INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

The photosynthetic purple bacterium Rhodospirillum rubrum is a free-living diazotroph, in which nitrogenase activity is regulated at the metabolic level by reversible covalent modification, ADP-ribosylation (17). Under nitrogen-fixing conditions, ammonium is assimilated via the glutamine synthetase-glutamate synthase pathway (2, 25), as in other diazotrophs. The regulation of glutamine synthetase in R. rubrum or other phototrophs has not been as extensively studied as regulation in the enteric bacteria, but interesting differences have been reported, indicating a more complex system.

In enteric bacteria, glutamine synthetase is one of the key enzymes in nitrogen assimilation, catalyzing the ATP-dependent formation of glutamine from NH₄⁺ and glutamate. The enzyme is a dodecamer of the glnA gene product and is subject to tight regulatory control at three different levels (22). In addition to feedback inhibition by several end products of nitrogen metabolism and transcriptional regulation by the Ntr system, enzyme activity is regulated by covalent modification. This process involves two bifunctional enzymes, uridylyltransferase (UTase) and adenylyltransferase (ATase), and the regulatory protein P_II, which works as a signal transducer between the two enzymes. UTase, encoded by glnD, senses the nitrogen level and catalyzes uridylylation and deuridylylation of P_II (14, 26). When the nitrogen level is high, P_II, a trimer of the glnB product (22, 31), is unmodified and stimulates ATase to catalyze the inactivation of glutamine synthetase by adenylylation. P_II-UMP is required for the deadenylylation activity of the enzyme, leading to activation of glutamine synthetase (a phosphorylysis reaction which requires P_i and produces ADP) (22). It has recently been demonstrated that ATase in Escherichia coli, encoded by glnE, consists of two homologous domains (8, 30). The N-terminal part is responsible for the deadenylylation activity which is activated by -ketoglutarate, dependent on P_II-UMP, and inhibited by P_II. The C-terminal part of ATase catalyzes the adenylylation of glutamine synthetase and is activated by glutamine and P_II (8).

Recently the glnK gene in E. coli, encoding a P_II homologue, was identified (29). The structure of GlnK is similar to that of P_II, and it can also be modified by reversible uridylylation catalyzed by UTase (29). GlnK can complement P_II and affects both ATase and NtrB (1). In several organisms, a second gene encoding a protein similar to P_II has now been identified (3, 4, 18, 19, 21), but in R. rubrum, it has not yet been found.

Among the anoxygenic photosynthetic bacteria, glutamine synthetase in R. rubrum has been most extensively characterized (6, 20, 25, 32). Studies at the metabolic level have demonstrated that the enzyme is regulated in a way partly similar to that of the E. coli enzyme; however, several interesting differences have also been observed (6, 20, 25, 32). Adenylylated glutamine synthetase from R. rubrum shows lower -glutamyltransferase activity in both the presence and absence of 60 mM Mg²⁺ (20), whereas for the adenylylated E. coli enzyme, a decrease in this activity is observed only in the presence of 60 mM Mg²⁺ (27).

In R. rubrum, glutamine synthetase can also be modified by ADP-ribosylation, but the effect of this modification has not yet been clarified and the enzyme(s) responsible for the modification has not yet been identified. Furthermore, each subunit can, at one time, be modified only by AMP or ADP-ribose (32). Glutamine synthetase also shows a switch-off behavior, which seems to be coordinated with nitrogenase activity in response to both ammonium and darkness (5, 20). Several in vivo studies have shown that the glutamine synthetase activity is recovered after ammonium switch-off and that this process is slow compared to inactivation (5, 12).

We have previously shown that P_II in R. rubrum is modified by reversible uridylylation in response to nitrogen status (13). P_II was demodified by the addition of ammonium, glutamine, or NAD⁺ and partially demodified by the addition of glutamate. Oxygen and/or darkness had no effect on uridylylated P_II. We also showed that the UTase in R. rubrum responds to the glutamine concentration (13).

In this communication, we report studies of the effects of P_II, P_II-UMP, glutamine, and -ketoglutarate on adenylyltransferase in extracts from R. rubrum cells grown under different conditions.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Purification of P_II and P_II-UMP. The glnB gene was PCR amplified with designed primers, including a NdeI site at the 5' end of the gene. The plasmid pMJET was constructed by using the NdeI and BamHI sites located downstream of glnB and cloning the product into the plasmid pET15b (Novagen). In this construct, the 5' end of glnB was fused in frame with a base sequence encoding a polyhistidine sequence at the 5' end and a thrombin restriction protease site (Leu-Val-Pro-Arg-Gly-Ser) at the 3' end. The construct was verified by sequencing.

