Article pubs.acs.org/jpr
Calcium-Dependent Regulation of Genes for Plant Nodulation in Rhizobium leguminosarum Detected by iTRAQ Quantitative Proteomic Analysis Giorgio Arrigoni,†,§ Serena Tolin,†,‡ Roberto Moscatiello,⊥ Antonio Masi,‡ Lorella Navazio,⊥ and Andrea Squartini*,‡ †
Proteomics Center of Padova University, Via G. Orus 2b, 35129 Padova, Italy Department of Agronomy, Food, Natural Resources, Animals and Environment, DAFNAE, Viale dell’Università 16, 35020 Legnaro, Padova, Italy § Department of Biomedical Sciences and ⊥Department of Biology, University of Padova, Via U. Bassi 58/B, 35131 Padova, Italy ‡
S Supporting Information *
ABSTRACT: Rhizobia, the nitrogen-fixing bacterial symbionts of legumes, represent an agricultural application of primary relevance and a model of plant−microbe molecular dialogues. We recently described rhizobium proteome alterations induced by plant flavonoids using iTRAQ. Herein, we further extend that experimentation, proving that the transient elevation in cytosolic calcium is a key signaling event necessary for the expression of the nodulation (nod) genes. Ca2+ involvement in nodulation is a novel issue that we recently flagged with genetic and physiological approaches and that hereby we demonstrate also by proteomics. Exploiting the multiple combinations of 4-plex iTRAQ, we analyzed Rhizobium leguminosarum cultures grown with or without the nod gene-inducing plant flavonoid naringenin and in the presence or absence of the extracellular Ca2+ chelator EGTA. We quantified over a thousand proteins, 189 of which significantly altered upon naringenin and/or EGTA stimulation. The expression of NodA, highly induced by naringenin, is strongly reduced when calcium availability is limited by EGTA. This confirms, from a proteomic perspective, that a Ca2+ influx is a necessary early step in flavonoid-mediated legume nodulation by rhizobia. We also observed other proteins affected by the different treatments, whose identities and roles in nodulation and rhizobium physiology are likewise discussed. KEYWORDS: calcium signaling, plant−microbe interactions, nitrogen fixation, Rhizobium leguminosarum, nod genes, NodA, naringenin, iTRAQ
■
INTRODUCTION
perceiving unit that is localized at the cytoplasmic bacterial membrane8 where it is postulated to bind with the incoming plant flavonoids.9−11 While the binding of NodD to the nod box promoter of the common nod genes is an ascertained event, the binding of NodD to flavonoids has not yet been demonstrated in vivo. Our recent work has put in evidence a role for calcium in the early signaling events on the bacterial partner side. By using the recombinant expression of the bioluminescent Ca2+ reporter aequorin, we showed that Mesorhizobium loti, which nodulates Lotus japonicus, senses host plant root exudates via transient intracellular Ca2+ elevations.12 Subsequently, in Rhizobium leguminosarum bv. viciae, the transient intracellular Ca2+ increase induced by the flavonoid naringenin was proven to be essential for the activation of nod gene expression, thus
The interaction between leguminous plants and symbiotic nitrogen-fixing bacteria entails a subject of paramount importance both in natural ecosystems’ productivity and in agricultural applications. The controlled invasion of host plants by specific bacteria is also a favorite theme for model-testing studies aiming at unravelling the stepwise molecular dialogue displayed by the two partners.1,2 The signal exchange starts with flavonoids,3 the secondary plant metabolites which trigger gene expression in the symbiotic bacteria.4,5 While initially acting as chemoattractants and stimulators of bacterial multiplication in the rhizosphere, flavonoids are also recognized as specific inducers of the common nodulation (nod) genes in rhizobia. Proteins codified by these genes lead to the synthesis of the chito-lipo oligosaccharide signal (Nod factor) that, in turn, triggers the nodule formation in plants.6,7 The long-standing model on nod gene induction sees the rhizobial NodD transcriptional activator protein as a flavonoid© 2013 American Chemical Society
Special Issue: Agricultural and Environmental Proteomics Received: June 30, 2013 Published: September 16, 2013 5323
dx.doi.org/10.1021/pr400656g | J. Proteome Res. 2013, 12, 5323−5330
Journal of Proteome Research
Article
Figure 1. Outline of the proteomics experimental workflow.
available.16 We could verify that differentially expressed, flavonoid-induced proteins are detectable, and we identified 47 proteins whose expression was increased by the flavonoid naringenin.17 In the present report, which is the second part of that investigation, we now show, among a number of other aspects, that the induction conferred by naringenin to the main nodulation gene nodA is heavily counteracted if Ca2+ availability is suppressed by the calcium chelator EGTA. Our data support the key role played by Ca2+ in the rhizobium nod gene circuitry and the paradigm shift on this aspect of the flavonoid−NodD interactive events.
