Paraburkholderia phymatum STM815^T Pectate Lyase Has a Negative Impact on Nitrogen-Fixing Symbiosis with Common Bean

Abstract

In the face of global challenges such as food insecurity, environmental degradation, and climate change, biological nitrogen fixation by rhizobia has become increasingly crucial for supporting sustainable agriculture and reducing reliance on synthetic fertilizers. Paraburkholderia phymatum STM815^T is a beta-proteobacterial rhizobium notable for its exceptionally broad host range, forming nitrogen-fixing symbioses with over 50 legume species. In this study, we identified pelB on the P. phymatum STM815^T symbiotic plasmid, which codes for a pectate lyase, whose expression is activated by the presence of pectin in the medium and during symbiosis with common bean. In the absence of pelB, P. phymatum STM815^T shows improved symbiotic performance with common bean. Plants infected with the pelB mutant developed fewer but larger nodules and exhibited a 43% increase in nitrogenase activity, suggesting that pelB in P. phymatum STM815^T may negatively affect nodulation efficiency and nitrogen fixation in common bean.

1. Introduction

Nitrogen is vital for all living organisms since it is a fundamental building block of proteins and nucleic acids [1,2]. Despite nitrogen gas (N₂) comprising roughly 78% of the atmosphere, it often remains a limiting nutrient in ecosystems, as most organisms are unable to use it directly in its atmospheric form [3]. Nitrogen fixation is the process that converts atmospheric N₂ into ammonia (NH₃), the biologically usable form of nitrogen [3]. Ammonia can be produced chemically via the Haber–Bosch process and biologically through biological nitrogen fixation (BNF). Although the Haber–Bosch process has alleviated major agricultural limitations over the past century, it remains a substantial contributor to climate change and to environmental issues such as water pollution, eutrophication, and disruptions of the global nitrogen cycle [4]. BNF accounts for half of the bioavailable nitrogen on Earth and offers a sustainable alternative to synthetic fertilizers [5]. BNF mainly relies on rhizobia, which enter a specialized symbiosis with legume plants, forming root nodules where they live intracellularly in organelle-like structures called symbiosomes. A single bacterium or few bacteria are encapsulated in one symbiosome and differentiate into bacteroids capable of reducing N₂ using their enzyme nitrogenase [6].

Until 2001, all reported rhizobia were classified within the alpha-proteobacterial group (alpha-rhizobia). Since then, the classification of rhizobia has been substantially revised following the discovery of two beta-proteobacterial species within the order Burkholderiales that possess nitrogen-fixing and symbiotic capabilities (beta-rhizobia) [7,8]. The strain we used in this study, Paraburkholderia phymatum STM815^T, was among the first of the beta-rhizobial strains isolated from root nodules [7]. One defining feature of P. phymatum is its broad host range, as it can nodulate more than 50 mimosoid and papilionoid legume species. In co-inoculation experiments, P. phymatum frequently outcompetes other rhizobia, highlighting its strong symbiotic competitiveness for root nodulation [9,10,11]. Originally classified within the genus Burkholderia, the species was reassigned to Paraburkholderia following taxonomic revisions that separated predominantly environmental and plant-associated bacteria from the pathogenic Burkholderia sensu stricto [12,13,14,15]. Consequently, P. phymatum has become a prominent model organism for investigating beta-rhizobial symbiosis. Its large multipartite genome [16] consists of two chromosomes and two plasmids. The smaller plasmid (0.59 Mb), known as the symbiotic plasmid, carries genes involved in symbiotic functions, including those required for nitrogen fixation (nif) and nodulation (nod).

