This study assessed the co-cultivation of Gluconacetobacter diazotrophicus and Paraburkholderia tropica, two diazotrophic species previously used in sugarcane, to evaluate their compatibility, phytohormone profiles, and effects on maize (Zea mays) root system architecture. Both strains were grown separately and together, followed by metabolomic analyses and plant bioassays under controlled and greenhouse conditions. The co-cultivation maintained stable populations and produced a wider range of phytohormones than individual cultures, including indole-3-acetic acid (IAA), indole-3-butyric acid (IBA), 4-Cl-IAA, gibberellins, cytokinins, and abscisic acid, with many of these compounds released into the culture medium. Short-term bioassays showed that auxins, particularly IAA, were the only phytohormones produced at concentrations high enough to cause quick and consistent changes in root length and surface area, with IAA levels increasing from 14.07 to 15.30 ng mL⁻¹ in monocultures to 30.68 ng mL⁻¹ under co-cultivation, indicating that auxin-driven signaling is the main mechanism behind early root responses. These findings suggest that co-cultivating compatible diazotrophic bacteria is a viable and practical strategy for developing advanced microbial inoculants to enhance maize root architecture and support sustainable crop production.

Background and aims
Biological nitrogen fixation (BNF) supplies much of soybean (Glycine max (L.) Merr.) nitrogen (N) demand, but reported fixation rates vary widely, leading to uncertainty in the soybean N cycle. We quantified whole-plant soybean BNF and N allocation using long-term 15N labeling to assess whether BNF can offset grain N removal in a high-yielding system, and to quantify root contributions to fixed N.

Methods
Field mesocosms in Iowa, USA received three legacy fertilizer N rates as 2 atom% 15N-enriched urea for four prior seasons of maize (Zea mays L.) production. In 2024, N-fixing and non-N-fixing soybeans were grown to physiological maturity. Grain, aboveground biomass, and roots were analyzed to quantify BNF and to estimate whole-plant N budget.

Results
Soybean derived 81.8% ± 1% (SE) of total plant N from BNF. Legacy N rate had no effect on whole-plant N accumulation, BNF, or N partitioning among tissues (p > 0.05). Roots contained ~2% of total fixed N, with most BNF-derived N allocated to grain. High grain N removal was largely offset by BNF (174.7 ± 8.9 kg N ha−1), resulting in a near-neutral partial N balance (−5.7 ± 2 kg N ha⁻1).

Conclusion
From this one-season mesocosm study, our finding of negligible fixed N in soybean roots indicates that neglecting root BNF does not necessarily bias conclusions about soybean N cycle. However, the combination of long-term isotope labeling and non-nodulating isolines yielded much less variable and higher BNF rates than previous studies, suggesting the possibility of underestimation for soybean BNF in historical studies.

Arbuscular mycorrhizal fungi (AMFs) have facilitated the colonization of land by plants some 470 million years ago. The vast majority of land plants have maintained a symbiotic association with these fungi to facilitate the uptake of mineral nutrients, such as phosphorus, at the cost of photosynthates delivered to the fungus in the form of lipids and sugars. Despite their importance for plant nutrient status, plants can refuse AMF if soil nutrient conditions are such that it is less costly for the plants to take up the nutrients by themselves or if environmental conditions are not appropriate. Recently, Hong et al. (2026) revealed how multiple signaling pathways in rice converge on a transcription factor complex involving the GRAS transcription factors NSP1 and NSP2 to control AM colonization. In this Spotlight, we highlight recent insights into the molecular mechanisms that control the permissiveness of plants to allow AMF entry into their roots (Figure 1).

