Accumulation of Inorganic Polyphosphate in phoU Mutants of Escherichia coli and Synechocystis sp. Strain PCC6803

Abstract

The biological process for phosphate (P_i) removal is based on the use of bacteria capable of accumulating inorganic polyphosphate (polyP). We obtained Escherichia coli mutants which accumulate a large amount of polyP. The polyP accumulation in these mutants was ascribed to a mutation of the phoU gene that encodes a negative regulator of the P_i regulon. Insertional inactivation of the phoU gene also elevated the intracellular level of polyP in Synechocystis sp. strain PCC6803. The mutant could remove fourfold more P_i from the medium than the wild-type strain removed.

Inorganic phosphate (P_i) is recognized as one of the major nutrients causing eutrophication of lakes, bays, and waterways (18). Considerable attention has been paid to effective P_i removal from wastewater (6). Many bacteria are capable of accumulating excess P_i in the form of inorganic polyphosphate (polyP), which is a linear polymer of hundreds of P_i residues linked by high-energy phosphoanhydride bonds (7, 9, 10). Improvement of the ability to accumulate polyP contributes to increased P_i removal from wastewater (18).

Previously, we have demonstrated genetic improvement of polyP accumulation in Escherichia coli (8). High levels of accumulated polyP were achieved by increasing the dosage of the E. coli genes encoding polyP kinase (ppk) and the P_i-specific transport system (pstSCAB). The E. coli recombinant accumulated approximately 16% of its dry weight as phosphorus (P) (49% as P_i) (8). Over 60% of cellular P was stored in the form of polyP in the genetically engineered E. coli strain. However, growth of the E. coli recombinant was severely limited in minimal medium (8). In addition, this recombinant released polyP back into the medium when it accumulated high levels of polyP. In this paper, we report that a mutation in the phoU gene, which encodes a negative regulator of the P_i regulon (3, 21), led to high levels of accumulated polyP in E. coli. phoU mutants could be easily screened on agar plates containing 5-bromo-4-chloro-3-indolylphosphate (XP) after N-methyl-N′-nitro-N-nitrosoguanidine mutagenesis. Therefore, isolating phoU mutants seems to be a useful way to improve the ability of bacteria to accumulate polyP. To show whether this method is effective in another bacterium, we performed insertional inactivation of the phoU gene in Synechocystis sp. strain PCC6803 and showed that the intracellular level of polyP increased in the mutant.

Isolation of E. coli mutants.

The levels of polyP in E. coli MG1655 were very low (less than 1 nmol of P_i residues/mg of protein) when the organism was grown on a rich medium (11). PolyP was recovered with silicate glass from cells lysed with guanidine isothiocyanate, and the polyP content was determined by a two-enzyme assay (4). PolyP was first converted to ATP by polyphosphate kinase, and then the amount of ATP was measured by a bioluminescence assay. We first selected alkaline phosphatase constitutive mutants, which could form blue colonies on agar plates containing XP (50 mg/liter) under P_i-sufficient conditions after N-methyl-N′-nitro-N-nitrosoguanidine mutagenesis (17). One of 150 mutants, designated MT4, accumulated a large amount of polyP (approximately 100 nmol of P_i residues/mg of protein). The levels of polyP were at least 100-fold higher than the levels in the wild-type strain, MG1655.

To assess P_i transport (14), the E. coli mutants were grown overnight in MOPS (morpholinepropanesulfonic acid) medium (12) containing either 0.1 mM P_i (P_i limiting) or 2 mM P_i (P_i sufficient). Cells were harvested by centrifugation at 8,000 × g for 5 min and washed with HEPES buffer (13). Assays were started by adding ³²P_i to a final concentration of 2.5 μM, and the amount of ³²P_i incorporated into the cells was determined by using a scintillation counter (Packard). Only MT4 exhibited P_i-specific transport even when the organisms were grown with excess P_i (Fig. 1).

FIG. 1.

