Ribosomal Protein S1 Specifically Binds to the 5′ Untranslated Region of the Pseudomonas aeruginosa Stationary-Phase Sigma Factor rpoS mRNA in the Logarithmic Phase of Growth

Abstract

The rpoS gene encodes the stationary-phase sigma factor (RpoS or σ^s), which was identified in several gram-negative bacteria as a central regulator controlling the expression of genes involved in cell survival in response to cessation of growth (stationary phase) and providing cross-protection against various stresses. In Pseudomonas aeruginosa, the levels of σ^s increase dramatically at the onset of the stationary phase and are regulated at the transcriptional and posttranscriptional levels. The P. aeruginosa rpoS gene is transcribed as a monocistronic rpoS mRNA transcript comprised of an unusually long 373-bp 5′ untranslated region (5′ UTR). In this study, the 5′ UTR and total protein extracts from P. aeruginosa logarithmic and stationary phases of growth were used in order to investigate the protein-RNA interactions that may modulate the translational process. It was observed that a 69-kDa protein, which corresponded to ribosomal protein S1, preferentially binds the 5′ UTR of the rpoS mRNA in the logarithmic phase and not in the stationary phase. This is the first report of a protein-rpoS mRNA 5′ UTR interaction in P. aeruginosa, and the possible involvement of protein S1 in translation regulation of rpoS is discussed.

Bacteria often encounter constantly changing nutrient availability and exposure to various forms of physical stress, including osmotic stress, oxidative stress, and temperature shock. These environmental conditions lead to a reduction in or cessation of growth, also known as the stationary phase, resulting in a major switch in gene expression that allows the cells to cope. The stationary-phase sigma factor σ^s (also called RpoS) has been identified in Escherichia coli as a central regulator during the stationary phase of growth; it is involved in regulating more than 100 genes involved in cell survival, cross-protection against various stresses, and virulence (14, 20). This set of genes is called the σ^s regulon and has been studied mainly in E. coli; recently, σ^s has also been described in the pseudomonads and has been shown to be a general stress regulator in Pseudomonas putida and Pseudomonas aeruginosa (25, 33). σ^s in Pseudomonas is also involved in regulation of the extracellular virulence products alginate and exotoxin A in the opportunistic pathogen P. aeruginosa (33); in Pseudomonas fluorescens, σ^s has been implicated in the production of antibiotics and in biological control by suppression of soilborne plant pathogens (30).

σ^s levels are carefully monitored within the bacterial cell and increase considerably at the onset of the stationary phase, when they reach 30% the level of the housekeeping σ⁷⁰ subunit, thus improving competition with other available σ subunits for core RNA polymerase (15). The mechanisms controlling σ^s levels have been and are currently extensively studied in E. coli. These studies have determined that this is one of the most complex regulatory mechanisms in bacteria; regulation takes place at the level of transcription, at the level of translation, and at the level of protein stability, all coordinated by the response to several stress signals (13).

The rpoS transcript originates in the nlpD gene located upstream, which produces a monocistronic mRNA with a long 5′ untranslated region (5′ UTR) consisting of 567 bp; the rpoS mRNA secondary structure, together with trans-acting factors and small regulatory RNAs, controls translation initiation under different environmental conditions (13). At a low temperature, an 87-nucleotide small regulatory RNA, called DsrA, is induced, which is partially complementary to a segment of the 5′ UTR of the rpoS mRNA disrupting intramolecular base pairing and promoting translational initiation (21). dsrA mutants can be complemented by the presence of multiple copies of another regulatory RNA consisting of 106 nucleotides called RprA through a mechanism which is not yet understood (22). Initiation of translation of rpoS is also positively controlled by an RNA binding protein called Hfq (24) and by the HU histone-like protein (1). By contrast, the histone-like protein H-NS represses rpoS translation (19). OxyS is a 109-nucleotide regulatory RNA that inhibits rpoS translation by binding Hfq in competition with the rpoS 5′ UTR (38). It therefore appears that many players involved in the control of rpoS translation initiation in E. coli have been determined, and future work should establish how these numerous trans-acting factors integrate into a working model.

Regulation of σ^s levels has also been addressed recently in the fluorescent pseudomonads (P. aeruginosa, P. fluorescens, and P. putida), and the results highlight the finding that this regulation is significantly different from the regulation in E. coli, with transcriptional regulation playing a major role (11, 17, 18, 35). There are indications that in Pseudomonas translational regulation must also be present, but no investigations have been carried out. Similar to the situation in E. coli, in P. aeruginosa and P. putida the rpoS transcript originates within the nlpD gene located upstream, producing a monocistronic mRNA with a long 5′ UTR consisting of 373 bp (17); this 5′ UTR could be involved in the regulation of rpoS translation by employing a trans-acting factor(s). In this study we determined that σ^s levels in P. aeruginosa increase significantly at the onset of the stationary phase, indicating that there is a translational level of control of σ^s also in Pseudomonas. In order to initiate investigations in this direction, the 5′ UTR of rpoS mRNA of P. aeruginosa was used to investigate the interactions of this mRNA with total protein cell extracts from the logarithmic and stationary phases of growth. These studies established that one protein, the ribosomal protein S1, binds preferentially to the 5′ UTR of P. aeruginosa rpoS mRNA in the logarithmic phase of growth and does not bind in the stationary phase. Possible involvement of ribosomal protein S1 in negative translational control of rpoS is discussed below.

