http://www.bioinf.uni-leipzig.de/Publications/SUPPLEMENTS/07-023/). As PhrS emerged from four different screens (Livny et al., 2006; González et al., 2008; Tables 2 and 3) and phrS overexpression resulted in changes in the protein pattern, it appears to be a prime candidate for an sRNA in Pseudomonas. All other sRNA candidates detected by RNomics, which mapped to intergenic regions (Table 2), were not found in the RNAz screen, indicating that they are species-specific. Likewise, none of them was identified by Livny et al. (2006) and González et al. (2008). A compilation of matching sRNAs/ncRNAs predicted and/or identified by RNomics, RNAz, and by Livny et al. (2006) and González et al. (2008) is shown in Supplementary Table S2.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: L. S. Frost
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Argaman, L., Hershberg, R., Vogel, J., Bejerano, G., Wagner, E. G., Margalit, H. & Altuvia, S. (2001). Novel small RNA-encoding genes in the intergenic regions of Escherichia coli. Curr Biol 11, 941–950.[CrossRef][Medline]
Blanchette, M., Kent, W. J., Riemer, C., Elnitski, L., Smit, A. F., Roskin, K. M., Baertsch, R., Rosenbloom, K., Clawson, H. & other authors (2004). Aligning multiple genomic sequences with the threaded blockset aligner. Genome Res 14, 708–715.
Ding, Y., Davis, B. M. & Waldor, M. K. (2004). Hfq is essential for Vibrio cholerae virulence and downregulates sigma expression. Mol Microbiol 53, 345–354.[CrossRef][Medline]
Drew, R. (1984). Complementation analysis of the aliphatic amidase genes of Pseudomonas aeruginosa. J Gen Microbiol 130, 3101–3111.
González, N., Heeb, S., Valverde, C., Kay, E., Reimmann, C., Junier, T. & Haas, D. (2008). Genome-wide search reveals a novel GacA-regulated small RNA in Pseudomonas species. BMC Genomics 9, 167[CrossRef][Medline]
Gottesman, S. (2005). Micros for microbes: non-coding regulatory RNAs in bacteria. Trends Genet 21, 399–404.[CrossRef][Medline]
Guillier, M., Gottesman, S. & Storz, G. (2006). Modulating the outer membrane with small RNAs. Genes Dev 20, 2338–2348.
Hendrick, J. P. & Hartl, F. U. (1993). Molecular chaperone functions of heat-shock proteins. Annu Rev Biochem 62, 349–384.[CrossRef][Medline]
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.
Holloway, B. W., Krishnapillai, V. & Morgan, A. F. (1979). Chromosomal genetics of Pseudomonas. Microbiol Rev 43, 73–102.
Hüttenhofer, A. & Vogel, J. (2006). Experimental approaches to identify non-coding RNAs. Nucleic Acids Res 34, 635–646.
Hüttenhofer, A., Cavaille, J. & Bachellerie, J. P. (2004). Experimental RNomics: a global approach to identifying small nuclear RNAs and their targets in different model organisms. Methods Mol Biol 265, 409–428.[Medline]
Janzon, L. & Arvidson, S. (1990). The role of the 
-lysin gene (hld) in the regulation of virulence genes by the accessory gene regulator (agr) in Staphylococcus aureus. EMBO J 9, 1391–1399.[Medline]
Kawamoto, H., Koide, Y., Morita, T. & Aiba, H. (2006). Base-pairing requirement for RNA silencing by a bacterial small RNA and acceleration of duplex formation by Hfq. Mol Microbiol 61, 1013–1022.[CrossRef][Medline]
Kay, E., Humair, B., Denervaud, V., Riedel, K., Spahr, S., Eberl, L., Valverde, C. & Haas, D. (2006). Two GacA-dependent small RNAs modulate the quorum-sensing response in Pseudomonas aeruginosa. J Bacteriol 188, 6026–6033.
Komine, Y., Kitabatake, M., Yokogawa, T., Nishikawa, K. & Inokuchi, H. (1994). A tRNA-like structure is present in 10Sa RNA, a small stable RNA from Escherichia coli. Proc Natl Acad Sci U S A 91, 9223–9227.
