P_II-like signalling molecules are trimeric proteins composed of 12–13 kDa polypeptides encoded by the glnB gene family. Heterologous expression of a cyanobacterial glnB gene in Escherichia coli leads to an inactivation of E. coli's own P_II signalling system. In the present work, we show that this effect is caused by the formation of functionally inactive heterotrimers between the cyanobacterial glnB gene product and the E. coli P_II paralogues GlnB and GlnK. This led to the discovery that GlnK and GlnB of E. coli also form heterotrimers with each other. The influence of the oligomerization partner on the function of the single subunit was studied using heterotrimerization with the Synechococcus P_II protein. Uridylylation of GlnB and GlnK was less efficient but still possible within these heterotrimers. In contrast, the ability of GlnB-UMP to stimulate the adenylyl-removing activity of GlnE (glutamine synthetase adenylyltransferase/removase) was almost completely abolished, confirming that rapid deadenylylation of glutamine synthetase upon nitrogen stepdown requires functional homotrimeric GlnB protein. Remarkably, however, rapid adenylylation of glutamine synthetase upon exposing nitrogen-starved cells to ammonium was shown to occur in the absence of a functional GlnB/GlnK signalling system as efficiently as in its presence.

Introduction

The glnB gene family encodes one of the most highly conserved signalling molecules characterized so far, the P_II protein. According to the rapidly expanding sequence information available, members of this family are found in all three domains of life and, as far as is known, they are all involved in nitrogen-related regulation (Merrick and Edwards, 1995; Hsieh et al., 1998). Moreover, genomic sequence information has revealed various bacterial and archaeal species containing multiple glnB-like sequences. Paralogues of P_II proteins were first identified in Escherichia coli and Azospirillum brasiliense (van Heeswijk et al., 1996; de Zamaroczy et al., 1996). In addition to the well-characterized P_II protein of Escherichia coli (hereafter termed GlnB for clarity) encoded by the glnB gene, the recently discovered paralogue GlnK is structurally very similar to GlnB, as revealed by crystallographic analysis of the two proteins (Jaggi et al., 1996; Xu et al., 1998). Both proteins are trimeric, and each subunit can be uridylylated at a tyrosyl residue (Tyr-51) (Son and Rhee, 1987) that is located at the tip of a solvent-exposed loop (T-loop) (Jaggi et al., 1996).

Signal transduction by GlnB has been studied for the last three decades (Shapiro, 1969; Brown et al., 1971; reviewed by Reitzer, 1996) and represents a paradigm for signal transduction cascades. Recently, a complete reconstitution of the GlnB signal transduction pathways with purified components was achieved (Jiang et al., 1998a, b, c). Depending on its uridylylation status, GlnB interacts with its receptors, NtrB and the adenylyltransferase/removase enzyme (ATase). NtrB is a sensor kinase that also transmits the GlnB signal to the response regulator NtrC; ATase regulates the activity of glutamine synthetase (GS) through adenylylation–deadenylylation. The P_II-modifying activity, a bifunctional uridylyltransferase/uridylyl-removing (UTase/UR) enzyme, responds to the cellular glutamine pool, which reflects the cellular N-status. High glutamine levels favour the uridylyl-removing activity, whereas at low glutamine levels, corresponding to nitrogen-poor conditions, the uridylyl-transferase activity dominates. In addition to signalling the glutamine status of the cell via uridylylation, GlnB and probably GlnK synergistically bind the small molecule effectors ATP and 2-oxoglutarate, thereby integrating an additional signal to the system (Kamberov et al., 1995; Xu et al., 1998; Jiang et al., 1998a). High 2-oxoglutarate concentrations, corresponding to relevant physiological conditions, can prevent the productive interaction of GlnB with its signal receptors (Jiang et al., 1998c). The precise function of the recently discovered GlnK protein has still to be established, although it has been shown that GlnK is able to interact with the GlnB receptors (van Heeswijk et al., 1996; Atkinson and Ninfa, 1998). In contrast to GlnB, which is constitutively produced, synthesis of GlnK is highly induced under nitrogen-poor conditions (van Heeswijk et al., 1996; Atkinson and Ninfa, 1998). GlnK of the related genus Klebsiella pneumoniae was recently shown to be involved in the control of nitrogen fixation (Jack et al., 1999). Other P_II-like proteins in proteobacteria have been shown to be modified by uridylylation, as in E. coli, and have been shown to be involved in different aspects of nitrogen regulation (de Zamaroczy, 1998).

A P_II-like protein with a different type of covalent modification has been discovered in cyanobacteria (Forchhammer and Tandeau de Marsac, 1994). The Synechococcus P_II protein (Sc P_II) can be phosphorylated at a seryl residue (Ser-49) located just two amino acids N-terminal to the conserved tyrosyl residue 51 (Forchhammer and Tandeau de Marsac, 1995a). Like the E. coli homologues, the cyanobacterial P_II protein is a trimer that binds ATP and 2-oxoglutarate in a mutually dependent manner (Forchhammer and Hedler, 1997). In contrast to E. coli, glutamine has no role in Sc P_II modification, but the phosphorylation state of Sc P_II depends primarily on the effector molecule 2-oxoglutarate (Irmler et al., 1997). The P_II protein was shown to be involved in the regulation of nitrate utilization (Forchhammer and Tandeau de Marsac, 1995b), presumably by controlling the nitrate uptake system (Lee et al., 1998).

In a previous investigation, we found that expression of the Synechococcus glnB gene in E. coli cells resulted in a profound perturbation of the P_II signalling system in E. coli (Forchhammer and Hedler, 1997). Cells synthesizing Sc P_II were unable to repress GS production under conditions of nitrogen excess. Phenotypically, these strains resembled a glnB null mutant. As this effect depended on NtrB, we suggested that Sc P_II either interacted directly with NtrB, thereby preventing regulation by the intrinsic E. coli P_II paralogues, or interacted with its E. coli homologues, thereby inactivating them. The fact that the cyanobacterial P_II protein interfered with the E. coli system indicated some fundamental conservation between these proteins from phylogenetically distinct organisms. Investigation of this process promised to provide new insights into P_II-mediated signalling systems.

Results

Investigation of a possible interaction between Sc P_II and NtrB

One possible explanation for the derepression of NtrC-dependent transcription caused by Sc P_II is the following: Sc P_II could bind to NtrB in such a manner that GlnB or GlnK are unable to induce NtrB-phosphatase activity. This possibility was tested by performing an NtrB-dependent phosphorylation/phosphatase assay using purified proteins. NtrB was incubated with the soluble receiver domain of NtrC (His₆–NtrC_R) together with radiolabelled [γ-³²P]-ATP until a constant level of phosphorylated NtrC_R was attained. Addition of purified GlnB led to the rapid dephosphorylation of phospho-NtrC_R, whereas the addition of Sc P_II had no effect on the level of phospho-NtrC_R (Fig. 1A). When Sc P_II was first added to the reaction mixture and GlnB was added subsequently, a rapid dephosphorylation of NtrC_R was observed, as in the absence of Sc P_II (Fig. 1B), indicating that Sc P_II was not able to inhibit NtrB-phosphatase activity directly. However, when GlnB was preincubated with Sc P_II for 20 min before addition to the phosphorylation reaction, dephosphorylation of NtrC_R-P induced by GlnB was partially inhibited (Fig. 1C).

