INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Prochlorococcus is a marine photosynthetic prokaryote ubiquitous in most intertropical areas of the oceans and is responsible for a significant part of the global primary production (for a review, see reference 27). Its unusual photosynthetic apparatus (12, 13, 15, 29) and wide genetic diversity have attracted increasing scientific interest in recent years. Its ability to tolerate a very wide light gradient has been linked to the co-occurrence in the field of two ecotypes with distinct irradiance optima for growth and photosynthesis (21, 40). While many reports have described these features in detail, little is known about other important aspects of Prochlorococcus metabolism, such as nutrient assimilation. In particular, the nitrogen assimilatory pathways in Prochlorococcus have not yet been studied. However, it is widely accepted that nitrogen is the main limiting nutrient in the upper layer of the oceans (27), and an understanding of the mechanisms involved in nitrogen assimilation could thus provide some keys to unveiling the remarkable ability of Prochlorococcus to colonize very oligotrophic regions. Nevertheless, such studies with Prochlorococcus face two problems: first, this organism is not easy to cultivate, and second, most isolated strains or clones contain contaminant heterotrophic bacteria. Only very recently, Rippka and coworkers described the first axenic strain, PCC 9511 (31), a typical high-light-adapted Prochlorococcus ecotype, allowing proper study of nonphotosynthetic metabolic pathways.

In the present work, we have studied the physiological response of glutamine synthetase (GS) in cultures of Prochlorococcus subjected to different conditions by measuring transferase and biosynthetic activities and antigen concentration. The standard nitrogen assimilatory pathway in non-nitrogen-fixing cyanobacteria is composed of a complex, highly modulated system of transporters, enzymes, and regulatory proteins (6). It allows the use of nitrate, nitrite, ammonia, or urea as a nitrogen source, among others. Urea and ammonia have routinely been used as nitrogen sources for Prochlorococcus (27, 31). GS catalyzes ammonium incorporation into glutamate and is responsible for most ammonium assimilation in photosynthetic organisms, including cyanobacteria (6). The GS-glutamate synthase cycle plays a crucial role in nitrogen assimilation, since most forms of nitrogen are first converted into ammonium before being further metabolized (6).

Here, we report on GS regulation in Prochlorococcus in order to understand its physiological behavior and how it is affected by nutrient limitation in the field. For this purpose, we subjected cultures of PCC 9511 to different conditions, focusing our attention on two parameters known to affect GS regulation: the nitrogen source and the energy state of cultures. We studied the effects of nitrogen and light limitations, as well as the effects of different nitrogen sources and of specific inhibitors blocking several metabolic steps of photosynthesis and nitrogen assimilation.

(Some of the results reported here were obtained during the Second International Workshop on Prochlorococcus, held in Roscoff, France, in 1999.)

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strains and culturing. Prochlorococcus sp. strains PCC 9511 (high irradiance adapted, axenic) and SS120 (low irradiance adapted) were routinely cultured in Nalgene polycarbonate flasks (10 liters) using PCR-S11 medium as described by Rippka and coworkers (31). The seawater used as a basis for this medium was kindly provided by the Station Biologique de Roscoff (Roscoff, France) and the Centro Oceanográfico de Fuengirola (Málaga, Spain). Cells were grown in a culture room set at 24°C under continuous blue irradiance (40 and 4 µE m² s¹ for PCC 9511 and SS120, respectively). All experiments were performed during the exponential phase of growth. Growth was determined by measuring the absorbance at 674 nm (A₆₇₄) of cultures.

In vivo experiments. Aliquots (250 ml) of cultures were taken at various times and centrifuged at 30,100 × g for 5 min in an Avanti J-25 Beckman centrifuge equipped with a JA-14 rotor. After most of the supernatant was poured off and the remaining medium was carefully pipetted out, the pellet was directly resuspended in 500 µl of cold 50 mM Tris-HCl (pH 7.5) and immediately frozen at 20°C until used for enzymatic or immunochemical analysis.

Enzymatic assays. GS transferase activity was determined as previously described (11) over 30 min at 37°C. The composition of the reaction mixture was changed slightly after characterization of the GS transferase activity in Prochlorococcus (S. El Alaoui, J. Diez, and J. M. García-Fernández, unpublished data) and contained the following: 100 mM glutamine, 10 mM sodium hydroxylamine, 50 µM manganese chloride, 10 µM ADP, and 50 mM sodium arsenate in 0.2 M morpholinepropanesulfonic acid (MOPS; pH 7). GS biosynthetic activity was determined basically as described by Marqués et al. (18). One unit of activity is the amount of enzyme that transforms 1 µmol of substrate/min.

