The Novel PII-Interacting Protein PirA Controls Flux into the Cyanobacterial Ornithine-Ammonia Cycle

The Novel P II -Interacting Protein PirA Controls Flux into the Cyanobacterial Ornithine-Ammonia Cycle Paul Bolay, a Rokhsareh Rozbeh, b M. Isabel Muro-Pastor, c Stefan Timm, d Martin Hagemann, d Francisco J. Florencio, c Karl Forchhammer, b Stephan Klähn a a Helmholtz Centre for Environmental Research, Department of Solar Materials, Leipzig, Germany b Interfaculty Institute for Microbiology and Infection Medicine, Organismic Interactions Department, Tübingen University, Tübingen, Germany c Instituto de Bioquímica Vegetal y Fotosíntesis, CSIC-Universidad de Sevilla, Sevilla, Spain d Department of Plant Physiology, University of Rostock, Rostock, Germany Paul Bolay, Rokhsareh Rozbeh, and M. Isabel Muro-Pastor contributed equally. The order was determined by mutual agreement. ABSTRACT Among prokaryotes, cyanobacteria have an exclusive position as they perform oxygenic photosynthesis. Cyanobacteria substantially differ from other bacteria in further aspects, e.g., they evolved a plethora of unique regulatory mechanisms to control primary metabolism. This is exemplified by the regulation of glutamine synthetase (GS) via small proteins termed inactivating factors (IFs). Here, we reveal another small protein, encoded by the ssr0692 gene in the model strain Synechocystis sp. PCC 6803, that regulates flux into the ornithine-ammonia cycle (OAC), the key hub of cyanobacterial nitrogen stockpiling and remobilization. This regulation is achieved by the interaction with the central carbon/nitrogen control protein P II , which commonly controls entry into the OAC by activating the key enzyme of arginine synthesis, N-acetyl-L-glutamate kinase (NAGK). In particular, the Ssr0692 protein competes with NAGK for P II binding and thereby prevents NAGK activation, which in turn lowers arginine synthesis. Accordingly, we termed it P II - interacting regulator of arginine synthesis (PirA). Similar to the GS IFs, PirA accumulates in response to ammonium upshift due to relief from repression by the global nitrogen control transcription factor NtcA. Consistent with this, the deletion of pirA affects the balance of metabolite pools of the OAC in response to ammonium shocks. Moreover, the PirA-P II interaction requires ADP and is prevented by P II mutations affecting the T-loop conformation, the major protein interaction surface of this signal processing protein. Thus, we propose that PirA is an integrator determining flux into N storage compounds not only depending on the N availability but also the energy state of the cell. IMPORTANCE Cyanobacteria contribute a significant portion to the annual oxygen yield and play important roles in biogeochemical cycles, e.g., as major primary producers. Due to their photosynthetic lifestyle, cyanobacteria also arouse interest as hosts for the sustainable production of fuel components and high-value chemicals. However, their broad application as microbial cell factories is hampered by limited knowledge about the regulation of metabolic fluxes in these organisms. Our research identified a novel regulatory protein that controls nitrogen flux, in particular arginine synthesis. Besides its role as a proteinogenic amino acid, arginine is a precursor for the cyanobacterial storage compound cyanophycin, which is of potential interest to biotechnology. Therefore, the obtained results will not only enhance our understanding of flux control in these organisms but also help to provide a scientific basis for targeted metabolic engineering and, hence, the design of photosynthesis-driven biotechnological applications. KEYWORDS nitrogen metabolism, cyanobacteria, small inhibitory proteins, P II protein Citation Bolay P, Rozbeh R, Muro-Pastor MI, Timm S, Hagemann M, Florencio FJ, Forchhammer K, Klähn S. 2021. The novel P II - interacting protein PirA controls flux into the cyanobacterial ornithine-ammonia cycle. mBio 12:e00229-21. https://doi.org/10.1128/mBio .00229-21. Invited Editor Ray Dixon, John Innes Centre Editor Eduardo A. Groisman, Yale School of Medicine Copyright © 2021 Bolay et al. This is an openaccess article distributed under the terms of the Creative Commons Attribution 4.0 International license. Address correspondence to Stephan Klähn, [email protected]. Received 28 January 2021 Accepted 16 February 2021 Published 23 March 2021 March/April 2021 Volume 12 Issue 2 e00229-21 ®mbio.asm.org 1 RESEARCH ARTICLE Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. Nitrogen (N) is one of the key elements of life and needs to be incorporated into biomolecules via assimilatory pathways. Despite being an ever-present resource in the atmosphere, only a few bacteria can fix dinitrogen (N 2 ), and the majority rely on the uptake and assimilation of combined N sources from their environment (1–3). To respond to fluctuations in the availability of combined N sources, bacteria possess complex regulatory networks to control N uptake as well as the activity of assimilatory enzymes (for reviews, see references 4–7). As a prime example, glutamine synthetase (GS), a key enzyme of bacterial ammonium assimilation, is tightly regulated in a variety of ways. In Escherichia coli and other proteobacteria, the expression of the GS-encoding glnA gene is controlled at the transcriptional level by the widespread NtrC/NtrB twocomponent system (7). Moreover, GS is controlled at the activity level via cumulative feedback inhibition from numerous metabolites related to N and energy metabolism as well as by covalent modification, i.e., adenylylation of the GS subunits. This modification system is operated by a bicyclic modification cascade involving the ubiquitous P II signal transducer protein as a regulatory element (reviewed in reference 8). However, striking differences compared to widely accepted