MPMI PhytoFrontiers Phytobiomes all journals
RESEARCHOpen Access icon OPENOpen Access license

PII Overexpression in Lotus japonicus Affects Nodule Activity in Permissive Low-Nitrogen Conditions and Increases Nodule Numbers in High Nitrogen Treated Plants

    Authors and Affiliations
    • Enrica D’Apuzzo1
    • Vladimir Totev Valkov1
    • Aurora Parlati1
    • Selim Omrane1
    • Ani Barbulova1
    • Maria Martha Sainz1
    • Marco Lentini2
    • Sergio Esposito2
    • Alessandra Rogato1
    • Maurizio Chiurazzi1
    1. 1Institute of Biosciences and Bioresources, CNR, Via P. Castellino 111, Napoli, Italy;
    2. 2Università degli Studi di Napoli Federico II, Dipartimento di Biologia, Via Cinthia 6, 80126, Napoli, Italy

    Published Online:


    We report here the first characterization of a GLNB1 gene coding for the PII protein in leguminous plants. The main purpose of this work was the investigation of the possible roles played by this multifunctional protein in nodulation pathways. The Lotus japonicus LjGLB1 gene shows a significant transcriptional regulation during the light-dark cycle and different nitrogen availability, conditions that strongly affect nodule formation, development, and functioning. We also report analysis of the spatial profile of expression of LjGLB1 in root and nodule tissues and of the protein’s subcellular localization. Transgenic L. japonicus lines overexpressing the PII protein were obtained and tested for the analysis of the symbiotic responses in different conditions. The uncoupling of PII from its native regulation affects nitrogenase activity and nodule polyamine content. Furthermore, our results suggest the involvement of PII in the signaling of the nitrogen nutritional status affecting the legumes’ predisposition for nodule formation.

    The superfamily of PII signal transduction proteins represents one of the most widely distributed signaling proteins in nature (Forchhammer 2004; Huergo et al. 2013). Many bacteria and archaea have multiple PII proteins, whereas a single copy is found in plants and most cyanobacteria. PII proteins are homotrimers composed of 12- to 13-kDa subunits, forming a compact cylinder from which three long loops (the T loops) protrude (Cheah et al. 1994; Vasudevan et al. 1994). The PII regulatory action in bacteria takes place through a protein–protein interaction mechanism controlling the activities of a wide range of targets, including enzymes, transcription factors, and membrane transporters mostly involved in nitrogen (N) metabolism. Thereafter, PII was also identified as a key C and N metabolic integrator in early studies examining glutamine synthetase (GS) GS regulation (Moorhead and Smith 2003) and, more recently, a biochemical characterization of bacterial PII proteins revealed allosteric complexes involving ATP or ADP and 2-oxoglutarate (2-OG) 2-OG, which function to sense cellular energy and carbon (C) levels (Jiang and Ninfa 2007). Layered on top of allosteric sensing of cellular C and energy availability is the covalent modification of PII in response to cellular N status. An ATPase activity has been formally proven in the PII paralog GlnK, leading to a model where the role of ATP or ADP binding is to effect a 2-OG-dependent molecular switch that drives a conformational change in the T loop (Radchenko et al. 2013). The convergence of these two types of input alters the ability of PII to interact with partner proteins which, in turn, controls the ability of these partners to modify various aspects of N metabolism, including gene transcription, membrane transporters, and metabolic enzymes (Forchhammer 2008; Moorhead and Smith 2003).

    C and N requirements of plant cells can vary between tissue types, requiring precise intertissue and, therefore, cellular cooperation. N taken up by roots, mainly in the form of nitrate and ammonium, can be either used in root N metabolism or transported to photosynthetic tissues for incorporation into amino acids. Conversely, C can be fixed locally by photosynthetic processes and synthesized into the necessary C substrates to supplement chloroplast-mediated N assimilation and amino acid biosynthesis, or translocated in the form of sucrose from photosynthetic tissues to provide energy and C skeletons for N assimilation in root tissues. In the case of leguminous plants, N-fixing root nodules are the organs where atmospheric N reduction and release take place and, at the same time, are optional C sink organs that need to assimilate an energy source for the formation of nodule primordium (Complainville et al. 2003) and to provide energy for the N fixation performed by the microsymbiont (Rhizobium) and assimilation of the produced ammonium and starch biosynthesis (Vance 2008).

    Despite the central role played as an integrator of C and N cellular metabolism in many prokaryotes, plant PII is a nuclear-encoded chloroplast protein (GLB1) that seems to have evolved secondary, tissue-specialized roles. Preliminary studies suggested a conserved role as an N or C sensor due to the lack of glutamine sensing in Arabidopsis thaliana PII overexpressors (Hsieh et al. 1998). Later, PII has been implicated in the control of the arginine (Arg) biosynthetic pathway through interaction with N-acetyl-L-glutamate kinase (NAGK) (Burillo et al. 2004; Ferrario-Méry et al. 2006; Sugiyama et al. 2004) to reduce feedback inhibition by high Arg concentrations (Chen et al. 2006; Ferrario-Méry et al. 2006; Maheswaran et al. 2004). The plant complex PII-NAGK only shows gradual inhibition of complex formation at 2-OG concentrations greater than 1 mM (Beez et al. 2009; Mizuno et al. 2007a). Most recently, major fluctuation of the GLB1 transcript during seed maturation in A. thaliana (Uhrig et al. 2009) was associated with a crucial role played by PII in the fine tuning of fatty acid biosynthesis and partitioning in Arabidopsis seed (Baud et al. 2010). In addition, two PII target proteins in A. thaliana that are biotin carboxyl carrier subunits of the plastidial acetyl-CoA-carboxylase (ACCase) (namely, BCCP1 and BCCP2), involved in the production of the precursor of lipid biosynthesis, malonyl-CoA, have been identified (Feria Bourrellier et al. 2010). ACCase activity was inhibited by PII in chloroplast extracts, in the presence of MgATP, while the addition of 2-OG, pyruvate, or oxalacetate was sufficient to release the inhibition, thus suggesting, for the first time, the involvement of PII in the regulation of an enzyme dedicated to C metabolism (Feria Bourrellier et al. 2010).

