INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Inorganic polyphosphates (polyP) are linear polymers in which inorganic orthophosphate (P_i) residues are linked by energy-rich phosphoanhydride bonds (21). Their chain lengths may vary from two to several hundred P_i residues. PolyP are widely distributed in nature and have been detected in bacteria, fungi, protozoa, plants, and mammals (22, 24); however, uncertainty still remains regarding their exact physiological role.

Depending on the species, the cellular localization, and the physiological conditions, polyP may have a variety of functions (19), including serving as an ATP substitute and energy source, a phosphate reservoir, and a chelator for divalent cations (16) and in the inhibition of RNA degradation (6) and regulatory responses to stresses and nutritional deficiencies (30). Moreover, it can be a structural element in DNA entry and transformation (32).

Several polyP-metabolizing enzymes have been purified, and their genes have been cloned. Accumulation of this polymer is known to occur in microorganisms catalyzed by the enzyme polyphosphate kinase (PPK); PPK catalyzes the reversible transfer of the terminal phosphate from ATP to polyP (2, 13, 15, 36). PolyP degradation and utilization is mediated by exopolyphosphatases (PPX) (3, 38), endopolyphosphatases (23), and specific kinases (37).

Extensive accumulation of polyP has been detected in the Pseudomonas aeruginosa mucoid strain 8830, particularly during the stationary phase (18) and in response to phosphate and amino acid limitations (5), yet essentially nothing is known about the genes and their regulation in the synthesis of polyP. We report here the cloning and sequencing of the P. aeruginosa ppk and ppx genes and the characterization of their promoters and gene products.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains and growth conditions. Strains and plasmids used in this work are described in Table 1. Escherichia coli and P. aeruginosa were routinely grown at 37°C on Luria-Bertani (LB) solid or liquid medium (Difco). For antibiotic selection of E. coli, ampicillin was used at 50 or 100 µg/ml, tetracycline was used at 12.5 µg/ml, kanamycin was used at 50 µg/ml, and chloramphenicol was used at 34 µg/ml. For selection of P. aeruginosa, carbenicillin, chloramphenicol, and tetracycline were used at 300 µg/ml in solid media and at 100 µg/ml in liquid media. Triparental mating was performed by using pRK2013 as the helper plasmid (12), and exconjugants were isolated on Pseudomonas isolation agar (Difco) with appropriate antibiotic selection. The broad-host-range plasmid pQF50 and its constructs were introduced into P. aeruginosa via electroporation (28).

DNA manipulations. Steps involved in the cloning of the P. aeruginosa 8830 ppk and ppx genes are described in the Results section. Genomic DNA was prepared with a GNome Kit (Bio 101, Inc.). Restriction endonucleases (Gibco BRL), T4 DNA ligase (Gibco BRL) and calf intestinal phosphatase (Pharmacia) were used as specified by the manufacturers. Plasmid DNA was isolated on a small scale as described by Zhou et al. (41) and on a large scale by using the Qiagen Plasmid Kit (Qiagen).

Cloning of ppk and ppx genes for overexpression and protein purification. The ppk gene was amplified by using oligonucleotides PPK-N and PPK-C corresponding to the 5' and 3' ends of the gene. The plasmid DNA extracted from PPKB was used as template DNA. EcoRI and HindIII sites were designed in the two oligonucleotides to allow the PCR product to be cloned into the expression vector pET24a(+) (Novagen). In PPK-C the stop codon at the 3' end of the ppk coding sequence has been omitted to allow an in-frame fusion with six histidine residues encoded in the vector.

Assay for PPK activity. PPK activity was measured in membrane preparations as the production of acid-insoluble ³²P-labeled polyP, as previously described (1). Membranes were prepared by centrifugation of total cell extracts in a Beckman ViTi65 rotor at 45,000 × g for 1 h after sonication.

Assay for PPX activity. The activity of purified PPX was tested in 20 µl of 50 mM Tris-HCl buffer (pH 8)-1 mM MgCl₂-175 mM KCl-250 µM polyP. The synthesis and purification of the [³²P]polyP substrate was done as described for the PPK assay. The PPX reaction was performed at 37°C. Samples (2 µl) were taken periodically and loaded on a polyethyleneimine plate. Polyphosphates remaining at the origin were separated from orthophosphate by thin-layer chromatography (TLC) with 0.75 M KH₂PO₄ (pH 3.5). Dried plates were visualized with a PhosphorImager (Molecular Dynamics).