MATERIALS AND METHODS

Bacterial strains, plasmids, and media.

The strains used in this study include E. coli DH5α (12) and P. aeruginosa PAO1 (B. Holloway collection). E. coli was grown in Luria-Bertani (LB) medium (23) at 37°C, whereas Pseudomonas was grown in LB medium or M9 minimal medium at 30°C (29). The following antibiotic concentrations were used: tetracycline, 10 μg/ml for E. coli and 300 μg/ml for P. aeruginosa; kanamycin, 100 μg/ml; nalidixic acid, 25 μg/ml; ampicillin, 100 μg/ml; gentamicin, 10 μg/ml for E. coli; and chloramphenicol, 25 μg/ml for E. coli and 500 μg/ml for P. aeruginosa. The 369-bp UTR (i.e., the region starting at the +1 initiation-of-transcription site until the position corresponding to the first codon of the σ^s protein) of rpoS mRNA from P. aeruginosa was amplified by PCR by using the following two synthetic oligonucleotides as primers: 5′UTR PAO1 carrying a SacI recognition site (boldface type) (5′-CGAGCTCGGCTGCGTCTGGTGGGAC-3′) and 3′UTR PAO1 carrying a XbaI recognition site (boldface type) (5′-GCTCTAGAGCCATGTCGTTATCCCTTGCATG-3′). The resulting 386-bp PCR product was cloned directly into the pMOS-blue vector (Amersham Pharmacia Biotech), yielding pMOS-UTR, and was sequenced to confirm its identity. The 5′ UTR of rpoS mRNA was then further cloned under T7 promoter control into pBluescript KS as a SacI-XbaI fragment, yielding pBS-UTR. Shorter 5′UTR rpoS fragments (UTRI, UTRII, and UTRIII) were cloned by using four additional primers, 3′UTRI (5′-GCTCTAGATGTGGAGCCCATCTCGGCAA-3′), 5′UTRII (5′-CGAGCTCGGAACCGATCGGGTGAAGCTGC-3′), 5′UTRIII (5′-CGAGCTCCGAGACCTACGTGAGTGCC-3′), and 3′UTRIII (5′-GCTCTAGACGAACTCCCGGTCAGCGACG-3′). The PCR-amplified DNA fragments were three overlapping fragments of the 369-bp UTR, designated UTRI, UTRII and UTRIII, that were 196, 212, and 214 bp long, respectively (see Fig. 6); these fragments were cloned under T7 promoter control into pBluescript KS as SacI-XbaI fragments, yielding pBS-UTRI, pBS-UTRII, and pBS-UTRIII. The P. aeruginosa 5′ UTR of the GDP-mannose 6-dehydrogenase algD mRNA and the 5′ UTR of lasI, encoding the autoinducer synthesis protein, were used in control experiments. The algD 5′ UTR RNA was prepared for in vitro transcription by PCR amplification by using the 5′ primer that contained the sequence of the T7 promoter (boldface type) and 16 bp from the +1 position of the algD mRNA (5′T7algD [5′-TACGTAATACGACTCACTATAGCGATGCCTATCGATAG-3′) and primer 3′algD (5′-TCGCATTCACCTCGATTG-3′), which contained 18 bp from the translational start codon, resulting in a PCR product that was 388 bp long. This amplification product was used directly for in vitro transcription. The same procedure was used for the synthesis of the 5′ UTR of the rhlI mRNA, and the primers contained 22 (boldface type) and 21 bp of the original sequence (5′T7rhlI [5′-TACGTAATACGACTCACTATAGCCTCATGTGTGTGCTGGTATGTC-3′] and 3′rhlI [5′-TGACCAAGTCCCCGTGTCGTG-3′]). The amplified product was 122 bp long. The 5′UTR region of the lasI gene with addition of the T7 promoter sequence (for a total of 48 bp) and the psrA gene UTR (60 bp long) were synthesized as sense and antisense strands by using primers lasI-S (5′-TACGTAATACGACTCACTATAGAGCTTCCTATTTGGAGGAAGTGAAGA-3′), lasI-AS (5′-TCTTCACTTCCTCCAAATAGGAAGCTCTATAGTGAGTCGTATTACGTA-3′), psrA-S (5′-TACGTAATACGACTCACTATAGGTATGTTTCAAACAAGTGTTTGTCAGGCGGAGAAACCA-3′), and psrA-AS (5′-TGGTTTCTCCGCCTGACAAACACTTGTTTGAAACATACCTATAGTGAGTCGTATTACGT A-3′), heat denatured, and annealed by slow cooling at room temperature. These two DNA fragments were used directly for in vitro transcription.

