Microbiome convergence enables siderophore-secreting-rhizobacteria to improve iron nutrition and yield of peanut intercropped with maize

Wang, Nanqi; Wang, Tianqi; Chen, Yu; Wang, Ming; Lu, Qiaofang; Wang, Kunguang; Dou, Zhechao; Chi, Zhiguang; Qiu, Wei; Dai, Jing; Niu, Lei; Cui, Jianyu; Wei, Zhong; Zhang, Fusuo; Kümmerli, Rolf; Zuo, Yuanmei

doi:10.1038/s41467-024-45207-0

Download PDF

Article
Open access
Published: 29 January 2024

Microbiome convergence enables siderophore-secreting-rhizobacteria to improve iron nutrition and yield of peanut intercropped with maize

Nature Communications volume 15, Article number: 839 (2024) Cite this article

8997 Accesses
13 Citations
35 Altmetric
Metrics details

Subjects

Abstract

Intercropping has the potential to improve plant nutrition as well as crop yield. However, the exact mechanism promoting improved nutrient acquisition and the role the rhizosphere microbiome may play in this process remains poorly understood. Here, we use a peanut/maize intercropping system to investigate the role of root-associated microbiota in iron nutrition in these crops, combining microbiome profiling, strain and substance isolation and functional validation. We find that intercropping increases iron nutrition in peanut but not in maize plants and that the microbiota composition changes and converges between the two plants tested in intercropping experiments. We identify a Pseudomonas secreted siderophore, pyoverdine, that improves iron nutrition in glasshouse and field experiments. Our results suggest that the presence of siderophore-secreting Pseudomonas in peanut and maize intercropped plays an important role in iron nutrition. These findings could be used to envision future intercropping practices aiming to improve plant nutrition.

Legume rhizodeposition promotes nitrogen fixation by soil microbiota under crop diversification

Article Open access 04 April 2024

Harnessing the plant microbiome for sustainable crop production

Article 15 August 2024

Superiority of native soil core microbiomes in supporting plant growth

Article Open access 04 August 2024

Introduction

One of humanity’s greatest challenges is how to sustainably feed a large population¹, especially in countries such as China, comprising almost a fifth of the world population but having limited available arable land. One solution to guarantee food security is to grow at least two crops simultaneously in the same field because intercropping often increases productivity, resource use efficiency, pest and pathology control, and ecological sustainability². Peanut (Arachis hypogaea L.) /maize (Zea mays) intercropping is common in China as it is more effective and ecologically more sustainable than peanut monocropping, particularly for small land holders³. Peanut is an important oilseed legume. China produces 40% and Northern China Plain approximately 20% of the global peanut production⁴ (Fig. 1a). However, peanut yield and quality are severely hampered in this area because of severe iron (Fe) deficiency prevailing in the alkaline and calcareous soils, making iron insoluble and bio-unavailable⁵. This problem is alleviated through intercropping with maize, which improves the iron nutrition and photosynthetic efficiency of peanut (Fig. 1b) and thereby increases yield and land-use efficiency⁵ (Fig. 1c). This intercropping is thought to offer an effective and sustainable way to ensure iron and zinc biofortification to prevent hidden hunger³.

**Fig. 1: Maize/peanut intercropping improves peanut iron nutrition and yield in calcareous soils.**

Here, we use peanut/maize intercropping as a model system to elucidate the exact mechanism leading to increased iron nutrition, particularly focusing on the yet unexplored role of the rhizosphere microbiome. It is known that peanut/maize intercropping improves iron nutrition via direct belowground facilitation through root exudates^5,6. Higher plants show two alternative iron-deficiency-induced molecular and physiological responses^7,8. Peanut follows strategy I (dicots and non-graminaceous monocots) and directly reduces Fe(III) to Fe(II) followed by Fe(II) absorption^6,9. In contrast, maize follows strategy II (graminaceous monocots) and secretes phytosiderophores of the mugineic acid family (MAs) to chelate insoluble Fe(III) followed by MAs-Fe(III) complex absorption^6,10. Maize outperforms peanut in iron acquisition in alkaline soils because strategy II is less sensitive to high pH⁷. It was proposed that intercropping benefits peanut because it enhances deoxymugineic acid (DMA) secretion from maize root, which solubilizes more Fe(III) in the soil that is then absorbed as DMA-Fe(III) complex by nearby peanut plants⁶.