Primer extension analysis. Total RNA was isolated by using TRIzol reagent (Gibco BRL) from P. aeruginosa 8830 culture at an OD₆₀₀ of 1.0. Samples of 30 µg of RNA were hybridized overnight with 10 pmol of P1 primer labeled with the MaxKinase kit (Ambion). After precipitation, RNA was resuspended in 10 µl of water, and the primer extension reaction was performed with the First Strand cDNA synthesis kit (Pharmacia) at 37°C. Sequencing was performed by use of the dideoxynucleotide chain termination reaction method with T7 DNA polymerase (Sequenase 2.0 kit; USB Corp.).

-Galactosidase assay. P. aeruginosa 8830 cells harboring the promoter probe constructs were grown overnight in LB with ampicillin (100 µg/ml). These overnight cultures were used to inoculate fresh LB (1%). P. aeruginosa 8830 cells were harvested in mid-log phase and stationary phase and then divided into two parts. One part was treated with NaCl (added to a 2 M final concentration) or H₂O₂, while the other was used as a control. Samples (50 ml) were centrifuged, and the cell pellets were resuspended in the assay buffer and sonicated. After centrifugation to eliminate cell debris, the cell extract was assayed for -galactosidase activity. Quantitative determination of -galactosidase activity was performed by the method of Miller (29). Each experiment was performed in triplicate.

Nucleotide sequence accession numbers. The nucleotide sequences of the P. aeruginosa 8830 ppk and ppx genes have been deposited in the GenBank nucleotide sequence database under accession numbers AF087931 and AF053463, respectively.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning of P. aeruginosa 8830 ppk and ppx genes. On the basis of the homology alignment of amino acid sequences of PPKs from Neisseria meningitidis, Klebsiella aerogenes, and E. coli, we designed degenerate primers that corresponded to highly conserved amino acid sequences. These oligonucleotides were used to perform PCR as described in Materials and Methods. The PCR product had the expected size of 1.7 kb. This product was sequenced, and the sequence homology to known ppk genes suggested it could be the amplification of a portion of the ppk gene.

View larger version (52K): [in this window] [in a new window]   FIG. 1.   (A) Organization of ppk and ppx genes. (B) Alignment of P. aeruginosa 8830 PPX with E. coli PPX (accession number L06129) and E. coli pppGpp-5'-phosphohydrolase (GppA) (accession number M83316) (CLUSTAL X). Sequences were retrieved from GenBank. *, amino acid identity; ":" and ".", limited conserved substitutions.

Construction of ppk mutants with reduced PPK activity. To isolate a knockout mutation in the ppk gene, the tetracycline cassette was inserted in the unique BamHI site of the 1.6-kb PCR fragment as described in Materials and Methods, and pMOB2 was inserted in the NotI site (33). The resulting plasmid (pNotK::Tc) was conjugated into P. aeruginosa 8830 by triparental mating, and several tetracycline-resistant but carbenicillin-sensitive colonies were obtained. Southern blotting of chromosomal DNA from one of these transformants digested with SphI and by using the 1.7-kb PCR probe showed two bands of ca. 2.3 and 3 kb, instead of a single 3.5-kb band. This finding was consistent with the insertion of the 1,720-bp Tc cassette which had one SphI restriction site and no KpnI site. Moreover, DNA digested with KpnI showed an increase in fragment size from 8 to ca. 9.7 kb (Fig. 2).

View larger version (60K): [in this window] [in a new window]   FIG. 2.   Southern blot analysis of ppk knockout mutant genomic DNA, determined by using the 1.7-kb PCR probe. Lanes: 1, P. aeruginosa 8830 chromosomal DNA digested with SphI; 2, ppk::Tc genomic DNA digested with SphI; 3, P. aeruginosa 8830 chromosomal DNA digested with KpnI; 4, ppk::Tc genomic DNA digested with KpnI. The different band patterns for the parental and mutant strains are consistent with the insertion of the Tc cassette of 1,720 bp having one SphI restriction site and no KpnI sites.