The rhizobium–legume symbiosis plays a vital role in the global nitrogen cycle. Although microplastics have been shown to affect this symbiotic system, the accumulation and impacts of nanoplastics (NPs) in rhizobia and their root nodules remain poorly understood, particularly regarding the interactive effects of NPs of different sizes on symbiotic nitrogen fixation. This study demonstrated that polystyrene (PS) NPs exhibited a significant size difference effect on rhizobia and their symbiotic nitrogen-fixing association with soybean (Glycine max). We found that both rhizobia and soybean nodules efficiently internalized PS NPs, with differently sized NPs showing mutual enhancement during the cellular uptake of rhizobia. 100 mg/kg of 20 nm PS NPs severely disrupted the symbiotic nitrogen fixation, reducing nitrogenase activity by 51.3% in single exposures and 28.6% in combined exposure to 200 nm PS NPs. This observed disruption caused by 20 nm PS NPs was associated with suppressed nodule formation (26.0% reduction in number, 50.4% decrease in fresh biomass), diminished leghemoglobin content (64.9% reduction), impaired nutrient acquisition (26.5% decrease in nodule Mo content), reduced rhizobia infection efficiency, impaired plant growth, and modified expression of nodulation- and nitrogen-fixation-related genes. These findings revealed that small-sized PS NPs posed a substantial threat to the rhizobium–legume symbiosis, underscoring the ecological risks of NP pollution in agricultural systems.

Nitrogenase is the only known enzyme that catalyzes the reduction of dinitrogen to ammonia. The most prevalent isozyme, molybdenum nitrogenase, comprises the catalytic molybdenum–iron protein (MoFeP) and the ATP-dependent reductase iron protein (FeP). Although Mo-nitrogenases are widespread across bacteria and archaea and appear to share conserved mechanistic and structural features, FeP and MoFeP show considerable sequence variability across diazotrophs. This raises questions about the conservation of chemomechanical mechanisms coupling FeP-dependent ATP hydrolysis and electron transfer to MoFeP, and about the functional compatibility of nitrogenase components from divergent species. Previous studies showed that some heterologous FeP−MoFeP pairs can functionally complement each other, whereas other pairs lack catalytic activity, but the absence of structural information on such heterologous pairs has limited mechanistic understanding. To this end, we investigated the functional and structural compatibility of FeP and MoFeP from Azotobacter vinelandii (Av) and Gluconacetobacter diazotrophicus (Gd), two phylogenetically and ecologically distinct species. Building on our prior work with Gd-nitrogenase and recently developed cryogenic electron microscopy (cryoEM) protocols, we determined the ADP·BeFx-trapped structure of the homologous GdFeP–GdMoFeP complex and showed that it adopted the same geometry as its Av counterpart. Activity measurements showed that heterologous Gd/Av combinations retained 60–80% of homologous catalytic activities despite 30–50% sequence divergence in FeP and MoFeP. High-resolution cryoEM structures of GdFeP–AvMoFeP and AvFeP–GdMoFeP corroborated these activities and revealed that functional complementation tolerates substantial sequence variation when the core structural elements supporting ATP binding/hydrolysis, protein–protein interaction, electron transfer, and substrate reduction are conserved.

Symbiotic nitrogen fixation reduces reliance on synthetic fertilizers and is central to agriculture and ecosystem functioning. In legumes, this process occurs in root nodules where rhizobia differentiate into bacteroids that either remain viable or undergo terminal differentiation, a strategy that can enhance nitrogen fixation but limits nodule life span. This irreversible program suits annual legumes and has evolved convergently across multiple lineages, including the inverted repeat-lacking clade (IRLC), where it is enforced by nodule-specific cysteine-rich (NCR) peptides. By contrast, perennial legumes with indeterminate nodules must sustain symbiosis over extended periods, which is incompatible with terminally differentiated bacteroids. Here, we identify a family of nodule-specific proline-glycine-rich peptides (NPGs) in Robinia that are induced upon rhizobial infection. NPGs are highly expressed in nodules, encode intrinsically disordered peptides, and accumulate in infected cells within the fixation zone. Exposure of Mesorhizobium robiniae to recombinant NPGs induces transcriptional changes associated with a fixation-related physiological state while preserving bacterial viability. These findings identify NPGs as candidate host effectors at the plant-microbe interface and point to an alternative mode of symbiont modulation in perennial legumes.