Lenz, D. H., Mok, K. C., Lilley, B. N., Kulkarni, R. V., Wingreen, N. S. & Bassler, B. L. (2004). The small RNA chaperone Hfq and multiple small RNAs control quorum sensing in Vibrio harveyi and Vibrio cholerae. Cell 118, 69–82.[CrossRef][Medline]
Lin-Chao, S. & Bremer, H. (1986). Effect of the bacterial growth rate on replication control of plasmid pBR322 in Escherichia coli. Mol Gen Genet 203, 143–149.[CrossRef][Medline]
Livny, J., Brencic, A., Lory, S. & Waldor, M. K. (2006). Identification of 17 Pseudomonas aeruginosa sRNAs and prediction of sRNA-encoding genes in 10 diverse pathogens using the bioinformatic tool sRNAPredict2. Nucleic Acids Res 34, 3484–3493.
Lung, B., Zemann, A., Madej, M. J., Schuelke, M., Techritz, S., Ruf, S., Bock, R. & Hüttenhofer, A. (2006). Identification of small non-coding RNAs from mitochondria and chloroplasts. Nucleic Acids Res 34, 3842–3852.
Lynch, M. J., Drusano, G. L. & Mobley, H. L. (1987). Emergence of resistance to imipenem in Pseudomonas aeruginosa. Antimicrob Agents Chemother 31, 1892–1896.
Massé, E. & Gottesman, S. (2002). A small RNA regulates the expression of genes involved in iron metabolism in Escherichia coli. Proc Natl Acad Sci U S A 99, 4620–4625.
Massé, E., Majdalani, N. & Gottesman, S. (2003). Regulatory roles for small RNAs in bacteria. Curr Opin Microbiol 6, 120–124.[CrossRef][Medline]
Mathews, D. H., Sabina, J., Zuker, M. & Turner, D. H. (1999). Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J Mol Biol 288, 911–940.[CrossRef][Medline]
McNealy, T. L., Forsbach-Birk, V., Shi, C. & Marre, R. (2005). The Hfq homolog in Legionella pneumophila demonstrates regulation by LetA and RpoS and interacts with the global regulator CsrA. J Bacteriol 187, 1527–1532.
Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Moll, I., Afonyushkin, T., Vytvytska, O., Kaberdin, V. R. & Bläsi, U. (2003a). Coincident Hfq binding and RNase E cleavage sites on mRNA and small regulatory RNAs. RNA 9, 1308–1314.
Moll, I., Leitsch, D., Steinhauser, T. & Bläsi, U. (2003b). RNA chaperone activity of the Sm-like Hfq protein. EMBO Rep 4, 284–289.[CrossRef][Medline]
Møller, T., Franch, T., Udesen, C., Gerdes, K. & Valentin-Hansen, P. (2002). Spot 42 RNA mediates discoordinate expression of the E. coli galactose operon. Genes Dev 16, 1696–1706.
Nakao, H., Watanabe, H., Nakayama, S. & Takeda, T. (1995). yst gene expression in Yersinia enterocolitica is positively regulated by a chromosomal region that is highly homologous to Escherichia coli host factor 1 gene (hfq). Mol Microbiol 18, 859–865.[CrossRef][Medline]
Pessi, G., Williams, F., Hindle, Z., Heurlier, K., Holden, M. T., Camara, M., Haas, D. & Williams, P. (2001). The global posttranscriptional regulator RsmA modulates production of virulence determinants and N-acylhomoserine lactones in Pseudomonas aeruginosa. J Bacteriol 183, 6676–6683.