paradigms of N assimilation have been revealed in other bacteria, e.g., cyanobacteria. Cyanobacteria are the only prokaryotes performing oxygenic photosynthesis and play a major role in global biogeochemical cycles (9–14). Presently they are receiving growing interest as biocatalysts in photobiotechnological applications, e.g., for the sustainable production of valued chemicals and fuels (15–19). To rationally engineer cyanobacteria, i.e., channeling metabolic fluxes to obtain the maximum yield of a desired chemical product, it is of paramount importance to fully comprehend underlying regulatory processes targeting primary metabolism. Although our overall understanding of cyanobacterial systems is still fragmentary compared to other well-established bacterial models, a few systems have been extensively investigated and include distinctive features. For instance, GS activity is controlled via the interaction with small, inhibitory proteins unique to cyanobacteria (20, 21). These GS-inactivating factors (IFs) exclusively control GS activity linearly with their abundance. Moreover, with the global nitrogen control protein NtcA, cyanobacteria use another type of transcription factor to control the expression of genes in response to N fluctuation (22). NtcA belongs to the CRP transcriptional regulator family and commonly works as an activator of N assimilatory genes (23–26). During N limitation, increasing levels of 2-oxoglutarate and the coactivator protein PipX stimulate DNA binding of NtcA (27–29). The interaction between NtcA and PipX is antagonized by the P II protein, which acts as a global multitasking sensor and regulator, adjusting the carbon-nitrogen homeostasis through versatile protein-protein interactions (30, 31). This, for instance, includes the key enzyme for arginine synthesis, N-acetyl glutamate kinase (NAGK), which is activated by complex formation with P II (32). In addition to the activation of N assimilatory genes, NtcA can also act as a repressor of genes under N limitation. The physiological consequences of simultaneous positive and negative transcriptional regulation are again exemplified by the well-investigated GS regulatory system. Under N-limiting conditions, NtcA activates the transcription of the glnA gene, thereby increasing GS abundance and the rate of ammonium assimilation. Simultaneously, enhanced DNA binding of NtcA represses the transcription of the genes gifA and gifB, encoding the two known IFs, IF7 and IF17 (33). GS activity thereby is tuned in a trade-off between cellular N demands and relief from the metabolic burden imposed by the glutamateand ATP-consuming GS-catalyzed reaction (for a review, see reference 88). Besides gifA and gifB, only a few other genes appear to be negatively regulated by NtcA. In an attempt to define the entire regulon of NtcA in the unicellular model strain Synechocystis sp. strain PCC 6803 (here called Synechocystis), Giner-Lamia et al. identified the gene ssr0692 as another NtcA-repressed candidate (23). It encodes a small protein consisting of 51 amino acids with a high portion of N-rich, positively charged residues that were shown to be indispensable for protein-protein interaction in the case of the GS IFs Bolay et al. ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 2 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. (21). These distinguishing traits point toward a vital function related to N control similar to the known GS IFs, e.g., as a regulator of a metabolic pathway. Here, we report on the functional analysis of the small protein Ssr0692 in Synechocystis. It accumulates in response to ammonium supply and fulfills crucial regulatory roles in cyanobacterial metabolism via interaction with the P II signaling protein. We show that it directly interferes with the P II -dependent activation of NAGK. Consistent with this, under fluctuating N regimes, ssr0692 mutant strains are impaired in balancing the synthesis of arginine and other amino acids associated with the cyanobacterial ornithine ammonia cycle identified recently (34). Therefore, we named Ssr0692 the P II -interacting regulator of arginine synthesis (PirA). RESULTS Homologs of the pirA gene of Synechocystis are frequently present in cyanobacterial genomes and show a high degree of sequence conservation at the amino acid level (Fig. 1A and B). With only a few exceptions, sequences similar to that of PirA are absent from genomes of other bacterial phyla (as of July 2020, exceptions are “Candidatus Gracilibacteria bacterium,”Chloroflexaceae bacterium,Flavobacterium sp. strain CLA17, and Methylacidiphilales bacterium). At first glance, this observation suggests a function associated with oxygenic photosynthesis. However, pirA has previously been identified as part of the NtcA regulon in Synechocystis (23), consistent with two putative NtcA binding motifs located upstream of the transcriptional start site (TSS) (Fig. 1C). In promoters that are activated by NtcA, the respective binding motifs are centered close to position 241.5 with regard to the TSS (+1), bringing NtcA into a favorable position to promote the binding of RNA polymerase (35). However, the location of both motifs present in the pirA promoter is compatible with a repressive role of NtcA in the transcription of this gene. The proximal site, which is centered at 233.5 bp upstream of the TSS, is in a position very similar to the binding sites described for the well-characterized NtcA-repressed gifA-B genes (23, 33). NtcA binding in close proximity to the TSS would mediate repression by steric hindrance of the RNA polymerase interaction. This assumption is consistent with pirA downregulation under N limitation, similar to the gifA-B genes and in contrast to NtcA-activated genes such as glnA or nrtA, encoding GS and a nitrate transporter component, respectively (Fig. 1D). Interestingly, the