In this study, we report the characterization of a gene located on the symbiotic plasmid that encodes a pectate lyase (PelB, Bphy_7394), an enzyme that cleaves de-esterified pectin (pectate) by breaking the α-1,4-glycosidic bonds between galacturonic acid residues via a beta-elimination mechanism. Pectins constitute a complex family of polysaccharides, including homogalacturonan, xylogalacturonan, apiogalacturonan, rhamnogalacturonan I, and rhamnogalacturonan II, and are a major component of the plant cell wall. Pectate lyases are used by plant-associated bacteria to penetrate or soften plant tissues, support internal colonization, and provide access to plant-derived nutrients [17,18,19]. Plant pathogens use pectate lyases to macerate plant tissues, leading to soft rot and plant disease. Moreover, a recent study showed that a plant nodule-specific pectate lyase is required throughout the infection process during symbiotic interactions [20]. Since P. phymatum enters the plant via infection threads formed in response to Nod factor production, we investigated the role of PelB during symbiosis. Paraburkholderia phymatum pelB was expressed in the presence of pectin in vitro and during the different symbiotic steps. Plant infection assays showed that the pelB mutant induced the formation of fewer nodules on common bean roots; however, these nodules exhibited a 43% increase in nitrogenase activity compared to wild-type nodules. This suggests that pectate lyase activity negatively affects P. phymatum symbiotic performance.

2. Results

2.1. In Silico Analysis of P. phymatum Pectate Lyase PelB

A gene potentially coding for a pectate lyase PelB (Bphy_7394) was identified on the P. phymatum symbiotic plasmid (NC_010627.1). Since pectate lyases are used by phytopathogens to degrade plant cell walls, we aimed to investigate the role of PelB in P. phymatum. A bioinformatic analysis revealed the presence of PelB orthologs in other Paraburkholderia strains. Moreover, the P. phymatum PelB sequence showed between 60% and 62% identity (on a query coverage between 82% and 87%) with pectate lyases from plant pathogens belonging to the Xanthomonadaceae family (Supplementary Table S1); however, no sequence similarity was detected with pectate lyase HrpW from the alpha-rhizobial species Rhizobium etli [21]. To visualize the similarities among these protein sequences, a phylogenetic tree of PelB sequences was constructed using protein sequences from Paraburkholderia species, Xanthomonas species, two Burkholderia strains (Burkholderia metallica A53 [22] and Burkholderia pyrrocinia JBC226 [23]), two Paenibacillus strains (Paenibacillus rigui JCM16352^T [24] and Paenibacillus caseinilyticus GW78^T [25]) and the strain Uilginosibacterium sediminicola M1-21^T [26]. Our phylogenetic analysis revealed a clear separation between the pectate lyases from rhizobial species (Paraburkholderia) and those from plant pathogens (Xanthomonas) (Figure 1). Interestingly, two species belonging to the Burkholderia sensu stricto group (B. metallica and B. pyrrocinia) also harbored genes encoding pectate lyases similar to P. phymatum PelB (68% identity, 86% query coverage) (Supplementary Table S1). However, the pectate lyases from these two species were clustered together and were phylogenetically distant from those found in the genus Paraburkholderia.

Paraburkholderia phymatum PelB belongs to the PL1 family (Polysaccharide Lyase Family 1), displays a typical β-helix structure (Supplementary Figure S1A) and contains an N-terminal Sec-type secretion signal peptide (Supplementary Figure S1B,C). Genes adjacent to pelB are potentially involved in the secretion of the pectate lyase (Supplementary Figure S1D). Upstream of pelB, a gene encoding a YecA-family protein (Bphy_7393) was predicted by AlphaFold3 to interact with the N-terminal signal peptide of PelB (Supplementary Figure S1E). Downstream, a single gene with homology to gspD (Supplementary Figure S1D) encoded one of the main structural components of the type 2 secretion system (T2SS). Notably, the GspD protein on the symbiotic plasmid (Bphy_7395) shared 74.5% sequence similarity with the GspD located in the T2SS cluster on the chromosome (Bphy_3070).

2.2. Paraburkholderia phymatum Is Able to Degrade Pectin

Next, the ability of wild-type P. phymatum and the pelB mutant to degrade pectin was tested. The strains were first grown on LB no salt medium plates for 48 h, after which apple pectin medium was overlaid on top, as described in the Section 4. Dickeya chrysanthemi DSM 4610, which possesses multiple copies of pectate lyase encoding gene in its genome [28], was used as positive control. After a 24 h incubation, the plates were flooded with 1% (w/v) cetyltrimethylammonium bromide. The formation of a clear zone around the colony spotted in the middle of the plate indicated pectinolytic activity of the tested strain. While D. chrysanthemi showed a strong pectin-degradation ability (large clear zone), P. phymatum exhibited weak activity in the wild type and no activity in the pelB mutant (Figure 2), suggesting that apple pectin degradation by P. phymatum in vitro is PelB-dependent.