This current plant-plant interaction model for iron utilization does not integrate root-microbiome interactions. This is surprising because it is well established that the rhizosphere microbiome is critical for both plant fitness and iron acquisition^11,12. Specifically, strategy I and II plants secrete different root exudates under iron-limited conditions, to recruit various rhizobacteria that improve plant nutrition in general and iron acquisition in particular^13,14,15,16. A common mechanism of iron acquisition in these beneficial bacteria operates through the secretion of siderophores to dissolve insoluble Fe(III)^11,17. For these reasons, we hypothesize that peanut/maize intercropping could modulate the rhizosphere microbiome through the proximity of plant roots, which in turn could promote the mixing of root exudates, microbiome members, their siderophores and thus improve iron acquisition opportunities.

To test our hypothesis, we first examined whether increased iron acquisition in intercropping depends on the presence of a functional rhizosphere microbiome. Subsequently, we combined microbiome profiling, functional strain characterization, substance identification, and greenhouse and field experiments to establish the mechanistic links between cropping systems, belowground root-microbiome interaction networks and iron acquisition. Our study design allows us to identify microorganisms and microbial metabolites that contribute to the increase in nutrient use efficiency and crop yield by intercropping.

Results

The rhizosphere microbiome contributes to iron nutrition improvement of peanut intercropped with maize

In the first experiment, we explored whether peanut/maize intercropping is beneficial under greenhouse conditions in naturally iron-limited soils. We found that intercropping indeed improved iron nutrition in peanut (Fig. 2a, b), but not in maize (Supplementary Fig. 1). Specifically, intercropping increased chlorophyll levels (Soil Plant Analysis Development (SPAD) value, 17.6% to 51.5%) and active iron concentration (23.7% to 60.3%, indicating Fe-deficiency extent) in young leaves from 53 days post sowing (dps) onwards (Fig. 2a, b). Compared to monocropping peanut, there was a significant increase in the available iron concentration in the rhizosphere (14.3%–21.4%, i.e., iron prevails in a form that can be absorbed by the plant) (Fig. 2b). In contrast, monocropping peanut started to show iron-deficiency chlorosis in young leaves at 53 dps, which progressively worsened over time (Fig. 2a, b).

**Fig. 2: Rhizosphere microbiome contributes to the iron nutrition improvement of peanut intercropped with maize.**

To test whether the rhizosphere microbiome is involved in the observed iron nutrition improvement in intercropping peanut, we repeated our experiments in normal (unsterilized) and sterilized soils (Fig. 2c). For all the three measured parameters – SPAD, active iron in leaves and available iron in the rhizosphere – there were significant statistical interactions between soil treatments (sterilized vs. normal) and cropping types (intercropping peanut vs. monocropping peanut), demonstrating that intercropping peanut was only beneficial and increased iron availability for peanut in the presence of a functional rhizosphere microbiome (Fig. 2d).

Pseudomonas spp. – cross-enriched from maize to peanut – are associated with improved iron nutrition

Intercropping alters rhizosphere microbiota

Given that the rhizosphere microbiome is essential for improving iron nutrition in intercropping peanut, we hypothesized that intercropping improves iron acquisition via modification of the peanut microbiome. To test our hypothesis, we used 16S rRNA amplicon sequencing to profile the rhizosphere microbiome of peanut and maize plants from monocropping and intercropping at four-time points (46, 53, 63 and 73 dps, Supplementary Fig. 2). Rhizobacterial communities formed four distinct clusters (monocropping peanut, monocropping maize, intercropping peanut, intercropping maize) at all stages and showed the greatest divergence at 73 dps (Fig. 3a). While bacterial α diversity in the rhizosphere bacterial communities was not affected by intercropping (Supplementary Fig. 3a), we observed that intercropping peanut and intercropping maize microbiomes are more similar than the monocropping peanut and monocropping maize microbiomes (Fig. 3a and Supplementary Fig. 3b). This pattern suggests that maize and peanut microbiome members are exchanged, possibly through the interwoven inter-species root network formed with intercropping (Supplementary Fig. 4). Altogether, our results reveal that intercropping induces both a change in microbiome composition and a convergence between the peanut and maize microbiomes.

