INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Cyanobacteria are photoautotrophic procaryotes that perform oxygen-evolving photosynthesis with carbon dioxide as the primary oxidizing agent and carbon source. Requiring only light, water, carbon dioxide, and inorganic salts, cyanobacteria have very simple nutrient requirements that allow these organisms to occupy highly diverse ecological niches. All cyanobacteria can utilize nitrate as their sole nitrogen source (30), although reduced nitrogen sources, such as ammonia and urea, can also be utilized (7). Since nitrogen is a major nutrient and accounts for about 11% of the dry weight of cyanobacterial cells (39), nitrate assimilation and reduction place a high demand for energy and electrons on the photosynthetic machinery. The estimated ratio of the maximal rates of carbon assimilation and nitrate assimilation is roughly 2 to 2.5 (6, 26) under optimized assay conditions. This suggests that up to 30% of the electrons generated by photosynthetic water oxidation are consumed during the reduction of nitrate to ammonia when cells are grown on nitrate.

Cyanobacterial cells preferentially utilize reduced nitrogen sources such as ammonia and urea. Nitrate consumption is completely inhibited within minutes of the addition of exogenous ammonia (5, 7, 19), and the nitrate transport system is thought to be the rate-limiting step for nitrate assimilation (7). It is believed that the nitrate transporter is rapidly inactivated in the presence of ammonia, although the molecular mechanism of this inhibition of nitrate assimilation by ammonia has not yet been elucidated (33).

It is important to know the structure of the nitrate transport system in cyanobacteria in order to analyze the regulation of nitrate assimilation by the availability of nitrogen sources as well as by photosynthesis and carbon assimilation. However, most recent molecular biological studies of the nitrate transport system have been limited to the freshwater cyanobacteria Synechococcus sp. strain PCC 7942 (21, 33) and Anabaena sp. strain PCC 7120 (2, 9). The genes for nitrite reductase (nirA), the nitrate transporter (nrtABCD) and nitrate reductase (narB) are transcriptionally regulated as an operon, nirA-nrtABCD-narB, in Synechococcus sp. strain PCC 7942 (for reviews, see references 20 and 21).

Nitrate assimilation is also very important in the low-temperature physiology of cyanobacteria. We have previously demonstrated that when cells are grown with nitrate as a nitrogen source, growth at a low temperature causes nitrogen limitation in the unicellular marine cyanobacterium Synechococcus sp. strain PCC 7002 (25). Cells of this cyanobacterium become chlorotic and grow arithmetically at 15°C in a medium containing nitrate as the sole nitrogen source. However, when cells are grown at 15°C on urea as the nitrogen source, cells grow exponentially and the symptoms of chlorosis are not observed (25). In the freshwater cyanobacterium Synechococcus sp. strain PCC 6301, nitrate transport also limits cell growth at low temperatures (26). The temperature-dependent decrease in the rate of nitrate consumption is similar to the temperature-dependent decrease in the growth rate, and both nitrate consumption and cell growth cease at 15°C in Synechococcus sp. strain PCC 6301. The marine cyanobacterium Synechococcus sp. strain PCC 7002 is somewhat more tolerant of low-temperature growth conditions, and both nitrate consumption and growth of this strain cease at 12°C (27). In order to understand better the important role of nitrate transport in defining the lower limits of cyanobacterial growth, we decided to characterize the nitrate transport system in the marine cyanobacterium Synechococcus sp. strain PCC 7002 in detail. These studies led to the discovery of a novel gene encoding a nitrate/nitrite permease, which was functionally characterized by reverse genetics and physiological studies.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Organisms and culture conditions. A laboratory wild-type strain of Synechococcus sp. strain PCC 7002, denoted PR6000, has been maintained at The Pennsylvania State University since 1981. Cells were grown photoautotrophically in medium A⁺ containing 12 mM NaNO₃ (31), in medium AU₁₀Ni containing 10 mM urea as a sole nitrogen source and supplemented with 5 µM NiSO₄ (28), or in medium A-HEPES containing 5 mM NH₄Cl as a nitrogen source and buffered with 25 mM HEPES-NaOH at pH 7.2 (28) under constant illumination from cool-white fluorescent lamps (250 µE m² s¹) with aeration from 1% (vol/vol) CO₂ in air at 38°C. The photon flux density was measured with a model QSL-100 light meter (Biospherical Instruments, San Diego, Calif.). For the selection of kanamycin-resistant mutants, kanamycin (50 µg ml¹) was added to liquid media and agar plates. To change the growth media, cells were collected by centrifugation at room temperature, washed with nitrogen-free medium, and inoculated in fresh medium. Liquid cultures (25 ml) were incubated in 22- by 175-mm glass culture tubes, and the initial cell density for growth curve experiments was adjusted to an optical density at 550 nm (OD₅₅₀) of 0.05 (90% light transmittance). Cell growth was monitored by the OD₅₅₀ with a model Spectronic 20 spectrophotometer (Bausch & Lomb [now Milton Roy], Rochester, N.Y.). A cell suspension from an exponential-phase culture grown at 38°C in medium A⁺ at a light intensity of 250 µE m² s¹ with an OD₅₅₀ of 1.0 contained 3.4 ± 0.3 µg of chlorophyll ml¹ and 1.0 × 10⁸ ± 0.2 × 10⁸ cells ml¹ as determined by microscopic count (25). To induce the nitrate consumption activity of cells grown with urea as a nitrogen source, cells grown overnight at 38°C in medium AU₁₀Ni were collected by centrifugation and washed twice with fresh medium A⁺; the cells were then resuspended in the fresh medium A⁺ at the original cell density and incubated for 1 h under the same growth conditions. To inhibit the glutamine synthetase-glutamine(amide)-2-oxoglutarate aminotransferase (GOGAT) cycle in Synechococcus sp. strain PCC 7002, 6-diazo-5-oxo-L-norleucine (DON) (D-2141; Sigma Chemical Co., St. Louis, Mo.), which is a specific inhibitor for GOGAT, was used when indicated for some experiments. L-Methionine sulfoximine (M-5379; Sigma Chemical Co.) cannot be used with Synechococcus sp. strain PCC 7002, since L-methionine sulfoximine was not effective in releasing nitrate consumption from the repression effects of ammonia (data not shown). Wild-type cells of Synechococcus sp. strain PCC 7002 grow in medium A⁺ containing 0.5 mM L-methionine sulfoximine, although 0.5 mM DON completely inhibited cell growth under otherwise standard growth conditions (data not shown).