    We report here the first characterization of the GLB1 gene in a leguminous plant, with a detailed analysis of the transcriptional regulation in different growth conditions metabolically linked to the symbiotic N fixation process. Phenotypic analysis of transgenic Lotus japonicus overexpressing lines revealed a putative novel role of PII in the control of nodule functioning phenotypes and involvement in the signaling pathways governing N-dependent nodulation competence.


    Identification of the L. japonicus sequence encoding for the PII protein and description of its structural features.

    Blast search for orthologs of the Medicago sativa PII protein (accession number AY027892) in the L. japonicus genome sequence database identified a single-copy gene (chr1.CM0122.1600.r2.m) coding for a 195-amino-acid (aa) protein with a molecular mass of 21.49 kDa sharing 70% aa identity with the M. sativa PII protein. The genomic locus consists of eight exons separated by seven introns (Supplementary Fig. S1) with a gene structure identical to that identified in A. thaliana (AT4G01900). Multiple sequence alignment recognized the region (positions 106 to 124) highly conserved between bacteria and plant PII proteins (the T-loop motif) that plays a key role in mediating interactions between PII and downstream-effector proteins (Mizuno et al. 2007b; Sant’Anna et al. 2009). In particular, amino acid residues involved in the 2-OG-dependent conformational change of the T-loop (Radchenko et al. 2013) are conserved in the L. japonicus sequence (Q108 and K128). Furthermore, highly conserved charged amino acids likely to be involved in homotrimer formation in A. thaliana are identical in the L. japonicus sequence corresponding to K72, E74, D101, R103, K130, and E132 (Smith et al. 2003). The N- and C-terminal plant PII signatures are also identified in the Lotus sequence (positions 69 to 81 and 173 to 188, respectively). The molecular phylogenetic tree (Fig. 1) based on the alignment of complete amino acid PII sequences and drawn using the neighbor-joining method (Saitou and Nei 1987) integrates the taxonomic analysis previously reported (Uhrig et al. 2009), indicating a close relationship among the PII legume sequences.

    Fig. 1.

    Fig. 1. Phylogenetic relationship of PII amino acid sequences. Twenty full length amino acid sequences were aligned with the ClustalW program. The optimal tree with the sum of branch length = 2.33470189 is shown. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Poisson correction method (Zuckerkandl and Pauling 1965) and are in the units of the number of amino acid substitutions per site. Phylogenetic analyses were conducted in MEGA5.2.2 (Tamura et al. 2007). Protein sequences were obtained from NCBInr and DOEJGI online databases using BLSTP algorithms. Sequences are as follows: Arabidopsis thaliana, Cajanus cajan, Capsella rubella, Chlamydomonas reinhardtii, Cicer arietinum, Escherichia coli, Fragaria vesca, Glycine max, Medicago sativa, M. truncatula, Oryza sativa, Phaseolus vulgaris, Physcomitrella patens, Populus trichocarpa, Porphyra purpurea, Prunus persica, Solanum lycopersicon, Vitis vinifera, and Synechococcus sp. PCC7002.

    Download as PowerPoint

    Transcriptional regulation of LjGLB1 in response to light-dark cycle and N supply.

    In the literature, most of the reported transcriptional analyses of the GLB1 gene use leaf tissues to investigate a possible mechanism of control associated with the dark-light cycle or C supply derived from photosynthesis. Our analysis of LjGLB1 was especially aimed to investigate a possible regulation of the transcriptional profile in both photosynthetic (leaves) and nonphotosynthetic organs (roots and nodules) under conditions that may affect symbiotic performances. Using quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR), we first analyzed the distribution of the LjGLB1 transcript in different organs of L. japonicus. The amount of LjGLB1 transcript didn’t change significantly between root and leaf tissues of plants grown on Gamborg-B5-derived media with 1 mM ammonium nitrate as N source or without N (Fig. 2A). In the latter conditions, plants were inoculated with Mesorhizobium loti to test GLB1 expression in mature nodules and a comparable level of transcript was detected in nodular tissue, suggesting that PII might be performing some roles in this organ (Fig. 2A).

    Fig. 2.

    Fig. 2. LjGLB1 transcriptional regulation in different organs and during day-night cycle. A, Expression in different organs. RNAs were extracted by 3-week-old plants grown on 1 mM NH4NO3 (+N) or without nitrogen (−N). White bars = roots, black bars = leaves, and gray bar = nodules. B and C, Expression during the day-night cycle. Plants were grown in hydroponic conditions (1 mM NH4NO3) and RNAs were extracted by roots (B) and leaves (C) of 3-week-old plants at different hours from the beginning of light (white) and dark (gray) periods. The presence of 3% sucrose in the growth medium is indicated. LjGLB1 expression levels were normalized with respect to the internal control ubiquitin (UBI) gene and plotted relative to the expression from roots in A and to the 7-h dark samples in B and C. Data bars represent the mean and standard deviations of data obtained with RNA extracted from three different sets of plants and three real-time quantitative polymerase chain reaction experiments. Asterisks indicate significant differences with the levels of expression under light conditions (P < 0.05).

    Download as PowerPoint

    The GLB1 transcriptional analysis during the dark-light cycle (8 and 16 h, respectively) in hydroponic cultures of L. japonicus plants is reported in Figure 2B and C. A rapid enhancement (approximately fourfold) of GLB1 transcription in roots was revealed at 3 h from the beginning of the light cycle, with the amount of transcript remaining constant till the end of the light period and starting to decay after 2 h until the end of the dark period (Fig. 2B). A similar trend of transcriptional regulation was detected in leaves under the same experimental conditions (Fig. 2C). In both root and leaf tissues, the presence of 3% sucrose in the growth medium doesn’t affect the level of expression during the light period. However, the effect of light on GLB1 mRNA could be mimicked in both root and leaf tissues at the end of the dark period by the addition of 3% sucrose, indicating that, in L. japonicus, the effect of light on GLB1 transcription is mediated by photosynthetic products (Fig. 2B and C).