FIG. 6.

Binding specificity of ribosomal protein S1 to the UTR of rpoS mRNA and further localization of the binding site. UV cross-linking experiments were performed as described in Materials and Methods by using three ³²P-labeled overlapping regions of the UTR (A) and total protein extract from logarithmic-phase cultures of P. aeruginosa PAO1. S SacI; X, XbaI. The signal that corresponded to the labeled ribosomal protein S1 was detected in the presence of UTRI in the reaction mixture (B). An additional UV cross-linking assay performed with ³²P-labeled antisense UTR RNA that was transcribed from plasmid pBS-UTR linearized with SacI (see Materials and Methods) and T3 RNA polymerase. The UV cross-linking assay yielded binding of some low-molecular-weight proteins to the target RNA, but not ribosomal protein S1 (C). (D) All four RNAs were transcribed in vitro and cross-linked to the logarithmic-phase total protein extract. Cross-linked material was separated by SDS—12% PAGE, and the gel was exposed to BioMax film overnight. There were no radioactively labeled proteins that were the molecular mass that corresponded to the molecular mass of ribosomal protein S1 when the lasI, rhlI, and psrA 5′ UTRs were used. A faint band near the molecular mass of protein S1 (69 kDa) was observed when the 5′ UTR of algD was used; it is not known whether this protein is S1. See text for more details. kD, kilodaltons.

Recombinant DNA techniques.

Digestion with restriction enzymes, agarose gel electrophoresis, purification of DNA fragments, ligation with T4 DNA ligase, the S1 reaction, Southern hybridization, DNA sequencing, transformation of E. coli, and sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) analysis were performed as described by Sambrook et al. (29). Analytical amounts of plasmids were isolated as described by Birnboim (4), whereas preparative amounts were purified with QIAGEN columns (QIAGEN, Hilden, Germany). Total DNA from Pseudomonas was isolated by Sarkosyl-pronase lysis as described by Better et al. (3). Triparental matings from E. coli to Pseudomonas were performed with the helper strain E. coli(pRK2013) (10).

Preparation of total cellular extracts of P. aeruginosa PAO1 and protein analysis.

P. aeruginosa PAO1 was grown in liquid LB medium at 37°C; cultures were collected for isolation of total proteins in the logarithmic growth phase at an optical density at 650 nm (OD₆₅₀) of 0.6, whereas for proteins expressed in the stationary growth phase an OD₆₅₀ of 2.0 was used. Each cellular pellet was resuspended in a solution containing phosphate-buffered saline (29) and lysozyme at a final concentration of 0.4 mg/ml and was sonicated on ice by using five 10-s bursts at the maximum power. After centrifugation, the supernatant fraction contained the total cellular proteins. When the amount of S1 ribosomal protein (free and ribosome bound) was investigated (see Fig. 6), an ultracentrifugation step (110,000 × g at 4°C for 3.5 h) was performed.

Total cell proteins for Western analysis were isolated by resuspending the bacterial pellet in SDS sample buffer (0.125 M Tris-HCl [pH 6.8], 4% SDS, 20% glycerol, 10% 2-mercaptoethanol), heated for 15 min at 95°C, and centrifuged at the maximum speed. The proteins were transferred onto a nitrocellulose membrane (Hybond-C extra; Amersham Pharmacia Biotech) by using a tank system according to the manufacturer's instructions. The membrane was subjected to Western blot analysis by using polyclonal antibodies against σ^s raised in rabbits (18) or anti-S1 antibodies (5, 9) and alkaline phosphatase-conjugated anti-rabbit immunoglobulin (IgG) (Sigma) or horseradish peroxidase-conjugated anti-rabbit IgG (DAKO). Membranes were either stained with the alkaline phosphatase substrates 5-bromo-4-chloro-3-indolylphosphate (BCIP) and nitroblue tetrazolium (Promega) or incubated with ECL detection reagents (Amersham) and exposed to Hyperfilm (Amersham) for between 30 s and 3 min.

Expression and purification of ribosomal protein S1 of P. aeruginosa PAO1.