Cloning and interposon mutagenesis of nrtP and narB. Genomic DNA of Synechococcus sp. strain PCC 7002 was partially digested with EcoRI and inserted into the EcoRI site of cosmid vector of pHC79 to construct a genomic DNA library as described previously (38). A cosmid clone was isolated from the genomic DNA library of Synechococcus sp. strain PCC 7002 by cross-hybridization with a 2.1-kb DNA fragment carrying the narB gene of Synechocystis sp. strain PCC 6803 (11). The screening probe was amplified by PCR with genomic DNA from Synechocystis sp. strain PCC 6803 as the template and the specific primers 5'-ATG GAT TCA CCA GCT ATT CTT (forward) and 5'-TTA AAC TGG TGT AAT ATT AAC (reverse). Heterologous hybridization experiments were performed as previously described (1). The complete nucleotide sequence of the 7.2-kb region from EcoRI to BamHI of the positive clone (Fig. 1A) was determined at the Nucleic Acid Facility of the Biotechnology Institute of the Pennsylvania State University. The nrtP and narB genes were insertionally inactivated by insertion of a 1.4-kb XbaI fragment carrying the aphII gene and derived from plasmid pRL170 (4) into a unique SpeI site within the nrtP gene (position 2432 in the nucleotide sequence of the 7.2-kb region (Fig. 1B) and into an XbaI site within the narB gene (position 4583) (Fig. 1C). Wild-type cells of Synechococcus sp. strain PCC 7002 (strain PR6000) were transformed with plasmid DNA containing the interrupted gene essentially as described by Stevens and Porter (32). Kanamycin-resistant transformants were selected on medium AU₁₀Ni supplemented with 50 µg of kanamycin ml¹. The aphII gene insertion into each gene was confirmed by PCR with total genomic DNA as the template and the following specific primers: forward primer for the nrtP gene, 5'-AAC TTT GGC TCG GCA TTT TC; reverse primer for the nrtP gene, 5'-CAG CAC TTG GAA GAA GCC A; forward primer for the narB gene, 5'-ATT CCA GCG CAT CAG TTG; and reverse primer for the narB gene, 5'-CGG AAC TGG GGA CAA TAG (Fig. 1).

View larger version (31K): [in this window] [in a new window]   FIG. 1.   (A) Physical map of the 7.2-kb genomic region of the nrtP and narB genes in Synechococcus sp. strain PCC 7002. Arrows indicate the direction of transcription for the following open reading frames: sll1258, which is similar to hypothetical gene in Synechocystis sp. strain PCC 6803; petM, encoding a subunit of b6f complex; psbW2, encoding a subunit of photosystem II; merA, encoding mercuric reductase; and tmk, encoding thymidylate kinase. (B) The construct for the insertional inactivation of the nrtP gene in Synechococcus sp. strain PCC 7002. The kanamycin resistance gene cartridge (aphII gene) of 1.4 kb was inserted at the SpeI site in the nrtP gene with the same transcription orientation. The EcoRI-to-BglII DNA fragment was cloned between the EcoRI and BamHI sites of pUC19. Arrowheads indicate the positions of the primers for the PCR analysis. (C) The construct for insertional inactivation of the narB gene in Synechococcus sp. strain PCC 7002. The kanamycin resistance gene cartridge (aphII gene) of 1.4 kb was inserted into the XbaI site in the narB gene with the same transcription orientation. The HindIII-to-BamHI DNA fragment was cloned into the HindIII and BamHI sites of pUC19. Arrowheads indicate the positions of the primers for the PCR analysis. (D) Evaluation of gene replacement in the chromosomal DNAs of the nrtP mutant and the narB mutant by PCR analysis. Genomic DNAs from the wild type (lanes 1 and 3), the nrtP mutant (lane 2), and the narB mutant (lane 4) were used as the template for PCR with primers specific for the nrtP gene (lane 1 and 2) and for the narB gene (lanes 3 and 4). PCR products were separated by agarose gel electrophoresis and detected by ethidium bromide staining.