Immunochemical techniques. Electroimmunoassay (EIA, or rocket immunoelectrophoresis) was performed essentially as described by Axelsen and Bock (1). Anti-GS rabbit antibodies (300 µl) from Synechocystis sp. strain PCC 6803 (produced as described previously [16]) were added to 15 ml of a solution containing 40 mM Tris (pH 8.6), 100 mM glycine, 600 mM calcium lactate, 1.5 mM sodium azide, and 1% agarose, previously melted and kept at 56°C for 15 min; this mixture was used to prepare the electrophoretic bed on a GelBond (FMC Bioproducts) film sheet. Due to the low GS concentration in crude extracts prepared for enzymatic activity determinations (see above), the extracts for EIA were concentrated fivefold (by resuspending the cells from 250 ml of culture in 100 µl of the same buffer instead of 500 µl). Ten microliters of extract prepared in this manner was loaded into each well. Electrophoresis was carried out overnight with a flat-bed apparatus (FBE 3000; Pharmacia) at 200 V and 10°C. Plates were washed with 15 mM NaCl for 48 h to remove non-cross-reacting proteins and then stained with Coomassie blue to detect immunoprecipitates. The chlorophyll concentration was used to standardize the measured values of the rocket areas.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Adaptation and setting of Prochlorococcus cultures. The Prochlorococcus strains used in this work were provided by the Station Biologique de Roscoff (Roscoff, France) and the Institut Pasteur (Paris, France). The considerable difficulty in obtaining laboratory cultures of Prochlorococcus resulted in the need for a long period of adaptation of the strains to the conditions of our culture room (several months) prior to experimental work. After an initial period of growth in the original medium (using seawater from Roscoff), the cells were progressively adapted to PCR-S11 medium prepared with oligotrophic seawater from the Mediterranean Sea (near Málaga, Spain), where the final cultures were established. The growth rate was measured by determining the A₆₇₄ (i.e., the red maximum in absorption spectra for Prochlorococcus), which was recently shown to be well correlated with the Prochlorococcus cell concentration (3); in fact, this correlation proved better than that of A₇₅₀ and cell concentration (data not shown). Under our culture conditions, a maximum A₆₇₄ of 0.2 was achieved in carboys of up to 20 liters, with a yield of ca. 100 mg of cells (fresh weight) per liter of culture; thus, physiological experiments could be performed with sufficient protein concentrations for enzymatic assays. Cultures with an A₆₇₄ of ca. 0.05 (exponential phase) were used to start all the experiments.

Nitrogen-mediated regulation. Exponentially growing Prochlorococcus strain PCC 9511 cultures in standard PCR-S11 medium (i.e., with ammonium as the sole nitrogen source) showed enzymatic activities of 2.5 to 5 U · mg of chlorophyll¹ for GS. Most of the total GS activity (>95%) in crude thawed extracts was detected in the supernatant after 5 min of centrifugation at 16,100 × g, indicating that GS is a soluble enzyme (data not shown). The biosynthetic activity was determined as well. A value of ca. 0.034 U · mg of chlorophyll¹ was obtained; this value is ca. 78-fold lower than the transferase activity.

View larger version (10K): [in this window] [in a new window]   FIG. 1.   Effect of nitrogen starvation on GS activity in Prochlorococcus strain PCC 9511. Samples were collected from a 10-liter culture growing on 400 µM ammonium at time 1, and the cells were resuspended in nitrogen-free PCR-S11 medium at time zero. Samples were taken at the indicated times for GS activity determinations. GS activity at time zero was 2.7 U · mg of chlorophyll1.

View larger version (14K): [in this window] [in a new window]   FIG. 2.   Long-term effect of N versus P starvation in Prochlorococcus. A Prochlorococcus strain PCC 9511 culture was centrifuged. The cells were washed with N- or P-free medium, divided into four fractions, and then resuspended in the following media: standard PCR-S11 (control for N starvation) (), N-free PCR-S11 (), standard PCR-S11 (control for P starvation) (), and P-free PCR-S11 (). GS activity at time zero was 1.22 U · mg of total protein1.

View larger version (13K): [in this window] [in a new window]   FIG. 3.   Time course of GS inactivation after the addition of MSX and azaserine to Prochlorococcus strain PCC 9511 cultures. Symbols: , control (no addition); , addition of 1 mM MSX; , addition of 100 µM azaserine. Inhibitors were added at time zero. GS activity at time zero was 3.27 U · mg of chlorophyll1.

Regulation through the energy state of cells. The GSs of most photosynthetic organisms are regulated by light (6). Hence, the effect of subjecting cultures to darkness or to the addition of specific inhibitors of electron transport, such as diuron [3-(3-4-dichlorophenyl)-1,1-dimethylurea; DCMU] and DBMIB, was investigated.

View larger version (12K): [in this window] [in a new window]   FIG. 4.   Effect of darkness on GS activity. A Prochlorococcus strain PCC 9511 culture growing in light was divided into two fractions; one was kept under the same irradiance (control, ), and the other was placed in darkness (). GS activity at time zero was 3.3 U · mg of chlorophyll1.