The rpsA gene encoding ribosomal protein S1 was amplified from chromosomal DNA of P. aeruginosa PAO1 by using primers 5′Start-BamHI-S1 (5′-CGGGATCCAGCGAAAGCTTCGCAGAACTC-3′) and 3′End-Eco-S1 (5′-GGAATTCTTAGCCCTGATTCTCCATCTGA-3′). The 1,679-bp PCR product was digested with BamHI and EcoRI and cloned at the corresponding sites into the pGEM-T easy vector (Promega), yielding pGEM-S1, by following the instructions of the supplier, and the DNA sequence was determined in order to confirm that no PCR errors took place. The rpsA gene was then removed as a BamHI-SphI fragment from pGEM-S1 and cloned into the corresponding sites in His₆-tagged expression vector pQE30 (QIAGEN), yielding pQE-S1. Expression and purification of His₆-tagged S1 were carried out in E. coli M15(pREP-4) according to the instructions of the supplier (QIAGEN).

UV cross-linking assay.

The 5′ UTR rpoS mRNA was prepared as follows. Plasmid pBS-UTR was linearized by digestion with BamHI, and transcription was performed by using T7 RNA polymerase (Stratagene) in the presence of [α-³²P]dUTP according to standard protocols; the product was purified on a Nick column (Amersham Pharmacia Biotech). The same procedure was used for in vitro transcription of UTRI, UTRII, and UTRIII RNAs. [α-³²P]dUTP-labeled RNAs (4 to 6 fmol) were incubated at 30°C for 15 min with 50 μg of P. aeruginosa total cellular proteins in a solution (final volume, 20 μl) containing 25 mM Tris-HCl (pH 7.5), 5 mM MgCl₂, 150 mM KCl, and5 μg of heparin per ml (27). Following incubation, the samples were exposed to UV light (254 nm, 100 W) at a distance of 5 cm for 10 min on ice. The RNA was then removed with 1 U of RNase A at 37°C for 30 min. ³²P-labeled proteins were separated by SDS—10% PAGE together with molecular weight standards. The gel was dried and exposed to BioMax film at −80°C. The same procedure was used for cross-linking of the purified His₆-tagged ribosomal protein S1 with [α-³²P]dUTP-labeled 5′ UTR rpoS mRNA.

In vitro transcription of UTR-AS (antisense) RNA was performed by using linearized plasmid pBS-UTR cut with SacI, followed by transcription with T3 RNA polymerase. The resulting [α-³²P]dUTP-labeled UTR-AS RNA was 483 bp long.

Affinity purification of the protein(s) that binds the 5′ UTR of the rpoS mRNA.

Cold 5′ UTR RNA (8 μg) was placed in a reaction mixture (400 μl) containing 0.1 M sodium acetate (NaOAc) (pH 5.0) with 5 mM sodium m-periodate (Sigma) and incubated for 1 h in the dark at room temperature. The RNA was then ethanol precipitated and resuspended in 100 μl of 0.1 M NaOAc (pH 5.0). Then 400 μl of adipic acid dehydrazide-agarose beads (50% slurry; Sigma) was washed four times in 10 ml of 0.1 M NaOAc (pH 5.0) and pelleted by centrifugation (3,000 rpm, 5 min) after each wash. At the end of the final washing step, the beads were mixed with 900 μl of 0.1 M NaOAc (pH 5.0) and with a periodate-treated RNA sample and rotated for 12 h in the dark at 4°C. The agarose beads were then washed twice with 2 M NaCl and twice with RNA wash buffer (52 mM HEPES-KOH [pH 7.5], 10 mM MgCl₂, 8 mM magnesium acetate, 5.2 mM dithiothreitol, 38% [vol/vol] glycerol). The beads were incubated with 0.3 mg of either logarithmic- or stationary-phase cellular extract for 30 min at room temperature in 500 μl (final volume) of the binding buffer (RNA wash buffer with 7.5 mM ATP, 10 mM GTP, and 5 mg of heparin per ml), pelleted by centrifugation (1,000 rpm, 3 min), and washed three times with 10 ml of binding buffer without dithiothreitol and heparin. The protein bound to the beads was boiled in 200 μl of SDS sample buffer for 5 min at 90°C and separated on an SDS—10% PAGE gel. An internal sequence analysis of the Coomassie blue-stained bands excised from the gel was performed by using an electrospray ionization mass spectrometer (LCQ DECA XP; ThermoFinnigam). The bands were digested with trypsin, and the resulting peptides were extracted with water and 60% acetonitrile-1% trifluoroacetic acid. The fragments were then analyzed by mass spectroscopy, and the proteins were identified by analysis of the peptides and by using the annotated P. aeruginosa genome (www.pseudomonas.com).

Immunoprecipitation of ribosomal protein S1 bound to the 5′ UTR rpoS mRNA.

UV cross-linking of the [α-³²P]dUTP-labeled RNA and cellular extracts from the bacteria in the logarithmic growth phase was performed as described above. Following UV cross-linking, two samples were incubated for 3 h at 4°C in 150 μl of IPP buffer (20 mM Tris-HCl [pH 8.0], 300 mM NaCl, 1 mM EDTA, 0.25% [vol/vol] NP-40) with 3 μl of anti-S1 antibodies (5, 9) and rabbit serum. Labeled protein-IgG complexes were affinity purified by using protein A/G PLUS-agarose (Santa Cruz Biotechnology). Immunoprecipitated labeled protein was incubated with 40 μl of the AG beads for 3 h at 4°C. The beads were washed three times with IPP buffer and mixed with SDS sample buffer. The mixture was then boiled for 5 min at 90°C, and the proteins were separated on an SDS—10% PAGE gel. The gel was dried and exposed to BioMAx film at −80°C. This experiment was done in duplicate.