2.3. PelB Is Expressed in the Presence of Pectin and During Symbiosis

To investigate pelB expression under free-living and symbiotic conditions, the promoter of pelB was fused to gfp, encoding a green fluorescence protein. The pelB-gfp reporter strain (WT-_ppelB) was first grown on a Phenotypic MicroArray^TM plate PM2A (Biolog, Inc., Hayward, CA, USA) for 62 h, during which cell density and fluorescence were recorded (Supplementary Table S2). The Phenotypic MicroArray^TM plate PM2A contained 95 carbon sources, including pectin, as well as a negative control lacking any carbon substrate. In the presence of pectin, D-arabinose, D-arabitol, L-arabitol, and D-raffinose, pelB expression increased overtime (Figure 3), with pectin inducing the highest expression levels (pink line, Figure 3). The induction of pelB expression in presence of 0.75% pectin and 0.2% arabinose was validated using the WT-_ppelB strain and a promoterless reporter strain (WT-pPROBE) as a negative control (Supplementary Figure S2).

Paraburkholderia phymatum pelB expression was also investigated during different stages of symbiotic development. First, the pelB-gfp reporter strain was grown for three days at 28 °C in contact with a germinated common bean (Phaseolus vulgaris cv. Negro Jamapa) seedling. As a positive control, the nodB-gfp reporter strain, which is activated by flavonoids secreted by the root, was used [29]. The promoterless reporter strain served as a negative control and showed no fluorescence under any tested condition [29]. The pelB reporter strain showed fluorescence around the radicle, confirming that pelB was expressed at the early stages of symbiosis (Figure 4A). Next, pelB expression was examined in planta 14 days post-inoculation (dpi) (Figure 4B–D). Common bean seeds were inoculated with the pelB-gfp reporter strain, and roots were examined for developing and mature nodules. Both nodule primordia (Figure 4B) and mature nodules (Figure 4C,D) displayed fluorescence, indicating that pelB was expressed during organogenesis as well as in mature nodules (Figure 4B–D). These results show P. phymatum pelB expression throughout the different nodule developmental stages.

2.4. The Lack of PelB Leads to the Formation of Fewer Nodules with Increased N₂-Fixation Ability

To evaluate if P. phymatum pelB plays a role during symbiosis, the symbiotic properties of the pelB mutant and a complemented strain were tested in planta using common bean and compared to plants inoculated with wild-type P. phymatum, as previously described [30]. After 21 days of incubation in a greenhouse, under controlled conditions, nodules induced by the pelB mutant showed significantly higher nitrogenase activity, fewer nodules and increased nodule weight on average compared to the wild type and the complemented strain (pelB-Comp) (Figure 5). Specifically, nodules formed by the pelB mutant exhibited 43% higher N₂ fixation than those induced by the wild type or the complemented mutant strain (Figure 5a), while the number of nodules was 29% lower (Figure 5b). Interestingly, nodules induced by the pelB mutant were approximately 23% heavier than those containing the wild-type strain (Figure 5c). Nodules formed on plants inoculated with the complemented strain (pelB-Comp) were similar to wild-type nodules in number, weight and nitrogenase activity (Figure 5a–c). These results indicate that the observed differences in symbiotic performance are specifically due to the presence and function of pelB.

Analysis of nodule occupancy in 21-day-old nodules revealed that wild-type P. phymatum and the pelB mutant occupied the nodules at comparable cell numbers (10⁹ cells). In contrast, nodules induced by the complemented strain contained significantly fewer cells (10^8.5 cells), suggesting that, similarly to its effect on nitrogenase activity, pelB negatively influences nodule occupancy (Figure 6).