distal site, centered at 248.5bp, could also interfere with the polymerase binding, specificallybypreventing the correct interaction of the carboxy-terminal domain of its alpha subunit with the promoter (36). The presence of two NtcA binding motifs probably contributes to a tighter control of pirA expression as a function of N conditions. PirA accumulates under N excess and is linked to a function in cyanobacterial N metabolism. Genes that are repressed by NtcA, such as the gifA-B genes, show low or even nondetectable transcription under N limitation but are highly expressed in response to excess N supply. To test whether this is also true for pirA,weprecultivated Synechocystis cells in the presence of nitrate and analyzed transcript levels after induction of N excess by adding 10 mM ammonium. As expected, the pirA mRNA strongly accumulated under these conditions (Fig. 1E). To investigate whether this regulatory pattern is conveyed to the protein level, we obtained an antibody specific to the PirA protein. Consistent with the observed transcriptional regulation, the PirA protein also accumulated in response to ammonium upshifts (Fig. 1F). Moreover, the protein appeared to have a high turnover because it eluded detection shortly after N was depleted (see Fig. S1 in the supplemental material). These observations clearly link PirA and its function to cyanobacterial N metabolism. To investigate the biological function of PirA, knockout and overexpression strains for the pirA gene were established in Synechocystis. The DpirA knockout mutant was generated by replacing the entire pirA open reading frame with a kanamycin resistance cassette via homologous recombination. In the case of the pirA 1 overexpression strain, a pVZ322 plasmid derivative harboring a transcriptional fusion of pirA with the Cu 21 -inducible petE promoter (PpetE) was transferred into the Synechocystis wild type (WT) (Fig. 2A). Full segregation of the mutant allele in the DpirA mutant as well as the Regulation of Cyanobacterial Arginine Synthesis ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 3 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. presence of the recombinant plasmid in the pirA 1 strain were verified by PCR (Fig. 2B). Subsequent Northern blot analyses with RNA isolated from cells grown in the presence of 1 m M CuSO 4 confirmed the generated mutants: the overexpression strain showed increased pirA mRNA levels compared to the WT, while in the knockout strain the pirA FIG 1 N-regulated gene pirA and its occurrence among cyanobacteria. (A) Amino acid alignment of randomly selected cyanobacterial PirA homologs. The alignment was made using ClustalW and visualized by using Jalview. (B) Phylogenetic tree of selected cyanobacteria based on 16S rRNA gene sequences. The tree was generated with the MEGA7 (83) software package and the neighbor-joining method. Note that we reused a calculated tree from our previous publication (38) and assigned the presence of genes in the corresponding genomes manually. Gene presence (illustrated by filled rectangles) was investigated using the BLASTP algorithm (84). As a reference, the amino acid sequences of PirA, IF7 (GifA, Ssl1911), and IF17 (GifB, Sll1515) from Synechocystis were used. (C) Overview of the promoter region upstream of the pirA gene in Synechocystis. Two putative NtcA binding sites are highlighted. The transcriptional start site (TSS; 11) and the location of the 210 element were extracted from differential transcriptome sequencing data (85). (D) Changes of mRNA levels for several Synechocystis genes in response to N limitation. Data were extracted and plotted from previously published microarray data (86). (E) Northern blot showing transcript accumulation of pirA in nitrate-grown Synechocystis cells upon addition of 10 mM ammonium chloride. 16S rRNA was used as a loading control. (F) Western blot showing changes in PirA protein levels in response to ammonium upshifts. For this, a specific, customized antibody against PirA was raised in rabbit. An antibody against thioredoxin (TrxA) was used to verify equal loading. Bolay et al. ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 4 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. transcript was absent (Fig. 2C). Interestingly, even though the mRNA was present and its abundance significantly increased in the pirA 1 strain due to the ectopic expression triggered by Cu 21 , the PirA protein could not be detected in nitrate-grown cells. However, after adding ammonium, which triggers the expression of the native pirA gene from the chromosome, increased PirA levels were detectable in the pirA 1 strain compared to the WT (Fig. 2D). This, in addition to the verified increase at the mRNA level, clearly confirmed that the overexpression construct is operative. Obviously, PirA abundance is not exclusively controlled at the transcriptional level. This observation was further supported by experiments using a pirA knockout mutant in which a PpetE-fused gene copy was introduced. As observed before, the PirA protein could not be detected after adding Cu 21 to nitrate-grown cells (Fig. 2E). Remarkably, its presence was still N dependent, similar to the WT, i.e., it was detectable only after adding ammonium (Fig. 2E), even though pirA transcription was controlled by PpetE and, hence, exclusively triggered by Cu 21 .Consequently, these data indicate an additional, posttranscriptional control mechanism, which obviously prevents stable PirA accumulation unless N availability suddenly increases. This again resembles the GS IFs encoded by the gifA-B genes, which are tightly regulated at the transcriptionalaswellasposttranscriptionallevel(37–39). PirA plays a critical role upon changes in the C/N balance. Under standard conditions, i.e., with nitrate as the sole N source and under ambient CO 2 , at which PirA is