RESULTS

σ^s is under translational regulation in P. aeruginosa.

Previous studies on the regulation of rpoS in Pseudomonas have concentrated at the level of transcription and have highlighted the important role played by a TetR family regulator called PsrA. A P. aeruginosa psrA knockout mutant displayed a 90% decrease in rpoS promoter activity and exhibited hardly any transcriptional induction at the onset of the stationary phase (17, 18). It was of interest to determine whether further regulatory control was present at the level of translation; thus, σ^s protein levels, as measured by Western analysis with anti-σ^s antibodies, were determined in P. aeruginosa and the psrA knockout derivative PAO1psrA::Tn5 at different growth phases (Fig. 1). The σ^s levels were considerably stimulated at the onset of the stationary phase in wild-type P. aeruginosa PAO1 and also in the psrA::Tn5 mutant, which, however, had less RpoS protein.

FIG. 1.

σ^s levels in P. aeruginosa PAO1 (A) and PAO1psrA::Tn5 (B). Cells were inoculated into LB medium and sampled periodically during the growth transition. Lane 1, OD₆₅₀ of <0.2; lane 2, OD₆₅₀ of 0.2 to 0.4; lane 3, OD₆₅₀ of 0.4 to 0.6; lane 4, OD₆₅₀ of 0.6 to 0.8; lane 5, OD₆₅₀ of 0.8 to 1.2; lane 6, OD₆₅₀ of 1.2 to 1.5; lane 7, OD₆₅₀ of >2. Extracts were normalized for OD₆₅₀, and total proteins corresponding to 2 × 10⁷ CFU were loaded in each lane and examined by Western analysis with anti-RpoS antiserum (see text for details). σ^s levels were estimated by scanning Western blots with a densitometer. The quantities of protein are indicated in the graphs on the right as a function of the signal intensity of lane 1 (culture at the beginning of growth). For gel A the intensities were 1× (lanes 1 and 2), 2.6× (lane 3), 6× (lane 4), 6.6× (lane 5), 7.6× (lane 6), and 8× (lane 7). For gel B the intensities were 1× (lanes 1 and 2), 2.5× (lane 3), 4.5× (lane 4), 6× (lane 5), 6.6× (lane 6), and 11× (lane 7).

Identification of a protein(s) that binds the 5′ UTR of the rpoS mRNA.

In order to identify trans-acting factors in P. aeruginosa that are able to bind to the 369-bp 5′ UTR of the rpoS mRNA and may be involved in the translational regulation, UV cross-linking assays were performed with in-vitro-transcribed 5′ UTR rpoS mRNA and with total protein extracts of P. aeruginosa from the logarithmic and stationary phases of growth. The ³²P-labeled 5′ UTR rpoS mRNA was cross-linked to a protein(s) by exposure to UV light, and the resulting ³²P-labeled proteins were separated by SDS-PAGE. A prominent band at approximately 69 kDa was observed in cross-linking experiments performed with total proteins from exponentially growing P. aeruginosa but not with proteins from stationary-phase cells (Fig. 2).

FIG. 2.

UV cross-linking to 5′ UTR of rpoS RNA with total protein extracts from P. aeruginosa. (A) A UV cross-linking assay was performed by using total cellular extracts of P. aeruginosa PAO1 in different growth phases (lanes 1, 3, and 5, logarithmic phase [Log.]; lanes 2, 4, and 6, stationary phase [Stat.]) and in-vitro-transcribed labeled 5′ UTR rpoS RNA. The total cellular proteins in the reaction mixtures in lanes 1, 2, 3, and 4 were isolated from the same PAO1 liquid culture; proteins from independent PAO1 cultures (lanes 5 and 6) were used for repeating the experiment. Heparin was added to the reaction mixtures to a final concentration of 5 mg/ml. Samples were separated on an SDS—10% PAGE gel, which was then vacuum dried and exposed to BioMax film after it was dried for 12 h at −80°C. The UV cross-linking reaction of an approximately 69-kDa protein with the 5′ UTR rpoS RNA with protein extracts from logarithmic-phase P. aeruginosa is indicated by an arrow. (B) Coomassie brilliant blue staining of SDS—12% PAGE-separated proteins from logarithmic- and stationary-phase cellular extracts used in the UV cross-linking assay, as shown in panel A. The amount of soluble protein used in the UV cross-linking assay corresponded to 3.5 × 10⁷ CFU. kD, kilodaltons.