P_i transport as assessed by using P_i-sufficient MG1655 (▪), P_i-limited MG1655 (•), P_i-sufficient MT4 (▵), and P_i-limited MT4 (○) cultures. Cells were grown in MOPS medium containing either 2 mM P_i (P_i sufficient) or 0.1 mM P_i (P_i limited). OD, optical density at 600 nm.

Analysis of the mutation of MT4.

An E. coli DNA library based on the SuperCos plasmid (Stratagene) was introduced into MT4 by transformation (17). Approximately 200 transformants were examined for the ability to accumulate polyP in 2×YT medium (17). One transformant reduced the levels of polyP to the level in the wild-type strain. This transformant carried a recombinant plasmid containing a 30-kb DNA fragment of the E. coli chromosome. Nucleotide sequence analysis of the 30-kb insert revealed that this fragment contained the pst and bgl operons which were located at 84 min on the E. coli linkage map (5). Subcloning and complementation analysis revealed that a 3.0-kb EcoRI fragment, carrying the entire phoU gene, could complement the mutation of MT4. The chromosomal phoU gene of MT4 was amplified by PCR with primers EU1 (5′-ATTGGGATTTGTCTGGTGAA-3′) and EU2 (5′-AGAAGACTACATCACCGGTC-3′) and cloned into the pGEM-T vector (Promega). Nucleotide sequence analysis showed that the 29th codon of the phoU gene was changed from a glycine codon to an aspartic acid codon in MT4. We again selected a polyP-accumulating mutant from the mutants producing alkaline phosphatase constitutively. In this mutant, the 83rd codon of the phoU gene was changed from an alanine codon to a threonine codon. These results indicated that mutation of the phoU gene resulted in elevated intracellular levels of polyP in E. coli.

To further confirm that phoU mutants could accumulate high levels of polyP, the chromosomal phoU gene of MG1655 was disrupted by inserting a kanamycin resistance (Km^r) gene cassette into the wild-type gene. A 0.8-kb DNA fragment containing the entire phoU gene was amplified by PCR with the EU1 and EU2 primers and inserted into the pGEM-T vector. The resultant plasmid was digested with ClaI and ligated with the Km^r gene cassette of pUC4K (Amersham-Pharmacia), and an EcoRI fragment containing a disrupted phoU gene was inserted into an EcoRI site of pGP704 containing the sacB gene (16). This plasmid was introduced into E. coli S17-1 and then transferred into MG1655 by transconjugation (22). Transconjugants were selected on agar plates containing 5% sucrose and kanamycin (50 mg/liter). Disruption of the chromosomal phoU gene was confirmed by Southern hybridization. As expected, the insertional phoU mutant accumulated high levels of polyP (approximately 400 nmol of P_i residues/mg of protein).

One might assume that ppk is a member of the P_i regulon in E. coli and that derepressed expression of this gene in the phoU mutants results in polyP accumulation. However, no significant increase in the level of polyphosphate kinase was observed under P_i-limited conditions in E. coli (Morohoshi and Kuroda, unpublished data). Furthermore, expression of a single-copy ppk-lacZ transcriptional fusion did not increase under P_i limitation conditions (L. Zhou and B. L. Wanner, personal communication). P_i transport is rate limiting for polyP accumulation in E. coli (8). A phoU mutant exhibited P_i-specific transport even under P_i-excess conditions (Fig. 1). We transferred a Δ(pstSCAB-phoU)::km mutation from BW17335 (20) to the wild-type strain, MG1655, by using P1 phage. This mutant failed to accumulate polyP (0.5 nmol/mg of protein). Consequently, it is likely that constitutive expression of P_i-specific transport (PstSCAB) is responsible for the elevated levels of polyP in the phoU mutants.