**Fig. 3: Intercropping induces shifts in rhizobacteria community composition and reveals *Pseudomonas* as one of the top candidate taxa involved in intercropping peanut iron nutrition improvement.**

Pseudomonas is the strongest biomarker closely associated with improved iron nutrition

To detect key bacterial taxa linked with improved iron nutrition, we employed the linear discriminant analysis effect size (LefSe). We identified 10 biomarkers at the genus level (absolute LDA-scores > 3.0 and p < 0.05) that were enriched in intercropping peanut compared to monocropping peanut (Supplementary Fig. 5) and were more abundant in monocropping maize compared to monocropping peanut (Fig. 3b). Hence, these taxa are the top candidates for being cross-enriched from maize to peanut in intercropping. We further found that the abundance of those 10 genera positively correlated with either the active iron in young leaves, the iron availability in the rhizosphere, or both metrics (Fig. 3b, Supplementary Fig. 6 showing positive correlation of the two iron metrics). While five genera (Pseudomonas, Pseudoxanthomonas, Luteolibacter, Allorhizobium-Neorhizobium-Pararhizobium-Rhizobium, Sphingobium) showed positive correlations with both iron metrics, Pseudomonas had the strongest association (Spearman’s ρ = 0.81) with rhizosphere iron availability (Fig. 3b and Supplementary Data 1). Thus, our analysis reveals five keystone taxa possibly guiding iron nutrition improvement in intercropping peanut, with Pseudomonas being the top candidate.

Moreover, we found that intercropping also leads to microbiome shifts in the maize rhizosphere. Particularly, we identified three genera (Azotobacter, Rubellimicrobium and Bryobacter) that were enriched in intercropping maize compared to monocropping maize (Supplementary Fig. 7).

Pseudomonas rhizobacteria from intercropping show high siderophore-secreting ability

Due to the high capacity of siderophores to solubilize iron¹¹, we hypothesized that siderophores secreted by rhizobacteria in intercropping could be responsible for the enhanced iron bioavailability in the peanut rhizosphere. To test this hypothesis, we isolated siderophore-secreting rhizobacteria from intercropping peanut using the chrome azurol S (CAS) plate assay (Fig. 4a). We isolated over 300 siderophore-secreting rhizobacteria strains, among which 46 strains had particularly high siderophore-secreting activity (halo diameter on CAS plates > 5 mm, Fig. 4b). Among the top producers, 58.7% were Pseudomonas spp. (according to their 16S rRNA sequence), and the majority of these strains (63.0%) had sequence identities > 97% with a representative amplicon sequence variant (ASV487) obtained by the DADA2 algorithm, capturing one particular clade of Pseudomonas spp. (Fig. 4b and Supplementary Table 1). Importantly, ASV487 was undetectable in monocropping peanut, but detectable in intercropping peanut and its relative abundance was higher in monocropping maize than intercropping maize (Fig. 4c). Additionally, ASV487-similar strains were isolated in the intercropping peanut rhizosphere at a key stage of the iron nutritional shift (46–53 dps) and at the stage where the iron nutrition gaps between monocropping and intercropping peanut were widest (73 dps) (Supplementary Table 1). These results suggest that ASV487 represents a group of high siderophore-secreting Pseudomonas strains that were cross-enriched from maize to peanut during intercropping and are associated with improved iron nutrition in peanut.