Characterization of the 69-kDa protein that binds the 5′UTR of rpoS mRNA only during the logarithmic phase of growth.

The protein that binds the 5′ UTR of rpoS mRNA during the logarithmic phase of growth was investigated by using an affinity purification procedure that involved cross-linking of 5′ UTR rpoS mRNA to adipic acid dehydrazide-agarose beads. As a control, we also performed the affinity purification procedure using total protein extracts from stationary-phase cells of P. aeruginosa. The bead preparation was therefore incubated independently with P. aeruginosa logarithmic-phase and stationary-phase cell extracts, and the proteins bound were separated on an SDS-PAGE gel and then analyzed by Coomassie blue staining. Figure 3 shows that there were clear differences between the binding patterns and the 5′ UTR rpoS mRNA incubated with proteins present during the logarithmic phase of growth, which resulted in specific pulling down of a 69-kDa protein, whose molecular mass was similar to that of the band observed in UV cross-linking assays (see above). Internal sequencing by mass spectroscopy of the 69-kDa band resulted in nine peptides whose sequences corresponded to residues 2 to 14, 143 to 151, 231 to 241, 243 to 256, 321 to 331, 342 to 347, 505 to 511, 514 to 520, and 585 to 599 of ribosomal protein S1 (68.7 kDa) of P. aeruginosa (PA3162; www.pesudomonas.com). Protein S1 in bacteria is the largest ribosomal protein of the small subunit of the 70S ribosome (32), and P. aeruginosa S1 is 559 amino acids long and displays 84% identity to the well-studied S1 protein of E. coli (data not shown).

FIG. 3.

Affinity purification of proteins that bind the 5′ UTR rpoS mRNA: SDS-PAGE separation of the proteins that bind the 5′ UTR RNA pulled down by using adipic acid dehydrazide-agarose beads. The gel was stained with Coomassie brilliant blue R-250, and protein bands were analyzed by mass spectrometry. The arrow indicates the protein band with a molecular mass that corresponded to the molecular mass of the protein band at 69 kDa that was UV cross-linked to the labeled 5′ UTR RNA. kD, kilodaltons.

In order to confirm that the protein that cross-linked to the 5′UTR of rpoS mRNA was ribosomal protein S1, immunoprecipitation studies with anti-S1 antibodies (which were raised against the E. coli S1 protein, which displayed 84% identity with the S1 protein of P. aeruginosa and cross-reacted very efficiently [data not shown]) were performed. The ³²P-labeled 5′ UTR rpoS mRNA was cross-linked to P. aeruginosa logarithmic-phase protein extracts by exposure to UV light and consequently was immunoprecipitated by using anti-S1 antibody (see Materials and Methods for details), and the proteins recovered were analyzed by SDS-PAGE. As shown in Fig. 4, this experiment further confirmed that the protein observed which cross-linked 5′ UTR rpoS mRNA was ribosomal protein S1.

FIG. 4.

Immunoprecipitation of ribosomal protein S1 bound to the 5′ UTR rpoS mRNA. A UV cross-linking assay was performed as described in Materials and Methods (lane 1), and samples were incubated with polyclonal anti S1-antibodies (antiS1-Ab′) (lane 2) or rabbit serum (lane 3), after which the immnunoprecipitates were affinity purified by using AG beads and separated on an SDS-PAGE gel. Labeled protein-IgG complexes were visualized by exposure of the dried gel to BioMax film for 12 h at −80°C. kD, kilodaltons.

Ribosome-free S1 protein binds the 5′ UTR of rpoS mRNA.

In gram-negative bacteria, protein S1 is part of the 30S ribosomal subunit and in this subunit plays the well-documented role of recognition and binding to the majority of mRNAs during the translational initiation process (26). Besides this function, S1 was shown to play a variety of roles (see below), specifically recognizing as a ribosome-free protein its own mRNA acting as an autogenous translational repressor (6). It was therefore of interest to establish if ribosome-free S1 could directly bind the 5′ UTR of rpoS mRNA. The P. aeruginosa rpsA gene encoding protein S1 was cloned, expressed, and purified as a His₆-tagged protein, and it was used in UV cross-linking experiments with 5′ UTR of rpoS mRNA. As shown in Fig. 5, these in vitro experiments demonstrated that free S1 protein can specifically bind the 5′ UTR of the rpoS mRNA.

FIG. 5.

Binding of the purified His₆-tagged S1 protein from P. aeruginosa PAO1 to labeled UTR RNA. The ³²P-labeled 5′ UTR rpoS mRNA transcribed in vitro was used for a UV cross-linking assay with purified ribosomal protein S1 from P. aeruginosa PAO1. Lanes 1, 2, and 3 contained samples with increasing concentrations of protein S1 (3.5, 7, and 15 ng, respectively). The weaker signal in lane 3 was due to the greater degradation of the S1 protein under the conditions used. kD, kilodaltons.