Cereal root microbiomes harbour diverse diazotrophic bacteria, yet the taxa capable of sustained nitrogen fixation in association with cereal roots remain poorly characterised. Here, two high-performing nitrogen-fixing strains (B6 and J2) were isolated from barley roots and identified as belonging to the family Rhizobiacae in the genus Paenirhizobium. Both strains possess plasmid-encoded canonical rhizobial nif and fix genes for nitrogen fixation but lack nodulation genes. Their genomes have a 5.7 Mb chromosome and four repABC plasmids. Unlike most nodulating rhizobia, strains B6 and J2 fixed nitrogen in laboratory culture on a range of carbon sources, achieving maximal activity on organic acids at low ammonium (<0.5 mM) and oxygen concentrations (1–3%). Both strains colonised the total root systems of barley plants, with population densities of 106 CFU g−1 fresh root weight. Strains fixed high levels of nitrogen on barley plants, similar to or greater than other known free-living diazotrophs. These findings expand the ecological context of rhizobial nitrogen fixation and identify cereal-associated Paenirhizobium as a previously unrecognised component of the diazotrophic cereal root microbiome.

Engineering a novel N2-fixing rhizobia symbiosis in cereal crops is a strategy being pursued to improve agricultural sustainability. However, if such a symbiosis were introduced, it would have to be economically viable in the context of plants’ existing nutrient acquisition strategies, including existing symbioses with arbuscular mycorrhizal fungi (AMF) that most plants already engage in. It is important to understand how the metabolic costs and benefits from these symbioses with overlapping functions might impact plant growth when evaluating the potential benefits of this engineering strategy. To address this, we developed metabolic models describing how the relative growth rate of Zea mays is impacted by the AMF Rhizophagus irregularis and a hypothetical N2-fixing symbiosis with Bradyrhizobium diazoefficiens in isolation and in tandem. The metabolic models of the plant-AMF symbiosis and plant-AMF-rhizobia symbiosis are the first of their kind. To validate the AMF component of our model, we conducted a field evaluation comparing AMF-compatible and mutant AMF-incompatible maize hybrids. The empirically measured AMF-mediated growth benefit agreed well with model predictions. Our model of the rhizobium symbiosis predicted that the lower N content of cereal crops makes the growth penalty associated with acquiring nitrogen from rhizobia smaller than in legumes. Finally, the model of the plant-AMF-rhizobia symbiosis predicted positive synergies between rhizobia and AMF under nutrient-limited conditions but negative synergies under phosphorus-replete conditions. This indicates that these bioengineering strategies could improve cereal crop yields and may achieve greater gains in tandem, but soil nutrient levels and plant nitrogen requirements should be considered.

The mutualistic symbiosis between plants and arbuscular mycorrhizal (AM) fungi is widespread in plants but has been abandoned several times. Abandonment is a two-step evolutionary process that proceeds from the AM state to full abandonment (a non-mycorrhizal (NM) state) via an intermediate AMNM state in which the presence of AM fungi in plants is dependent on ecological context. We tested the hypothesis that the evolution of AM fungal symbiosis abandonment is caused by colder and drier climates as well as elevated soil nutrients, which reduce the benefit-to-cost ratio of the symbiosis from the plant perspective. Using phylogenetic logistic multiple regression and georeferenced environmental data for more than 4000 species, we found that colder temperature was the strongest predictor of evolutionary transitions from the AM to AMNM states, but higher inorganic soil P was the strongest predictor of the evolution of full symbiosis abandonment from the AMNM to NM states. Our findings are consistent with observations that many NM clades evolved during the past 50 million years, as global temperature dropped during the Cenozoic. Furthermore, increased phosphorus deposition from anthropogenic sources could cause the evolution of symbiosis abandonment, particularly in lineages that are in the intermediate AMNM state.