not FIG 2 Properties and expression profiles in DpirA and pirA 1 recombinant strains. (A) Schematic view of the pirA locus in the WT and in the DpirA knockout strain as well as of a pVZ322 plasmid derivative harboring a pirA gene copy under the control of the Cu 21 -inducible promoter PpetE that is present in the pirA 1 overexpression strain. In the DpirA knockout strain, pirA was replaced by a kanamycin resistance cassette (Km r ) via homologous recombination. The plasmid enabling ectopic pirA expression was introduced into Synechocystis WT. The arrows labeled with asterisks indicate the binding sites for primers used to verify the mutants. (B) PCR verification of the genotype of independently obtained mutant strains. In each case three clones were tested using primer combinations Ssr0692_KO-seg_fw/Ssr0692_KOseg_rev (in case of DpirA strain) or PpetE_fw(XhoI) and Toop_rev(AseI) (in case of pirA 1 strain). M, marker; bp, base pairs; cl., clone; 2, negative control (water as the template); 1, positive control (purified plasmid as the template). (C) Relative abundance of the pirA mRNA, measured via Northern blotting using sequence-specific 32 P-labeled ssRNA probes. In all cases, RNA was isolated from cells grown in the presence of 1 m M CuSO 4 . (D) Western blot showing PirA protein levels in cells of the WT and pirA 1 strains, treated with 1 m MCu 21 for 3 h and afterwards with 10mM ammonium. Thioredoxin (TrxA) levels verify equal loading. (E) PirA levels relative to WT. Data were obtained by densitometric evaluation of respective bands using the ImageJ software (87). Data are mean 6standard deviation (SD) values obtained from two independent Western blots, i.e., two biological replicates (independent clones). (F) PirA accumulation in a DpirA strain that was complemented with a pirA gene fused to the petE promoter. Note that the data shown here were obtained using a mutant in which the PpetE-pirA construct was integrated into the chromosome, i.e., this strain does not harbor the plasmid derivative shown in panel A. Regulation of Cyanobacterial Arginine Synthesis ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 5 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. detectable in the WT, the pirA-manipulated recombinant strains grew similarly to the WT, as expected (Fig. 3A and C). Given that PirA rapidly accumulated in response to increasing N availability, which suggests a function related to these conditions, it was tempting to speculate whether both recombinant strains show a phenotype, e.g., an affected pigment synthesis/degradation, when the N concentration is altered. To test this, we cultivated the WT, DpirA, and pirA 1 strains under N oscillating conditions. We inoculated cultures in nitrate-free BG11 and cultivated for 3 days, which was accompanied by pigment degradation (Fig. 3B and C), causing nitrogen starvation-induced chlorosis (40). Cultures of both recombinant strains showed the same behavior as the WT and did not show a nonbleaching (nbl) phenotype, as is known for nbl mutants that are affected in phycobilisome degradation (41). Consistent with this, the phycocyanin content was strongly reduced in all cells, measured by the diminished absorption at 630 nm (Fig. 3D, day 3). The similar bleaching kinetics of all strains is consistent with the fact that PirA is not detectable under N limitation. Afterward, the fully chlorotic cells were exposed to consecutive pulses of limited amounts of ammonium (1 mM) to simulate conditions under which PirA is rapidly accumulating and likely important. The regreening process was monitored by measuring growth as well as wholecell absorption spectra at wavelengths in a range between 400 and 750 nm. While growth recovery was rather similar in all three strains, a clearly altered pigmentation was observed in the pirA 1 strain after iterated ammonium pulses (Fig. 3C and D). Consistent with the visible difference, a lower absorption at 630 nm was detected, resulting from reduced phycocyanin content. These data indicate that the cells are impeded in coping with fluctuating N concentrations and struggle to recover from chlorosis when PirA accumulation is not correctly balanced. This supports the assumption that this small protein plays a crucial role and participates in regulatory processes that control N metabolism. Altered PirA abundance affects metabolites of N metabolism. To further examine a potentially regulatory function of PirA, time-resolved quantification of selected metabolites was performed for nitrate-grown cells of the WT and both mutants after addition of 10 mM ammonium. Interestingly, perturbation of PirA levels had a distinct FIG 3 Growth and pigmentation of the WT and the DpirA and pirA 1 mutant strains when N is oscillating. (A and B) Growth under standard conditions and when ammonium is consecutively added to N-starved cultures. Arrows indicate time points at which 1 mM NH 4 Cl was added. Data are the means 6SD from three independent cultures (including three independent clones of each mutant). (C) Representative photographs of cultures used in the experiment. Ammonium was added after day 3 and again after days 5 and 6. (D) Whole-cell absorption spectra. Values were normalized to A 750 values. Bolay et al. ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 6 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. impact on the accumulation of several key metabolites in Synechocystis. Most intriguingly, the kinetics of metabolites that are part of or are associated with the recently discovered ornithine-ammonia cycle (34) were strongly affected in DpirA and pirA 1 strains compared to the WT (Fig. 4; an extended data set is shown in Fig. S2). In general, N upshift triggered a transient accumulation of citrulline, ornithine, arginine, and aspartate in WT cells, similar to previous reports (34). Interestingly, the absence of PirA intensified and prolonged the accumulation of these metabolites, while the overexpression of pirA prevented or delayed their accumulation to a significant extent (Fig. 4). Moreover, kinetics of glutamine and glutamate, both key amino acids in N metabolism, showed striking differences between the tested strains. For instance, glutamate, which represents the main amine donor in a plethora of pathways, was significantly decreased in the pirA 1 strain throughout the experiment. FIG 4 Kinetics of metabolites linked to the OAC cycle in response to ammonium addition. (A) Simplified overview of metabolic pathways associated with ammonium assimilation and a possible regulatory impact of PirA on certain enzymatic reactions. 2-OG, 2-oxoglutarate; CP, carbamoyl phosphate; GS, glutamine synthetase; GOGAT, glutamine oxoglutarate aminotransferase; NAG, Nacetyl-glutamate; NAGK, N-acetyl glutamate kinase; OAC, ornithine-ammonia cycle. (B to G) Kinetics of selected metabolites after adding 10 mM ammonium to nitrate-grown cells in the exponential phase. Metabolites were determined by ultrahigh-performance liquid chromatography-tandem mass spectrometry after ethanol extraction from cells of the WT, DpirA,andpirA 1 strains. Data are the means 6SD from two independent experiments, each conducted with three biological replicates (independent clones). Significant differences in the mutant strains compared to WT at each time point are labeled and were revealed by one-way analysis of variance (ANOVA; *,P,0.05; **, P,0.01; ***,P,0.001). Regulation of Cyanobacterial Arginine Synthesis ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 7 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. The data clearly indicate that PirA plays a pivotal role in balancing fluxes through or into key amino acids, such as arginine. In cyanobacteria, the rate-limiting step of arginine synthesis is controlled by a well-investigated regulatory mechanism through complex formation of the key enzyme NAGK with the P II protein (31, 32, 42). Moreover, a P II variant with highly increased affinity toward NAGK (P II -I86N) causes constitutive NAGK activation and, hence, arginine accumulation (43), which was at least transiently observed in cells of the DpirA strain. Thus, it was tempting to speculate that PirA interferes at this regulatory node. PirA interacts with the signaling protein P II in an ADP-dependent manner. Recently, PirA was found enriched in pulldown experiments of the signaling protein P II (44). This indicated that PirA directly interacts with the P II protein and thereby exercises a regulatory function similar to other small P II interacting proteins, such as PipX (27) or CfrA/PirC, which has recently been discovered by two independent laboratories (45, 46). To verify the interaction between PirA and P II from Synechocystis,in vitro binding experiments were performed using biolayer interferometry (BLI). To this end, recombinant protein variants were expressed in and purified from E. coli. His 8 -tagged P II protein was immobilized on a nickel-nitrilotriacetic acid (Ni-NTA)-coated sensor tip, and a glutathione S-transferase (GST)-tagged PirA variant was used as the analyte in the presence or absence of various effector molecules (Fig. 5A). Indeed, complex formation was detected in the presence of ADP in a clear concentration-dependent manner (Fig. 5B). In contrast, no interaction was observed in the presence of ATP, mixtures of ATP and 2-oxoglutarate (2-OG), or when no effector molecule was present. These data unambiguously revealed ADP-dependent interaction between P II and GST-tagged PirA. To test the specificity of the interaction, we performed similar measurements using FIG 5 Determination of complex formation between PirA and the P II protein, measured by biolayer interferometry (BLI). (A) Schematic view of the measuring principle. (B) Representation of the maximum binding response of P II (WT)-His and GST-PirA interaction in the presence of different concentrations of ADP, ATP, or ATP/2-OG. (C) The maximum binding response at different protein concentrations of GST-PirA, tag-free PirA, or free GST in the presence of 2 mM ADP. As the binding response is a function of the mass of bound interactor, the response with GST-tagged PirA is correspondingly higher than that with isolated PirA peptide. (D) Representation of the maximum binding response at increasing concentrations of GST-PirA in the absence of effector molecules or in the presence of 2 mM ADP with three different T-loop variants of P II . Data are the means 6SD from triplicate measurements. Bolay et al. ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 8 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. PirA variants, where the GST tag was removed by proteolytic cleavage. The small PirA peptide yielded a binding signal that was about 6-fold lower than the signal observed for the GST fusion protein (Fig. 5C). This agrees well with the expected signal, since the BLI response depends on the mass changes at the sensor tip (GST-PirA versus PirA; 31.8 kDa/5.8 kDa = 5.5). Furthermore, BLI experiments with only the GST tag (26 kDa) did not result in any detectable signal (Fig. 5C), which clearly confirms that P II specifically interacts with PirA in these binding studies. Since the GST fusion protein is easier to handle accurately and the signal is superior to that of the isolated PirA peptide, further experiments were performed with GST-tagged PirA. To further study the P II -PirA interaction, different P II variants were examined. In most cases, interaction of proteins with P II involves the highly flexible T-loop structure that can adopt a multitude of conformations (30, 47). Accordingly, a P II variant lacking the T-loop (P II (DT)-His 8 ) was also tested. As expected, no response was observed, confirming interaction with PirA