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

. Dephosphorylation of NtrC-P by Sc P_II and GlnB. His₆–NtrC_R (1 μM) was phosphorylated in the presence of NtrB (100 nM) and [γ–³²P]-ATP (0.5 mM) at 37°C. After 23 min (A and B) or 25 min (C), the reaction was split and 7 μl of the following samples was added to each 63 μl sample. A. Sc P_II (10 μM) to a final concentration of 1 μM (▾), 3 μM GlnB to a final concentration of 300 nM (▴). B. Buffer containing 0.1 mg ml⁻¹ BSA (○); (▾) to 126 μl of the reaction mixture 14 μl of 10 μM Sc P_II was added (final concentration 1 μM) and, after incubation for another 20 min, 10 μl of 3 μM GlnB (to a final concentration of 300 nM) (▴). At the indicated time, 10 μl aliquots were removed, spotted onto glassfibre filters and analysed as described in *Experimental procedures*. C. Buffer containing 0.1 mg ml⁻¹ BSA (○),10 μM Sc P_II and 10 μM GlnB, incubated for 20 min at 37°C in buffer P containing 0.5 mM ATP before addition to the reaction mixture, to a final concentration of 1 μM each (◊).

Heterotrimerization of Sc P_II with GlnB and GlnK from E. coli

To analyse whether Sc P_II might interfere with the E. coli P_II system by the formation of non-functional complexes, an immunological study was performed. Previously, we raised antibodies against the Sc P_II protein that did not cross-react significantly with the homologous proteins from E. coli. To be able to detect the E. coli proteins, we now raised antibodies against a highly purified GlnK preparation (see Experimental procedures). The resulting antiserum was highly specific for both E. coli GlnK and GlnB but did not react with Sc P_II. With these tools, we performed co-immunoprecipitation experiments with α-GlnK serum using extracts derived from the E. coli glnB glnK double mutant HS9060 carrying plasmids that led to the co-expression of Synechococcus glnB with either glnB or glnK from E. coli or, as controls, strains expressing only one of these genes. The immunoprecipitates were analysed by SDS–PAGE and immunoblotting with α-GlnK or α-Sc P_II IgGs. If Sc P_II formed heteromers with the E. coli homologues, co-immunoprecipitation in extracts from strains co-expressing the respective genes should be expected. Figure 2 shows that this was indeed the case. Whereas no Sc P_II can be detected from the immunoprecipitate of the Synechococcus GlnB-overproducing strain, this protein immunoprecipitated with the GlnK/GlnB-specific serum when overproduced together with GlnB or GlnK. A reciprocal result was obtained when immunoprecipitation was performed with the α-Sc P_II IgGs: GlnB and GlnK were immunoprecipitated only when they were co–expressed with Sc P_II (data not shown).

To prove that the different P_II homologues form heterotrimeric structures, the migration of these proteins was analysed by immunoblot analysis after non-denaturing PAGE. This approach separates proteins of equal size according to charge differences and was used successfully to resolve different phosphorylated species of Sc P_II (Forchhammer and Tandeau de Marsac, 1994), uridylylated forms of GlnB (Atkinson et al., 1994) and heterotrimers composed of wild-type and mutant GlnB polypeptides (Jiang et al., 1997). First, extracts were analysed from cells that were grown under nitrogen-excess conditions (no P_II uridylylation) and that synthesized Sc P_II together with GlnB. In addition to the unmodified homotrimeric forms, there appeared two novel species, which migrated between the slower migrating homotrimeric Sc P_II and the faster migrating GlnB homotrimer (Fig. 3A). These intermediate forms reacted both with the α-Sc P_II IgGs and with the α-GlnK serum, which indicated that these species were indeed heterotrimers. From their electrophoretic property, the slower migrating novel species corresponds to a heterotrimer containing two subunits of Sc P_II and one subunit of GlnB, and the faster migrating species corresponds to a heterotrimer composed of one Sc P_II subunit and two GlnB subunits. A similar observation was made by analysing the Sc P_II and GlnK-overproducing strain (Fig. 3B). The heterotrimers appear as bands with an intermediate mobility between the two homotrimeric forms. A plasmid was constructed that leads to the production of all three homologues. In this case, all possible intermediate forms could be detected, including a putative Sc P_II/GlnB/GlnK heterotrimer (Fig. 3C).

Uridylylation of heterotrimers

In vivo modification of the different heterotrimeric molecules was analysed by non-denaturing gel electrophoresis of extracts from cells producing the molecular species of interest, followed by immunodetection of the different forms using a mixture of α-Sc P_II and α-GlnK IgGs to visualize all possible molecular species concomitantly. 4Figure 4A shows an experiment in which modifications of GlnB/Sc P_II heterotrimers were analysed. Extracts containing unmodified homotrimeric GlnB (lane 1), partially uridylylated GlnB (showing all possible modified species, lane 2) and the same extract after phosphodiesterase treatment (lane 3) were used as controls. Phosphodiesterase partially removed the faster migrating forms, which corresponded to trimeric GlnB containing one, two or three uridylylated subunits (Atkinson et al., 1994). Modification of GlnB/Sc P_II heterotrimers was analysed in HS9060 cells carrying plasmid pMP4 (Synechococcus glnB and E. coli glnB). In nitrogen-rich medium, the GlnB/Sc P_II heterotrimers were unmodified (lane 4), but 2 h after transfer to nitrogen-poor conditions (lane 5) and even more pronounced 16 h after the transfer (lane 6), faster migrating forms appeared that were as sensitive to phosphodiesterase as uridylylated GlnB (lane 7). Note that the heterotrimeric species were apparently modified more slowly than homotrimeric GlnB. A similar experiment was conducted to reveal modification of GlnK/Sc P_II heterotrimers (Fig. 4B). The control (lanes 1–3) shows modification and phosphodiesterase sensitivity of homotrimeric GlnK. Modification of GlnK/Sc P_II heterotrimers was analysed using cells of HS9060 carrying pMP5 (Synechococcus glnB and E. coli glnK ). Modification of these heterotrimers was already visible after 1 h in nitrogen-poor medium (lane 5) and was more pronounced after 2 h (lane 6). Note that the putative twofold uridylylated species of the (Sc P_II)₂(GlnK)₁ heterotrimer migrates at nearly the same position as unmodified GlnK (lane 6) but can be removed by phosphodiesterase treatment (lane 7). The same type of experiment was carried out in the UTase-deficient strain YMC26. There, the faster migrating bands did not appear during nitrogen starvation, further confirming that these bands correspond to uridylylated forms (data not shown).