Preparation of extracts. R. rubrum S1 cultures were grown diazotrophically to an OD₆₀₀ of 1.5 (13). Extracts were produced from cultures grown and treated in either of the following ways: (i) grown with N₂ as N source (N₂ grown), (ii) grown with N₂, then switched off with 10 mM NH₄Cl 30 min prior to harvest (switch off), or (iii) grown with N₂, followed by nitrogen starvation under argon for 6 h prior to harvest (N starved). Cultures were harvested by centrifugation, washed once anaerobically in buffer A (1 mM dithiothreitol, 1 mM MnCl₂, 100 mM Tris-HCl [pH 7.8]), and then frozen in liquid nitrogen in pellet form. Crude extracts were prepared anaerobically by osmotic shock (16). Unbroken cells were removed by centrifugation at 6,000 × g and the extract containing the soluble fraction and chromatophores was frozen in liquid nitrogen. In some experiments, chromatophores were removed by centrifugation at 100,000 × g for 90 min. Low-molecular-mass molecules were removed from this supernatant by gel filtration with PD10 columns (Pharmacia).

Enzyme assays. ATase activity was determined as the change in glutamine synthetase activity after incubation under different conditions. A 50-µl extract with a protein concentration of 13 mg/ml was mixed with effectors and diluted to a final volume of 100 µl. Ten-microliter samples were removed after 0, 10, 30 and 60 min, and glutamine synthetase activity was determined by the transferase assay described previously (13). In the inactivation (adenylylation) experiments, the mixture contained final concentrations of 2 mM ATP and 6 mM MgCl₂. The activation (deadenylylation) mixtures instead contained 10 mM P_i. Final concentrations of effectors like glutamine, P_II, P_II-UMP, and -ketoglutarate are given in the figure legends.

Immunoblotting of glutamine synthetase and P_II. Extracts were analyzed by SDS-10% PAGE, and glutamine synthetase was detected by immunoblotting (28), with a polyclonal antibody raised against the purified protein. P_II was analyzed by SDS-18% PAGE and detected with an antibody against a peptide identical to the N-terminal part of P_II, as described previously (13).

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

Purification of P_II and P_II-UMP. Extracts from 100 ml of E. coli Bl21(DE3)pLysS cells, harboring the pMJET plasmid, generated 2 mg of purified histidine-tagged P_II protein after one Ni²⁺-chelating affinity column step. The pure protein, before and after digestion with thrombin, was analyzed by SDS-18% PAGE (Fig. 1A). The identity of the protein was also verified with our P_II antibody (data not shown).

View larger version (68K): [in this window] [in a new window]   FIG. 1.   SDS-PAGE analysis of R. rubrum PII purified from E. coli. Lanes: A1, E. coli extract with induced PII; A2, purified histidine-tagged PII; A3, the same as A2, but with PII after thrombin digestion; B1, histidine-tagged PII-UMP; B2, mixture of histidine-tagged PII and PII-UMP; B3, histidine-tagged PII. PII-UMP was purified from E. coli treated with glutamine, methionine sulfoximine, and UTP, as described in the text. M, molecular weight markers. His-tag, histidine-tagged.

Effects on glutamine synthetase activity in extracts. Although some studies of the regulation of glutamine synthetase activity by reversible adenylylation in photosynthetic bacteria have been reported (11, 32), the role of P_II and the effectors has not been clearly established. The importance of determining the involvement of P_II, P_II-UMP, and other effectors in anoxygenic photosynthetic bacteria is further emphasized by the situation in Azospirillum brasilense, where neither P_II nor P_Z is essential for the reversible adenylylation of glutamine synthetase, although the degree of adenylylation responds to the status of nitrogen in that organism (3). In R. rubrum, a glnB mutant would be preferable to fully investigate the in vivo functions of P_II. Repeated attempts to generate glnB mutants have been unsuccessful, indicating that P_II could be essential for growth. Therefore, we have now used an in vitro system to study the actions of P_II, P_II-UMP, and other effectors on the adenylylation and deadenylylation of glutamine synthetase. The methods are similar to those used for extracting cells from enteric bacteria (23, 24); however, no glutamine synthetase was added, since the level of enzyme activity was sufficient in the R. rubrum extracts.

Adenylylation of glutamine synthetase in extracts. Changes in glutamine synthetase activity were studied in extracts from three different growth conditions (Fig. 2). Adenylylation was dependent on the addition of ATP and MgCl₂ and showed essentially the same characteristics in all three extracts used in the results depicted in Fig. 2, although the initial activity of glutamine synthetase was about half of that in the extract from switch-off cells. Adenylylation was stimulated by the addition of glutamine or by addition of (unmodified) P_II but was totally inhibited by -ketoglutarate; instead, a small increase in glutamine synthetase activity was observed. On the other hand, the addition of P_II counteracted the effect of -ketoglutarate. The addition of purified uridylylated P_II had neither a stimulating nor an inhibiting effect on the adenylylation reaction in any extract examined (data not shown). The observation that no inactivation was obtained when ATP and MgCl₂ were excluded from the extract is in agreement with the hypothesis that adenylylation is the cause of the lower activity of glutamine synthetase. Further support was obtained from SDS-PAGE and Western blotting of the incubation mixtures from the experiments shown in Fig. 2. In the extracts from N₂-grown and N-starved cells, glutamine synthetase was essentially in the unadenylylated, faster-migrating form (Fig. 3, lanes A1 and B1). After incubation with glutamine and MgATP, the adenylylated, slower-migrating form could be demonstrated (Fig. 3, lanes A3 and B3). In the switch-off extract, both forms were present in about equal amounts; after incubation in the presence of glutamine and MgATP, the adenylylated form increased (Fig. 3, lanes C1 and C3). Previous reports have shown that glutamine synthetase from R. rubrum can be modified by adenylylation and ADP-ribosylation (32), but we found no effect on glutamine synthetase activity when ATP was replaced with NAD⁺, the donor of the ADP-ribose moiety, with or without the addition of glutamine or P_II (data not shown).