    GLB1 transcription was further examined in roots of L. japonicus plants grown in hydroponic cultures for 2 weeks in N-sufficient conditions (1 mM ammonium nitrate) and then transferred to fresh nutrient solution containing no N source. To minimize the diurnal changes in the expression level, all samples were taken in the middle of the light time (Fig. 2, 6 h) and RNA was extracted from roots at time 0, 24, 48, and 72 h and 5 days after the shift. The LjGLB1 transcription showed a rapid downregulation (fourfold) that was maintained until at least 72 h after the shift (Fig. 3). Interestingly, the amount of LjGLB1 transcript in roots increased again after 5 days under N-starvation conditions, to the basal level observed at 0 h (Fig. 3). As a control of the experimental conditions, we analyzed the expression profile of the high-affinity ammonium transporter LjAMT1;1 that, as expected, was strongly induced after the shift from N-sufficient to N-starvation conditions (Fig. 3) (D’Apuzzo et al. 2004). Therefore, our results indicated a significant regulation of the GLB1 expression at the transcriptional level that takes place in either leaves or root tissues.

    Fig. 3.

    Fig. 3. Effect of N deficiency treatment on LjGLB1 and LjAMT1;1 expression. RNAs were extracted from roots of plants grown in presence of 1 mM NH4NO3 (T0) and after shifting to −N conditions (24 h = T1, 48 h = T2, 72 h = T3, and 5 days = T4). Expression levels were normalized with respect to the internal control ubiquitin (UBI) gene and plotted relative to the expression of T0. White bars = LjGLB1 and black bars = LjAMT1;1. Data bars represent the mean and standard deviations of data obtained with RNA extracted from three different sets of plants and three real-time polymerase chain reaction experiments. Asterisks indicate significant differences with T0 conditions (P < 0.05).

    Download as PowerPoint

    PII localization and spatial profile of expression.

    Unlike other plant PII proteins, the Lotus sequence didn’t show a clear-cut N-terminal transit peptide for chloroplast targeting according to the prediction program CHLOROP. Therefore, in order to check the PII protein subcellular localization, we fused its C-terminal end without a stop codon to the green fluorescent protein (GFP) gene, and placed the fusion downstream of the Cauliflower mosaic virus (CAMV)-35S promoter. Confocal laser-scanning fluorescence microscopy in stable transgenic hairy roots indicated unambiguously a plastid localization of GFP in roots, confirming previous data obtained in Arabidopsis and rice (Baud et al. 2010; Hsieh et al. 1998; Sugiyama et al. 2004). The specificity of the plastids as target was indicated by colocalization with the red fluorescence of the AtrecA-DsRed2 fusion (Supplementary Fig. S2 A to C) (Köhler et al. 1997).

    To gain further information about the profile of LjGLB1 expression in roots and nodules, we isolated the 5′ region of the gene to obtain a T-DNA construct carrying a promoter-gusA fusion. A PCR fragment extending up to 980 bp upstream of the ATG of LjGLB1 and including the first 21 PII codons was subcloned in the pBI101.1 binary vector to obtain a translational fusion with the gusA reporter gene (Jefferson 1987). Lotus composite plants obtained upon transformation with Agrobacterium rhizogenes (Martirani et al. 1999) were used to analyze the expression of the translational pLjGLB1-gusA fusions in a hairy root system. In the root tissues, β-glucuronidase (GUS) activity was substantially limited to the stele, where it was detected in 100% of the stained roots (Fig. 4A). In some cases, a longer incubation time allowed the detection of the blue staining in cortical cells and cap cells, whereas no activity could be observed in meristematic regions (Fig. 4B and C). Root cross sections allowed the definition of a more precise pattern showing a distribution of GUS activity inside the endodermis cell layer, with the main staining apparently located in the primary phloematic space (Fig. 4D). The promoter activity of the Arabidopsis thaliana GLB1 gene has been described in leaf vascular structures only in an overexpressing Arabidopsis WRl1 genotype (Baud et al. 2009, 2010). However, analysis of the 980-bp 5′ untranslated region LjGLB1 region exploited for driving the gusA expression in transgenic hairy roots didn’t reveal any of the 15-bp boxes (cAAAAGtAggggttT) reminiscent of the consensus sequence required for WRl1 induction of glycolitic and fatty acid biosynthetic genes (Baud et al. 2010), while only one putative AW-box motif (CnTnGn7CG; Maeo et al. 2009) was found at position –231 relative to the translational start codon (data not shown).

    Fig. 4.

    Fig. 4. Spatial profile of expression of the LjGLB1 gene in root and nodular tissues. A to C, Whole-mount staining of Lotus japonicus hairy roots transformed with the pLjGLB1-gusA construct. A, β-Glucuronidase (GUS) activity in the root vascular bundle; B and C, GUS activity in the root vascular bundle, cortical cells, and root cap (arrows); D, cross section of atransgenic root with GUS activity detected into the root stele; E, longitudinal section of a transgenic mature representative nodule. GUS activity is detected in the nodule vascular bundle (arrows) and parenchyma (arrowhead). Bars on the left = 50 μm.

    Download as PowerPoint

    The profile of LjGLB1 promoter activity was also analyzed during different steps of nodule development or function. GUS activity was never induced during the initial stages of M. loti infection whereas the analysis in young and mature nodules allowed the detection of a strong blue staining in the nodule vascular bundles, with a weaker activity in the nodule parenchyma (Fig. 4E). The absence of LjGLB1 transcriptional regulation early after inoculation was also confirmed by the qRT-PCR analysis shown in Supplementary Figure S3. In addition, the observed N-dependent transcriptional regulation together with the reported impact of the photosynthate supply on the LjGLB1 transcriptional profile (Figs. 2 and 3) prompted us to investigate the relationship between nodular LjGLB1 expression and N fixation activity. Therefore, the pattern of GUS activity was evaluated in young nodules obtained with an M. loti fix mutant. The transgenic nodules obtained in hairy roots inoculated with either the wild type or an nifH mutant didn’t show any change in the quantitative and qualitative GUS activity distribution (data not shown).

    Generation of transgenic L. japonicus plants ectopically expressing the GLB1 gene.