Uridylylation of Sc P_II in E. coli

In a previous study, we have shown that Sc P_II, which was expressed in E. coli strain RB9060 (ΔglnB), could be uridylylated in a glnD-dependent manner under nitrogen-deficient conditions (Forchhammer and Hedler, 1997). The results shown above suggested that Sc P_II might have formed heterotrimers with GlnK, which were uridylylated on the GlnK subunit. Using the analytical method employed above, we could show that, indeed, heterotrimers formed between the chromosomally encoded GlnK and the plasmid-borne Sc P_II when cells were grown under nitrogen-deficient or nitrogen-poor conditions (Fig. 5A). This result emphasizes the dependence of glnK expression on the nitrogen status reported previously (van Heeswijk et al., 1996; Atkinson and Ninfa, 1998). To clarify whether homotrimeric Sc P_II can be uridylylated in E. coli, the GlnB/GlnK-deficient strain HS9060 carrying pMP1 was incubated in nitrogen-depleted medium or in nitrogen-poor GArg medium. Only after 3 h incubation in the nitrogen-free medium did a faster migrating species of Sc P_II appear (Fig. 5B, lane 2). This species disappeared after phosphodiesterase treatment (lane 5), which was in accordance with being a uridylylated form of Sc P_II. As a final proof that the Sc P_II protein was modified by uridylylation, an in situ [α-³²P]-UTP labelling experiment was performed. HS9060 cells carrying plasmids pMP2, pMP3 or pBluescript were labelled as controls. Cell lysates were separated by SDS–PAGE and blotted onto nitrocellulose. The blot was first autoradiographed (Fig. 5C, top), and then the position of GlnB, GlnK and Sc P_II was determined immunologically using a mixture of α-Sc P_II and α-GlnK IgGs (Fig. 5C, bottom). As shown in 5Fig. 5C, a weak label appeared at the position of Sc P_II, whereas the control reactions with GlnB or GlnK were much more strongly labelled, supporting the conclusion that Sc P_II is uridylylated, but with a very low efficiency.

Heterotrimerization of E. coli GlnB with GlnK

The results obtained in this study suggested that GlnB and GlnK could form heterotrimeric species in E. coli. To examine this prediction, a plasmid (pBK1) carrying both the glnK and the glnB gene was introduced in the UTase-deficient strain YMC26. This strain was chosen to avoid uridylylation, which could interfere with the interpretation of multiple bands in non-denaturing gels. As controls, YMC26 cells carrying pMP2 or pMP3 were used. Extracts of these cells were analysed by immunoblot analysis of non-denaturing gels (Fig. 6). Co-expression of glnK and glnB indeed resulted in the appearance of bands of intermediate mobility between homotrimeric GlnK and GlnB (lane 2). The mobility of these species was clearly different from the uridylylated GlnK species that were loaded in lane 4. Interestingly, YMC26 cells expressing only glnK from the plasmid (lane 3) exhibit a GlnK/GlnB cross-reactive material migrating like a putative (GlnB)₁(GlnK)₂ heterotrimer in addition to the dominant homotrimeric GlnK species. This species could be produced by heterotrimerization of GlnK with the chromosomally encoded, constitutively synthesized GlnB (Liu and Magasanik, 1993).

Effect of the P_II heterotrimers on glutamine synthetase adenylylation/deadenylylation

Heterotrimerization of chromosomally encoded GlnB or GlnK with overproduced Sc P_IIin vivo offers a new approach to investigate the regulation of ATase by P_II, in particular the role of trimerization. Under these conditions, the E. coli paralogues will be distributed preferentially into heterotrimers with Sc P_II. In a first series of experiments, we studied the deadenylylation of GS upon shifting cells from a nitrogen-excess medium (LBGln) to nitrogen-depleted (G-N) conditions. Cells that were grown in nitrogen-excess medium until mid-exponential phase (OD₆₀₀ = 0.4) were rapidly transferred to the N-depleted medium by filtration, and aliquots were removed at different times and assayed for GS adenylylation. The strains carrying the vector control showed, as expected, that GS deadenylylation depended on GlnB. As shown in Fig. 7, wild-type cells deadenylylated GS almost to completion within the first 5 min. In contrast, the glnB mutant RB9060 exhibited only a very slow deadenylylation of GS, and the double mutant HS9060 was apparently unable to activate GS by deadenylylation. Sc P_II could not complement the deadenylylation deficiency of this strain. On the contrary, expression of Synechococcus glnB slowed deadenylylation further in RB9060. The most pronounced effect of Synechococcus glnB expression was observed in the wild-type strain: GS deadenylylation was as slow as in the glnB-deficient strain. This suggested that GlnB-UMP, which was trapped in heterotrimers with Sc P_II, was unable to stimulate the adenylyl-removing activity of ATase.

By analogy with the above study, the effect of P_II heterotrimerization on GS adenylylation was investigated. As some of the strains used do not grow in the nitrogen-poor GArg medium, cells were first grown in LB medium and then transferred to nitrogen-free medium and incubated under these non-growing conditions. During prolonged nitrogen starvation, the adenylylation state of GS gradually decreased, even in those strains that were shown above to be unable to deadenylylate GS rapidly. After 3 h under nitrogen-depleted conditions, the extent of GS deadenylylation allowed a subsequent analysis of GS adenylylation. Adenylylation of GS was triggered by the addition of 5 mM ammonium chloride to the cultures and, at different times, the adenylylation state of GS was examined. In the wild type, GS was almost completely adenylylated within a few minutes after ammonium shock (Fig. 8A). Surprisingly, expression of Synechococcus glnB did not alter this response. Similarly, the glnB deletion strain responded like the wild type, regardless of whether Synechococcus glnB was expressed or not. This was a first indication that this process proceeded independently of a P_II system. To test this possibility further, the experiment was performed with the glnB glnK double mutant (Fig. 8B). In this strain, the initial GS adenylylation was significantly higher than in the other strains, probably because of the requirement of P_II for GS deadenylylation. Consequently, the amplitude of the increase in GS adenylylation was lower than in the other strains. Nevertheless, the response to ammonium was unambiguous: adenylylation of GS increased to maximal levels within the same time period as in the other strains. As a further confirmation that the observed rapid adenylylation of GS is independent of the P_II system, the ammonium shock experiment was carried out with a glnD mutant strain that lacks the P_II-modifying UTase/UR enzyme. Upon addition of ammonium, GS adenylylation increased as rapidly as in the wild type (Fig. 8C). Again, expression of Synechococcus glnB had no effect on this response.