via the T-loop (not shown). Moreover, we tested the variant P II (I86N), where a single amino acid replacement, Ile86 to Asn86, locks the T-loop in a conformation that promotes constitutive NAGK binding (48, 49). Strikingly, this variant was not able to bind PirA, even in the presence of 2 mM ADP, which otherwise promotes binding to the native P II (Fig. 5D). In contrast, the phosphomimetic variant P II (S49E), which does not interact with NAGK (50), shows unaffected complex formation with PirA (Fig. 5D). The affinity of P II (S49E) to PirA was even slightly higher than that observed for the native variant [K D values, 2.9 60.34 m M for P II (WT) and 2.5 60.27 m M for P II (S49E)]. Obviously, a conformation of the T-loop that mediates a high P II affinity to NAGK prevents its interaction with PirA. Together with the metabolite profiles showing dysregulated arginine synthesis in the DpirA mutant, the present data implicate the interference of PirA with NAGK regulation through interaction with P II . PirA antagonizes P II -dependent activation of arginine-inhibited NAGK. To further demonstrate that PirA interferes with the P II -NAGK complex, an enzyme assay with purified components was conducted. Using the standard NAGK assay, where ADP formation is coupled to pyruvate-kinase and lactate-dehydrogenase activity (48), thereby keeping ADP concentrations at zero, no effect of PirA on P II -promoted activation of NAGK could be observed (not shown). Therefore, an assay was employed where NAG phosphorylation was coupled to subsequent NADPH-dependent N-acetylg -glutamyl5-phosphate reduction (51), allowing the addition of ADP. In this assay, the presence of P II protects NAGK from arginine inhibition, with 100 m M arginine fully discriminating free NAGK from P II -complexed NAGK (Fig. 6A). When the assay was performed in the presence of 0.1 mM arginine and 1 mM ATP, again no effect of adding increasing PirA concentrations could be detected. When, however, the same assay was performed in the presence of 1 mM ATP and 1 mM ADP, a concentration-dependent inhibition of NAGK activity of up to 50% could be detected, in accord with the ADP requirement for P II -PirA interaction (Fig. 6B). Altogether, the data point to a PirA-mediated disaggregation of the P II -NAGK complex that subsequently leads to NAGK inhibition by arginine. FIG 6 Impact of PirA on P II -dependent NAGK activity in vitro. (A) Inhibition of NAGK by arginine in the presence or absence of 2.4 m gP II . (B) NAGK activity as a function of increasing PirA concentration in the presence or absence of P II and 1 mM ADP. The assay otherwise contained 0.1 mM arginine and 1 mM ATP. Data are the means 6SD from triplicate measurements. Regulation of Cyanobacterial Arginine Synthesis ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 9 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. 26. Vega-Palas MA, Flores E, Herrero A. 1992. NtcA, a global nitrogen regulator from the cyanobacterium Synechococcus that belongs to the Crp family of bacterial regulators. Mol Microbiol 6:1853–1859. https://doi.org/10 .1111/j.1365-2958.1992.tb01357.x. 27. Espinosa J, Forchhammer K, Burillo S, Contreras A. 2006. Interaction network in cyanobacterial nitrogen regulation: PipX, a protein that interacts in a 2-oxoglutarate dependent manner with P II and NtcA. Mol Microbiol 61:457–469. https://doi.org/10.1111/j.1365-2958.2006.05231.x. 28. Forcada-Nadal A, Llácer JL, Contreras A, Marco-Marín C, Rubio V. 2018. The P II -NAGK-PipX-NtcA regulatory axis of Cyanobacteria: a tale of changing partners, allosteric effectors and non-covalent interactions. Front Mol Biosci 5:91. https://doi.org/10.3389/fmolb.2018.00091. 29. Vázquez-Bermúdez MF, Herrero A, Flores E. 2002. 2-Oxoglutarate increases the binding affinity of the NtcA (nitrogen control) transcription factor for the Synechococcus glnA promoter. FEBS Lett 512:71–74. https://doi.org/10.1016/ s0014-5793(02)02219-6. 30. Forchhammer K, Lüddecke J. 2016. Sensory properties of the P II signalling protein family. FEBS J 283:425–437. https://doi.org/10.1111/febs.13584. 31. Forchhammer K, Selim KA. 2020. Carbon/nitrogen homeostasis control in cyanobacteria. FEMS Microbiol Rev 44:33–53. https://doi.org/10.1093/ femsre/fuz025. 32. Heinrich A, Maheswaran M, Ruppert U, Forchhammer K. 2004. The Synechococcus elongatus P II signal transduction protein controls arginine synthesis by complex formation with N-acetyl-l-glutamate kinase. Mol Microbiol 52:1303–1314. https://doi.org/10.1111/j.1365-2958.2004.04058.x. 33. García-Domínguez M, Reyes JC, Florencio FJ. 2000. NtcA represses transcription of gifA and gifB, genes that encode inhibitors of glutamine synthetase type I from Synechocystis sp. PCC 6803. Mol Microbiol 35:1192–1201. https:// doi.org/10.1046/j.1365-2958.2000.01789.x. 34. Zhang H, Liu Y, Nie X, Liu L, Hua Q, Zhao G-P, Yang C. 2018. The cyanobacterial ornithine–ammonia cycle involves an arginine dihydrolase. Nat Chem Biol 14:575–581. https://doi.org/10.1038/s41589-018-0038-z. 35. Llácer JL, Espinosa J, Castells MA, Contreras A, Forchhammer K, Rubio V. 2010. Structural basis for the regulation of NtcA-dependent transcription by proteins PipX and P II . Proc Natl Acad Sci U S A 107:15397–15402. https://doi.org/10.1073/pnas.1007015107. 36. Browning DF, Busby SJW. 2016. Local and global regulation of transcription initiation in bacteria. Nat Rev Microbiol 14:638–650. https://doi.org/ 10.1038/nrmicro.2016.103. 37. Galmozzi CV, Fernández-Avila MJ, Reyes JC, Florencio FJ, Muro-Pastor MI. 2007. The ammonium-inactivated cyanobacterial glutamine synthetase I is reactivated in vivo by a mechanism involving proteolytic removal of its inactivating factors. Mol Microbiol 65:166–179. https://doi.org/10.1111/j .1365-2958.2007.05773.x. 38. Klähn S, Bolay P, Wright PR, Atilho RM, Brewer KI, Hagemann M, Breaker RR, Hess WR. 2018. A glutamine riboswitch is a key element for the regulation of glutamine synthetase in cyanobacteria. Nucleic Acids Res 46:10082–10094. https://doi.org/10.1093/nar/gky709. 