    In order to test whether the GLB1 gene in legumes plays any role in the formation or functioning of N-fixing-nodules, a process strictly dependent on a correct C/N balance, we cloned the LjGLB1 coding sequence between a CaMV-35S promoter sequence and a tNOS terminator sequence (Hajdukiewicz et al. 1994) to obtain transgenic PII-overexpressing Lotus plants. Primary transformed plants were selected on hygromycin medium and allowed to self-pollinate. Independent T1 lines, germinated on hygromycin (hygromycin resistance segregation = 3:1), were tested by semiquantitative RT-PCR to analyze the level of GLB1 transcript, and transgenic lines with various levels of overexpression were identified. T2 homozygous plants 7-13 and 8-9, showing 5- and 10-fold increase, respectively, of the LjGLB1 transcript, were selected for further phenotypic analysis (Fig. 5A). To verify proper protein synthesis of the ectopically expressed gene, Western blot analysis was performed on crude protein extracts from the transgenic and wild-type plants, and PII was detected using a polyclonal antiserum raised against recombinant A. thaliana PII protein (Ferrario-Méry et al. 2005). A linear increase of the PII protein in the transformants, proportional to the observed induced amount of transcript when compared with the wild-type plant, is shown in Figure 5B.

    Fig. 5.

    Fig. 5. Molecular characterization of 35S-LjGLB1 overexpressing lines. A, Semiquantitative analysis of GLB1 transcript level in roots of the wild type (wt) (white bar) and the two overexpressing lines 7-13 and 8-9 (T2 homozygous plants; black bars). Expression levels were normalized with respect to the internal control ubiquitin (UBI) gene and plotted relative to the expression in wild-type plants. Data bars represent the mean and standard deviations of data obtained with RNA extracted from two different sets of plants and three reverse-transcriptase polymerase chain reaction experiments. B, Corresponding analysis of PII protein levels by western blot using A. thaliana PII specific antibody. The lanes contain 20 μg of soluble proteins extracted from three Lotus plants. Asterisks indicate significant differences with the level of expression in wild-type plants (P < 0.05).

    Download as PowerPoint

    Phenotypic characterization of PII-overexpressing plants.

    The spatial profile of GLB1 promoter activity in the nodular tissue prompted us to test whether the PII protein could play a role as an integrator of the C, N, and energy levels that are critical for a correct nodule organogenesis and functioning. A comparison of the nodulation response in Lotus wild-type and PII-overexpressing seedlings grown with low N supply (10 μM NH4NO3) and inoculated with M. loti indicated equivalent phenotypes in terms of nodule numbers (Fig. 6A), nodule mass (data not shown), and structural organization (Supplementary Fig. S4A). On the other hand, analysis of N fixation activity at 28 days after inoculation indicated a significant 30% reduction in nodules of PII-overexpressing plants (Fig. 6B). However, this deficiency was not sufficient to cause a significant difference in terms of fresh shoot weight that was comparable in the two plant genotypes (Supplementary Fig. S4B). Because Ferrario-Méry and associates (2006) reported a reduction of several products of the ornithine/arginine biosynthetic pathway in A. thaliana PII knock-out mutants in response to ammonium resupply after N starvation, we tested the content of polyamines, which are the final products of this biosynthetic pathway and have been involved in the regulation of symbiotic efficiency, in PII-overexpressing plants. Polyamines were extracted in mature nodules of plants grown in 10 μM NH4NO3 and their relative amounts compared in wild-type and 8-9 plants. The content of spermidine and spermine were much higher than that of citrulline and didn’t change significantly between wild-type and overexpressing plants, whereas the citrulline level was strongly increased (6.5-fold) in nodules of the 8-9 plants (Fig. 6C and D).

    Fig. 6.

    Fig. 6. Nodular phenotypical analysis of wild-type (wt) and PII-overexpressing plants grown in low N conditions. A, Number of nodules per plant. Plants were germinated on H2O agar, and 7-day-old seedlings were transferred on Gamborg B5 derived medium with 10 μM NH4KNO3 as sole N source and inoculated with Mesorhizobium loti. Nodules were scored at 4 weeks after inoculation; wt (white bars), 8.9 (black bars), and 7.13 (gray bar). Data represent the mean and standard error obtained from three independent experiments (15 plants per experiment). Asterisks indicate significant differences with the number of nodules observed without putrescine. B, Acetylene reduction activity per nodule weight of wt and 8-9 plants. Data bars indicate the mean and standard error of three independent experiments (n = 2 plants per experiment). Asterisk indicates significant difference (P < 0.02). C, Spermine (left) and spermidine (right) content in nodules of wt and 8-9 plants. Polyamines were extracted from 4-week-old nodules. Data represent the mean and standard error obtained from two independent experiments (20 plants per experiment). Asterisk indicates significant difference (P < 0.01). D, Citrulline content in nodules of wt and 8-9 plants. Polyamines were extracted from 4-week-old nodules. Data represent the mean and standard error obtained from two independent experiments (20 plants per experiment). Asterisk indicates significant difference (P < 0.01).

    Download as PowerPoint

    N depletion in the soil is a prerequisite for nodule development and function, and high concentrations of N as nitrate and ammonia might abolish nodulation. We reported a link between the L. japonicus N nutritional status and nodule capacity predisposition, showing that the competence for nodulation of L. japonicus plants preincubated, prior to the M. loti inoculation, on high N (10 mM NH4NO3) conditions was significantly reduced (approximately 50%) when compared with plants maintained in low N (10 μM NH4NO3) permissive conditions (Omrane et al. 2009). The inhibitory effect was maintained for at least 6 days in Lotus plants preincubated on high N, indicating that it was correlated with a systemic change of the general N nutritional state of the plants that was transduced to the root nodulation machinery (Omrane et al. 2009). Therefore, in order to analyze any possible involvement of the PII protein in the signal transduction route connecting N nutritional status and nodulation capacity, we compared the nodule formation phenotype of wild-type and PII-overexpressing plants following the experimental scheme reported in Omrane and associates (2009). Wild-type L. japonicus plants preincubated on 10 mM ammonium nitrate showed a reduced number of nodules when compared with plants maintained on 10 μM ammonium nitrate (Fig. 7A; 3.37 versus 1.75 nodules/plant). Strikingly, both overexpressing 7-13 and 8-9 PII lines were able to rescue this inhibitory effect, showing the same level of nodulation capacity in low and high N conditions (Fig. 7A). In L. japonicus, a central role of the CLAVATA3/ESR-related glycopeptide (CLE) CLE glycopeptide LjCLERS2 in the nitrate-induced systemic suppression of nodulation through interaction with the hypernodulation aberrant root formation (HAR1) receptor kinase has been reported (Okamoto et al. 2009, 2013). In order to test whether PII overexpression could induce a deregulation of LjCLERS2, we compared the amount of transcript in wild-type and PII-overexpressing plants incubated for 10 days in low and high NH4NO3 conditions. Consistently with data reported in the literature, LjCLERS2 was induced in Lotus wild-type plants incubated on higher N concentrations when compared with plants grown on 10 μM NH4NO3 permissive conditions, and a similar pattern was observed in the 8-9 plants (Fig. 7B).