3. Discussion

In this study, we report the identification and characterization of a P. phymatum gene located on the symbiotic plasmid that is homologous to pelB, which encodes a pectate lyase. Pectate lyases are a class of pectin-degrading enzymes found in organisms from all domains of life [17,31]. These enzymes have been extensively characterized in plant pathogens, which often carry multiple pectate lyases with distinct properties and auxiliary pectinases such as pectin methyl-, acetyl- or feruloyl-esterases [32]. The combined activity of these enzymes enhances pectin degradation, leading to soft rot of plant tissues [17]. While pectate lyases are considered virulence factors in phytopathogens, contributing to plant invasion [17], their function in plant growth promoting rhizobia, which form intracellular N₂-fixing symbiosis with legumes, remains largely unknown. Rhizobia usually colonize plant cells through the formation of a plant-derived tubular structure called infection thread (IT), which forms at the root hair tip and progresses transcellularly through the root cortex, inducing locally confined cell wall modifications [20]. Bacteria are subsequently released into plant cells, where they are surrounded by a symbiosome membrane. Pectins are key components of the IT, and in Pisum sativum, the pectin rhamnogalacturonan-II accumulates in young, non-fixing symbiosomes but is absent in old symbiosomes [33]. Plants also possess nodule-specific pectate lyase enzymes (NODULE PECTATE LYASE NPL) that degrade de-esterified pectins in the legume cell wall, facilitating root-hair formation and nodule organogenesis, as shown in Lotus japonicus, Medicago truncatula and Glycine max [20,34,35]. Very little is known, however, about rhizobial pectate lyases. In Rhizobium etli, a gene in the type III secretion system cluster (hrpW) shows homology to PL3-type pectate lyases and exhibits pectinolytic activity. Although hrpW expression is specifically induced at the plant root surface, deletion of the gene does not affect symbiosis with common bean [21]. Our bioinformatic analysis reveals that P. phymatum PelB is not similar to R. etli HrpW but has close homologs in several Paraburkholderia genomes, seemingly restricted to N₂-fixing South American beta-rhizobia such as P. phenoliruptrix, P. atlantica and P. mimosarum [14]. Outside the Paraburkholderia genus, P. phymatum PelB shows similarity to the pectate lyases of B. pyrrocinia and B. metallica, two known endophytes belonging to the Burkholderia cepacia complex (Bcc) that promote plant growth and suppresses bacterial and fungal pathogens [22,23]. We also observed some similarity between the Paraburkholderia and plant-pathogenic Xanthomonas PelB sequences. Our phylogenetic analysis (Figure 1) suggests that PelB from these beneficial and pathogenic strains may have originated from a distant common ancestor, potentially reflecting evolutionary divergence among symbiotic and pathogenic lineages associated with plant host colonization [31,36].