Discussion

A surprising result of this investigation was the finding that the rapid adenylylation of GS upon exposing nitrogen-starved cells to ammonium did not require the GlnB/GlnK signalling system, which seems to disagree with the established model of GS regulation (Reitzer, 1996). Previously, it has been recognized that regulated adenylylation of GS was not completely impaired in a glnB-deficient mutant (Bueno et al., 1985; van Heeswijk et al., 1996); however, the remaining regulation was attributed to the recently discovered GlnB paralogue GlnK (Atkinson and Ninfa, 1998). Furthermore, it was known that the bifunctional ATase has a bias to adenylylate GS in the absence of P_II (Rhee et al., 1985; Jiang et al., 1998a); this intrinsic activity is responsible for the constitutively high degree of GS adenylylation in the glnB glnK double mutant, but it was not considered to be of physiological relevance with respect to rapid adaptation to ammonium shock. ATase was recently characterized in vitro (Jaggi et al., 1997) and was shown to be composed of two homologous domains, one of which catalyses the deadenylylation reaction and the other the adenylylation of GS. The isolated deadenylylation domain was shown to be regulated by the uridylylation status of P_II, whereas the isolated adenylylation domain was only regulated by glutamine but not by P_II. However, adenylylation of GS by the entire ATase required P_II, which was taken as evidence that the activity of the adenylylation domain is controlled by the binding of P_II to the deadenylylation domain. The in vitro study by Jiang et al. (1998c) concluded that GS adenylylation can be triggered by glutamine alone but is 15-fold accelerated in the presence of P_II and low concentrations of 2-oxoglutarate. We identify here for the first time a physiological condition under which the GlnB/GlnK signalling system is not required for the regulated adenylylation of GS, as shown by three lines of evidence. (i) The regulated GS adenylylation upon ammonium shock was not affected in a strain in which GlnB function was impaired through heterotrimerization with Sc P_II. (ii) The time course of GS adenylylation was not altered in the glnB /glnK double mutant strain. (iii) GS adenylylation was unimpaired in a UTase/UR-deficient strain, in which P_II cannot be uridylylated and, thus, cannot transmit the nitrogen signal to ATase. The fact that, in a glnD mutant strain, GS will be partially deadenylylated during prolonged nitrogen starvation has not been reported before but can be explained by the glutamine dependence of ATase activity (Rhee et al., 1978). As the internal glutamine pools drop during nitrogen starvation (Ikeda et al., 1996), ATase activity will decrease, and de novo synthesized GS escapes from adenylylation. After ammonium exposure of these cells, the deadenylylated GS could be adenylylated as fast as the wild type, although the nitrogen-sensing enzyme UTase/UR was absent. It is tempting to speculate that GS adenylylation was triggered directly by increased glutamine levels, as suggested by in vitro data (Jiang et al., 1998c). Depending on the particular growth conditions, there might be P_II-dependent and -independent pathways leading to GS adenylylation. Further studies are required to elucidate the relevant physiological conditions under which this occurs.

The present study also reveals another novel aspect of P_II-mediated signalling systems: the formation of heterotrimers between different P_II-like proteins. Heterotrimerization between P_II-like proteins from phylogenetically distinct organisms reveals that the structure of the oligomerization domain of these proteins is highly conserved. The high conservation of subunit interaction is probably an indication that it is of functional importance and may be involved in allosteric regulation, as discussed below. In a previous study, we have shown that the stoichiometry of metabolite binding increases with the metabolite concentration (Forchhammer and Hedler, 1997), suggesting antico-operative interactions between the subunits. A recent re-examination of the metabolite binding characteristics of GlnB reported strong anticooperativity between unmodified subunits of homotrimeric GlnB upon 2-oxoglutarate binding (Jiang et al., 1998a). The recently resolved crystal structure of E. coli GlnK with bound ATP revealed that the ATP binding sites were located in the clefts between adjacent subunits. The high conservation of quaternary structure between Sc P_II and GlnB/GlnK could thus be related to the highly conserved metabolite binding characteristics of these proteins.

A previous approach using mutant GlnB heterotrimers aimed to characterize single subunit interactions and the role of the T-loop in GlnB function (Jiang et al., 1997). In that study, the wild-type GlnB polypeptides that were distributed in heterotrimers with mutant polypeptides (defective T-loop structures) were not impaired in their activity towards UTase/UR and the P_II receptor proteins NtrB and ATase, suggesting that the P_II receptors interact with single T-loop structures. In contrast, the activity of the GlnB or GlnK polypeptides towards the P_II receptors NtrB and ATase (deadenylylating) was abolished in heterotrimers with Sc P_II. Heterotrimer formation with the evolutionary distinct Synechococcus P_II homologue could affect the T-loop structure of GlnB via allosteric interactions, thus preventing productive interaction with its receptors. If this were the case, the uridylylation of GlnB or GlnK subunits in heterotrimers with Sc P_II would indicate that UTase is less specific towards subtle changes in the T-loop structure. Indeed, Jiang et al. (1998a) showed that UTase, in contrast to NtrB or ATase, does not discriminate between the GlnB trimer liganded to one, two or three 2-oxoglutarate molecules. A lower specificity of UTase towards its substrate is also plausible, as this enzyme can to some extent uridylylate the only distantly related cyanobacterial P_II protein.

Evidence presented in this study show that GlnB and GlnK have the potential to form heterotrimers. If the formation of such heterotrimers occurs in the same way with the chromosomally encoded gene products, one has to conclude that the synthesis of homotrimeric GlnB is inversely correlated to the synthesis of GlnK, which is strongly induced under nitrogen-poor conditions. This conclusion is consistent with the recently reported dependence of GS deadenylylation on growth conditions. GlnK is much less efficient in stimulating GS deadenylylation than GlnB (van Heeswijk et al., 1996; Atkinson and Ninfa, 1998). Under growth conditions resulting in high-level expression of glnK (low N), the deadenylylation of GS in vivo is much slower than under conditions of glnK repression (N excess) (van Heeswijk et al., 1996). According to our interpretation, under low-nitrogen conditions, the highly active GlnB protein could be diluted into heterotrimers with the less efficient GlnK protein, resulting in GlnB/GlnK heterotrimers in which the GlnB subunit is less active, similar to the GlnB/Sc P_II heterotrimers. Consequently, organisms with multiple P_II paralogues could exhibit an additional level of regulation. Through differential expression of the genes encoding P_II-like proteins resulting in the formation of different heterotrimers, the efficiency of the P_II signalling system could be adjusted to various environmental conditions.

Transformation of E. coli cells with the Synechococcus glnB gene is an example of horizontal gene transfer, which results in novel regulatory capabilities of the recipient organism. By expressing the Synechococcus glnB gene, the cells are able to inactivate or modulate their P_II signalling system selectively. Such a process might have been a driving force in the evolution of organisms with multiple P_II proteins.