**Fig. 4: Functional siderophore-secreting rhizobacteria and siderophore characterization.**

Next, we focused on Pseudomonas sp. 1502IPR-01, a strain that is 97.7% identical to ASV487 and had the highest siderophore-secreting ability among the isolated strains (Fig. 4b and Supplementary Fig. 8). We used Pseudomonas sp. 1502IPR-01 as a representative strain to experimentally explore its role in iron nutrition improvement in intercropping peanut. We sequenced the whole-genome of Pseudomonas sp. 1502IPR-01 and found that 1502IPR-01 is similar to P. extremorientalis (Fig. 4d and Supplementary Fig. 9). By searching its genome through cluster analysis, we detected the pyoverdine biosynthetic and secretion cluster sharing 51.4–84.1% protein sequence identity with homologous genes of P. aeruginosa PAO1, the model strain for pyoverdine research (Fig. 4e and Supplementary Table 2).

To confirm that Pseudomonas sp. 1502IPR-01 indeed secretes pyoverdine as its primary siderophore, we isolated the iron-chelating compound (Supplementary Fig. 10) from the supernatant, after cultivation in the iron-limited medium. Our assay revealed that 90.5% ± 0.9% of the iron-chelating ability of supernatants was attributable to siderophores (Supplementary Table 3). Structure elucidation and chemical analysis (Fig. 4f, Supplementary Figs. 11–21 and Supplementary Tables 4 and 5, details in “Methods” section) yielded a chemical formula of C₅₇H₈₅N₁₆O₂₅, and the chemical structure confirms that the primary siderophore secreted by Pseudomonas sp. 1502IPR-01 is a pyoverdine (Fig. 4f).

Next, we assessed the activity of the detected pyoverdine to bind Fe(III) from an insoluble Fe(OH)₃ source in suspension. Fe(III) chelation is indicated by a color change from clear to yellow to brown (Fig. 4f). We found that the chelated iron concentration in the solution increased with an equimolar amount of the added pyoverdine, matching the color change of the pyoverdine-Fe(III) control complex (Fig. 4f and Supplementary Fig. 10c). In sum, our results identify pyoverdine as the primary siderophore secreted by Pseudomonas sp. 1502IPR-01, and show its high capacity to dissolve Fe(III) from insoluble Fe(OH)_3, which corresponds to the main form of iron in calcareous soil¹⁸.

Pseudomonas sp. 1502IPR-01 and its pyoverdine improve peanut iron nutrition in greenhouse and field experiments

To experimentally validate that Pseudomonas is a keystone taxon improving iron nutrition through pyoverdine secretion, we conducted greenhouse and field experiments. For the greenhouse experiments, we grew peanut in monocropping and intercropping peanut in sterilized calcareous soil either without treatment as control, or with Pseudomonas sp. 1502IPR-01 or its iron-free pyoverdine as treatments. Both treatments significantly increased SPAD values, active iron concentration in young leaves and in the rhizosphere, and improved peanut biomass in monocropping and intercropping peanut (Fig. 5a, b). We observed the same significant improvements when repeating the experiment in normal non-sterile soil in monocropping peanut (Fig. 5b). Thus, Pseudomonas sp. 1502IPR-01 and pyoverdine treatments can compensate for iron deficiency in peanut in the absence of a functional microbiome in intercropping and monocropping peanut in normal soil. In contrast, the iron nutrition of maize did not improve when treated with Pseudomonas sp. 1502IPR-01 or pyoverdine (Supplementary Fig. 22). However, maize biomass significantly increased with both treatments in sterilized and normal soil in intercropping maize (Supplementary Fig. 22), suggesting that maize benefits from intercropping with peanuts by other means than improved iron nutrition.