Paraburkholderia phymatum’s pectate lyase activity was tested using apple pectin, a common source of pectin (15–18% of its content) extracted from apple waste [37]. Wild-type Paraburkholderia phymatum grew and produced a degradation halo, which was absent in the pelB mutant (Figure 2). However, the halo of P. phymatum was smaller than the one observed for D. chrysanthemi DSM 4610, likely due to the presence of multiple pectate lyase-encoding genes in D. chrysanthemi [28]. Interestingly, pelB expression was induced by the presence of pectin and other pectin precursors such as arabinose, which constitute the side chains of rhamnogalacturonan I and II [37] and are involved in the pentose and glucuronate interconversions pathway (Figure 3, Supplementary Figure S2). This observation is consistent with previous studies on the plant-growth-promoting soil bacterium Bacillus velezensis GA1, where pelA and pelB gene expression increased in the presence of pectic homogalacturonan and oligogalacturonides, the fragments generated by pectin cleavage [38]. This suggests that the presence of the substrate and its precursors activate pelB expression. In planta, pelB was expressed throughout all stages of symbiosis, from initial P. phymatum attachment to the root (Figure 4A) to fully developed nodules (Figure 4B–D). It should be noted that the use of the pelB promoter fusion strain provides an indirect assessment of the pelB expression through the GFP fluorescent activity; thus, complementary transcript or protein quantification approaches may be used to confirm pelB expression. Plant infection experiments revealed that the absence of pelB altered P. phymatum behavior during symbiosis. Surprisingly, common bean inoculated with the pelB mutant developed fewer nodules on the root, but these nodules exhibited a 43% increase in nitrogenase activity compared to nodules formed by the wild-type strain (Figure 5a). Introduction of a functional pelB copy into the pelB mutant restored the wild-type behavior in the complemented strain (Figure 5). This suggests that pelB may negatively affect symbiosis and that its substrate is important for P. phymatum performance inside root nodules. A recent study on Medicago truncatula nodules occupied by Sinorhizobium meliloti 2011 showed a correlation between the accumulation of unesterified pectins within symbiosomes and bacterial N₂ fixation activity [39]. Indeed, oligogalacturonides released during cellular damage and microbial infection activate transient cytosolic Ca²⁺ changes and oxidative burst in plant cells and act as damage-associated molecular patterns (DAMPs) that trigger plant defense responses [40,41,42]. It would therefore be interesting to compare the effects of P. phymatum wild type and pelB mutant strain on immune responses within the nodules, as this may explain the reduced nitrogenase activity observed in nodules occupied by the wild type compared to those occupied by the pelB mutant. Nevertheless, it should be noted that pathogenic bacteria typically exhibit higher pectate lyase activity to induce plant host cell death, whereas beneficial rhizobacteria release pectate lyases in a highly localized manner, producing low concentrations of oligogalacturonides that do not elicit a strong immune response [31]. Oligogalacturonides also affect beneficial rhizobacteria such as Rhizobium leguminosarum bv. viciae 3841 by inducing intracellular reactive oxygen species (ROS) accumulation, increasing cytosolic Ca²⁺ levels, and repressing expression of naringenin-induced nodD and nodC genes [43]. In addition, B. velezensis GA1 uses PelA and PelB to generate specific short oligogalacturonides patterns that induce systemic immunity against pathogens in tomato leaves, while eliciting only weak pattern-triggered immunity (PTI) responses in roots, thereby facilitating root colonization in Arabidopsis [38]. Together, these results suggest an additional layer of complexity in the early rhizobium–legume recognition, enabling plants to distinguish between pathogenic and mutualistic organisms. However, the specific substrate of P. phymatum pectate lyase during symbiosis remains unknown. Whether the PelB substrate accumulates in nodules induced by the pelB mutant and is depleted in nodules infected by the wild-type and complemented strains could be addressed by metabolomics analyses of nodules formed by different strains. Moreover, immunofluorescence and immunogold labelling experiments using monoclonal antibodies recognizing different pectin epitopes (e.g., LM19 for de-esterified pectins [20]) could determine which pectins accumulate in common bean nodules occupied by the pelB mutant compared to wild-type nodules.

In conclusion, the absence of pelB homologs in the genome of S. meliloti, as in most alpha-rhizobia, led us to speculate that the presence of pelB in many beta-rhizobia of the genus Paraburkholderia may result in reduced nitrogenase activity but enhanced competitiveness for root infection, a key trait of P. phymatum. Future experiments aimed at assessing the competitive and nodulation abilities of P. phymatum wild-type and pelB mutant strains on different host and non-host plants would be valuable to further investigate the role of the functional PelB encoded on the symbiotic plasmid.

4. Materials and Methods

4.1. Bacterial Strains, Plasmids, and Growth Conditions

The bacterial strains, plasmids, and primers employed in this work are listed in Supplementary Table S3. Escherichia coli and P. phymatum strains were routinely maintained on Luria–Bertani (LB) [44] and modified LB without salt (LB-NaCl: 10 g tryptone and 5 g yeast extract per liter) plates, respectively. When applicable, the following antibiotic concentrations were used: chloramphenicol (20 µg/mL for E. coli and 80 µg/mL for P. phymatum) and kanamycin (25 µg/mL for E. coli and 50 µg/mL for P. phymatum).

4.2. Construction of the P. phymatum pelB Mutant, Complemented, and Reporter Strains

Genomic DNA of P. phymatum STM815^T was isolated by using a GenElute^TM Bacterial Genomic DNA Kit (Sigma-Aldrich, St. Louis, MO, USA). Plasmid DNA from E. coli strains was obtained by using the QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany). To generate a pelB mutant, an internal fragment of Bphy_7394 was PCR amplified using primers Bphy7394_IM_F_EcoRI and Bphy7394_IM_R_XbaI and cloned into a pSHAFT2 plasmid between the EcoRI and XbaI restriction sites. The plasmid pSHAFT-pelB was then transferred into the wild-type P. phymatum to generate the Bphy_7394 insertion mutant (pelB mutant). Genomic integration of the plasmid was confirmed with PCR using primers Bphy7394_veri_R and pSHAFT_R1. The gene sequence of Bphy_7394, together with its native promoter, was amplified by PCR with primers Bphy7394_comp_F_XbaI and Bphy7394_comp_R_EcoRI and was then cloned into pBBR1MCS-2 between the XbaI and EcoRI sites. The constructed plasmid pBBR1MCS-2-pelB was transferred into pelB mutant to complement the strain, resulting in the complemented strain (pelB-Comp), in which the expression of pelB is driven by its native promoter. To study the expression of the pelB gene, a promoter fusion strain was constructed. The promoter was cloned using Bphy7394_comp_F_XbaI and Bphy7394_GFP_R_EcoRI into a pPROBE-NT vector between the XbaI and EcoRI sites. The resulting plasmid was transconjugated into wild-type P. phymatum to create the promoter reporter (WT-_ppelB). After each cloning step, the cloned sequences were confirmed at Microsynth AG (Balgach, St. Gallen, Switzerland).