    Fig. 7.

    Fig. 7. N-dependent nodule formation phenotypes in wild-type (wt) and Lotus japonicus PII-overexpressing 7-13 and 8-9 plants. A, White bars represent number of nodules observed in plants preincubated for 10 days on 10 μM NH4NO3 and then shifted on the same condition and inoculated with Mesorhizobium loti. Black bars represent nodule numbers observed in plants preincubated for 10 days on 10 mM NH4NO3 and then shifted onto 10 μM NH4NO3 and inoculated with M. loti. Nodules were counted at 4 weeks post inoculation. Data represent the mean and standard error obtained from three independent experiments (20 plants per experiment). Asterisks indicate significant differences from the number of nodules observed in wt plants preincubated on 10 mM NH4NO3 (P < 0.05). B, Relative quantification of LjCLERS2 mRNA measured by quantitative reverse-transcriptase polymerase chain reaction in roots of wt and 8-9 plants incubated for 10 days with 10 mM NH4NO3 or 10 μM NH4NO3 conditions. L. japonicus UBI was used as reference gene to normalize the expression of LjCLERS2. Data bars represent means and standard deviation of data obtained with RNA extracted from two different sets of plants and three technical repeats. Asterisks indicate significant differences with the level of expression in 10 mM conditions (P < 0.05).

    Download as PowerPoint


    How PII is regulated in higher plants and, hence, how this transduction protein can sense the C/N status of a plant is still a matter of debate. A posttranslational control of its activity similar to that of PII from prokaryotes has not yet been demonstrated (Smith et al. 2004). In oxygenic photosynthetic organisms, the PII interaction with NAGK, the key enzyme of the arginine biosynthesis pathway, enhances the catalytic kinase activity and leads to the relief of the feedback inhibitory effect by the final pathway product, arginine (Chen et al. 2006; Ferrario-Méry et al. 2006). PII in the ATP-ligated state has a very high affinity for NAGK, whereas ADP and 2-OG display a nonantagonizing inhibitory effect on the formation of the PII-NAGK complex which, therefore, seems not governed by the intrinsic ATPase activity of PII (Maheswaran et al. 2004; Zeth et al. 2014). 2-OG also has been proposed as the main signal controlling the PII inhibition of ACCase in chloroplast extracts, thereby expanding plant PII function beyond N regulation (Feria Bourrellier et al. 2010).

    With regard to the role played by a transcriptional regulation on the GLB1 gene, recently, a striking observation reporting the upregulation of the PII transcript (10-fold) in the early- to midstages of A. thaliana developing seed strongly supported an involvement of such a mechanism of regulation in the control of plant PII function (Uhrig et al. 2009). Partially inconsistent data were reported for the transcriptional level of regulation of the GLB1 gene expression in leaves (Ferrario-Mèry et al. 2005; Hsieh et al. 1998). A strong and rapid activation of GLB1 transcription was first reported in dark-adapted Arabidopsis plants after a shift in light conditions and this derived, at least in part, from photosynthetic products (Hsieh et al. 1998). On the other hand, a stable amount of the GLB1 transcript was observed in Arabidopsis rosette leaves during most of the day-night cycle, with a slight increase only at the end of the day (Ferrario-Mèry et al. 2005). We analyzed the profile of LjGLB1 expression in both Lotus root and leaf tissues and found a rapid induction of the LjGLB1 transcript during the light period. This effect could be mimicked, during the dark period, by the addition of 3% sucrose, indicating that induction of the transcription is mediated by the sensing of photosynthetic products (Fig. 2). We also observed an N-dependent regulation of transcription with a rapid and transient decrease of the amount of LjGLB1 transcript followed by a progressive increase in roots of plants transferred into N-starvation conditions (Fig. 3). This latter result might be consistent with data reported in A. thaliana, where the AtGLB1 transcript level was analyzed only 5 days after a shift to an N-deficiency condition and showed no significant variation (Ferrario-Mèry et al. 2005). Thus, our results indicate that a significant part of the regulation of the LjGLB1 profile of expression takes place at the transcriptional level.

    The growth conditions tested in this work (availability of different N and C sources) significantly affecting the regulation of LjGLB1 gene expression (Figs. 2 and 3) might strongly impact the efficiency of symbiotic N-fixation processes (nodule formation, development, and function). Photosynthate resources must be allocated to the nodules where C derived from the metabolism of sucrose is used for several physiological processes, including plant and bacterial respiration, assimilation of fixed N2, and starch and cellulose biosynthesis (Tsikou et al. 2013). Furthermore, a strict interdependence exists between the level of nitrogenase activity and the efficiency of C source flux providing energy for the bacteroid-mediated N2 reduction (Pathirana et al. 1992; Schulze et al. 1998). On the other hand, limitation of combined N in the soil is a prerequisite for initiation, development, and functioning of N-fixing nodules (Barbulova et al. 2007; Bisseling et al. 1978; Caetano-Anollés and Gresshoff 1991; Matamoros et al. 1999), and legumes employ a number of regulatory mechanisms to avoid nodulation under N-replete conditions, when sufficient mineral or organic N is available in the soil, restricting nodulation to sustainable levels only when soil N is limiting (Omrane and Chiurazzi 2009).

    As preliminary tool to initiate studies on the putative role played by the PII protein during the nodulation process, we exploited two transgenic overexpressing lines in which LjGLB1 expression is uncoupled from its native regulation (Fig. 5). PII-overexpressing plants grown under low N permissive conditions do not show differences in terms of nodule formation capacity, whereas a significant 30% reduction of nitrogenase activity measured as acetylene reduction activity (ARA) was observed when compared with wild-type plants (Fig. 6A and B). However, in the exploited experimental system where 10 μM KNO3 is present as an N source, this nodular activity defect is not sufficient to affect normal shoot growth phenotype (Supplementary Fig. S4B).