Repoila, F., Majdalani, N. & Gottesman, S. (2003). Small non-coding RNAs, co-ordinators of adaptation processes in Escherichia coli: the RpoS paradigm. Mol Microbiol 48, 855–861.[CrossRef][Medline]
Rist, M. & Kertesz, M. A. (1998). Construction of improved plasmid vectors for promoter characterization in Pseudomonas aeruginosa and other Gram-negative bacteria. FEMS Microbiol Lett 169, 179–183.[CrossRef][Medline]
Robertson, G. T. & Roop, R. M., Jr (1999). The Brucella abortus host factor I (HF-I) protein contributes to stress resistance during stationary phase and is a major determinant of virulence in mice. Mol Microbiol 34, 690–700.[CrossRef][Medline]
Rose, D., Hertel, J., Reiche, K., Stadler, P. F. & Hackermüller, J. (2008). NcDNAlign: plausible multiple alignments of non-protein-coding genomic sequences. Genomics 92, 65–74.[CrossRef][Medline]
Schnider-Keel, U., Seematter, A., Maurhofer, M., Blumer, C., Duffy, B., Gigot-Bonnefoy, C., Reimmann, C., No, R., Défago, G. & other authors (2000). Autoinduction of 2,4-diacetylphloroglucinol biosynthesis in the biocontrol agent Pseudomonas fluorescens CHA0 and repression by the bacterial metabolites salicylate and pyoluteorin. J Bacteriol 182, 1215–1225.
Segal, R. & Ron, E. Z. (1996). Regulation and organization of the groE and dnaK operons in Eubacteria. FEMS Microbiol Lett 138, 1–10.[CrossRef][Medline]
Shevchenko, A., Jensen, O. N., Podtelejnikov, A. V., Sagliocco, F., Wilm, M., Vorm, O., Mortensen, P., Boucherie, H. & Mann, M. (1996). Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two dimensional gels. Proc Natl Acad Sci U S A 93, 14440–14445.
Sittka, A., Pfeiffer, V., Tedin, K. & Vogel, J. (2007). The RNA chaperone Hfq is essential for the virulence of Salmonella typhimurium. Mol Microbiol 63, 193–217.[CrossRef][Medline]
Sledjeski, D. D., Whitman, C. & Zhang, A. (2001). Hfq is necessary for regulation by the untranslated RNA DsrA. J Bacteriol 183, 1997–2005.
Sonnleitner, E., Hagens, S., Rosenau, F., Wilhelm, S., Habel, A., Jager, K. E. & Bläsi, U. (2003). Reduced virulence of a hfq mutant of Pseudomonas aeruginosa O1. Microb Pathog 35, 217–228.[CrossRef][Medline]
Sonnleitner, E., Schuster, M., Sorger-Domenigg, T., Greenberg, E. P. & Bläsi, U. (2006). Hfq-dependent alterations of the transcriptome profile and effects on quorum sensing in Pseudomonas aeruginosa. Mol Microbiol 59, 1542–1558.[CrossRef][Medline]
Sorger-Domenigg, T., Sonnleitner, E., Kaberdin, V. R. & Bläsi, U. (2007). Distinct and overlapping binding sites of Pseudomonas aeruginosa Hfq and RsmA proteins on the non-coding RNA RsmY. Biochem Biophys Res Commun 352, 769–773.[CrossRef][Medline]
Trias, J. & Nikaido, H. (1990). Protein D2 channel of the Pseudomonas aeruginosa outer membrane has a binding site for basic amino acids and peptides. J Biol Chem 265, 15680–15684.
Ulanova, M., Petersen, T. D., Ciofu, O., Jensen, P., Hahn-Zoric, M., Hanson, L. A. & Hoiby, N. (1997). The clonal antibody response to Pseudomonas aeruginosa heat shock protein is highly diverse in cystic fibrosis patients. APMIS 105, 449–456.[Medline]
Valverde, C., Heeb, S., Keel, C. & Haas, D. (2003). RsmY, a small regulatory RNA, is required in concert with RsmZ for GacA-dependent expression of biocontrol traits in Pseudomonas fluorescens CHA0. Mol Microbiol 50, 1361–1379.[CrossRef][Medline]
Van Delden, C. (2004). Virulence factors in Pseudomonas aeruginosa. In Pseudomonas, vol. 2, pp. 3–45. Edited by J. L. Ramos. New York: Kluwer Academic/Plenum Publishers.