4.3. Phylogenetic Analysis

The amino acid sequence of P. phymatum PelB was used as query in BLASTP (NCBI, https://blast.ncbi.nlm.nih.gov/, accessed on 22 December 2025) with the RefSeq Select protein (refseq_select) database. Only full-length proteins with at least 60% protein-level identity were included. This dataset was supplemented with another P. phymatum PelB BLASTP query using the non-redundant protein sequence (nr) database to increase resolution in the Burkholderiaceae family. Here, only one full-length protein sequence with at least 65% protein-level identity was included for each species. The full list of sequences including BLAST outputs are available in Supplementary Table S1. Protein sequences were aligned with ClustalW [45] and the phylogenetic analysis was conducted using the Jones–Taylor–Thornton model (+Freq) and a discrete gamma distribution across five categories with invariant sites. MEGA12 was used for alignment and reconstruction of the maximum likelihood phylogeny with 500 bootstrap replicates [46]. In silico predictions on the PelB sequences were performed using the following on-line tools: AlphaFold3 [47] for the structure, SignalP-6.0 [48] for the signal peptide location and DeepLocPro-1.0 [49] for the subcellular location.

4.4. Pectin Degradation Assay

To evaluate the ability to degrade pectin, cultured bacterial cells were washed twice with 10 mM MgSO₄ and adjusted to reach an OD₆₀₀ reading of 0.5. A total of 20 µL of adjusted cell suspension were spotted onto LB-NaCl agar plates. After incubating the plates at 28 °C for 2 days, 10 mL of the melted pectin medium (1.5% apple pectin, 100 mM Tris-HCl, pH 8.5, 1% agar) at 40 °C were poured on top and plates were incubated at 28 °C for another 24 h. The plates were then flooded with 1% (w/v) cetyltrimethylammonium bromide (HCMABr, Darmstadt, Germany, Merck) and incubated at room temperature until clear zones were formed [50].

4.5. Metabolic Profiling Using Biolog Phenotypical Microarray^TM Plates and In Vitro Validation

Biolog Phenotype MicroArray^TM (PM) plates were used to assess the metabolic profile of the pelB reporter strain P. phymatum WT-_ppelB. The protocol provided by Biolog Inc. (Hayward, CA, USA) was followed. In short, the WT-_ppelB strain was plated on R-2A agar two days before the experiment. Colonies from the grown plate were diluted in IF-0 (Inoculation Fluid) provided by Biolog Inc., and the bacterial solution was adjusted and diluted to a final transmission of 85%. A total of 100 μL of the bacterial solution was inoculated in each well of the PM2A plates, which were incubated at 28 °C for 62 h. The OD₆₀₀ and the fluorescence (excitation at 488/9 nm and emission at 520/20 nm) of each well was measured every 12–16 h. Data were normalized to the first OD₆₀₀ measurement at timepoint of zero hours for each compound. The results were evaluated with Microsoft Excel (Microsoft 365, 2021). To validate the results obtained using Biolog PM plates, precultures of the pelB reporter and WT-pPROBE promoterless strains were washed twice with 10 mM MgSO₄ and bacterial suspensions were adjusted to reach an OD₆₀₀ reading of 0.5. A total of 10 µL of the cultures were spotted on plates (10% LB without salt and 0.75% agarose) supplemented with 0.75% citrus pectin (25 mM Tris pH 8.5) or 0.2% arabinose. Plates were incubated at 28 °C for one to three days and images were taken with a custom-built fluorescence imaging device (Infinity 3 camera, Lumenera, Ottawa, ON, Canada) using an excitation/emission wavelength of 490 nm/510 nm for GFP fluorescence. A minimum of one biological replicate was performed.