**Fig. 5: *Pseudomonas* sp. 1502IPR-01 and its siderophore (pyoverdine) prevent iron-deficiency chlorosis and improve peanut growth and yield.**

Next, we investigated whether the application of Pseudomonas sp. 1502IPR-01 and its iron-free pyoverdine are also effective treatments in the field. With intercropping peanut in normal soils, the application of Pseudomonas sp. 1502IPR-01 and its iron-free pyoverdine had no effect on iron nutrition and biomass (Supplementary Fig. 23), probably because intercropping itself suffices to mitigate iron deficiency in peanut. Hence, we focused exclusively on monocropping peanut and treated plants with either Pseudomonas sp. 1502IPR-01 or pyoverdine (both applied via root watering) at two field sites in North China Plain, characterized by calcareous soils. At both sites, treatments significantly increased SPAD values (increase: 45.9–67.6%), active iron concentration in young leaves (78.2–107.5%), iron availability in the rhizosphere (53.6–73.2%) and crop yield (44.8–89.8%), compared to the control plants that received no treatment (Fig. 5c, d). We further applied a traditional treatment used by local farmers, which involved foliar spraying with EDTA-Fe, a synthetic iron chelator. We observed that treating plants with either Pseudomonas sp. 1502IPR-01 or its pyoverdine increased peanut iron nutrition and yield to a similar extent and in one case even more than the traditional EDTA-Fe treatment (Fig. 5d). These results indicate that secreted or added pyoverdine of Pseudomonas sp. 1502IPR-01 could effectively chelate Fe(III) in calcareous soils in the field, and this mechanism is responsible for improved peanut iron nutrition and crop yield. Thus, pyoverdine could be considered as an iron fertilizer for sustainable agriculture without the need for external iron supply.

Validation experiments reveal that pyoverdine directly affects plant iron metabolism

To functionally validate that pyoverdine contributes to improved peanut iron nutrition, we conducted experiments with a defined pyoverdine null mutant (Fig. 6a). Both the lab strain P. aeruginosa PAO1 and the natural isolate Pseudomonas sp. 1502IPR-01 significantly improved monocropping peanut iron nutrition and biomass in normal soil. In stark contrast, treatment with the siderophore-deficient mutant PAO1 ΔpvdDpchEF (Fig. 6b) had no beneficial effects on plants, showing that pyoverdine is required to improve peanut iron nutrition.

**Fig. 6: Pyoverdine secreted by *Pseudomonas* spp. is essential to improve peanut iron nutrition and biomass.**

We then asked whether pyoverdine secretion feeds back on the activity of the plants’ own iron acquisition systems. Indeed, we observed that the ferric-chelate reductase (FCR) activity of peanut is significantly upregulated in the presence of the siderophore-deficient mutant (Fig. 6b), but down-regulated in the presence of pyoverdine-producing wildtype strains and when pyoverdine is added (Supplementary Fig. 24). The same responses occurred in maize, which down-regulated its phytosiderophore DMA in the presence of Pseudomonas sp. 1502IPR-01 or its pyoverdine (Supplementary Fig. 24). These experiments highlight that pyoverdine directly affects the iron metabolism of plants.

Discussion

Intercropping is important for sustainable agriculture and food security¹⁹. However, the precise mechanisms by which intercropping increases yield often remain unclear. In this study, we find that improved iron nutrition is mediated by the bacterial rhizosphere microbiome and arises through microbiome convergence between the two plant species. Specifically, we find that intercropping enables beneficial bacteria such as Pseudomonas spp. to translocate from the maize to the peanut rhizosphere, where these bacteria increase the amount of bioavailable iron through the secretion of their siderophore pyoverdine. The iron chelated to pyoverdine increases iron nutrition in peanut plants and negates iron-deficiency chlorosis in leaves (Figs. 4–7). Siderophores (e.g., pyoverdine) have previously been associated with beneficial effects for plants^20,21,22,23, and our study now provides evidence linking bacterial pyoverdine production in planta to improved plant iron nutrition (Fig. 6). This insight closes a major knowledge gap in the research field. Crucially, we show that the uncovered microbe-plant interaction is essential to improve iron nutrition, as intercropping in the absence of a functional microbiome is not beneficial (Figs. 2, 5). Our results further show how detailed knowledge of inter-kingdom interaction mechanisms helps to develop ecologically sustainable agriculture and food security.