1

1

1

1

Purification and activity of recombinant PPK. The His-tagged protein was purified by metal-chelate affinity chromatography on an Ni-iminodiacetic acid column. Eluted protein (3.5 µg) was loaded onto an SDS-PAGE gel and stained with Coomassie blue R250. Protein purification was homogeneous since a single 75-kDa band was detected. The purified protein was tested for its ability to synthesize polyP in vitro. For this purpose PPK was incubated with [-³²P]ATP, and the product of the in vitro reaction was purified with Glassmilk and loaded onto a TLC plate. After TLC analysis, the product of the in vitro reaction was retained at the origin (Fig. 3, lane 2), but after digestion for 1 h with an excess of recombinant yeast PPX (rPPX1), as described by Wurst and Kornberg (38), it was completely degraded to free P_i (Fig. 3, lane 3). Similar results were observed when purified E. coli PPK (Fig. 3, lanes 4 and 5) was used. When the same product was incubated with an ATPase purified in our laboratory from Mycobacterium bovis (40), no release of free P_i was detected (Fig. 3, lanes 7 and 8). However, the ATPase activity of this enzyme was confirmed by digestion of ATP to free P_i in the PPK reaction mixture (Fig. 3, lane 6).

View larger version (141K): [in this window] [in a new window]   FIG. 3.   TLC analysis of PPK enzymatic activity product. Lanes: 1, P. aeruginosa PPK reaction mixture; 2, polyP purified with Glassmilk from the reaction mixture shown in lane 1; 3, P. aeruginosa polyP digested with rPPX1; 4, purified polyP synthesized by E. coli PPK; 5, E. coli polyP digested with rPPX1; 6, reaction mixture in lane 1 digested with an M. bovis ATPase; 7, E. coli polyP digested with M. bovis ATPase; 8, P. aeruginosa polyP digested with M. bovis ATPase.

Purification of recombinant PPX and polyP degradation. PPX with an oligohistidine domain at the N terminus was purified to homogeneity as visualized by Coomassie blue R250 staining of 6 µg of elution fraction run on an SDS-PAGE gel. The molecular mass of the denatured histidine-tagged protein was determined by SDS-PAGE to be 60 kDa, with a linear regression curve of the standard low-molecular-mass protein (14,400 to 97,400 Da) (Bio-Rad). The theoretical molecular mass calculated from the amino acid sequence encoded by the ppx gene, including six histidine terminal residues, was 58.7 kDa. The hydrolytic activity of this enzyme toward polyP was demonstrated in vitro by using polyP as a substrate (Fig. 4). Its specific activity, measured with E. coli polyP as a substrate, was 7 × 10⁶ pmol of P_i produced min¹ mg of protein¹.

View larger version (87K): [in this window] [in a new window]   FIG. 4.   TLC analysis of P. aeruginosa PPX enzymatic activity. [32P]polyP (267 µM) was incubated for various time periods as specified with 200 ng of PPX protein in 20 µl of total reaction mixture. Samples (1 µl) were taken periodically and spotted onto a polyethyleneimine plate.

ppk and ppx promoter investigation. The region upstream of the ppk gene was investigated for its promoter activity through construction of a promoter-reporter gene fusion. For this purpose, a 376-bp XhoI fragment including 251 bp upstream of the GTG translational start site was cloned into the SalI site of the promoterless lacZ gene of the pQF50 plasmid in both orientations. These constructs were electroporated into P. aeruginosa 8830, and the -galactosidase activity was tested at various stages of growth. When the fragment was cloned in the inverted orientation, the activity was 8.7 times lower (37.27 ± 1.4 Miller units) than in the direct orientation (327.4 ± 12.87 Miller units) (pQF-376K). This suggested that the promoter could be located upstream of the ppk coding sequence. Primer extension analysis identified three putative transcriptional start sites (Fig. 5A) differing by one or two bases. The start site farthest upstream is a C residue that produces the most intense transcript band. This start site is accompanied by 35 and 10 promoter sequences, which resemble the consensus promoter sequence for ^E-dependent promoters as exemplified by E. coli htrA and rpoH or P. aeruginosa algT, algR1, or algD (10, 14) (Fig. 5B). A 649-bp SphI fragment upstream of the ppx gene was cloned in the direct orientation into pQF50 upstream of the promoterless lacZ gene (pQF-649X). When two strains of P. aeruginosa 8830, which contained the promoter probe construct (pQF-649X) and the control plasmid (pQF50), respectively, were grown to log phase (OD₆₀₀ = 1.50) and the -galactosidase activity was measured, it was 7.7 times greater in pQF-649X (93 ± 0.4 Miller units) than in the control (12 ± 0.2 Miller units). To investigate a putative coregulation at the transcriptional level between ppx and ppk, the -galactosidase activity was measured in P. aeruginosa 8830 strains containing pQF-376K or pQF-649X in response to stress conditions such as a high concentration of NaCl or oxidative stress, which are known to influence polyP synthesis in E. coli (30). -Galactosidase levels were measured after 30 min of exposure to 2 M NaCl. Compared to the activity measured in each control, a modest increase was seen for both ppk and ppx promoters (Table 3). The increase became more significant after 2 h, but the response of the ppk promoter to such a stress was greater (3.2 times) in the stationary phase than in the log phase (Table 3). In contrast, ppx promoter activation was stronger in the log phase. The response to oxidative stress was evaluated by using 53 mM H₂O₂ (final concentration) and measuring the -galactosidase activity after 10 min. Stress-induced ppx promoter activity decreased slightly during both the log and the stationary phases, while a small increase of ppk promoter activity was measured during the stationary phase.