Nitrate consumption by whole cells (ammonia evolution). The rate of nitrate consumption by whole cells was measured by ammonia release to the assay medium under illumination in the presence of the GOGAT inhibitor DON. Cells were collected by centrifugation and washed twice with 25 mM HEPES-NaOH buffer (pH 7.0) containing 1 mM KCl; the cells were resuspended at a chlorophyll concentration of 25 or 50 µg ml¹ in 25 mM HEPES-NaOH buffer (pH 7.0) containing 1 mM KCl, 10 mM NaHCO₃, and 1 mM DON. The 2.5-ml sample in a 12- by 75-mm test tube was illuminated at 250 µE m² s¹ provided from a halogen floodlight after passage through a 5-cm layer of water as a heat filter, and the reaction temperature was maintained at 38°C by using a refrigerated, circulating water bath. The samples were continuously mixed with a magnetic stirrer. The reaction was initiated by the addition of 12 mM NaNO₃ to the assay medium. Aliquots of 200 µl were taken from the reaction mixture at 15-min intervals up to 60 min, and the formation of ammonia was determined. Ammonia concentrations were determined by the indophenol method. After cells were removed by centrifugation, 120 µl of the supernatant was transferred to a test tube, and 200 µl of phenol-nitroprusside solution (640-1; Sigma Chemical Co.) and 200 µl of alkaline hypochlorite solution (640-3; Sigma Chemical Co.) were added to the sample. The sample mixture was incubated at room temperature for approximately 20 min to allow color development. Distilled water (1.0 ml) was added, and the absorbance at 570 nm was measured with a spectrophotometer (model 14R; Cary, San Fernando, Calif.) modified for computerized data acquisition by On-Line Instrument Systems (Bogart, Ga.). The ammonia concentration was determined from a standard curve constructed with known NH₄Cl concentrations (0.1 to 1 mM) in the assay medium.

Nitrate consumption by whole cells (high-affinity uptake assay). High-affinity nitrate uptake was measured by monitoring the disappearance of nitrate with a nitrate-specific electrode as described previously (26). Cells were collected by centrifugation and washed twice with 25 mM HEPES-NaOH buffer (pH 7.0) containing 1 mM KCl. The cells were resuspended at a chlorophyll concentration of 10 µg ml¹ in 25 mM HEPES-NaOH buffer (pH 7.0) containing 1 mM KCl, 10 mM NaHCO₃, and 0.5 mM DON. Illumination (250 µE m² s¹) was provided from a halogen floodlight after passage through a 5-cm layer of water, and the temperature of the samples was maintained at 38°C by using a refrigerated, circulating water bath. The 25-ml sample was placed in a 50-ml jacketed beaker and was preilluminated for 15 min before the reaction was started. The assay was started by the addition of NaNO₃ to give an initial concentration of 100 µM in the assay medium. The disappearance of nitrate from the medium was directly monitored in real time with a model PHM240 ion meter (Radiometer, Westlake, Ohio) equipped with a model 9746 BN combination nitrate electrode (Orion, Beverly, Mass.). Changes in the electrode potential, and thus in the nitrate concentration, were monitored with a chart recorder.

Nitrite consumption by whole cells. The rate of nitrite consumption by whole cells was measured by the rate of disappearance of nitrite from the assay medium under illumination. Since nitrite can be taken up by passive diffusion of nitrous acid at neutral pH (7), assays performed at pH 7 reflect the in vivo activity of nitrite reductase, while assays performed at pH 10 reflect the active uptake of nitrite by cells. Cells were collected by centrifugation and washed twice with 25 mM HEPES-NaOH buffer (pH 7.0) containing 1 mM KCl; the cells were then resuspended at a chlorophyll concentration of 5 or 10 µg ml¹ in 25 mM HEPES-NaOH buffer (pH 7.0) or 25 mM CAPS [3-(cyclohexylamino)-1-propanesulfonic acid]-NaOH buffer (pH 10.0) containing 1 mM KCl, 10 mM NaHCO₃, and 0.5 mM DON. Illumination (250 µE m² s¹) was provided from a halogen floodlight, and the reaction temperature was maintained at 38°C by using a refrigerated, circulating water bath. The 5.0-ml samples in 13- by 100-mm test tubes were preilluminated for 15 min before the reaction was started. The assay was started by the addition of NaNO₂ to give an initial concentration of 100 µM in the assay. Aliquots of 350 µl were taken from the reaction mixture at 5-min intervals up to 20 min, and nitrite concentration in the medium was determined by the diazo coupling method. After cells were removed by centrifugation, 250 µl of the supernatant was transferred to a test tube, and 250 µl of 1% (wt/vol) sulfanilamide (S-9251; Sigma Chemical Co.) in 3 M HCl, 250 µl of 0.02% (wt/vol) N-(1-naphthyl)ethylenediamine (N-5889; Sigma Chemical Co.), and 0.4 ml of distilled water were added to the sample. After incubation at room temperature for color development, the absorbance at 540 nm was measured. The concentration of nitrite was determined from a standard curve constructed with NaNO₂ concentrations ranging from 1 to 100 µM.