Experimental procedures

Bacterial strains and growth conditions

The bacterial strains and plasmids used in this work are listed in Table 1. The construction of HS9060 will be described in detail elsewhere. Briefly, in HS9060, glnK was deleted between the EcoRI site (located 22 bp upstream of the start codon for glnK ) and the ClaI site of glnK (bp 220), such that the σ⁵⁴-dependent promoter for the glnKamtB operon was retained. The chloramphenicol resistance gene was inserted between basepairs 3354 and 3365 of mdl (Allikmetz et al., 1993). Nitrogen-excess medium (LBGln) was Luria–Bertani medium (Sambrook et al., 1989) supplemented with 0.2% (w/v) glutamine; nitrogen-poor GArg medium was a 0.4% (w/v) glucose-containing minimal medium supplemented with 0.2% arginine, and GGln medium contained 0.2% glutamine, instead, as described by Bueno et al. (1985). The nitrogen-depleted medium (G-N) was the same glucose-containing minimal medium but lacking a nitrogen source. For maintenance of plasmids, all media were supplemented with 100 μg ml⁻¹ ampicillin. Cells were grown routinely at 37°C under aerobic conditions. Nitrogen stepdown experiments were performed by growing cells to an optical density (OD) at 600 nm of the culture of approximately 0.4. Cells from 20 ml of culture were harvested by filtering through a 0.45 μm polyvinylidene difluoride (PVDF) membrane filter (Millipore; type HVLP), washed with 20 ml of nitrogen-free G-N medium on the filter and finally resuspended in 20 ml of G-N medium, transferred back to sterile conical flasks and incubated under the previous conditions. The whole procedure was completed within 1 min and is thus more rapid than centrifugation and washing of cell sediments.

Table 1. . Escherichia coli strains and plasmids used in this study (Backman et al. (1981); Bueno et al. (1985); Forchhammer and Hedler (1997)).

Construction of plasmids

For the construction of plasmids pMP2, pMP3, pMP4, pMP5 and pMP6, the glnB and glnK coding regions together with the respective ribosomal binding sites were amplified by polymerase chain reaction (PCR) from E. coli chromosomal DNA using the following oligonucleotides: glnB5′ (5′-GAGGATCC AGATCAAAAGACAGGCGAC-3′); glnB3′ (5′-GCTCTAGAT CCGGCAACCCTTGACGC-3′); glnK5′ (5′-CTGGATCCATT ACCGAATTCTGACCG-3′); glnK3′ (5′-GTGGATCCTGTTG CTGTGTGCCAG-3′).

The E. coli glnB–Synechococcus glnB-carrying plasmid pMP4 was constructed by restricting the 0.4 kb glnB amplification product with BamHI and XbaI and cloning it upstream of the Synechococcus glnB gene of pMP1B into the BamHI/XbaI restriction sites, such that both genes are arranged in the same orientation. Transcription is driven from an inducible lac promoter, which is located in front of the multiple cloning site of the vector pBluescript SK (Stratagene) on which pMP1B is based. Similarly, the glnK–Synechococcus glnB-carrying plasmid pMP5 was constructed by restricting the 0.4 kb glnK amplification product with BamHI and cloning into the BamHI site of pMP1B, upstream of the Synechococcus glnB gene. Constructs with the glnK gene in the same orientation as the Synechococcus glnB gene were identified by restriction analysis. Plasmid pMP2, carrying only the E. coli glnB sequence, was derived from pMP4 by deleting the Synechococcus glnB gene. pMP4 was restricted with StyI (cuts in the 5′ terminal end of the Synechococcus DNA insert) and EcoRV. The 1.3 kb fragment of cloned Synechococcus DNA was eliminated, and the remaining 3.3 kb fragment containing E. coli glnB and the pBluescript sequences were religated. pMP3 was constructed by cloning the BamHI-restricted 0.4 kb glnK amplification product directly into the BamHI restriction site of pBluescript and screening for constructs with the glnK gene in the same orientation of transcription as the lac promoter. Plasmid pMP6 was derived from pMP4 by ligating the 0.4 kb BamHI-restricted glnK fragment from pMP3 to BamHI-restricted pMP4 and screening for constructs in which glnK has the same orientation as glnB. Plasmid pBK1 was derived from pMP6 by deleting the original Synechococcus DNA insert containing the Synechococcus glnB gene by StyI/XhoI digestion and religation of the remaining plasmid DNA. All constructed plasmids were checked by DNA sequencing. Standard cloning techniques were performed according to Sambrook et al. (1989).

For the construction of pHMS1, glnB was amplified by PCR from plasmid pGS27 using the primers 5′-CGGAAGCTTCATATGAAAAAGATTGA TGCGATT-3′ and 5′-CGGAATTCG GGCAACCCTTGACGC-3′. The fragment was cut with NdeI and EcoRI and ligated to the large EcoRI/NdeI fragment of pCYTEXP1 (Belev et al., 1991).

pFI27 (U. Fiedler) carries ntrC (bp 1–369) cloned into the BamHI site of pDS56/RII,6Hi (B. Bukau). The encoded His₆–NtrC_R comprises amino acids 1–123 from NtrC fused to an N-terminal His-tag with the sequence MRGSHHHHH HGSTQ.

Purification of GlnK and generation of a GlnK-specific antiserum

For the purification of GlnK, E. coli cells RB9060 carrying pMP3 were grown in 2 l of Luria broth supplemented with 100 μg ml⁻¹ ampicillin overnight at 37°C under aerobic conditions. All subsequent steps were performed at 0–5°C. Cells were harvested by centrifugation and washed once with 50 mM Tris-Cl, pH 7.4, containing 4 mM EDTA. The cell paste was resuspended with 20 ml of buffer A [50 mM Tris-Cl, pH 7.4, 50 mM KCl, 5 mM MgCl₂, 2 mM EDTA, 2 mM benzamidine and 1 mM Pefablock (Boehringer Mannheim)], and the cells were broken by three consecutive passages through a French press cell (110 kPa in⁻¹). Unbroken cells and debris were removed by centrifugation at 5000 × g and 20 000 × g respectively. The lysate was clarified by protamine sulphate precipitation (0.2% w/v) and was then fractionated by ammonium sulphate precipitation. The fraction between 30% and 50% ammonium sulphate saturation was collected and dissolved in 9 ml of buffer B [10 mM HEPES, pH 7,0, 50 mM NaCl, 0.4 mM EDTA, 1 mM MgCl₂, 1 mM dithiothreitol (DTT)] and dialysed against the same buffer. After centrifugation at 10 000 × g for 10 min, the soluble material was passed through a 10 ml heparin–agarose column (Bio-Rad). The column was developed with a 200 ml 50–400 mM NaCl gradient in buffer B, and the fractions containing GlnK were detected by immunoblot analysis using antibodies raised against Azospirillum brasiliense GlnB-derived oligopeptides (de Zamaroczy et al., 1996). Fractions containing GlnK were pooled, and the proteins were precipitated by the addition of ammonium sulphate to 50% saturation. The precipitate was collected, dissolved in buffer C (10 mM Tris-Cl, pH 7.4, 50 mM NaCl, 0.4 mM EDTA. 1 mM MgCl₂, 1 mM DTT) and passed through a Superdex 75 HR 16/60 FPLC gel filtration column (Pharmacia) equilibrated with buffer C. Fractions containing GlnK were detected as described above, pooled and further chromatographed on a 5 ml High Q anion exchange column (Bio-Rad) that was equilibrated with buffer C. The column was developed with a 100 ml gradient (50–400 mM NaCl). Fractions containing GlnK were pooled, and the material was finally purified by chromatography through a 5 ml High S cation-exchange column (Bio-Rad), which was equilibrated in buffer D (20 mM potassium phosphate, pH 7.4, 0.4 mM EDTA, 1 mM DTT). Elution was performed in the same buffer with a NaCl gradient of 0–500 mM. The final GlnK preparation was electrophoretically homogeneous according to a silver-stained SDS–PAGE analysis. The protein was dialysed against buffer D and concentrated by ultrafiltration. Approximately 0.4 mg of pure GlnK protein was sent to Eurogentec (Seraing) to raise a polyclonal serum in rats against GlnK.