    The spatial profile of the LjGLB1 promoter activity shown in Figure 4 which, to our knowledge, represents the first information reported in the literature about the PII distribution in root tissues, and indicates a main localization in vascular bundles (Fig. 4). Root vascular structures mediate long-distance transport of compounds such as N metabolites and sucrose from root to shoot and vice versa. Hence, this spatial profile of PII localization might be consistent with both transcriptional patterns shown in Figures 2 and 3, where regulation of the LjGLB1 expression was reported to be controlled by photosynthetic products and N supply conditions.

    However, the expression pattern of LjGLB1 promoter activity in nodule vascular bundles and parenchima (Fig. 4E) might provide an additional clue about a possible link between PII function and the observed ARA defective phenotype, because this partially overlaps those of L. japonicus genes encoding spermidine synthase, spermine synthases, arginine decarboxylase, and ornithine decarboxylase (LjSPDS, LjSPMS, LjADC, and LjODC, respectively), involved in the synthesis of polyamines spermidine and spermine from putrescine and mainly expressed in the nodule parenchima and vascular bundles (Efrose et al. 2008; Flemetakis et al. 2004). A high content of putrescine, spermine, and spermidine in mature nodules, with an increased amount compared with other legume organs, has been already reported (Flemetakis et al. 2004; Fujihara et al. 1994). The LjODC gene encodes ornithine decarboxylase, which catalyzes the synthesis of ornithine to putrescine, the obligate precursor of spermidine and spermine (Fig. 8), and a root- and nodule-specific expression profile for LjODC has been detected, with a strong induction observed during nodule development (Flemetakis et al. 2004). Our analysis of polyamine content in wild-type nodules with the lower level of citrulline compared with spermine or spermidine (Fig. 6C and D) suggests the occurrence in nodular tissue of a metabolic shunt of ornithine into putrescine for polyamine biosynthesis through the action of ornithine decarboxylase, rather than the alternative pathway catalyzed by ornithine transcarbamylase that utilizes the same substrate (ornithine) for the synthesis of citrulline (Fig. 8), and this hypothesis is consistent with the previously reported LjODC profile of expression (Flemetakis et al. 2004). The high level of spermine and spermidine was not affected in nodules of PII-overexpressing plants (Fig. 6C), where we observe a clear-cut 10-fold increase of the citrulline content when compared with the wild type (Fig. 6D). The biosynthetic pathway involving the enzymes ornithine decarboxylase, arginine decarboxylase, spermidine synthase, and spermine synthases is controlled by the major plant PII-interacting protein NAGK (Fig. 8) (Burillo et al. 2004; Ferrario-Méry et al. 2006; Sugiyama et al. 2004) and the increased level of citrulline observed in nodules of the PII-overexpressing plants (Fig. 6D) is consistent with the reduced ornithine, citrulline, and arginine accumulation observed in Arabidopsis PII knock-out mutants (Ferrario-Méry et al. 2006). The reason we didn’t observe a parallel increase of the spermine and spermidine levels could be a consequence of the hypothesized metabolic shunt leading to a high basal level of these polyamines in wild-type nodules. Polyamines are important components for the establishment of optimal symbiotic performance, and a negative effect of high concentration of exogenous putrescine on ARA activity has been reported (Wisniewski and Brewin 2000). Therefore, the reduced N-fixation activity observed in PII-overexpressing plants could be related to the increase of nodular polyamines through the ornithin biosynthetic pathway.

    Fig. 8.

    Fig. 8. Pathways of polyamines biosynthesis. Dotted lines indicate the putative nitric oxide synthase (NOS)-dependent pathway for nitric oxide (NO) biosynthesis. ADC = arginine decarboxylase, NAGK = N-acetyl-L-glutamate kinase, ODC = ornithine decarboxylase, and OTC = ornithine transcarbamylase.

    Download as PowerPoint

    The other clear-cut phenotype scored in the PII-overexpressing lines is the capacity of responding very effectively to a preliminary high N treatment, prior to M. loti inoculation, that causes an approximately 50% reduction in the nodulation rate of wild-type L. japonicus plants, completely rescuing this deficient phenotype (Fig. 7A) (Omrane et al. 2009). Therefore, our results suggest that the PII protein in legumes may serve as part of a complex signal transduction network involved in perceiving the N status of the plant and regulating nodulation capacity predisposition. In legumes, a systemic adaptation mediated by plant N status regulating nodule formation has been demonstrated (Jeudy et al. 2010) and Affymetrix GeneChip analyses indicate a strong systemic influence of the N supply on the profile of gene expression in roots before and after inoculation (Omrane et al. 2009; Ruffel et al. 2008). The reported effect of PII overexpression on the ornithine/arginine biosynthetic pathway (Fig. 6D) also could be correlated with the improved nodulation response of Lotus plants to high N conditions (Fig. 7A) because polyamines might also affect nodule formation capacity and, in particular, a positive effect of citrulline supply on the formation of infection threads structures in L. japonicus nodules has been reported (Mishima et al. 2008). Another intriguing correlation to the increased level of citrulline could be a positive effect of PII overexpression on the arginine-dependent synthesis of nitric oxide (Fig. 8) that plays a direct role on the nodule formation process, because a depletion of this signal causes downregulation of plant genes involved in nodule development and formation (Boscari et al. 2013; del Giudice et al. 2011) and a significant decrease of nodule number (del Giudice et al. 2011; Pauly et al. 2011).

    However, our data indicate that one of the actors playing a crucial role in the nitrate-dependent control of nodule initiation, LjCLERS2 (Okamoto et al. 2009; Okamoto et al. 2013), is likely not to be a target of the PII action, because its transcriptional profile did not change in overexpressing plants (Fig. 7B). Nitrate may affect lateral root development (Zhang et al. 1999) as well as nodule formation (Omrane and Chiurazzi 2009) by acting both as a nutrient through the assimilation pathways and consequent change of the systemic general nutritional status and as a local signal independent of its assimilation. LjCLERS2 is a root-specific gene and its nitrate-dependent induction could be mediated by a local signaling pathway followed by the systemic transmission through xylem of the arabinosylated form of the peptide to the shoot where the interaction with the HAR1 receptor kinase takes place, controlling the process of nodule formation autoregulation (Okamoto et al. 2013). The LjGLB1 gene is ubiquitously expressed in Lotus plants (Fig. 2A) and PII protein likely acts by perceiving the general N nutritional change determined by nitrate supply conditions. Therefore, LjCLERS2 and PII might be part of two different nitrate-dependent regulatory pathways controlling the nodule formation capacity; this would explain the conserved LjCLERS2 profile of expression observed in wild-type and PII-overexpressing plants (Fig. 7B).