RNA blot hybridization analyses, RT-PCR analysis, and 5' end mapping of nrtP mRNA. RNA blot hybridization analyses were carried out as described by Sakamoto and Bryant (24). Total RNA (10 µg) was fractionated by agarose gel electrophoresis and transferred onto a nylon membrane (Nytran+; Schleicher & Schuell, Keane, N.H.). A 1.3-kb DNA probe specific for the nrtP gene (positions 1641 to 2995 of the nucleotide sequence of the 7.2-kb region described above) was prepared by restriction digestion with NdeI and HindIII (Fig. 1A), and a 1.2-kb probe for the narB gene (positions 4468 to 5659) was prepared by restriction digestion with EcoRV (Fig. 1A). RNA blots were hybridized with the ³²P-labeled probes. The membranes were exposed to Kodak X-Omat AR X-ray film with an intensifying screen (Cronex Lightning-Plus; DuPont, Wilmington, Del.) for 1 day to detect the nrtP transcripts and for 5 days to detect the narB transcripts. The sizes of the transcripts were estimated from the sizes of the rRNAs, i.e., 23S rRNA (2.8 kb), 16S rRNA (1.5 kb), and 23S rRNA in vivo cleavage products (2.3 and 0.5 kb). Genomic DNA Southern blot hybridization analyses were performed under the same stringency conditions as for the RNA blot hybridization analyses to verify the specificity of these probes; the probes hybridized only with the DNA fragments with the size expected from the restriction map (data not shown).

Nucleotide sequence accession number. The nucleotide sequence of a 7,210-bp region carrying the nrtP and narB genes in Synechococcus sp. strain PCC 7002 has been submitted to GenBank (accession no. AF089813).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning and insertional inactivation of the nrtP and narB genes. We initially tried to isolate the nrtA and nrtC genes, encoding subunits of an expected ABC-type nitrate transporter from Synechococcus sp. strain PCC 7002. However, probes for nrtA and nrtC from Synechococcus sp. strain PCC 7942 and from Synechocystis sp. strain PCC 6803 did not hybridize with genomic DNA fragments of Synechococcus sp. strain PCC 7002 under low-stringency conditions (data not shown). These results suggested either that the nrtA and nrtC genes of Synechococcus sp. strain PCC 7002 were unexpectedly divergent in sequence from those of these related cyanobacteria or that Synechococcus sp. strain PCC 7002 in fact does not possess the genes encoding an ATP-binding cassette-type nitrate transporter. As described below, the latter appears to be the case. Since nitrate permease/transporter genes are rather poorly conserved in sequence but are often linked to the gene encoding nitrate reductase, we thus used the highly conserved narB gene of Synechocystis sp. strain PCC 6803 as a heterologous hybridization probe to isolate a cosmid clone carrying the narB gene and its flanking genes from Synechococcus sp. strain PCC 7002. A 7.2-kb region of this cosmid was sequenced, and the physical map of the sequenced region is shown in Figure 1A.

View larger version (100K): [in this window] [in a new window]   FIG. 2.   Sequence alignment of the NrtP protein of Synechococcus sp. strain PCC 7002 to other Nrt2-type nitrate transporter sequences. Amino acids identical and similar to those in the NrtP protein of Synechococcus sp. strain PCC 7002 are marked by shading. Amino acids that are identical in more than five of the nine proteins are shown in boldface. Twelve putative hydrophobic transmembrane regions are indicated by the numbered lines. Osci, partial sequence of the open reading frame in the 5' upstream region of the narB gene in Oscillatoria chalybea (accession number of the protein sequence, S57965); NrtP, Synechococcus sp. strain PCC 7002 (this study); Chlamy, Chlamydomonas reinhardtii (S40142); Rice, Oryza sativa (BAA33382); Arabi, Arabidopsis thaliana (CAB09794); Ynt1, Pichia angusta (CAA93631); CrnA, Aspergillus nidulans (P22152); NarK, Bacillus subtilis (P46907); NasA, B. subtilis (P42432).

Growth of the nrtP and narB mutants. Figure 3 shows the growth profiles of wild-type cells, the nrtP mutant cells and the narB mutant cells on urea and after transfer to nitrate growth conditions. Wild-type cells of Synechococcus sp. strain PCC 7002 could grow on either nitrate or urea as the sole nitrogen source, with a doubling time of 4 h on nitrate and 3.5 h on urea at 38°C under saturating light and carbon dioxide conditions. The growth rates of the nrtP and narB mutant strains on urea were identical to that of the wild-type strain (Fig. 3). However, when the nrtP mutant cells grown on urea were transferred to fresh medium A⁺ containing 12 mM nitrate as the sole nitrogen source, the growth rate decreased and reached steady-state conditions with a doubling time of 9 h (Fig. 3) approximately 4 to 8 h after the nitrogen source shift from urea to nitrate. The nrtP mutant cells continued to grow exponentially on nitrate with a doubling time of 9 h in the second subculture (Fig. 3), but the mutant cells had a chlorotic yellow-green coloration (data not shown). When the nitrate concentration in the growth medium was increased to 50 mM NaNO₃, the nrtP mutant cells still exhibited a chlorotic yellow-green coloration, and the doubling time (9 h) was identical to that of cells grown on 12 mM NaNO₃. These results indicate that the nrtP mutant has a much lower capacity for nitrate utilization for cell growth and that this limited ability to use nitrate limits the growth of the nrtP mutant cells. In the narB mutant cells, the growth rate dramatically decreased after the nitrogen source was changed from urea to nitrate (Fig. 3). The narB mutant cells could not grow at all after subculturing of the narB mutant cells a second time into fresh medium A⁺ (Fig. 3). These results indicate that the narB mutant cells are unable to utilize nitrate and that cell growth ceases after some initial scavenging of nitrogenous storage materials has occurred in the cells. The narB mutant cells could grow with a rate similar to that of the wild type in a medium containing 5 mM NaNO₂ as the sole nitrogen source. Both the wild-type and the narB mutant cells had a lower growth rate on nitrite than on nitrate, with a doubling time of about 7 h; moreover, a lag phase occurred after transfer of cells from urea growth conditions to nitrite growth conditions (data not shown). These results suggest that the wild-type and narB mutant cells of Synechococcus sp. strain PCC 7002 have similar capacities for nitrite utilization.