**Fig. 7: Model of how peanut/maize intercropping enriches for *Pseudomonas* spp. in the rhizosphere to improve iron nutrition via secreting siderophore pyoverdine.**

Two lines of evidence suggest that the microbiome interactions observed in our greenhouse experiments occur in the field. First, the peanut and maize roots can easily connect across at least 60 cm in the field, matching the results from our pots experiments²⁴. Furthermore, root connections across similar distances also occur between maize/faba bean and maize/soybean^25,26. Thus, the rhizosphere microbiome convergence based on root connections between peanut and maize observed in the greenhouse likely also occurs in the field. Second, our field experiments themselves demonstrated that the addition of Pseudomonas sp. 1502IPR-01 or its pyoverdine improved peanut iron nutrition, showing that peanut plants can benefit from iron made available by Pseudomonas spp. Hence, pyoverdine-mediated iron mobilization during microbiome convergence could be a major component of improved iron nutrition and yield of peanut intercropped with maize.

Our work strengthens the view that Pseudomonas is a keystone taxon involved in plant iron nutrition. We show that Pseudomonas sp. 1502IPR-01 and its pyoverdine improve peanut iron nutrition in calcareous soils in both greenhouse and field conditions through microbiome convergence between peanut and maize (Figs. 3–6). Similar findings regarding the enrichment of Pseudomonas in peanut intercropped with maize have been reported in acidic soil²⁷. However, important to note is that our analysis revealed additional members of the rhizosphere microbiome that were also associated with iron nutrition improvement, for example, members of the genera Sphingobium and Luteolibacter. Therefore, it would be important to conduct experimental work with characterized strains of these taxa in the future, to assess their role more closely. The observed beneficial effects in intercropping are likely a bacterial community effect with Pseudomonas spp. possibly taking the lead in the process (Figs. 3 and 4). Attributing positive effects to a single species is clearly advantageous when aiming at developing bio-fertilizer treatments for agricultural applications.

The positive effects of pyoverdines on iron nutrition suggest that these molecules take on a yet undescribed role for rhizosphere communities under field conditions. Previous research revealed that specific bacterial siderophores, including pyoverdines, can be beneficial for plants because they suppress plant pathogens in the rhizosphere²³. The proposed mechanism entails that pyoverdines (and other siderophores) are strong iron chelators, and because they are taxon-specific that can lock iron away from pathogens that do not possess the matching receptors, hence acting as ‘public bads’²⁸. Competition for iron can therefore drive phytopathogen control and crop protection by natural rhizosphere microbiomes²³. In intercropping, pyoverdine seems to primarily serve as a ‘public good’ that solubilizes iron for the plant and possibly also for rhizosphere microbiome members possessing matching receptors²⁸. In principle, pyoverdine could exert a double role in intercropping by suppressing phytopathogens (public bad effect) and by improving iron nutrition of plants (public good effect). Important to note is that plants and microbes may also compete for iron. Siderophores like pyoverdines are primarily produced by bacteria to satisfy their own need for iron. But given that siderophores are highly diffusible they might become available and utilizable for plants²⁹ (Figs. 5–6). The loss of siderophores to plants could exert positive feedback on bacterial siderophore production as compensation. Conversely, microbial siderophores may also feedback on plant iron-uptake systems. For example, microbial siderophores may chelate iron from phytosiderophore-Fe(III) complexes or Fe(III) chelated by other root exudates, rendering these uptake systems less efficient. Thus, it remains unclear whether the observed iron nutrition improvement in intercropping occurs because plants efficiently exploit bacterial siderophores or whether there is a mutualistic interaction between the two interacting partners.