Purification of GlnB, His₆–NtrC_R and NtrB

For the purification of GlnB, E. coli cells RB9060 carrying pHMS1 were grown at 28°C in LB medium supplemented with 0.2% (w/v) glutamine and 100 mg ml⁻¹ ampicillin to an OD₅₇₈ of 0.9. Overproduction of GlnB was induced by increasing the temperature rapidly to 42°C. After 3 h of incubation at 42°C, cells were harvested, resuspended in 50 mM Tris-Cl, pH 7.5, 1 mM phenylmethylsulphonyl fluoride (PMSF), disrupted in a French pressure cell and centrifuged at 16 000 × g. Streptomycin sulphate was added to the supernatant to a final concentration of 2% (w/v), followed by centrifugation at 12 000 × g for 30 min. Protein was precipitated from the resulting supernatant with ammonium sulphate at 50% saturation. After centrifugation, the precipitated protein was dissolved in 20 mM Tris-Cl, pH 7.5, containing 1 mM PMSF and applied to a heparin column (Heparin Sepharose CL6-B; Pharmacia) equilibrated in 50 mM Tris-Cl, pH 7.5, 50 mM NaCl and 1 mM EDTA. GlnB was eluted with a linear gradient from 50 mM NaCl to 600 mM NaCl. Pooled fractions were concentrated by precipitation with ammonium sulphate at 75% saturation and dialysed overnight against 30 mM Tris-Cl, pH 7.5, 70 mM KCl, 0.1 mM EDTA, 2 mM MgCl₂, 1 mM DTT and 1 mM PMSF.

For purification of His₆–NtrC_R, FI8202 (Fiedler and Weiss, 1995) carrying pFI27 was grown at 37°C in LB medium supplementedwith 0.2% (w/v) glutamine and 100 mg ml⁻¹ ampicillin. At an OD₅₇₈ of 0.6, overexpression of His₆–NtrC_R was induced with 1 mM IPTG. After 2 h at 37°C, cells were harvested, washed with 50 mM NaH₂PO₄, pH 8.0, 20 mM imidazole, 300 mM NaCl, 5 mM MgCl_2, 1 mM PMSF, disrupted in a French pressure cell and centrifuged at 4000 × g. The supernatant was applied to a Ni²⁺-NTA-agarose column (Qiagen). His₆–NtrC_R was eluted with a linear gradient from 20 to 250 mM imidazole. Pooled fractions were concentrated by precipitation with ammonium sulphate at 40% saturation. Purified His₆–NtrC_R was dialysed overnight against 50 mM Tris-Cl, pH 8.0, 300 mM NaCl and 5 mM MgCl₂.

NtrB was purified as described previously (Fiedler and Weiss, 1995).

Protein concentrations were determined from their absorbances at 280 nm using A^1% = 2.04 for GlnB, A^1% = 2.1 for Sc P_II, A^1% = 3.79 for NtrB and A^1% = 10.06 for His₆–NtrC_R, as calculated by the amino acid composition (Gill and von Hippel, 1989), and are expressed in the molarity of the trimer for GlnB and Sc P_II, in the molarity of the dimer for NtrB and in the molarity of the monomer for His₆–NtrC_R.

Immunological methods

Immunoprecipitation experiments were carried out as described previously (Forchhammer and Tandeau de Marsac, 1994). Analysis of P_II migration in non-denaturing gels was performed as described previously (Forchhammer and Tandeau de Marsac, 1994), except that the acrylamide concentration was 7.5% (w/v) instead of 10% (w/v). The proteins were blotted onto nitrocellulose (Optitran BA-S 83; Schleicher & Schuell), and the blots were probed with α-GlnK antibodies, which were subsequently visualized using an α-rat IgG peroxidase conjugate and a chemoluminescent substrate (Boehringer Mannheim). Subsequently, the bound antibodies were stripped from the membrane by incubating it in Tris-Cl, pH 6.8, 100 mM β-mercaptoethanol, 1% SDS (50 mM) for 20 min at 37°C. Then, the blot was washed twice in each case for 10 min in TBS and reprobed with the Sc P_II-specific antibodies.

In situ labelling of cells with [α³²P]-UTP

In situ labelling of cells with [α-³²P]-UTP was performed according to the procedure of Colonna-Romano et al. (1993) with the exception that the labelling buffer contained 1 mM 2-oxoglutarate and 0.2 mM ATP. After SDS–PAGE separation of the labelled extracts, proteins were blotted on nitrocellulose, and the blot was exposed to an X-ray film, followed by immunodetection of GlnB/GlnK and Sc P_II.

Glutamine synthetase (GS) assays

To follow precisely the rapid adenylylation/deadenylylation kinetics of GS, 2 ml samples were withdrawn from the shifted cultures into microcentrifugation tubes and chilled for 7 s in liquid nitrogen. Then, the cells were spun down, and the sediment was frozen in liquid nitrogen until the enzyme assay was conducted. GS activity and adenylylation state was determined principally as described by Pahel et al. (1982). The adenylylation state of GS was calculated from the quotient (Q) of GS (transferase) activity in the presence or absence of 67 mM MgCl₂ in the assay according to the equation given by Rhee (1984): n (number of adenylylated subunits of GS) = 12–13.95 × Q.

Coupled NtrB/NtrC phosphorylation-phosphatase assay

The coupled NtrB/NtrC phosphorylation-phosphatase assay was performed according to Ninfa and Magasanik (1986). His₆–NtrC_R (1 μM) was phosphorylated at 37°C in the presence of NtrB (100 nM) and [γ-³²P]-ATP (0.5 mM; specific activity 1 Ci mmol⁻¹) in buffer P (40 mM Tris-Cl, pH 7.5, 50 mM NaCl, 1 mM EDTA, 10 mM MgCl₂, 50 mM 2-oxoglutarate and 16% glycerol). After 25 min, the reaction was split, and samples containing Sc P_II, GlnB or BSA were added to the reaction mixtures, as described in the figure legends. At the indicated time, aliquots were removed and spotted on glassfibre paper (Whatman), immediately immersed in ice-cold 10% TCA containing 1% sodium pyrophosphate, and washed for 4 h in 5% TCA containing 1% sodium pyrophosphate. The amount of phosphorylated His₆–NtrC_R was determined by liquid scintillation counting.