S1 protein binds within the first 78 nucleotides of the 5′ UTR.

In order to determine if the binding of the ribosomal S1 protein to the rpoS 5′ UTR was specific and to further localize its position, binding studies were performed with three subclones of the 369-bp 5′ UTR. The three subclones overlapped, as shown in Fig. 6; subclone UTRI contained the region from position 1 to position 178, UTRII contained the region from position 178 to position 369, and UTRIII contained the region from position 78 to position 274 (position 1 was the position that was farthest away from the ATG translational start codon). UV cross-linking assays were performed with in-vitro-transcribed 5′ UTRI, UTRII, and UTRIII rpoS mRNAs and total protein extracts of P. aeruginosa from the logarithmic phase of growth. The ³²P-labeled 5′ UTR rpoS mRNA was cross-linked to a protein(s) by exposure to UV light, and the resulting ³²P-labeled proteins were separated by SDS-PAGE. Interestingly, the 69-kDa S1 protein specifically cross-linked only to UTRI and not to UTRII or UTRIII (Fig. 6). These results indicate that S1 specifically bound within the first 78 nucleotides of the 5′ UTR rpoS mRNA (Fig. 6A and B). As a further control experiment, we performed a similar in vitro cross-linking experiment with the opposite strand of the complete 5′ UTR; this did not result in any S1 binding (Fig. 6C). In addition, we also performed in vitro cross-linking experiments using the 5′ UTRs of four other genes of P. aeruginosa, the alginate biosynthesis 366-bp 5′ UTR of the algD gene (37), the 25-bp 5′ UTR of the lasI quorum-sensing auotinducer synthase (31), the 122-bp 5′ UTR of the rhlI autoinducer synthase (8), and the 60-bp 5′ UTR of the transcriptional regulator psrA (18). In all cases when total protein from logarithmic-phase bacteria was used, there was no clear protein binding in the 69-kDa region which corresponded to protein S1 (Fig. 6D). A faint band was observed when the 5′ UTR of algD was used (Fig. 6D).

DISCUSSION

In this study, we began investigations of the translational regulation of the stationary-phase alternative sigma factor σ^s of the opportunistic human pathogen P. aeruginosa and the possible role played by the rpoS mRNA 5′ UTR. In E. coli and Pseduomonas sp. it was established that σ^s levels are carefully controlled at various levels, increasing dramatically at the onset of the stationary phase, thus increasing the ability to compete with other available σ subunits for core RNA polymerase (13, 17, 18; this study). The regulation of σ^s translation has been studied mainly in E. coli, and translation control involves the secondary structure of the 5′ UTR of rpoS mRNA together with small regulatory RNAs and two RNA binding proteins that affect the access of ribosomes to the initiation codon and ribosome binding sites (Shine-Dalgarno [SD]sequence) (reference 13 and references therein). Here we demonstrated that in P. aeruginosa, rpoS is under translational control since protein levels increased considerably at the onset of the stationary phase (Fig. 1). This could have been due in part to the increase in rpoS transcription which takes place upon entry into the stationary phase (17, 18); however, a considerable increase in σ^s levels was also observed in the psrA knockout mutant, which is known to have a 90% reduction in rpoS promoter activity. The increase in the RpoS protein in Pseudomonas in the stationary phase could also be due in part to protein stability, as recently demonstrated (2). Experiments reported here were aimed at identifying trans-acting factors that bind the 5′ UTR of the P. aeruginosa 5′ UTR rpoS mRNA. Using radioactively labeled 373-bp 5′ UTR RNA and total protein extracts from different growth phases in cross-linking experiments, we demonstrated that ribosomal protein S1 preferentially specifically binds in the logarithmic phase and not in the stationary phase. This was confirmed by affinity purification, immunoprecipitation, and binding of purified ribosome-free S1 to the 5′ UTR. In addition, the specific S1-5′ UTR rpoS mRNA interaction was further localized in the first 78 nucleotides of the 5′ UTR farthest away from the translational start codon. No S1-5′ UTR interaction was observed in experiments in which four 5′ UTRs of other P. aeruginosa genes were used. However, the possibility that ribosome-free S1 binding to UTRs in P. aeruginosa is a more general phenomenon cannot be excluded, and what functional role this could play is unknown.