    In conclusion, our data suggest a possible novel role of the legume PII protein in the control of nodule functioning and signaling pathways linking N nutritional status and nodulation competence. Ultimately, a global metabolomic analysis in the PII-overexpressing plants and the use of RNAi constructs driven by the LjGLB1 own promoter, vascular bundle-specific, or nodule-specific promoters will allow us to gain further insights into such functional roles of PII.


    Plant material and growth conditions.

    All experiments were carried out with Lotus japonicus ecotype B-129 F12 GIFU. Sterilized seed were sown on H2O agar plates and left over night at 4°C cap-side down. After 24 h in the dark in the growth chamber, Petri dishes were exposed to light and kept in a vertical position. Care was taken to maintain the young emerging roots in contact with the filter paper. For the analysis of nodulation capacities, unsynchronized seedlings were discarded at this stage. Plants were cultivated in a growth chamber with a light intensity of 200 μmol m−2 s−1 at 23°C with a day and night cycle of 16 and 8 h, respectively. Solid growth substrate had the composition of B5 medium (Gamborg 1970), except that, when needed, (NH4)2SO4 and KNO3 were omitted or replaced by ammonium nitrate (NH4NO3). KCl was added to the medium to replace the potassium source. The media containing vitamins (Duchefa catalog G0415) were buffered with 2.5 mM 2-(N-morpholino)-ethanesulfonic acid (MES) (Duchefa, MIS03.0250) and pH was adjusted to 5.7 with KOH.

    Experimental conditions for the analysis of the high N preincubation effect on nodulation have been described by Omrane and associates (2009). Seedlings (3 to 5 days old) were transferred and grown for 10 days on B5 derived medium where (NH4)2SO4 and KNO3 were omitted and replaced by 10 mM (High-N) or 10 μM (Low-N) NH4NO3. KCl was added to the medium to replace the potassium source. During these 10 days of preincubation, plants were transferred twice on fresh media to avoid nutrient depletion. After 10 days, both High-N and Low-N plants were transferred onto fresh 10 μM NH4NO3 medium and inoculated with M. loti. Six days after inoculation, plants were transferred onto 10 μM NH4NO3 medium with addition of cefotaxime at 50 mg liter1.

    Hydroponic culture conditions for the analysis of N- and light-dark cycle-dependent LjGLB1 regulation have been described by D’Apuzzo and associates (2004). One-week-old plants germinated on solid medium were transferred into vessels harboring eight plants in approximately 100 ml of the 1 mM NH4NO3 B5 derivative medium. In the N-free solution, the NH4NO3 was omitted. The pH of the media was adjusted to 5.7 with MES. To avoid depletion, the nutrient solution was renewed every 3 days during the 10 days of growth in hydroponic cultures. The pH of the medium was checked daily and it was maintained within close limits (5.8 to 5.6) in all the conditions of hydroponic growth. Plants were cultivated in a growth chamber with a light intensity of 200 μmol m−2 s−1 at 23°C with a day and night cycle of 16 and 8 h, respectively.

    M. loti strain R7A was used for the inoculation experiments and was grown in liquid Luria Broth (LB) medium supplemented with rifampicin (20 mg/liter). The M. loti nifH strain was kindly provided by Dr. Clive Ronson (University of Otago, New Zealand) and was grown in the same medium supplemented with rifampicin and gentamicin (10 mg/liter). The inoculation procedure (107 cells per root tip) for the in vitro nodulation assay has been reported elsewhere (Barbulova et al. 2005).

    L. japonicus transformation procedures.

    Agrobacterium tumefaciens and rhizogenes–mediated L. japonicus transformations were performed as described by Lombari and associates (2003) and Martirani and associates (1999), respectively.

    T-DNA constructs preparation.

    To obtain the PII-overexpressing construct, the LjGLB1 cDNA was amplified with the two oligonucleotides 5′-GAAGATCTATGGCGATTGCGAGAACGCAC-3′ (including a BglII site) and 5′-ACGCGTCGACTCATACAGTAGATAATATGTC-3′ (including a SalI site) and subcloned as a BglII-SalI fragment into pCAMBIA1300 BglII-SalI double digested.

    The 35S-GLB1-GFP fusion was prepared in the following way: the LjGLB1 cDNA sequence was amplified with the two oligonucleotides 5′-GAGGATCCATGGCGATTGCGAGAACGCAC-3′ (including a BamHI site) and 5′-GCGGTACCTACAGTAGATAATATGTCAGT-3′ (including a KpnI site) and cloned into the double-digested BamHI-KpnI β-GFP plasmid (Duby et al. 2001), and the correct sequence for the LjGLB1-GFP translational fusion was verified by sequencing. The LjGLB1-GFP cassette was then cloned as a BamHI-SacI fragment into the double-digested BglII-SacI pCAMBIA1300 vector.

    The pLjGLB1-gusA fusion was prepared in the following way: a PCR-amplified fragment was obtained on genomic DNA with two specific oligonucleotides—5′-GCGTCGACACCGTTTTTCCCCAGTAACCG-3′ (containing a SalI site) and 5′-CGCGGATCCCATTGGAGCTTCGTTGAGCTG-3′ (containing the BamHI site) —and subcloned as a SalI-BamHI fragment into the pBI101.1 vector to obtain a translational fusion.

    Confocal analysis.

    Confocal microscope analyses were performed using a Nikon PCM2000 (Bio-Rad, Germany) laser-scanning confocal imaging system. For GFP and red fluorescent protein detection, excitation was at 488 nm and detection between 515 and 530 nm. The images acquired from the confocal microscope were processed using ImageJ bundle software.

    Quantitative real-time RT-PCR.