Footnotes

†Institut für Mikrobiologie und Molekularbiologie, Justus-Liebig-Universität Giessen, Frankfurter Str. 107, D-35392 Giessen, Germany

‡Département de Biochimie et Génétique Moléculaire, Institut Pasteur, 28 Rue du Dr Roux, F-75724 Paris Cedex 15, France

Acknowledgements

We would like to thank U. Fiedler for plasmid pFI27, and G. Sawers for critical reading of the manuscript. This work was carried out in the laboratories of A. Böck and W. Boos and was supported by grants from the Deutsche Forschungsgemeinschaft to K.F. and to V. Weiss and H. Bujard.

References

1 Allikmets, R., Gerrard, B., Court, D., Dean, M. (1993) Cloning and organization of the abc and mdl genes of Escherichia coli: relationship to eukaryotic multidrug resistance. Gene 136: 231–236.
10.1016/0378-1119(93)90470-N
CAS PubMed Web of Science® Google Scholar
2 Atkinson, M.R. & Ninfa, A.J. (1998) Role of the GlnK signal transduction protein in the regulation of nitrogen assimilation in Escherichia coli. Mol Microbiol 29: 431–447.
10.1046/j.1365-2958.1998.00932.x
CAS PubMed Web of Science® Google Scholar
3 Atkinson, M.R., Kamberov, E.S., Weiss, R.L., Ninfa, A.J. (1994) Reversible uridylylation of the Escherichia coli P_II signal transduction protein regulates its ability to stimulate the dephosphorylation of the transcription factor nitrogen regulator I (NRI or NtrC). J Biol Chem 269: 8288–28293.
Google Scholar
4 Backman, K., Chen, Y.-M., Magasanik, B. (1981) Physical and genetic characterization of the glnA–glnG region of the Escherichia coli chromosome. Proc Natl Acad Sci USA 78: 3743–3747.
10.1073/pnas.78.6.3743
CAS PubMed Web of Science® Google Scholar
5 Belev, T.N., Singh, M., McCarthy, J.E. (1991) A fully modular vector system for the optimization of gene expression in Escherichia coli. Plasmid 26: 147–150.
10.1016/0147-619X(91)90056-3
CAS PubMed Web of Science® Google Scholar
6 Brown, M.S., Segal, A., Stadtman, E.R. (1971) Modulation of glutamine synthetase adenylylation and deadenylylation is modulated by metabolic transformation of the P_II-regulatory protein. Proc Natl Acad Sci USA 68: 2949–2953.
10.1073/pnas.68.12.2949
CAS PubMed Web of Science® Google Scholar
7 Bueno, R., Pahel, G., Magasanik, B. (1985) Role of glnB and glnD gene products in the regulation of the glnALG operon of Escherichia coli. J Bacteriol 164: 816–822.
CAS PubMed Web of Science® Google Scholar
8 Colonna-Romano, S., Patricia, E.J., Amar, M., Bernard, P., Manco, G., Lamberti, A., et al. (1993) Uridylylation of the P_II protein in Rhizobium leguminosarum. FEBS Lett 330: 95–98.
10.1016/0014-5793(93)80927-M
CAS PubMed Web of Science® Google Scholar
9 Fiedler, U. & Weiss, V. (1995) A common switch in activation of the response regulators NtrC and PhoB: phosphorylation induces dimerization of the receiver modules. EMBO J 14: 3696–3705.
10.1002/j.1460-2075.1995.tb00039.x
CAS PubMed Web of Science® Google Scholar
10 Forchhammer, K. & Hedler, A. (1997) Phosphoprotein P_II from cyanobacteria: analysis of functional conservation to the P_II signal transduction protein from Escherichia coli. Eur J Biochem 244: 869–875.
10.1111/j.1432-1033.1997.00869.x
CAS PubMed Web of Science® Google Scholar
11 Forchhammer, K. & Tandeau de Marsac, N. (1994) The P_II protein in the cyanobacterium Synechococcus sp. strain PCC 7942 is modified by serine phosphorylation and signals the cellular N-status. J Bacteriol 176: 84–91.
10.1128/jb.176.1.84-91.1994
CAS PubMed Web of Science® Google Scholar
12 Forchhammer, K. & Tandeau de Marsac, N. (1995a) Phosphorylation of the P_II protein (glnB gene product) in the cyanobacterium Synechococcus sp. strain PCC 7942: analysis of in vitro kinase activity. J Bacteriol 177: 5812–5817.
10.1128/jb.177.20.5812-5817.1995
CAS PubMed Web of Science® Google Scholar
13 Forchhammer, K. & Tandeau de Marsac, N. (1995b) Functional analysis of the phosphoprotein P_II from the cyanobacterium Synechococcus sp. strain PCC 7942. J Bacteriol 177: 2033–2040.
10.1128/jb.177.8.2033-2040.1995
CAS PubMed Web of Science® Google Scholar
14 Gill, S.C. & Von Hippel, P.H. (1989) Calculation of protein extinction coefficients from amino acid sequence data [published erratum appears in Anal Biochem 189: 283]. Anal Biochem 182: 319–326.
10.1016/0003-2697(89)90602-7
Web of Science® Google Scholar
15 Van Heeswijk, W.C., Hoving, S., Molenaar, D., Stegeman, B., Kahn, D., Westerhoff, H.V. (1996) An alternative P_II protein in the regulation of glutamine synthetase in Escherichia coli. Mol Microbiol 21: 133–146.
10.1046/j.1365-2958.1996.6281349.x
CAS PubMed Web of Science® Google Scholar
16 Hsieh, M.-H., Lam, H.-M., Van De Loo, F.J., Coruzzi, G. (1998) A PII-like protein in Arabidopsis: putative role in nitrogen sensing. Proc Natl Acad Sci USA 95: 13965–13970.
10.1073/pnas.95.23.13965
CAS PubMed Web of Science® Google Scholar
17 Ikeda, T.P., Shauger, A.E., Kustu, S. (1996) Salmonella typhimurium apparently perceives external nitrogen limitation as internal glutamine limitation. J Mol Biol 259: 589–607.
10.1006/jmbi.1996.0342
CAS PubMed Web of Science® Google Scholar
18 Irmler, A., Sanner, S., Dierks, H., Forchhammer, K. (1997) Dephosphorylation of the phosphoprotein P_II in Synechococcus PCC 7942: identification of an ATP and 2-oxoglutarate-regulated phosphatase activity. Mol Microbiol 26: 81–90.
10.1046/j.1365-2958.1997.5521918.x
CAS PubMed Web of Science® Google Scholar
19 Jack, R., De Zamaroczy, M., Merrick, M. (1999) The signal transduction protein GlnK is required for NifL dependent nitrogen control of nif gene expression in Klebsiella pneumoniae. J Bacteriol 181: 1151–1162.
Google Scholar
20 Jaggi, R., Ybarlucea, W., Chea, E., Carr, P.D., Edwards, K.J., Ollis, D.L., et al. (1996) The role of the T-loop of the signal transduction protein P_II from Escherichia coli. FEBS Lett 391: 223–228.