39. Klähn S, Schaal C, Georg J, Baumgartner D, Knippen G, Hagemann M, Muro-Pastor AM, Hess WR. 2015. The sRNA NsiR4 is involved in nitrogen assimilation control in cyanobacteria by targeting glutamine synthetase inactivating factor IF7. Proc Natl Acad Sci U S A 112:E6243–E6252. https:// doi.org/10.1073/pnas.1508412112. 40. Görl M, Sauer J, Baier T, Forchhammer K. 1998. Nitrogen-starvationinduced chlorosis in Synechococcus PCC 7942: adaptation to long-term survival. Microbiology 144:2449–2458. https://doi.org/10.1099/00221287 -144-9-2449. 41. Collier JL, Grossman AR. 1994. A small polypeptide triggers complete degradation of light-harvesting phycobiliproteins in nutrient-deprived cyanobacteria. EMBO J 13:1039–1047. https://doi.org/10.1002/j.1460-2075.1994.tb06352.x. 42. Llácer JL, Fita I, Rubio V. 2008. Arginine and nitrogen storage. Curr Opin Struct Biol 18:673–681. https://doi.org/10.1016/j.sbi.2008.11.002. 43. Watzer B, Engelbrecht A, Hauf W, Stahl M, Maldener I, Forchhammer K. 2015. Metabolic pathway engineering using the central signal processor P II . Microb Cell Fact 14:192. https://doi.org/10.1186/s12934-015-0384-4. 44. Watzer B, Spät P, Neumann N, Koch M, Sobotka R, Macek B, Hennrich O, Forchhammer K. 2019. The signal transduction protein P II controls ammonium, nitrate and urea uptake in cyanobacteria. Front Microbiol 10:1428. https://doi.org/10.3389/fmicb.2019.01428. 45. Muro-Pastor MI, Cutillas-Farray Á, Pérez-Rodríguez L, Pérez-Saavedra J, Vega-de Armas A, Paredes A, Robles-Rengel R, Florencio FJ. 2020. CfrA, a novel carbon flow regulator, adapts carbon metabolism to nitrogen deficiency in cyanobacteria. Plant Physiol 184:1792–1810. https://doi.org/10 .1104/pp.20.00802. 46. Orthwein T, Scholl J, Spät P, Lucius S, Koch M, Macek B, Hagemann M, Forchhammer K. 2020. The novel PII-interacting regulator PirC (Sll0944) identifies 3-phosphoglycerate mutase (PGAM) as central control point of carbon storage metabolism in cyanobacteria. bioRxiv https://doi.org/ https://doi.org/10.1101/2020.09.11.292599. 47. Forchhammer K. 2008. P(II) signal transducers: novel functional and structural insights. Trends Microbiol 16:65–72. https://doi.org/10.1016/j.tim .2007.11.004. 48. Fokina O, Chellamuthu V-R, Zeth K, Forchhammer K. 2010. A novel signal transduction protein P II variant from Synechococcus elongatus PCC 7942 indicates a two-step process for NAGK–P II complex formation. J Mol Biol 399:410–421. https://doi.org/10.1016/j.jmb.2010.04.018. 49. Zeth K, Fokina O, Forchhammer K. 2012. An engineered P II protein variant that senses a novel ligand: atomic resolution structure of the complex with citrate. Acta Crystallogr D Biol Crystallogr 68:901–908. https://doi .org/10.1107/S0907444912016447. 50. Burillo S, Luque I, Fuentes I, Contreras A. 2004. Interactions between the nitrogen signal transduction protein P II and N-acetyl glutamate kinase in organisms that perform oxygenic photosynthesis. J Bacteriol 186:3346–3354. https://doi.org/10.1128/JB.186.11.3346-3354.2004. 51. Fokina O, Herrmann C, Forchhammer K. 2011. Signal-transduction protein P II from Synechococcus elongatus PCC 7942 senses low adenylate energy charge in vitro. Biochem J 440:147–156. https://doi.org/10.1042/BJ20110536. 52. Charlier D, Bervoets I. 2019. Regulation of arginine biosynthesis, catabolism and transport in Escherichia coli. Amino Acids 51:1103–1127. https:// doi.org/10.1007/s00726-019-02757-8. 53. Cunin R, Glansdorff N, Piérard A, Stalon V. 1986. Biosynthesis and metabolism of arginine in bacteria. Microbiol Rev 50:314–352. https://doi.org/10 .1128/MR.50.3.314-352.1986. 54. Chellamuthu V-R, Ermilova E, Lapina T, Lüddecke J, Minaeva E, Herrmann C, Hartmann MD, Forchhammer K. 2014. A widespread glutamine-sensing mechanism in the plant kingdom. Cell 159:1188–1199. https://doi.org/10 .1016/j.cell.2014.10.015. 55. Sugiyama K, Hayakawa T, Kudo T, Ito T, Yamaya T. 2004. Interaction of Nacetylglutamate kinase with a P II -like protein in rice. Plant Cell Physiol 45:1768–1778. https://doi.org/10.1093/pcp/pch199. 56. Li Y, Liu W, Sun L-P, Zhou Z-G. 2017. Evidence for P II with NAGK interaction that regulates Arg synthesis in the microalga Myrmecia incisa in response to nitrogen starvation. Sci Rep 7:16291. https://doi.org/10.1038/ s41598-017-16644-3. 57. Selim KA, Ermilova E, Forchhammer K. 2020. From cyanobacteria to Archaeplastida: new evolutionary insights into P II signalling in the plant kingdom. New Phytol 227:722–731. https://doi.org/10.1111/nph.16492. 58. Leigh JA, Dodsworth JA. 2007. Nitrogen regulation in bacteria and archaea. Annu Rev Microbiol 61:349–377. https://doi.org/10.1146/annurev .micro.61.080706.093409. 59. Muro-Pastor MI, Reyes JC, Florencio FJ. 2001. Cyanobacteria perceive nitrogen status by sensing intracellular 2-oxoglutarate levels. J Biol Chem 276:38320–38328. https://doi.org/10.1074/jbc.M105297200. 60. Battchikova N, Vainonen JP, Vorontsova N, Keranen M, Carmel D, Aro E-M. 2010. Dynamic changes in the proteome of Synechocystis 6803 in response to CO 2 limitation revealed by quantitative proteomics. J Proteome Res 9:5896–5912. https://doi.org/10.1021/pr100651w. 61. Orf I, Klähn S, Schwarz D, Frank M, Hess WR, Hagemann M, Kopka J. 2015. Integrated analysis of engineered carbon limitation in a quadruple CO 2 / HCO 3 2 uptake mutant of Synechocystis sp. PCC 6803. Plant Physiol 169:1787–1806. https://doi.org/10.1104/pp.15.01289. 62. Zeth K, Fokina O, Forchhammer K. 2014. Structural basis and target-specific modulation of ADP sensing by the Synechococcus elongatus P II signaling protein. J Biol Chem 289:8960–8972. https://doi.org/10.1074/jbc .M113.536557. 