Our study reveals that pyoverdine can dissolve insoluble iron and make it bioavailable for plants (Figs. 4f, 5, 6). Earlier work has also reported the ability of several plant species to obtain iron from iron-loaded pyoverdines^11,20,21. However, the exact biochemical mechanism of how plants retrieve iron from pyoverdines remains unclear. One option is that plants take up the entire Fe(III)–pyoverdine complex and retrieve iron through the degradation of pyoverdine. The pyoverdine backbone consists of a peptide chain and could thus be degraded by proteases. Another option is that plants induce the reduction of Fe(III) to Fe(II) through their ferric-chelate reductase (FCR) to then directly absorb Fe(II). While FCR certainly plays a role, we observed that pyoverdine supplementation led to the down-regulated FCR activity in peanuts and DMA production in maize, which represent the regular iron-deficiency responses of these two plant species⁸ (Supplementary Fig. 24). These changes indicate that pyoverdine becomes the preferred route of iron acquisition. In line with this hypothesis, we observed that the down-scaling of the plants’ own iron-uptake systems goes along with biomass increases for both peanut and maize plants (Fig. 5 and Supplementary Figs. 22–24), suggesting that both plants benefit and save resources by taking up iron chelated to pyoverdines. This form of nutrient cycling mediated by plant-microbiome interactions is destroyed in sterilized soil, resulting in a decreased SPAD value in intercropping peanut, even lower than in monocropping peanut (Figs. 2d and 5b). However, intercropping can still increase the active iron in young leaves of peanut (Figs. 2d and 5b), suggesting that DMA secreted by maize can directly be utilized by intercropping peanut⁶ in sterilized soil. Therefore, our findings indicate that multiple iron sources are tapped during intercropping, including siderophores from bacteria, DMA from maize and possibly also flavins or coumarins from peanut root exudates, to mitigate the plants’ iron-deficiency. However, due to the large differences in iron-chelating affinity between pyoverdine and root exudates^10,30, the possibility is that plants balance the iron-uptake strategies between themselves systems and pyoverdine-based ones in intercropping, but future research is needed to test this hypothesis.

While we primarily focused on how iron nutrition is improved in peanut when intercropped with maize, we also detected benefits for maize in a microbiota-dependent manner, suggesting that intercropping benefits both plant species. Maize and peanut could engage in a ‘giving-receiving feedback’ interaction through the convergence of their microbiomes in intercropping. Peanut plants are legumes and host nitrogen-fixing bacteria in their root nodules³¹, such that peanut could improve nitrogen acquisition of maize^27,32. We indeed found evidence for this hypothesis. Specifically, we detected an enrichment of members of the Allorhizobium-Neorhizobium-Pararhizobium-Rhizobium genus and Devosia genus from maize to peanut (Fig. 3b). Increased abundance of nodule-forming Allorhizobium-Neorhizobium-Pararhizobium-Rhizobium and nonrhizobial nodule-inducing Devosia³³ bacteria could increase nodule activity and N₂ fixation in peanut, which in return could enhance nitrogen acquisition by nearby maize through reduced competition for soil and fertilizer nitrogen between the two plant species³¹. Furthermore, we found Azotobacter, a typical plant beneficial non-symbiotic nitrogen-fixing bacterial genus, to be cross-enriched from peanut to maize, possibly also contributing to improved nitrogen nutrition in maize. Our results are in line with previous findings on microbiota-dependent improvement of nitrogen nutrition^{31,32,34,35,36}. Overall, legume/gramineae combinations are promising candidates for intercropping and are therefore the subject of intensive research in soil science and plant nutrition.

Bioavailability of iron is particularly low in alkaline calcareous soils. That is the main reason why we focused on bacterial members of the rhizosphere microbiome, because bacteria are known for producing a diversity of siderophores, with some of them (e.g., pyoverdine) having a high affinity for iron^22,29,30. However, important to note is that other factors might also play a role in determining iron availability and benefits of intercropping. For instance, soil nutrients like NH₄⁺ and NO₃⁻ (the main forms of nitrogen in the rhizosphere) can impact pH and therefore iron availability. Long-term application of NH₄⁺ decreases rhizosphere pH, while NO₃⁻ increases rhizosphere pH^37,38. However, in our intercropping system, the pH remained stable across nutrient stages (Supplementary Fig. 6b), implying that the nitrogen cycle may have no direct effect on iron nutrition improvement in our system. The reason may be that calcareous soils have strong pH buffering capacity⁵. Moreover, soil properties, like NH₄⁺, NO₃⁻, or organic carbon, can potentially affect microbial community structure and function, meaning that beneficial taxa like Pseudomonas might not be equally abundant in all soils. Finally, while we focused on bacteria other members of the microbiota might also be involved in intercropping benefits. Top candidates are the nitrogen-fixing rhizobium (as found in our study for maize) and arbuscular mycorrhizal fungi. Understanding the effects of such multi-kingdom plant-bacteria-fungi interactions across soils with different properties and geographical scales for intercropping will be the challenge of future work.