10.1016/0014-5793(96)00737-5
CAS PubMed Web of Science® Google Scholar
21 Jaggi, R., Van Heeswijk, W.C., Westerhoff, H.V., Ollis, D.L., Vasudevan, S.G. (1997) The two opposing activities of adenylyltransferase reside in distinct homologous domains, with intramolecular signal transduction. EMBO J 16: 5562–5571.
10.1093/emboj/16.18.5562
CAS PubMed Web of Science® Google Scholar
22 Jiang P., Zucker, P., Ninfa, A. (1997) Probing interactions of the homotrimeric P_II signal transduction protein with its receptors by use or P_II heterotrimers formed in vitro from wild-type and mutant subunits. J Bacteriol 179: 4354–4360.
10.1128/jb.179.13.4354-4360.1997
CAS PubMed Web of Science® Google Scholar
23 Jiang, P., Peliska, J.A., Ninfa, A. (1998a) Enzymological characterization of the signal-transducing uridylyltransferase/uridylyl-removing enzyme (EC 2.7.7.59) of Escherichia coli and its interaction with the P_II protein. Biochemistry 37: 12782–12794.
10.1021/bi980667m
CAS PubMed Web of Science® Google Scholar
24 Jiang, P., Peliska, J.A., Ninfa, A. (1998b) Reconstitution of the signal-transduction bicyclic cascade responsible for the regulation of Ntr gene transcription in Escherichia coli. Biochemistry 37: 12795–12801.
10.1021/bi9802420
CAS PubMed Web of Science® Google Scholar
25 Jiang, P., Peliska, J.A., Ninfa, A. (1998c) The regulation of Escherichia coli glutamine synthetase revisited: role of 2-ketoglutarate in the regulation of glutamine synthetase adenylylation state. Biochemistry 37: 12802–12810.
10.1021/bi980666u
CAS PubMed Web of Science® Google Scholar
26 Kamberov, E.S., Atkinson, M.A., Ninfa, A.J. (1995) The Escherichia coli P_II signal transduction protein is activated upon binding 2-oxoglutarate and ATP. J Biol Chem 270: 17797–17807.
10.1074/jbc.270.30.17797
CAS PubMed Web of Science® Google Scholar
27 Lee, H.-M., Flores, E., Herrero, A., Houmard, J., Tandeau de Marsac, N. (1998) A role for the signal transduction protein P_II in the control of nitrate/nitrite uptake in a cyanobacterium. FEBS Lett 427: 291–295.
10.1016/S0014-5793(98)00451-7
CAS PubMed Web of Science® Google Scholar
28 Liu, J. & Magasanik, B. (1993) The glnB region of the Escherichia coli chromosome. J Bacteriol 175: 7441–7449.
CAS PubMed Web of Science® Google Scholar
29 Merrick, M.J. & Edwards, R.A. (1995) Nitrogen control in bacteria. Microbiol Rev 59: 604–622.
10.1128/MMBR.59.4.604-622.1995
CAS PubMed Web of Science® Google Scholar
30 Ninfa, A.J. & Magasanik, B. (1986) Covalent modification of the glnG product, NRI, by the glnL product, NRII, regulates the transcription of the glnALG operon in Escherichia coli. Proc Natl Acad Sci USA 83: 5909–5913.
10.1073/pnas.83.16.5909
CAS PubMed Web of Science® Google Scholar
31 Pahel, G., Rothstein, D.M., Magasanik, B. (1982) Complex glnA-glnL-glnG operon of Escherichia coli. Proc Natl Acad Sci USA 76: 4544–4548.
10.1073/pnas.76.9.4544
PubMed Web of Science® Google Scholar
32 Reitzer, L.J. (1996) Ammonia assimilation and the biosynthesis of glutamine, glutamate, aspartate, asparagine, L-alanine and D-alanine. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. Neidhardt, F.C., Curtiss, III, R., Ingraham, J.L., Lin, C.C., Low, K.B., Magasanik, B. et al. (eds). Washington, DC: American Society for Microbiology Press, pp. 391–407.
Google Scholar
33 Rhee, S.G. (1984) 5′-Nucleotidyl-O-tyrosine bond in glutamine synthetase. Methods Enzymol 107: 183–200.
10.1016/0076-6879(84)07012-9
CAS PubMed Web of Science® Google Scholar
34 Rhee, S.G., Park, R., Chock, P.B., Stadtman, E. (1978) Allosteric regulation of monocyclic interconvertible enzyme cascade systems: use of Escherichia coli glutamine synthetase as an experimental model. Proc Natl Acad Sci USA 75: 3138–3142.
10.1073/pnas.75.7.3138
CAS PubMed Web of Science® Google Scholar
35 Rhee, S.G., Chock, P.B., Stadtman, E.R. (1985) Glutamine synthetase from Escherichia coli. Methods Enzymol 113: 213–241.
10.1016/S0076-6879(85)13032-6
CAS PubMed Web of Science® Google Scholar
36 Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Google Scholar
37 Shapiro, B.M. (1969) The glutamine synthetase deadenylylating enzyme system in Escherichia coli. resolution into two components, specific nucleotide stimulation, and cofactor requirement. Biochemistry 8: 659–670.
10.1021/bi00830a030
CAS PubMed Web of Science® Google Scholar
38 Son, H.S. & Rhee, S.G. (1987) Cascade control of Escherichia coli glutamine synthetase. J Biol Chem 262: 8690–8695.
CAS PubMed Web of Science® Google Scholar
39 Xu, Y., Chea, E., Carr, P.D., Van Heeswijk, W.C., Westerhoff, H., Vasudevan, S.G., et al. (1998) GlnK, a P_II-homologue: structure reveals ATP binding site and indicates how the T-loop may be involved in molecular recognition. J Mol Biol 282: 149–165.
10.1006/jmbi.1998.1979
CAS PubMed Web of Science® Google Scholar
40 De Zamaroczy, M. (1998) Structural homologs P_II and Pz of Azospirillum brasiliense provide intracellular signalling for selective regulation of various nitrogen-dependent functions. Mol Microbiol 29: 449–463.
10.1046/j.1365-2958.1998.00938.x
CAS PubMed Web of Science® Google Scholar
41 De Zamaroczy, M., Paquelin, A., Peltre, G., Forchhammer, K., Elmerich, C. (1996) Coexistence of two structurally similar but functionally different P_II proteins in Azospirillum brasiliense. J Bacteriol 178: 4143–4149.
CAS PubMed Web of Science® Google Scholar

Citing Literature

Volume33, Issue2

July 1999

Pages 338-349

Heterotrimerization of P_II-like signalling proteins: implications for P_II-mediated signal transduction systems

Abstract

Introduction

Results

Investigation of a possible interaction between Sc P_II and NtrB