View larger version (19K): [in this window] [in a new window]   FIG. 3.   Growth curves for the wild-type strain, the nrtP mutant strain, and the narB mutant strain on urea and after transfer to nitrate growth conditions. Cells grown on medium AU10Ni containing urea as a nitrogen source were collected by centrifugation and washed with nitrogen-free medium. The cells were then inoculated into fresh medium A+ containing nitrate as a nitrogen source (open circles) and into medium AU10Ni containing urea as a nitrogen source (closed circles) and were cultured at 38°C under 250 µE m2 s1 with aeration with 1% CO2-enriched air. After growth in medium A+ for 24 h, the cells of the nrtP and narB mutants were diluted into fresh medium A+ and cultured under identical growth conditions (squares). The data shown are derived from a single experiment; however, this experiment was repeated three times, and essentially identical results were obtained in each case.

Nitrate and nitrite assimilation rates in the nrtP and narB mutants. Table 1 shows nitrate and nitrite assimilation rates in whole cells of wild-type, nrtP mutant, and narB mutant cells of Synechococcus sp. strain PCC 7002. Wild-type cells grown on nitrate had the maximum activities for ammonia evolution from nitrate, high-affinity nitrate uptake, and nitrite consumption. When cells were grown on urea, no nitrate consumption occurred and the rate of nitrite consumption was very slow in wild-type cells of Synechococcus sp. strain PCC 7002. These results suggest that the expression of the enzymes for nitrate uptake and reduction are suppressed when cells are grown on urea. When urea-grown cells were treated in A⁺ medium for 1 h under 250 µE m² s¹ at 38°C with aeration of 1% CO₂-enriched air, all activities of ammonia evolution, high-affinity nitrate uptake, and nitrite consumption were induced in the wild-type cells. In the nrtP mutant cells after a 1-h induction of these activities by incubation of cells under nitrate growth conditions, the reduction of nitrate to ammonia was detectable, but the rate of ammonia evolution was approximately 4.5-fold lower than that observed for wild-type cells. However, high-affinity nitrate uptake was not detectable in the nrtP mutant cells. These results indicate that the nrtP mutant cells cannot utilize nitrate as efficiently as the wild-type cells. Thus, the nrtP mutant can still take up and reduce nitrate when it is present at the concentration in the growth medium (12 mM), but the nrtP mutant does not exhibit nitrate uptake and consumption when the external concentration is only 100 µM. These data for the nitrate consumption rates in the nrtP mutant are consistent with the data for growth of this mutant on nitrate. The nitrite consumption rate of the nrtP mutant cells at neutral pH was similar to that of the wild-type cells, and this indicates that the in vivo activity of nitrite reductase in the mutant strain is similar to that in the wild type. In the narB mutant cells, no nitrate reduction occurred as assayed either by ammonia evolution or by the high-affinity uptake assay. These results again indicate that cells lacking nitrate reductase cannot utilize nitrate at all. The narB mutant cells did not consume nitrite as efficiently as the wild type, and the in vivo level of nitrite reductase in the narB mutant cells after a 1-h induction under nitrate growth conditions was similar to the level found in wild-type cells grown on urea. These results indicate that the narB mutant has a significantly lower nitrite reductase activity than the wild type or the nrtP mutant. Since the data for the nitrite consumption rate in the narB mutant was inconsistent with the data for growth of this mutant on nitrite, it was hypothesized that the product of nitrate reductase, nitrite, is a positive effector of the activity of nitrite reductase (see below).

High-affinity nitrite uptake in the nrtP mutant. Since nitrite can be taken up passively by the diffusion of nitrous acid at neutral pH (7), high-affinity uptake of nitrite at pH 10 in the nrtP mutant was measured to test the possible role of NrtP in nitrite uptake (Fig. 4). Wild-type cells consumed nitrite at both pH 7 and pH 10, with essentially identical rates of approximately 70 µmol mg of chlorophyll (Chl)¹ h¹. The nrtP mutant cells could consume nitrite with a rate nearly identical to that of the wild-type cells at pH 7 (Fig. 4), but at pH 10 the nrtP mutant cells could consume nitrite only very slowly (Fig. 4). The rate of nitrite consumption at pH 10 in the nrtP mutant was approximately 10 µmol mg of Chl¹ h¹ and was sevenfold lower than that in the wild type under these conditions. Although the nrtP mutant cells had a much lower rate of nitrite consumption at pH 10, nitrite was eventually completely consumed (data not shown). These results suggest that NrtP plays a specific role in the uptake of nitrite as well as nitrate. Moreover, these results indicate that nitrite uptake by Synechococcus sp. strain PCC 7002 takes place (i) by diffusion of nitrous acid at neutral pH; (ii) by high-efficiency uptake via the NrtP permease, and (iii) by low-efficiency but high-affinity uptake through another, unidentified transport mechanism(s).