63. Lüddecke J, Forchhammer K. 2013. From P II signaling to metabolite sensing: a novel 2-oxoglutarate sensor that details P II –NAGK complex formation. PLoS One 8:e83181. https://doi.org/10.1371/journal.pone.0083181. 64. Helling RB. 1998. Pathway choice in glutamate synthesis in Escherichia coli.J Bacteriol 180:4571–4575. https://doi.org/10.1128/JB.180.17.4571-4575.1998. 65. Picossi S, Flores E, Herrero A. 2014. ChIP analysis unravels an exceptionally wide distribution of DNA binding sites for the NtcA transcription factor in a heterocyst-forming cyanobacterium. BMC Genomics 15:22. https://doi .org/10.1186/1471-2164-15-22. 66. Hanson TE, Forchhammer K, de Marsac NT, Meeks JC. 1998. Characterization of the glnB gene product of Nostoc punctiforme strain ATCC 29133: glnB or the P II protein may be essential. Microbiology 144:1537–1547. https://doi.org/10.1099/00221287-144-6-1537. Bolay et al. ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 16 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32. 67.PalinskaKA,LalouiW,BéduS,Loiseaux-deGoerS,CastetsAM,RippkaR, Tandeau de Marsac N. 2002. The signal transducer P II and bicarbonate acquisition in Prochlorococcus marinus PCC 9511, a marine cyanobacterium naturally deficient in nitrate and nitrite assimilation. Microbiology 148:2405–2412. https://doi.org/10.1099/00221287-148-8-2405. 68. Flores E, Arévalo S, Burnat M. 2019. Cyanophycin and arginine metabolism in cyanobacteria. Algal Res 42:101577. https://doi.org/10.1016/j.algal .2019.101577. 69. Steglich C, Futschik ME, Lindell D, Voss B, Chisholm SW, Hess WR. 2008. The challenge of regulation in a minimal photoautotroph: non-coding RNAs in Prochlorococcus. PLoS Genet 4:e1000173. https://doi.org/10.1371/journal .pgen.1000173. 70. Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY. 1979. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Microbiology 111:1–61. https://doi.org/10.1099/00221287-111-1-1. 71. García-Domínguez M, Lopez-Maury L, Florencio FJ, Reyes JC. 2000. A gene cluster involved in metal homeostasis in the cyanobacterium Synechocystis sp. strain PCC 6803. J Bacteriol 182:1507–1514. https://doi.org/ 10.1128/jb.182.6.1507-1514.2000. 72. Elhai J, Wolk CP. 1988. A versatile class of positive-selection vectors based on the nonviability of palindrome-containing plasmids that allows cloning into long polylinkers. Gene 68:119–138. https://doi.org/10.1016/0378 -1119(88)90605-1. 73. Pinto FL, Thapper A, Sontheim W, Lindblad P. 2009. Analysis of current and alternative phenol based RNA extraction methodologies for cyanobacteria. BMC Mol Biol 10:79. https://doi.org/10.1186/1471-2199-10-79. 74. Hein S, Scholz I, Voß B, Hess WR. 2013. Adaptation and modification of three CRISPR loci in two closely related cyanobacteria. RNA Biol 10:852–864. https://doi.org/10.4161/rna.24160. 75. Reyes JC, Crespo JL, Garcia-Dominguez M, Florencio FJ. 1995. Electron transport controls glutamine synthetase activity in the facultative heterotrophic cyanobacterium Synechocystis sp. PCC 6803. Plant Physiol 109:899–905. https://doi.org/10.1104/pp.109.3.899. 76. Laemmli UK. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4 5259. Nature 227:680–685. https://doi.org/ 10.1038/227680a0. 77. Navarro F, Martín-Figueroa E, Florencio FJ. 2000. Electron transport controls transcription of the thioredoxin gene trxA in the cyanobacterium Synechocystis sp. PCC 6803. Plant Mol Biol 43:23–32. https://doi.org/10 .1023/a:1006472018601. 78. Reinholdt O, Schwab S, Zhang Y, Reichheld J-P, Fernie AR, Hagemann M, Timm S. 2019. Redox-regulation of photorespiration through mitochondrial thioredoxin o1. Plant Physiol 181:442–457. https://doi.org/10.1104/ pp.19.00559. 79. Scholl J, Dengler L, Bader L, Forchhammer K. 2020. Phosphoenolpyruvate carboxylase from the cyanobacterium Synechocystis sp. PCC 6803 is under global metabolic control by P II signaling. Mol Microbiol 114:292–307. https://doi.org/10.1111/mmi.14512. 80. Harper S, Speicher DW. 2011. Purification of proteins fused to glutathione S-transferase. Methods Mol Biol 681:259–280. https://doi.org/10.1007/ 978-1-60761-913-0_14. 81. Maheswaran M, Urbanke C, Forchhammer K. 2004. Complex formation and catalytic activation by the P II signaling protein of N-acetyl-l-glutamate kinase from Synechococcus elongatus strain PCC 7942. J Biol Chem 279:55202–55210. https://doi.org/10.1074/jbc.M410971200. 82. Takahara K, Akashi K, Yokota A. 2007. Continuous spectrophotometric assays for three regulatory enzymes of the arginine biosynthetic pathway. Anal Biochem 368:138–147. https://doi.org/10.1016/j.ab.2007.06.032. 83. Kumar S, Stecher G, Tamura K. 2016. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol Biol Evol 33:1870–1874. https://doi.org/10.1093/molbev/msw054. 84. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol 215:403–410. https://doi.org/10.1016/ S0022-2836(05)80360-2. 85. Mitschke J, Georg J, Scholz I, Sharma CM, Dienst D, Bantscheff J, Voss B, Steglich C, Wilde A, Vogel J, Hess WR. 2011. An experimentally anchored map of transcriptional start sites in the model cyanobacterium Synechocystis sp. PCC 6803. Proc Natl Acad Sci U S A 108:2124–2129. https://doi .org/10.1073/pnas.1015154108. 86. Krasikov V, Aguirre von Wobeser E, Dekker HL, Huisman J, Matthijs HCP. 2012. Time-series resolution of gradual nitrogen starvation and its impact on photosynthesis in the cyanobacterium Synechocystis PCC 6803. Physiol Plant 145:426–439. https://doi.org/10.1111/j.1399-3054.2012.01585.x. 87. Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9:671–675. https://doi.org/10.1038/ nmeth.2089. 88. Bolay P, Muro-Pastor MI, Florencio FJ, Klähn S. 2018. The distinctive regulation of cyanobacterial glutamine synthetase. Life 8:52. https://doi.org/ 10.3390/life8040052. Regulation of Cyanobacterial Arginine Synthesis ® March/April 2021 Volume 12 Issue 2 e00229-21 mbio.asm.org 17 Downloaded from https://journals.asm.org/journal/mbio on 23 May 2022 by 150.214.182.32.