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Adenylylation of glutamine synthetase in extracts. Extracts were supplemented with 2 mM ATP, 6 mM MgCl2, and other effectors (final concentrations as follows). , no addition; , 5 mM -ketoglutarate; , 5 mM -ketoglutarate and 0.6 µM PII; , 0.6 µM PII; , 10 mM glutamine and incubated at 30°C. Extracts are from an N-starved culture (A), an N2-grown culture (B), and a switch-off culture (C). The experiments were carried out several times. The data are from one representative experiment with duplicate samples. Glutamine synthetase activity is given in micromoles of -glutamyl hydroxamate/minute · milligrams of protein.

View larger version (34K): [in this window] [in a new window]   FIG. 3.   The two forms of glutamine synthetase in extracts, shown by SDS-10% PAGE and Western blotting. (A) N-starved cells; (B) extract from N2-grown cells; (C) extract from switch-off cells. Lanes: 1, no addition; 2, Pi added and extract incubated at 30°C for 1 h; 3, glutamine, MgCl2, and ATP incubated at 30°C for 1 h. GS, unmodified glutamine synthetase; GS-AMP, adenylylated glutamine synthetase.

View larger version (22K): [in this window] [in a new window]   FIG. 4.   The effect of -ketoglutarate on glutamine synthetase activity. The extract from a N2-grown culture was supplemented with 2 mM ATP, 6 mM MgCl2, and -ketoglutarate (final concentrations as follows). , no addition; , 0.2 mM; , 0.5 mM; , 1 mM; , 2.5 mM; , 5 mM. Extracts were incubated at 30°C. The experiments were carried out several times. The data are from one representative experiment with duplicate samples. Glutamine synthetase activity is given in micromoles of -glutamyl hydroxamate/minute · milligrams of protein.

View larger version (21K): [in this window] [in a new window]   FIG. 5.   The effects of PII and glutamine on glutamine synthetase activity. The extracts were supplemented by 2 mM ATP and 6 mM MgCl2. (A) Extract from N2-grown cells with 1 mM -ketoglutarate and PII (final concentrations as follows). , no addition; , 0.09 µM; , 0.17 µM; , 0.43 µM; , 0.85 µM; , 1.31 µM; , 1.97 µM; , 2.56 µM. The boxed insert shows glutamine synthetase activity after 10 min at different PII concentrations. (B) Extract from N-starved cells with 2 mM -ketoglutarate and glutamine (final concentrations as follows). , no addition; , 1 mM; , 5 mM; , 10 mM; , 20 mM. The boxed insert shows glutamine synthetase activity after 10 min at different glutamine concentrations. Experiments were carried out several times. The data are from one representative experiment with duplicate samples. Glutamine synthetase activity is given in micromoles of -glutamyl hydroxamate/minute · milligrams of protein.

Deadenylylation of glutamine synthetase in extracts. To study the deadenylylation (activation) of glutamine synthetase in R. rubrum, extracts from switch-off cultures were prepared and incubated in the presence of 10 mM phosphate. The addition of phosphate to the extracts was required, indicating that deadenylylation occurs as a phosphorylysis reaction, as in E. coli. The effects on deadenylylation by different effectors are shown in Fig. 6. Interestingly, the reaction was much slower than the adenylylation, -ketoglutarate had no stimulatory effect, and glutamine did not inhibit the reaction at the concentrations used. The addition of P_II-UMP or P_II-UMP together with -ketoglutarate did not change the rate of activation. It should be emphasized that there is no endogenous P_II-UMP in these extracts, as shown by SDS-PAGE and Western blotting.

View larger version (14K): [in this window] [in a new window]   FIG. 6.   Activation of glutamine synthetase in extracts. (A) Extract from switch-off culture supplemented by 10 mM Pi and effectors (final concentrations as follows). , no addition; , 10 mM -ketoglutarate; , 10 mM -ketoglutarate and 0.85 µM PII-UMP; , 0.85 µM PII-UMP; , 10 mM glutamine. (B) Extract from an N2-grown culture supplemented by 2 mM ATP and 6 mM MgCl2. After 20 min, the extract was divided into tubes containing 10 mM Pi and other effectors as in (A). The experiments were carried out several times. The data are from one representative experiment with duplicate samples. Glutamine synthetase activity is given in micromoles of -glutamyl hydroxamate/minute · milligrams of protein.