View larger version (45K): [in this window] [in a new window]   FIG. 5.   (A) Primer extension analysis of the transcription start sites of the ppk gene. P. aeruginosa 8830 total RNA was annealed to oligonucleotide P1 and extended with murine reverse transcriptase. Lanes G, A, T, and C represent the dideoxy sequencing ladder carried out with the same primer. Lane 1, primer extension products. (B) Comparison of 35 and 10 motifs identified upstream of the transcription start site (the C residue that gave the most intense band) with E consensus sequences.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Accumulation of polyP has been detected in P. aeruginosa in response to nutrient limitation (5, 18), but the implications of this physiological response to such stress conditions remain unclear. The pathway of synthesis and degradation of polyP in P. aeruginosa has never been investigated. We isolated the P. aeruginosa genes encoding the major enzymes involved in polyP metabolism, PPK and PPX. In contrast to E. coli, where the ppk and ppx genes are organized in an operon, in P. aeruginosa the two genes are transcribed divergently (Fig. 1A). Investigation of the two promoters did not show any coregulation between the expression of these two genes, at least under the conditions we analyzed, such as low or high P_i levels (data not shown), osmotic shock, or oxidative stress. This observation is in agreement with the observation that polyP accumulation is regulated at the enzymatic level through PPX inhibition by the stress response nucleotides ppGpp and pppGpp (25) without any modulation of the transcription rate of these two genes.

After ppk gene inactivation, the ppk knockout mutant does not show a growth defect compared to the parental strain. However, a residual low PPK activity, measured as polyP synthesis, in membrane preparations has been detected in the mutant. This may represent a separate mechanism of polyP synthesis, as suggested also by the observations that a small amount of polyP remains in the cells of E. coli (8) and N. meningitidis (36) where the ppk gene has been inactivated.

Downstream of the ppk gene in P. aeruginosa is an ORF encoding a putative thioredoxin. The ppx gene product has been shown to be able to hydrolyze polyP to P_i with a specific activity comparable to that of E. coli PPX (3). Moreover, this enzyme exhibits significant sequence similarity with E. coli PPX and E. coli GppA. Both of these enzymes belong to a large superfamily that includes sugar kinases, actin, and hsp70 proteins (31). Since we have now cloned both the genes with their respective promoters, we should be able to define various environmental parameters that might modulate the expression of either or both of the genes that allow the cells to accumulate polyP.

An interesting observation from our studies is that the ppk gene of P. aeruginosa appears to be expressed from a ^E-responsive promoter (Fig. 5B). The ^E-responsive promoters have been reported to regulate alginate synthesis through activation of critical alginate biosynthetic genes such as algR1, algT, algD, and algC (34). It is interesting to note that both polyP accumulation and alginate synthesis are coregulated in P. aeruginosa such that alginate-positive mucoid cells accumulate much more polyP than their nonmucoid counterparts (18). Indeed, a mutation in the algR2 gene that results in reduced alginate production also results in reduced polyP accumulation (18). Whether a common regulator controls the expression of key ^E-responsive promoters such as algT/algR1 and ppk is not known at present. Further investigations looking for a common link in the genetic expression of two stress-inducible polymers, such as polyP and alginate, presumably through ^E-responsive promoters, may provide important insights into the mode of regulation of the biosynthesis of these two polymers.