View larger version (13K): [in this window] [in a new window]   FIG. 5.   Effect of the addition of DCMU and DBMIB on the GS activity of Prochlorococcus strain PCC 9511. Samples were taken at the indicated times from cultures with no addition (control, ) or after 0.3 µM DCMU () or 0.06 µM DBMIB () addition. Inhibitors were added at time zero. GS activity at time zero was 3.66 U · mg of chlorophyll1.

EIA determination of GS concentration. It has been previously demonstrated that GS can be regulated at the level of enzyme concentration in other photosynthetic microorganisms, such as algae (9, 10), but it seems that GS concentration in cyanobacteria remains unchanged (19), regulation affecting only its activity and glnA expression. We have performed initial studies measuring the GS antigen concentration in Prochlorococcus samples by EIA in order to assess whether GS is regulated at this level. Table 2 shows the effects of different conditions on the GS protein concentration in Prochlorococcus strains PCC 9511 and SS120, as measured using antibodies raised against the GS of Synechocystis strain PCC 6803. The only clear change was observed after the addition of azaserine, which induced significant increases in the GS protein concentration in both strains. The effect of DBMIB on activity (48% reduction in SS120; data not shown) was not reflected in the enzyme concentration, which did not change.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Prochlorococcus is rapidly becoming a model organism because of its outstanding ecological success in oligotrophic areas of the oceans and its many unusual features (for a review, see reference 27). The U.S. Department of Energy has sequenced the genome of strain MED4, which is genetically very close, if not identical, to that of strain PCC 9511, which was used in the present study. In spite of this growing interest, basic knowledge of important metabolic pathways in Prochlorococcus is very limited. However, among the major remaining mysteries of this organism are the physiological mechanisms that it has developed to be able to proliferate in environments with extremely low concentrations of nutrients (27).

Here we have focused on the regulation in Prochlorococcus of GSa pivotal enzyme in the nitrogen assimilatory pathway. Although most of the presented results are for strain PCC 9511, the same experiments were performed with strain SS120 (data not shown) as a representative of low-irradiance-adapted ecotypes. In all experiments, the results were similar to those obtained with PCC 9511, except for the effect of DBMIB, which provoked a 48% reduction in GS activity in SS120 after 24 h (data not shown). However, since SS120 is nonaxenic, this possible regulatory difference between high-light- and low-light-adapted ecotypes requires further confirmation after SS120 or another, equivalent strain is made axenic.

GS in eubacteria is a well-studied enzyme, constituting a classical and well-known example of complex regulatory cascades that inactivate the enzyme by adenylylation, depending on the carbon-nitrogen balance of the cells (34). Feedback inhibition by some amino acids and nucleotides has also been shown for cyanobacteria (5, 19, 23, 35). In addition, darkness (19) and ammonium (20) provoke the inactivation of GS in Synechococcus strain PCC 6301. A new kind of GS inactivation by protein-protein interactions that mediate the ammonium-induced inactivation of GS has been very recently observed in Synechocystis strain PCC 6803 (7); this inactivation is controlled by the NtcA system (8).

The first striking result of our studies is that GS activity is slightly decreased in Prochlorococcus strain PCC 9511 cells grown on ammonium and then transferred to PCR-S11 medium containing either nitrate, nitrite, or no nitrogen (Table 1). The standard response in most photosynthetic organisms (including cyanobacteria [20]) is an increase in enzymatic activity under conditions of nitrogen starvation. This unusual behavior could represent a specific adaptation of Prochlorococcus for colonizing very oligotrophic environments (27) in order to avoid the expensive production of GS protein when there is no nitrogen to assimilate. If continued N depletion occurs in nature, such a response could represent a selective advantage.

The fact that the studied Prochlorococcus strain did not grow on nitrate or nitrite (31; this work) is also a very uncommon and surprising situation in cyanobacteria, since in most cases both oxidized and reduced N sources can be assimilated through the pathway constituted by nitrate reductase, nitrite reductase, GS, and glutamate synthase (6). Our results (Table 1) suggest that transfer to nitrate or nitrite is in fact equivalent to nitrogen starvation in Prochlorococcus strain PCC 9511. A positive correlation has been shown between Prochlorococcus abundance and nitrate concentration in temperate areas (22), and nitrate addition was found to stimulate cell cycling of Prochlorococcus in the Mediterranean Sea in winter (42). Although nitrate could be reduced by coexistent bacteria prior to assimilation, as suggested by Rippka and coworkers (31), the possibility of nitrate assimilation in some Prochlorococcus strains remains openin particular with regard to low-light-adapted ecotypes, which inhabit an environment not limited in nitrate, and in view of the genetic diversity (13, 33, 40) and wide distribution (26) of this cyanobacterium.