View larger version (21K): [in this window] [in a new window]   FIG. 4.   Nitrite consumption by whole cells of the wild-type strain and the nrtP mutant strain of Synechococcus sp. strain PCC 7002. The disappearance of nitrite from the assay medium was measured for wild-type cells at pH 7, wild-type cells at pH 10, nrtP mutant cells at pH 7, and nrtP mutant cells at pH 10. The samples were preilluminated for 15 min under 250 µE m2 s1 at 38°C in the presence of the GOGAT inhibitor DON. At time zero, 100 µM NaNO2 was added to initiate the reaction. The assay medium contained 25 mM HEPES-NaOH (pH 7.0) or 25 mM CAPS-NaOH (pH 10.0), 1 mM KCl, 10 mM NaHCO3, 0.5 mM DON, 100 µM NaNO2, and cells equivalent to 10 µg of chlorophyll ml1. The data shown are the average values from two independent experiments.

Effect of nitrite on the in vivo activity of nitrite reductase in the narB mutant. In the narB mutant cells, the in vivo activity of nitrite reductase was as low as that of wild-type cells grown on urea (Table 1). Since the narB mutant cells could grow on nitrite with a 7-h doubling time, similar to that of the wild-type cells, the narB mutant was expected to have a nitrite reductase activity equivalent to that of the wild-type cells. It has been reported that nitrite positively regulates the transcription of the nirA-nrtABCD-narB operon in Synechococcus sp. strain PCC 7942 (12). Thus, the effect of exogenously added nitrite on the level of the nitrite reductase activity was tested with the narB mutant strain of Synechococcus sp. strain PCC 7002, in which no endogenous nitrite can be produced from nitrate reduction. Table 2 shows the effect of 1 mM NaNO₂ on the in vivo activity of nitrite reductase. When the narB mutant cells were grown on urea and incubated in a medium containing 1 mM NaNO₂ in the absence of other reduced nitrogen sources, the activity of nitrite reductase in the narB mutant was enhanced. After a 1-h treatment in medium A⁺, the activity of nitrite reductase was slightly enhanced, but the activity level was lower than that in cells treated with 1 mM nitrite. Only a basal level of nitrite reductase activity was detected in the narB mutant cells when they were treated with AU₁₀Ni medium for 1 h, and the addition of nitrite with urea did not increase the activity of nitrite reductase. These results indicate that nitrite plays a role as a positive effector of nitrite reductase in the absence of urea but that the presence of urea suppresses the activity of nitrite reductase.

Transcripts of the nrtP and narB genes. To investigate the nitrogen source dependence of expression of the nrtP and narB genes, total RNAs were isolated from wild-type cells of Synechococcus sp. strain PCC 7002 that had been grown on nitrate, urea, and ammonium, and these RNA samples were subjected to RNA blot hybridization analyses (Fig. 5). When a nrtP hybridization probe was used, a major and very strong hybridization signal corresponding to transcripts of about 2.0 kb was detected in RNA extracted from nitrate-grown cells (Fig. 5, panel nrtP, lane 1). However no transcripts hybridizing to the nrtP probe were detected in the RNA of cells grown on urea or ammonium (Fig. 5, panel nrtP, lanes 2 and 3). When the narB hybridization probe was used, a hybridization signal from transcripts of approximately 3.5 to 4.8 kb and a smear of hybridization to transcripts smaller than 2 kb (Fig. 5, panel narB, lane 1) were detected only in the RNA sample prepared from nitrate-grown cells and not in the RNAs of cells grown on urea or ammonium (Fig. 5, panel narB, lanes 2 and 3). Based on the signal intensities and the exposure times, it was estimated that the signal intensity of the 2.0-kb transcripts hybridizing to the nrtP probe was more than eightfold stronger than that of the 4.8-kb transcript hybridizing with the narB probe. To test further the possibility of dicistronic transcription of the nrtP and narB genes, an RT-PCR analysis was performed in an effort to detect RT-PCR products derived from the noncoding region between the nrtP and narB open reading frames. However, no amplification of this region was detected with an RNA template prepared from nitrate-grown cells (data not shown). The 5' endpoint of the mRNA product of the nrtP gene was determined by sequencing the RT-PCR product from cDNA that was synthesized with a specific oligonucleotide primer for the nrtP gene and that was subsequently tailed with a poly(C) tract by terminal transferase as described in Material and Methods (Fig. 6). The 5' end of the nrtP mRNA was mapped at position 1371 in the sequence under GenBank accession no. AF089813, which is located 226 nucleotides upstream of the putative initiation codon of the nrtP gene. The upstream flanking sequence of the 5' mRNA endpoint closely resembles a consensus, NtcA-dependent promoter sequence (8). This result strongly suggests that the nrtP gene is transcribed from an NtcA-dependent promoter.

View larger version (74K): [in this window] [in a new window]   FIG. 5.   Nitrogen source-dependent expression of the nrtP and narB genes in Synechococcus sp. strain PCC 7002. Total RNA (10 µg per lane) was isolated from cells grown on nitrate (lane 1), urea (lane 2), or ammonium (lane 3) as a sole nitrogen source and transferred to nylon membrane filters. The membranes were hybridized with probes specific for the nrtP or narB genes. After washing to remove the excess probe DNA, the membranes were exposed to X-ray films for 1 day for the nrtP gene and for 5 days for the narB gene. The 23S (2.8-kb) and 16S (1.5-kb) rRNA bands and the in vivo cleavage products of the 23S rRNA (2.3 and 0.5 kb) were detected as nonspecific hybridization background (arrows). The data shown are from a single experiment which was repeated twice, and identical results were obtained in each experiment.