Van Delden, C. & Iglewski, B. H. (1998). Cell-to-cell signaling and Pseudomonas aeruginosa infections. Emerg Infect Dis 4, 551–560.[Medline]
Vecerek, B., Moll, I., Afonyushkin, T., Kaberdin, V. & Bläsi, U. (2003). Interaction of the RNA chaperone Hfq with mRNAs: direct and indirect roles of Hfq in iron metabolism of Escherichia coli. Mol Microbiol 50, 897–909.[CrossRef][Medline]
Vogel, J., Bartels, V., Tang, T. H., Churakov, G., Slagter-Jager, J. G., Hüttenhofer, A. & Wagner, E. G. (2003). RNomics in Escherichia coli detects new sRNA species and indicates parallel transcriptional output in bacteria. Nucleic Acids Res 31, 6435–6443.
Wadler, C. S. & Vanderpool, C. K. (2007). A dual function for a bacterial small RNA: SgrS performs base pairing-dependent regulation and encodes a functional polypeptide. Proc Natl Acad Sci U S A 104, 20454–20459.
Washietl, S. (2006). RNAz 1.0. Department for Theoretical Chemistry, University Vienna, 3, http://www.tbi.univie.ac.at/wash/RNAz.
Washietl, S., Hofacker, I. L. & Stadler, P. F. (2005a). Fast and reliable prediction of noncoding RNAs. Proc Natl Acad Sci U S A 102, 2454–2459.
Washietl, S., Hofacker, I. L., Lukasser, M., Hüttenhofer, A. & Stadler, P. F. (2005b). Mapping of conserved RNA secondary structures predicts thousands of functional noncoding RNAs in the human genome. Nat Biotechnol 23, 1383–1390.[CrossRef][Medline]
Wassarman, K. M., Repoila, F., Rosenow, C., Storz, G. & Gottesman, S. (2001). Identification of novel small RNAs using comparative genomics and microarrays. Genes Dev 15, 1637–1651.
Wilderman, P. J., Sowa, N. A., FitzGerald, D. J., FitzGerald, P. C., Gottesman, S., Ochsner, U. A. & Vasil, M. L. (2004). Identification of tandem duplicate regulatory small RNAs in Pseudomonas aeruginosa involved in iron homeostasis. Proc Natl Acad Sci U S A 101, 9792–9797.
Williams, K. P. & Bartel, D. P. (1996). Phylogenetic analysis of tmRNA secondary structure. RNA 2, 1306–1310.[Abstract]
Wilson, S. A. & Drew, R. E. (1995). Transcriptional analysis of the amidase operon from Pseudomonas aeruginosa. J Bacteriol 177, 3052–3057.
Worhunsky, D. J., Godek, K., Litsch, S. & Schlax, P. J. (2003). Interactions of the non-coding RNA DsrA and RpoS mRNA with the 30 S ribosomal subunit. J Biol Chem 278, 15815–15824.
Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119.[CrossRef][Medline]
Zhang, A., Wassarman, K. M., Ortega, J., Steven, A. C. & Storz, G. (2002). The Sm-like Hfq protein increases OxyS RNA interaction with target mRNAs. Mol Cell 9, 11–22.[CrossRef][Medline]
Zhang, A., Wassarman, K. M., Rosenow, C., Tjaden, B. C., Storz, G. & Gottesman, S. (2003). Global analysis of small RNA and mRNA targets of Hfq. Mol Microbiol 50, 1111–1124.[CrossRef][Medline]
Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31, 3406–3415.
Received 16 April 2008;
revised 27 June 2008;
accepted 4 July 2008.
This article has been cited by other articles:
|
|
 |
|
  A. Marchais, M. Naville, C. Bohn, P. Bouloc, and D. Gautheret Single-pass classification of all noncoding sequences in a bacterial genome using phylogenetic profiles Genome Res., June 1, 2009; 19(6): 1084 - 1092. [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 | |
) and PAO1(pME9651-1) (phrS-ORFAUG
) were determined as described by Miller (1972)