4.6. Common Bean Seed Sterilization and Inoculation with P. phymatum Strains

Phaseolus vulgaris (common bean) cv. Negro Jamapa seeds were surface sterilized first with absolute ethanol, and then with 35% H₂O₂ for 5 min for each treatment, followed by 10 washes with sterile deionized water, as previously described [51]. Seeds were then placed on 0.8% agarose plates and incubated in the dark for 48 h at 28 °C. Once germinated, seedlings were transferred into yogurt jars filled with nitrogen-free Jensen media [52], vermiculite (VTT-Group, Muttenz, Switzerland) and sand. Bacterial precultures were washed twice with modified nitrogen-free (A)B media [53] and adjusted to reach an OD₆₀₀ reading of 0.025. A total of 1 mL of the adjusted precultured was inoculated onto each seedling. Common beans were incubated for 21 days in a greenhouse with a temperature cycle of 25 °C day/22 °C night, with a constant humidity of 60% and a light intensity of 200 µM for 16 h per day.

4.7. In Planta Expression Analysis

To assess the expression pattern of the pelB promoter, the WT-_ppelB reporter strain was used for the expression analysis on seedlings and in planta. A positive (WT-_p7722, nodB promoter fusion) and a negative control (P. phymatum wild type with empty pPROBE) were included in the experiment [53]. As previously described [54], sterile common bean seedlings were placed in (A)BS (15.13 mM Na₂SO₄, 42.25 mM Na₂HPO₄, 22 mM KH₂PO₄, 51.33 mM NaCl, 2 mM MgCl₂, 0.1 mM CaCl₂, 3 µM FeCl₃, and 15 mM succinate) soft-agar plates inoculated with the reporter strains and incubated for three days. Fluorescence was measured using a Leica M205 FCA fluorescent stereo microscope equipped with a DFC 7000 T CCD camera and the relative fluorescent signal was acquired through an ET GFP filter set (470/40 nm excitation, 525/50 nm emission). Nodules colonized by WT-_ppelB were harvested after 14 dpi and pictures of the surface of nodule primordia and mature nodules were taken using the same settings above. Next, developed nodules were sectioned and stained with 10 µL of Direct Red 23 [55] (1 mg/mL) for five minutes, followed by three washes with deionized H₂O. The relative fluorescent signal was monitored using the ET GFP filter settings described above, and the absorption spectrum of Direct Red 23 dye was captured with TXR filter at an excitation wavelength of 580/60 nm and emission of 610/30 nm. For each strain, three independent biological replicates were prepared.

4.8. Characterization of Symbiotic Properties In Planta

Previously established methods were used to evaluate the symbiotic properties in planta [11,30]. For each strain tested, 15 plants were used, divided in 3 biological replicates. After 21 days of incubation, plants were brought back to the laboratory to perform the Acetylene Reduction Assay (ARA). Briefly, each single root was separated from its stem and was incubated with acetylene in glass tubes for 18 h. The conversion of acetylene to ethylene was measured using gas chromatography (Agilent, Santa Clara, CA, USA). Nodules were then harvested, counted and dried overnight at 60 °C to calculate the dry weight for each nodule. The bacteroid isolation from the nodules was performed by sterilizing at least four nodules per biological replicate, as previously described [11]. Surface-sterilized nodules were then crushed in LB-NaCl with 25% glycerol and serial dilutions were performed and plated in LB-NaCl agar plates to obtain the colony-forming units (CFUs).

4.9. Statistical Analysis

As previously stated, three biological replicates were used for bacterial experiments. Symbiotic properties were analyzed using one-way ANOVA, with strain as a single factor, to compare the means among three different groups (nodules induced by the wild-type P. phymatum, pelB mutant and pelB-Comp strains). Tukey HSD (Tukey Honest Significant Differences) post-hoc test was applied to perform multiple pairwise-comparisons between groups. Statistical analyses and plots were generated using GraphPad Prism v8.4.2 (GraphPad Software^®).