individuating a novel early event in the signaling cascade leading to nodulation.13 We then showed that naringenin induction of the nod genes can be inhibited by oligogalacturonides, which qualify as novel signals in rhizobium− legume interactions.14 Having addressed genetically the necessity of a calcium influx as a prerequisite for Nod factor production, we shifted the perspective to the proteomic standpoint to test such a novel and critical aspect of Ca2+-dependent gene expression by an independent approach. We took advantage of the iTRAQ technology as isobaric tagging protein quantitation strategy15 combined with LC-MS/ MS analysis performed with a high-resolution mass spectrometer. The iTRAQ tags are a set of 4 (4-plex) or 8 (8-plex) different isobaric tags that covalently react with free amine groups, therefore chemically modifying the N-terminus of each peptide and the side chains of the lysine residues. These chemicals are designed in such a way that, when they label a specific peptide, the total mass of the peptide after the modification is the same regardless of the specific tag that has been used. However, when subjected to fragmentation in the mass spectrometer, the different tags give rise to different specific fragments that can be monitored and quantified. Using this technique, peptide sequence identification and quantification are both achieved at the MS/MS level because the relative intensities of the reporter ions derived from the fragmentation of the tagged peptides are proportional to the relative amount of the peptides (and therefore of the proteins) present in the original samples. First we set up the methodology and tested it on Rhizobium leguminosarum bv. viciae 3841, whose sequenced genome is
■
MATERIALS AND METHODS
Bacterial Growth Conditions and Treatments
R. leguminosarum bv. viciae strain 3841, a spontaneous streptomycin-resistant mutant of field isolate 300, was grown in 200 mL of tryptone-yeast (TY) medium (5 g/L tryptone, 3 g/L yeast extract, and 1.3 g/L CaCl2·6H2O) containing 500 μg/mL streptomycin at 28 °C with shaking (170 rpm). For Ca2+ measurement assays, R. leguminosarum bv. viciae strain 3841 transformed with the pAEQ80 plasmid, carrying the cDNA for aequorin under the control of the strong isopropyl βD-thiogalactopyranoside (IPTG)-inducible synthetic promoter Psyn,14 was used. Bacterial suspension cultures were grown to midexponential phase (OD600 = 0.4) and then incubated in control conditions or with 10 μM naringenin for 6 h. For experiments performed in the presence of the Ca2+ chelator ethylene glycol tetraacetic acid (EGTA), exponentially growing cultures were washed three times with Ca2+-free culture medium (TY medium 5324
dx.doi.org/10.1021/pr400656g | J. Proteome Res. 2013, 12, 5323−5330
Journal of Proteome Research
Article
modification. False discovery rates (FDR) of 5 and 1% were used to filter low confidence data. Proteins were considered as positive hits (both for identification and quantification) if at least two independent peptides were matched with a FDR of 5%. The principle of maximum parsimony was used to group into families of proteins that were identified and quantified separately for each technical replicate. Data from all replicates were then merged to obtain a single list of peptides and proteins (Supporting Information S1 table, showing all the proteins and peptides found, featuring accession numbers to the database, scores, statistics, coverage, and number of peptides of each assigned protein). The list of quantified proteins was further filtered by removing all the proteins that showed a discordant trend in different replicates, and the final fold change was calculated as the average value obtained from all replicates only for the proteins found to be altered in both biological replicates (Supporting Information S2 table, showing the differentially expressed proteins under the different treatments and their ratios, with color-coded assigned classes of behavior with respect to the different incubation conditions). iTRAQ ratios ≥1.5 or ≤0.7 were set as the threshold for overexpression and under-expression, respectively.
deprived of CaCl2) and then resuspended in the same medium supplemented with 1 mM EGTA 10 min before the onset of the incubation in the absence or presence of naringenin. Stock solutions of 50 mg/mL naringenin (Sigma-Aldrich, St. Louis, MO, USA) were prepared in ethanol. Aequorin-Based Ca2+ Measurement Assays
Induction of aequorin expression and measurements of cytosolic Ca2+ concentrations were carried out as previously described.13 Total Protein Extraction
Bacterial cells were harvested by centrifugation at 3500g for 30 min at 4 °C. The cell pellet was washed three times with phosphate-buffered saline (PBS) and then resuspended in 2 mL of PBS and 0.1% (v/v) Triton X-100 supplemented with protease inhibitors (0.5 mM phenylmethylsulfonyl fluoride, 1 mM benzamidine, and 1 μM leupeptin). Bacterial cells were lysed by three cycles of 30 s each of sonication at 35 Hz (Fisher Sonic, Artek Farmingdale, NY, USA), each cycle followed by 30 s incubation on ice. Clear lysates obtained after centrifugation at 1600g for 15 min at 4 °C were used for total protein analysis.