    Total RNA was prepared from Lotus tissues using the procedure of Kistner and Matamoros (2005). The samples were treated with DNAse I (Ambion) to remove contaminating DNA, the absence of which was subsequently confirmed by PCR. Total RNA (1 μg) was annealed to random decamers and reverse-transcribed with reverse transcription (Ambion) to obtain cDNA. Real-time PCR was performed with a DNA Engine Opticon 2 System (MJ Research, MA, U.S.A.) using SYBR to monitor double-stranded DNA synthesis. The ubiquitin (UBI) gene (AW719589) was used as an internal standard. The concentration of primers was optimized for each PCR reaction and each amplification was carried out in triplicate. The PCR program used was as follows: 95°C for 13 min and 39 cycles of 94°C for 15 s, 60°C for 15 s, and 72°C for 15 s. Data were analyzed using Opticon Monitor Analysis Software (version 2.01; MJ Research). The qRT-PCR data were analyzed using the comparative cycle threshold (Ct) method. The relative level of expression was calculated with the following formula: relative expression ratio of the gene of interest is 2−ΔCT, with ΔCT = CtAMT1 − CTUBI. Analysis of the melting curve of PCR product at the end of the PCR run revealed a single narrow peak for each amplification product, and fragments amplified from total cDNA were gel purified and sequenced to assure accuracy and specificity. The oligonucleotides used for the qRT-PCR were the following: PII-forw 5′-GCAGAGGAAATGCCATGATT-3′, PII-rev 5′-CACCACGGATTCCCATATTC-3′, CLERS2-forw 5′-GCTCGTAATCTCCAAATCATTCACA-3′, and CLERS2-rev 5′-GGTGAGAGTCTTTGCTGTTGATATCC-3′

    Sodium dodecyl sulfate polyacrylamide gel electrophoresis and Western blot analysis.

    Tissue powder from 0.5-g aliquots was prepared by grounding fresh tissue in liquid N. The powder was then resuspended in 1 ml of 50 mM potassium phosphate buffer, 10 mM 2-mercaptoethanol, and 1 mM phenylmethylsulfonyl fluoride PMSF, pH 7.2, prechilled at 4°C. The cell-free crude extract samples were centrifuged for 15 min at 12,000 × g to remove insoluble material. Aliquots containing 20 μg of protein from a given supernatant preparation were separated by electrophoresis on a 12% (wt/vol) sodium dodecyl sulfate polyacrylamide gel. The separated proteins were then electroblotted onto polyvinylidene fluoride membranes and rinsed with Tris-buffered saline (20 mM Tris-HCl and 0.5 mM NaCl, pH 7.5). The membrane was stained with red Ponceau to check for an equally loaded amount of protein. PII protein was detected using the polyclonal antiserum raised against recombinant Arabidopsis thaliana PII protein (Ferrario-Méry et al. 2005). Western blot analysis was done according to the manufacturer’s instruction (ECL Plus Western Blotting Detection System, Amersham, Uppsala, Sweden).

    Histochemical GUS analysis.

    Histochemical staining of whole-plant and section material was performed as described by D’Apuzzo and associates (2004).

    Determination of ARA.

    Detached roots with comparable numbers of nodules were placed in glass vials. The vials were filled with an acetylene-air mixture (C2H2–air = 1:9 [vol/vol]). After 1 h of incubation at 25°C, the amount of ethylene in the gas phase was determined using a gas chromatograph (PerkinElmer Clarus 580).

    Measurement of polyamine levels.

    For polyamine extraction and high-performance liquid chromatography (HPLC) analysis, a benzoylation method was performed as described previously (Flores and Galston 1982; Smith and Davies 1985), with some modifications. Fresh tissue (1 g) was homogenized in 10 ml of cold 0.2 N per-chloride acid containing 1 μmol of hexanediamine as an internal standard. The samples are incubated on ice for 40 min, then centrifuged at 4°C for 20 min. Aliquots of 0.5 ml of supernatant were added to 1 ml of 2 N NaOH with 10 μl of benzoyl chloride. The mixtures were incubated at room temperature for 20 min, and the reaction was terminated by the addition of 2 ml of saturated NaCl. Benzoylamines were extracted with 2 ml of diethyl ether. After centrifugation, the ether layer was collected and dried under N gas. The powdered samples were stored at −20°C until HPLC measurements. For HPLC analysis, the residues were redissolved in 120 μl of methanol. Standards were treated in a similar way, with 1 μmol of putrescine, cadaverine, hexanediamine, spermidine, and spermine in the reaction mixture. HPLC analysis was performed with a programmable Kratos dual-pump liquid chromatograph with detector 773. The solvent system consisted of methanol and water, run at 65% (vol/vol) methanol at a flow rate of 0.7 ml/min. Benzoylated extract (5 μl) was eluted at room temperature through a 4.6 x 250 mm, 5-mm particle size reverse-phase (C18) column (Varian, Walnut Creek, CA) and detected at 254 mm. The peak areas were recorded on a pen recorder and calculated by a computer with NIH Image software (National Institutes of Health, Bethesda, MD).

    Phylogenetic studies.

    The evolutionary history was inferred using the neighbor-joining method (Saitou and Nei 1987). The evolutionary distances were computed using the Poisson correction method (Zuckerkandl and Pauling 1965) and are measured in units of the number of amino acid substitutions per site. The analyses involved 20 amino acid sequences. All positions containing gaps and missing data were eliminated. Phylogenetic analyses were conducted in MEGA5.2.2 (Tamura et al. 2007).

    Statistical analysis.

    Statistical analyses were performed using the VassarStats analysis of variance program.


    This work was supported by grants from the Italian Ministry of Education (Progetti di Rilevanza Nazionale, PRIN 2010/2011, PROROOT, Prot. 20105XLAXM); Progetto CISIA, Integration of Knowledge for sustainability and Innovation in the Agrofood Made in Italy; and Progetto POR CAMPUS, Qualità delle Produzioni Tipiche Campane ed il suo Territorio: Approcci Innovativi ed Integrati per rafforzare la Competitività del Sistema Agroalimentare—QUARC. We thank the facility of Integrated Microscopy of the Institute of Genetics and Biophysics for supporting the microscopy analysis; S. Salvia and D. Maiello for technical assistance; and S. Ferrario-Méry and M. Hayashi for providing PII antibody and AtrecA-DsRed2 construct, respectively.

    E. D’Apuzzo and V. T. Valkov contributed equally to this work.



    Current address for A. Barbulova: Arterra Bioscience Srl, Via B. Brin 69, 80142, Napoli, Italy.
    Current address for M. M. Sainz, Departamento de Nutrition Vegetal, Estacion Experimental de Aula Dei, Consejo Superior de Investigationes Cientificas, 50080, Zaragoza, Spain.