Rhizobia are soil bacteria that establish nitrogen-fixing symbioses with legumes. While many rhizobia use a Type III Secretion System to deliver “Nodulation Outer Protein” (Nop) effectors, some uniquely use these proteins to initiate nodule organogenesis, bypassing classical signalling. The molecular functions of these effectors remain largely unknown due to extreme sequence divergence. Using AlphaFold2-mediated structural proteomics, we identified a modular architecture in rhizobial effectors composed of 22 distinct structural units. We reveal that many Nop effectors are cryptic transcriptional or post-transcriptional regulators, harbouring unrecognised nucleic acid–binding modules and RNA-dependent RNA polymerase domains. Crucially, these modules are conserved in specific plant pathogens, such as gall-inducing Pantoea, where our predicted structural units align with experimentally validated DNA-binding domains. Furthermore, we discovered the BPN (B3 and PUA-like nucleic acid binding) domain as a structural mimic of plant B3-domain transcription factors, pointing to a direct mechanism for hijacking legume development. Our findings strongly suggest that rhizobia employ a modular domain-fusion strategy to act as direct genetic modulators, uncovering a conserved mechanism used by both symbionts and pathogens to hijack host developmental programmes.

Bacterial extracellular vesicles (bEVs) are emerging as key players in interkingdom communication, yet their role in delivering functional proteins to host cells during symbiosis remains unexplored. This study shows that Sinorhizobium fredii HH103 packages a PIN-like auxin transporter, AuxT, into bEVs that traffic within the peribacteroid space of soybean nodules. AuxT is chromosomally encoded and constitutively expressed, genetically uncoupled from the flavonoid-inducible auxin biosynthesis machinery located on the symbiotic plasmid. Structural prediction reveals that AuxT adopts an eight-transmembrane-helix architecture with striking homology to plant PIN auxin exporters, despite negligible sequence identity. Molecular docking demonstrates that AuxT binds indole-3-acetic acid within a central cavity, with dimerization inducing ligand-specific conformational changes consistent with transport activity. The auxT mutant exhibits significant symbiotic defects including reduced shoot biomass, nodule number, and nodule mass that are fully restored by complementation. Critically, AuxT-enriched bEVs contain elevated auxin levels, and nodules colonized by the complemented strain accumulate more auxin specifically within the bEVs peribacteroid space, while bacteroids themselves show no auxin retention. We postulate that bEV-associated AuxT mediates localized auxin export into the symbiosome, modulating the host hormonal environment to optimize symbiotic development. This work reveals a previously unrecognized mechanism of interdomain hormonal modulation, where a bacterium uses a structurally convergent transporter and vesicular delivery to actively shape host physiology and to improve the symbiotic performance.

Legume nodule formation is induced by rhizobia secreted Nod factors (NFs). It has been shown that NF receptors also accumulate in the apex of Medicago truncatula nodules. However, the NF signaling induced transcriptional changes in there have never been studied.

Here, we studied this by using NF signaling mutant TE7, a weak allele of IPD3, blocked in rhizobial release. Nodule apices were isolated with laser microdissection and used for transcriptional analysis.

We identified 1655 NF signaling controlled genes in nodule apex. By comparing this with the transcriptome data from VAMP721d&e RNAi nodule apices, we identified a subset of 445 genes whose expression depends on NF signaling and rhizobial release. Further, we compared the set of genes controlled by NF signaling in nodule apices with that controlled in root epidermis, and these showed only a small overlap. NIN is induced by NF signaling both in the root epidermis and in the nodule. By overexpression of NIN in TE7 and knock down of NIN in wildtype nodules we showed that NF signaling controlled rhizobial release depends on NIN.

NF signaling controls a distinct set of genes in nodules, the function of which depends at least in part on NIN.