■
Sample Preparation and LC-MS/MS Analysis
RESULTS In parallel to the proteomic analysis, we ran an in vivo physiological assay to ascertain that (a) the flavonoid induction would display a peak of calcium influx in the strain under investigation, and (b) EGTA would efficiently block such intracellular Ca2+ elevation. Results are shown in Figure 2. The
All details regarding protein quantification, trypsin digestion, iTRAQ labeling, sample fractionation, LC-MS/MS analysis, and data analysis, including bioinformatics, are described in ref 17. The experimental workflow and the labeling strategy are presented in Figure 1. Briefly, a 4-plex iTRAQ Reagents Multiplex Kit (AB Sciex, MA, USA) was used for the labeling. The quantitative analysis was carried out on two completely independent biological replicates, obtained from two separate exponentially growing bacterial cell populations subjected to independent treatment with different stocks of naringenin and EGTA. The experiments were performed on different days. Two technical replicates with tag swapping were also conducted for each biological replicate. For each replicate, eight fractions were obtained by strong cation exchange chromatography (SCX), by eluting the labeled peptides in a stepwise manner with the following concentrations of KCl: 20, 40, 60, 80, 100, 150, 200, and 350 mM. Samples were desalted using C18 cartridges (Sep-Pack, C18, Waters, Milford, MA, USA), dried under vacuum, and dissolved in H2O/0.1% formic acid. All LC-MS/MS analyses were conducted with a LTQ-Orbitrap XL mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) coupled online with a nano-HPLC Ultimate 3000 (Dionex−Thermo Fisher Scientific, Waltham, MA, USA) using a 10 cm homemade pico-frit column. A Top3 CID/HCD acquisition method with repeated injections and the application of excluding lists was used for the analysis. All details regarding chromatographic conditions and MS analyses are reported in ref 17. Raw files were analyzed using Proteome Discoverer 1.2 (Thermo Fisher Scientific, Waltham, MA, USA) connected to Mascot (version 2.2.4, Matrix Science, London, UK). MS/MS spectra were matched against a Rhizobium leguminosarum bv. viciae 3841 database (July 2011, 7254 sequences, 2 270 557 residues) concatenated with a database of common contaminant proteins, using trypsin with up to two missed cleavages as digesting enzyme. Peptide and fragment tolerance were set to 10 ppm and 0.6 Da, respectively. Methylthiocysteine, 4-plex iTRAQ at the N-terminus, and Lys were set as fixed modifications and oxidation of methionine as variable
Figure 2. Monitoring of cytosolic Ca2+ concentration ([Ca2+]cyt) in R. leguminosarum bv. viciae strain 3841 in response to naringenin. Where indicated (arrow, 100 s), aequorin-expressing rhizobia were challenged with 10 μM naringenin in the absence (black trace) or presence (gray trace) of 1 mM EGTA. The Ca2+ traces are representative of three independent experiments.
Ca2+ measurements carried out in R. leguminosarum bv. viciae strain 3841 expressing the Ca2+-sensitive photoprotein aequorin demonstrated, in response to the plant flavonoid naringenin (10 μM), the induction of a rapid and transient cytosolic Ca2+ elevation, which peaked at ∼1.30 μM after about 100 s and dissipated within 15 min. When the rhizobial cultures were pretreated with the extracellular Ca2+ chelator EGTA (1 mM) in Ca2+-free medium, the naringenin-induced Ca2+ response was completely abolished, as previously observed in R. leguminosarum bv. viciae strain 300.13 This result indicates that the naringenin-dependent Ca2+ signal derives from a Ca2+ uptake from the extracellular medium. Moreover, these data suggest that EGTA may be a useful tool to discriminate 5325
dx.doi.org/10.1021/pr400656g | J. Proteome Res. 2013, 12, 5323−5330
Journal of Proteome Research
Article
Table 1. Sorting of the Significantly Enhanced (plus) or Repressed (minus) Proteins within Groups Displaying a Given Expression Pattern Ratio When Comparing Pairwise the Four Incubation Conditions (Signs in Bold Indicate a Strong Effect, i.e., Ratios >3 for Upregulation or