View larger version (7K): [in this window] [in a new window]   FIG. 6.   The 5' endpoint of the nrtP mRNA and the putative NtcA-dependent promoter sequence of the nrtP gene in Synechococcus sp. strain PCC 7002. The 5' endpoint of the mRNA was determined by the sequence of RT-PCR products as described in Materials and Methods. The numbering for the sequence corresponds to the nucleotide numbering of the sequence under GenBank accession no. AF089813. The consensus sequence for an NtcA-dependent promoter is that described by Flores et al. (8).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Three distinct nitrate transport systems have been identified so far (for reviews, see references 3 and 20): the CHL1-type (35), the NRT2-type (36), and the ABC-type (22) nitrate transport systems. Based on their primary structures, these different nitrate transport systems are not evolutionarily related to one another (3). The CHL1 (NRT1) gene was originally identified in a chlorate-resistant mutant of Arabidopsis thaliana and encodes the dual-affinity nitrate transporter in plants (14). Expression of the CHL1 protein in Xenopus oocytes directly demonstrated the function of this gene product in nitrate transport (35). The NRT2-type transporters are high-affinity nitrate transporters that occur in wide range of organisms, including bacteria (17), fungi (23, 36), algae (10), and plants (3, 34). One transporter of this group, the Nrt2;1 transporter of Chlamydomonas reinhardtii, has been reported to be bispecific for nitrate and nitrite (10). Until this study, the ABC-type nitrate transporter was the only known high-affinity nitrate transporter in cyanobacteria. In the study presented here, the nrtP gene of Synechococcus sp. strain PCC 7002 was shown to encode a bispecific nitrate/nitrite permease of the NRT2 class. Since a partial open reading frame with very high sequence similarity to NrtP was found in the 5' upstream region of the narB gene in the cyanobacterium Oscillatoria chalybea (37), the NrtP permease will almost certainly prove to be more widely distributed among cyanobacteria. An important question that remains to be answered in future studies concerns the distribution of these two types of transporters and the possible physiological implications (if any) of this difference.

Based on the data presented here, the NrtP permease is involved in the active uptake of nitrate in Synechococcus sp. strain PCC 7002. Presumably, the energy for nitrate uptake can be provided by the proton gradient or proton motive force, as is the case in plants (3). It has been proposed that nitrate is taken up by a 2H⁺/NO₃ symport in plants (3), but the energy source (e.g., H⁺ versus Na⁺) for the nitrate uptake in cyanobacteria has not yet been directly characterized. In the ABC-type nitrate transporter in cyanobacteria, hydrolysis of ATP is presumed to provide the energy for the active uptake of nitrate (20), although coupling of ATP hydrolysis to transport has not been directly demonstrated yet. The discovery of the NrtP permease in Synechococcus sp. strain PCC 7002 provides an excellent model system for a combined biochemical and genetic analysis of the molecular mechanism of nitrate transport by the NRT2-type permeases that occur in a broad range of organisms, including higher plants.

Based on the genomic Southern hybridization analysis and the insertional inactivation of the nrtP gene, the nrtP gene appears to be a single-copy gene in Synechococcus sp. strain PCC 7002. Additionally, the data presented here suggest that the NrtP permease is the only high-efficiency transport system for both nitrate (Fig. 4; Table 1) and nitrite (Fig. 4) in this cyanobacterium. Since the in vivo activity of nitrite reductase was identical in both the wild-type and the nrtP mutant cells (Table 1; Fig. 4, pH 7), the results in Fig. 4 clearly show that NrtP is involved in the high-efficiency and high-affinity transport of nitrite at alkaline pH. Since the nrtP mutant could reduce nitrate to ammonia (Table 1) and grow on nitrate (Fig. 3), both nitrate and nitrite can also be taken up by low-efficiency transport mechanisms that have not yet been identified. Nonetheless, the data presented here clearly demonstrate that the NrtP permease is the primary, bispecific transporter for nitrate and nitrite in Synechococcus sp. strain PCC 7002.

Contradictory results have been reported concerning the specificity and mechanism of nitrite transport in Synechococcus sp. strain PCC 7942. Luque et al. (15) reported that insertional inactivation of the nrtD gene completely abolished active nitrite transport at pH 9.6. However, Maeda and Omata (16) reported that mutant cells harboring a deletion of the entire nrtABCD coding region could still consume nitrite at pH 9.6, albeit at a lower rate than wild-type cells. Our results show that a low rate of nitrite uptake (approximately 15% of the uptake rate observed for wild-type cells under the same assay conditions [Fig. 4]) occurs in the nrtP mutant cells of Synechococcus sp. strain PCC 7002 at pH 10.

The nrtP mutant cells could grow exponentially on nitrate but with a significantly lower rate than wild-type cells (Fig. 3), and whole cells of the nrtP mutant exhibited a lower rate of nitrate reduction to ammonia (Table 1). Although no direct experimental data have thus far been reported, it is likely that both nitrite and nitrate can also be taken up slowly by a transport system(s) for other ions (e.g., the bicarbonate transport system). Similar to the results presented here, Synechococcus sp. strain PCC 7942 mutants lacking a functional NrtABCD nitrate transporter could still grow on nitrate as the sole nitrogen source, and at high nitrate concentrations and under relatively low light intensity conditions, the mutants actually grew as rapidly as the wild-type strain (22). For a Synechococcus sp. strain PCC 7942 nrtD mutant, the growth rate increased as the concentration of nitrate was increased in the medium over the range of 2 to 40 mM KNO₃ (22). However, the growth rate of the Synechococcus sp. strain PCC 7002 nrtP mutant did not increase when the nitrate concentration in the growth medium was increased from 12 to 50 mM (at saturating light and CO₂ conditions) (data not shown). These results suggest that for Synechococcus sp. strain PCC 7002 the nitrate concentration (12 mM) in standard A⁺ medium is sufficient to saturate whatever transporter(s) are used. Moreover, this level of nitrate is sufficient, even in the absence of NrtP, to support exponential growth of the mutant cells that are otherwise supplied with sufficient light and other nutrients, although the nrtP mutant cells were chlorotic and had a lower growth rate under such conditions. The differences in the cell growth and saturation behavior in the various nitrate uptake mutants suggest that Synechococcus sp. strain PCC 7942 and Synechococcus sp. strain PCC 7002 may have different types of nitrate and nitrite transport systems for this slower, lower-efficiency uptake.