These results suggest that the studied Prochlorococcus strain (and probably ecotype) has experienced pressure inducing the loss of unnecessary genes that became useless in an environment where oxidized forms of nitrogen (nitrate or nitrite) were extremely scarce. Hence, the lack of genes encoding a nitrate-assimilating system in certain strains could save energy, leading to some evolutionary advantages and contributing to the reported compactness of the Prochlorococcus genome (36). This fact could be directly related to the very small size of Prochlorococcus cells, which has been proposed to be one of the critical factors in its ecological success, since its high surface area-to-volume ratio represents an advantage for the uptake of nutrients in oligotrophic areas of the oceans (27, 36).

Some recent reports have pointed to phosphorus limitation in some oceanic regions (41) where low chlorophyll concentrations are detected in spite of the abundance of nitrogen (2). The finding of a stronger effect on GS activity of P starvation than of N starvation (Fig. 2) was unexpected, since it can be assumed that the lack of nitrogen should mainly affect enzymes involved in nitrogen assimilation. This result could indicate that P limitation is more stressing to Prochlorococcus than the absence of N. The effect of P starvation on the Prochlorococcus cell cycle is much more marked than is that of N depletion, since prolonged P starvation provokes a block in all phases of the cell cycle rather than at a specific point (25). Furthermore, upon P addition, cells which are blocked in DNA synthesis cannot restart cycling and die. The presence of the pstS gene (expressed only under conditions of P depletion [27]), however, demonstrates that Prochlorococcus possesses mechanisms of adaptation to P limitation.

The effect of MSX on GS regulation has been studied with other unicellular organisms, such as cyanobacteria (24) and green algae (9). This inhibitor provokes rapid inactivation in Anabaena sp. strain PCC 7120 (24) at concentrations lower than those required for inactivation in vitro (1 µM MSX induces a drastic decrease in GS activity in a few hours in vivo, whereas 100 µM is necessary for in vitro inactivation). This result is probably due to transport mechanisms producing a very high intracellular concentration of MSX (3 mM). However, we observed that the addition of 1 mM MSX to cultures induced only partial inactivation in Prochlorococcus strain PCC 9511 (Fig. 3). This result could be the consequence of inefficient transport of MSX into these cells.

Inhibition of glutamate synthase had the opposite effect; GS activity (Fig. 3) and GS protein level (Table 2) increased quickly in ammonium-grown Prochlorococcus cultures after the addition of 100 µM azaserine. Very similar results were found for Synechocystis strain PCC 6803 after the addition of azaserine to ammonium-grown cultures (20). Still, the overall situation is different, since GS from Synechocystis grown on nitrate is inactivated by ammonium (20), so that the addition of azaserine induces reactivation of the enzyme. These results strongly suggest that (i) the glutamine concentration is not involved in the regulation of GS in Prochlorococcus, as occurs in other cyanobacteria (20) but not in other organisms (4, 9), and (ii) the mechanism inducing the upregulation of GS after the addition of azaserine is present in Prochlorococcus (as evidenced by the similar responses of Prochlorococcus strain PCC 9511 and Synechocystis strain PCC 6803), although other regulatory systems are clearly different.

The small change induced by darkness in GS activity from Prochlorococcus (Fig. 4) represents another main difference in the regulation of this enzyme compared to its regulation in most photosynthetic organisms, the usual situation being that darkness promotes a marked decrease in GS activity (10, 19, 32, 37, 38). Prochlorococcus can grow at a depth of 200 m, receiving less than 0.1% of the surface irradiance. The extremely low irradiance reaching this habitat has induced low-light-adapted ecotypes to develop a variety of adaptation mechanisms, including a high chlorophyll b/chlorophyll a ratio (28), large amounts of very efficient antenna proteins (15, 29), and the multiplication of the pcb genes encoding such antennae (12). Other modes of acclimation to low light could be expected for Prochlorococcus metabolic pathways directly related to photosynthesis, such as nitrogen assimilation.

Our results obtained using antagonistic inhibitors of photosynthetic electron transport show that the addition of DCMU induced a reduction in GS activity, while DBMIB provoked no clear change. Since both products block electron flow (30, 39), inducing the plastoquinone pool to be fully oxidized (DCMU; blocking between the Photosystem II complex and the plastoquinone pool) or fully reduced (DBMIB; blocking between the plastoquinone pool and the cytochrome b₆f complex), we propose that the redox state of the plastoquinone pool (and not the electron flow in a general sense) acts as a physiological sensor for GS regulation in Prochlorococcus.

In conclusion, we have found some evidence in our studies on GS regulation that suggests the occurrence in Prochlorococcus of adaptation mechanisms in the nitrogen assimilation pathway which may play a role in its ability to survive in conditions of strong nutrient limitation.