Ribosomal protein S1, encoded by the rpsA gene, is essential for cell viability and is part of the 70S ribosome in bacteria; it is the largest ribosomal protein and promotes binding and recognition of the 30S ribosomal subunit to mRNA during the initiation process (26). The N-terminal domain is involved in protein-protein interactions (including binding to the ribosome), whereas the central domain and the C terminus are comprised of four similar RNA binding motifs which are known to preferentially recognize single-stranded AU- or U-rich regions in mRNA ladders (6). In addition, in E. coli, S1 is a multifunctional protein and has been shown to have other functions, including a variety of roles during phage infections (28) and as an autogenous translational repressor when an excess is present as a ribosome-free protein (6). The autocontrol was shown to be mediated through specific binding of ribosome-free S1 to single-stranded RNA regions of the 90-bp translation initiation region (TIR) of the rpsA gene. This TIR, defined as a region 90 bp upstream of the rpsA start codon, specifically folds into three hairpins and can be bound by the 30S subunit, resulting in translational initiation, or alternatively, it can be bound by free S1 protein, causing disruption of its conformation and leading to translational repression (5, 6). It therefore appears that in this autogenous control of its own translation, sequence-specific binding of RNA by ribosome-free S1 protein may provide a general mechanism to regulate gene expression. As protein S1 is able to undergo protein-protein interactions with a variety of proteins, the possibility that S1 may be a convenient connection between a variety of proteins and RNA substrates cannot be excluded. The approach used here to identify trans-acting factors which bind to the rpoS ′ UTR does not detect small regulatory RNAs which are known to have an important role in translational regulation in E. coli (13). To our knowledge, there have been no reports of the involvement of regulatory RNAs in translational regulation of rpoS in Pseudomonas; in the future experiments need to address this possibility. Interestingly, it was recently observed that in E. coli, the DsrA regulatory RNA of rpoS translation exerts its positive effect on translation via the interaction between the rpoS UTR and the 30S ribosomal subunit (36).

In this study we determined that the rpoS 5′ UTR preferentially binds in vitro S1 protein during the logarithmic phase of growth but not in the stationary phase; the reason for this and the effect that this might have on translation are currently not known. Considering the role that S1 plays in E. coli as an autogenous translational repressor, it is tempting to speculate that S1 might have a similar role in regulating rpoS translation. Interestingly, the predicted fold and sequence of the TIR of the rpsA gene of P. aeruginosa is clearly different from the predicted fold and sequence of the TIR of the E. coli counterpart, and it appears that S1 is not under autogenous translational control in Pseudomonas (6, 34). In addition, in the rpsA TIR of E. coli there is no clear SD ribosome binding sequence, and the recruitment of ribosomes occurs mainly through a specifically folded TIR (see above), whereas in the rpsA gene of P. aeruginosa there is a strong SD sequence very close to the ACCUCCUUA consensus sequence. These elements (different folding of the TIR and presence of the SD sequence) could indicate that rpsA in Pseudomonas might be regulated differently than the E. coli gene. Interestingly, the rpoS gene of P. aeruginosa has a clear SD ribosome binding consensus sequence just upstream of the initiation codon (data not shown).

The observations made in these studies raise the important question of why the ribosomal protein S1 preferentially binds to the 5′ UTR of the rpoS mRNA when the gene product and consequently its translation are not maximal (Fig. 1). Since S1 is known to be able to bind as a ribosome-free protein negatively influencing its own translation and since it is now more evident that there is a hierarchy of RNA targets with respect to the affinity for S1 (5), future work should focus on the precise positioning of the S1 binding region within the 5′ UTR and the role that this might play in the overall rpoS translation. Interestingly, it was observed that there is more ribosomal S1 protein (both ribosome free and ribosome bound) present in logarithmic-phase P. aeruginosa cells than in stationary-phase cells (Fig. 7). This could favor more binding to the rpoS mRNA 5′ UTR in logarithmic-phase cells than in stationary-phase cells and could have an effect on its translation. Interestingly, recently it was reported that in Sinorhizobium meliloti, ribosomal protein S1 was 40% less abundant in the stationary phase of growth than in the logarithmic phase (7).

FIG. 7.

Ribosomal S1 protein levels in P. aeruginosa. Cells were inoculated into LB medium and sampled in the logarithmic phase (OD₆₅₀, 0.9) and in the stationary phase (OD₆₅₀, 3.7). For the extract preparation procedure see Materials and Methods. Lanes 1 and 3 contained soluble fractions after ultracentrifugation, and these fractions probably represented ribosome-free S1 protein. Lanes 2 and 4 contained pellet fractions after ultracentrifugation, and these fractions probably represented ribosome-bound S1 protein. The total amount of protein extract used corresponded to the amount of protein from 3.5 × 10⁷ CFU, which was the amount used for the UV cross-linking assays shown in Fig. 2. Log, logarithmic phase; Stat., stationary phase.

To our knowledge, this is the first report regarding proteins that bind the 5′ UTR of rpoS mRNA and may be involved in rpoS translation in Pseudomonas, and it is only the second report of putative involvement of ribosomal protein S1 in translation regulation in bacteria. Finally, understanding the regulation of translation of rpoS in the opportunistic pathogen P. aeruginosa might prove to be important in revealing the mechanisms of pathogenicity as σ^s has specific roles related to virulence, colonization, and stress survival (16, 33).