In Synechococcus sp. strain PCC 7942, the nirA-nrtABCD-narB genes are linked and regulated as a transcriptional operon, and the nirB-ntcB genes are linked to but divergently transcribed from the nirA gene (21). In Synechocystis sp. strain PCC 6803, the nrtABCD and narB genes are linked, but the nirA, ntcA, and ntcB genes are scattered about the genome (11). In Synechococcus sp. strain PCC 7002, the genes for nitrate assimilation are arranged quite differently. The nrtP and narB genes are linked, although the flanking genes (Fig. 1) are unlike those reported for Synechococcus sp. strain PCC 7942 (20) or Synechocystis sp. strain PCC 6803 (11). Although the genes involved in nitrate or nitrite uptake and assimilation seem to be scattered in the Synechococcus sp. strain PCC 7002 genome (29), the regulation of expression of these genes is apparently similar to that demonstrated in Synechococcus sp. strain PCC 7942 (21). The nrtP and narB genes were transcribed in nitrate-grown cells but not in urea- or ammonium-grown cells of Synechococcus sp. strain PCC 7002 (Fig. 5). These results strongly suggest that the nrtP and narB genes are probably coordinately controlled at the transcriptional level by the DNA-binding protein NtcA, which is the global transcriptional activator for the regulation of the genes for nitrogen metabolism in cyanobacteria (7). Consistent with the assumption of transcriptional regulation via NtcA, potential NtcA-dependent promoter sequence is found in the 5' flanking regions of the 5' end of the mRNA product of the nrtP gene (Fig. 6) and also in the 5' flanking regions of the narB gene (GTAATGGCGTTTTAC; nucleotides 3250 to 3264 in the sequence under GenBank accession no., AF089813). The presence of these two putative NtcA binding sites is consistent with our suggestion above that the nrtP and narB genes are independently transcribed from separate promoters. The activity of nitrite reductase is subject to dual-mode control. The absence of reduced nitrogen sources causes increased expression of nitrite reductase, but nitrite appears to enhance the expression level of nitrite reductase specifically as well (Table 2). These observations are consistent with a model that has been proposed for Synechococcus sp. strain PCC 7942 (12). We are currently isolating the genes involved in nitrate and nitrite assimilation, as well as the genes encoding the control elements for the expression of those genes involved in global nitrogen metabolism, from Synechococcus sp. strain PCC 7002 (29). Since the genes encoding the nitrate permease, nitrate reductase, urease (27, 28), and glutamine synthetase (38) are transcribed from independent promoters, it may be easier to analyze the transcriptional regulation of these genes than has been the case with some other cyanobacteria. Such comparative studies of the transcriptional regulatory circuitry in different cyanobacteria should provide insights into those mechanisms which are common to all cyanobacteria and those which may be distinctive for a given species.

It is thought that nitrate uptake is the rate-limiting step in the overall rate of nitrate assimilation, which includes nitrate uptake, nitrate reduction, and nitrite reduction to ammonia (7). This has specifically been shown to be true when Synechococcus sp. strain PCC 6301 is grown at low temperature (26). It has been shown that exogenously added ammonium ion arrests nitrate consumption within minutes in Anabaena cylindrica (19), Synechococcus sp. strain PCC 7942 (21), Synechococcus sp. strain PCC 6301 (5, 26), and Synechococcus sp. strain PCC 7002 (27). Thus, although the nitrate transport system for Synechococcus sp. strain PCC 7002 differs significantly from those of these other cyanobacteria, exogenously added ammonia appears to control its activity in a very similar fashion. Ohmori and Hattori (18) also demonstrated that the addition of ammonium ions caused a rapid and transient decrease in the ATP level within a few minutes, and those authors suggested that this could be due to the rapid consumption of ATP by glutamine synthetase because of the increased substrate level for this enzyme. Although this rapid decrease in the ATP pool could be a circumstantial event induced by the exogenously added ammonia, the resulting decrease in the ATP level could nevertheless have a major impact on decreasing active transport of nitrate and in the early events in the repression of nitrate assimilation by ammonia (18). No biochemical model that satisfactorily explains the rapid effect of exogenous ammonia in inhibiting nitrate consumption has yet been proposed (33). However, it was recently shown that the glnB gene product, the P_II protein, is involved in the regulation circuitry for the inhibition of nitrate or nitrite uptake by ammonium (13). It is assumed that such regulation of the nitrate transport system takes place to facilitate rapid responses to changes in nitrogen sources that are available in the environment. The discovery of the NrtP nitrate permease provides an important new tool for molecular biological approaches to analyses of the control mechanism(s) of the nitrate transport and reduction systems of cyanobacteria.