Although MT4 grew relatively slowly, it removed twofold more P_i from the medium than the wild type removed (Fig. 2). Introduction of multicopy plasmid pBC29, which contains the E. coli ppk gene (1), into the wild-type strain, MG1655, did not increase P_i removal significantly (Fig. 2). By contrast, MT4(pBC29) removed about twofold more P_i than MT4 removed (Fig. 2). The P content of MT4(pBC29) reached approximately 9% on a dry weight basis (27% as P_i) and was about sixfold greater than that of the wild-type strain, MG1655. PolyP granules were detected in MT4(pBC29) when the cells were observed by fluorescent microscopy after they were stained with 4′,6′-diamidino-2-phenylindole (DAPI) (2) (Fig. 3).

FIG. 2.

Growth (A) and P_i uptake (B) of E. coli MG1655, MG1655(pBC29), MT4, and MT4(pBC29). Strains MG1655 (▪), MG1655(pBC29) (•), MT4 (▵), and MT4(pBC29) (○) were grown in MOPS medium containing 2 mM P_i at 28°C. OD₆₀₀, optical density at 600 nm.

FIG. 3.

PolyP granules in MT4(pBC29) detected by DAPI fluorescence. Strain MT4(pBC29) was grown overnight on 2×YT medium, stained with DAPI at a final concentration of 50 μg/ml, and observed with a fluorescence microscope (Olympus BX-40).

phoU mutant of Synechocystis sp. strain PCC6803.

A chromosomal phoU mutant of Synechocystis sp. strain PCC6803 was also constructed by inserting a Km^r gene cassette into the wild-type gene. A 2.0-kb DNA fragment, which carried the entire phoU gene, was amplified by PCR with primers SU1 (5′-GGTACCATCAACCTGATCGCCTAT-3′) and SU2 (5′-GCTACTGCTCCAGTCGACCCGAGT-3′) and cloned into pUC119. The resultant plasmid was digested with BglII, ligated to the Km^r gene cassette of pUC4K, and introduced into Synechocystis sp. strain PCC6803 by electroporation (17). Disruption of the chromosomal phoU gene was confirmed by Southern hybridization (data not shown). The Synechocystis phoU mutant, designated SPU101, accumulated about 15% as much polyP (on a dry weight basis) as P_i. As in E. coli MT4(pBC29), polyP particles were detected in SPU101 by DAPI fluorescence (Fig. 4). The total P content of the mutant strain reached a maximum of 6% on a dry weight basis. P_i uptake experiments were performed with growing cultures of PCC6803 and SPU101 (Fig. 5). Cultures were grown in BG-11 medium (19) with 1% CO₂ at 30°C under light (5,000 lx). Strain SPU101 removed about fourfold more P_i from the medium than the parental strain removed. The growth of SPU101 was almost equivalent to that of PCC6803.

FIG. 4.

PolyP granules in Synechocystis sp. strain PCC6803 and phoU mutant SPU101 detected by DAPI fluorescence. Strains PCC6803 and SPU101 were grown in BG-11 medium at 30°C under light (5,000 lx) with 1% CO₂ for 3 days and stained with DAPI. Cells were observed with either visible light (A and B) or UV light (C and D).

FIG. 5.

Growth (A) and P_i uptake (B) of Synechocystis sp. strain PCC6803 and phoU mutant SPU101. Strains PCC6803 (•) and SPU101 (○) were grown in BG-11 medium at 30°C under light (5,000 lx) with 1% CO₂. OD₇₅₀, optical density at 750 nm.

We showed that a mutation of the phoU gene, which encodes a negative regulator of the P_i regulon, led to high levels of accumulated polyP in both E. coli and Synechocystis sp. Rao et al. previously reported that a phoU mutation had no effect on polyP accumulation (15). The phoU mutant of these authors probably had a secondary mutation in the Pst system, as described previously (20).

In enhanced biological P removal, sludge microorganisms accumulate 3 to 5% of their dry weight as P (18). Similar levels of accumulated polyP were observed with the phoU mutant of Synechocystis sp. strain PCC6803. In general, the levels of carbon sources in wastewater are relatively low. This is likely to limit P_i removal from wastewater by sludge microorganisms (18). Use of the Synechocystis mutant, which is able to accumulate polyP if light is present, may improve biological removal of P_i from wastewater.