Our results show that intercropping in natural soils is sufficient to improve iron nutrition in calcareous soils (Supplementary Fig. 23). However, intercropping might not always be possible. For those situations, the insights from our study provide a simple practicable treatment, which consists of the addition of Pseudomonas sp. 1502IPR-01 or its pyoverdine to the rhizosphere. Our treatment has several advantages over the use of traditional iron fertilizers such as synthetic EDTA-Fe with low efficiency and high environmental risk in soil^39,40. These issues do not apply to the pyoverdine treatment because it directly solubilizes iron from natural stocks in soil, pyoverdine itself is biodegradable^22,29,30, and P. extremorientalis (closest relative of Pseudomonas sp. 1502IPR-01) is classified as a harmless biosafety risk 1 organism that is unlikely to cause disease in humans⁴¹. Hence, Pseudomonas sp. 1502IPR-01 and its pyoverdine can be considered ecosystem-friendly biofertilizers. Additionally, pyoverdines can also enhance copper bioavailability and thus have the capacity to improve plant nutrition for other microelements⁴². Given its dual role in microelements (especially iron) fertilization and phytopathogen control^{22,23,30,42,43}, pyoverdine could simultaneously reduce the use of synthetic microelements-fertilizers and pesticides, which could present a critical step forward toward sustainable agriculture, improved food security and environmental conservation.

In summary, we elucidated a mechanism of how intercropping benefits plants through rhizosphere microbiome convergence. Our study highlights that the benefit of intercropping from belowground facilitation leads to a gained function for peanut through a cross-enrichment of rhizosphere microbiome members obtained from maize (Fig. 7). Our findings offer applied opportunities for a probiotic treatment to improve plant health and crop yield in situations where intercropping is not possible. While we have identified a probiotic strain and its secreted beneficial compound, there is great potential to improve the process through bioengineering. Our mechanistic insights on iron nutrition improvement offer perspectives and should trigger interest in promoting other intercropping systems with other plant combinations to increase both yield and sustainability.

Methods

NE and LER calculation

To quantify the benefits of peanut/maize intercropping, we calculated the land equivalency ratio (LER) and the net effect (NE) of this intercropping system for various field sites and years. LER > 1 indicates that intercropping has an advantage over monocropping in land-use efficiency. NE > 0 indicates that intercropping has more absolute yield than monocropping¹.

The NE is defined as the difference between the observed yield and the expected yield.

$${{{{{\rm{NE}}}}}} = ({{{{{{\rm{Y}}}}}}}_{1}+{{{{{{\rm{Y}}}}}}}_{2})-({{{{{{\rm{EY}}}}}}}_{1}+{{{{{{\rm{EY}}}}}}}_{2})$$

(1)

where Y₁ and Y₂ are observed yields of species 1 and 2 in the intercropping, and EY₁ and EY₂ are expected yields of the two species, calculated as the product of the monocropping yield and the land share¹.

$${{{{{{\rm{EY}}}}}}}_{1} = {{{{{{\rm{M}}}}}}}_{1}\times {{{{{{\rm{LS}}}}}}}_{1}$$

(2)

$${{{{{{\rm{EY}}}}}}}_{2} = {{{{{{\rm{M}}}}}}}_{2}\times {{{{{{\rm{LS}}}}}}}_{2}$$

(3)

where M₁ and M₂ are yields per unit area of species 1 and 2 in monocropping, and LS₁ and LS₂ are the land shares of species 1 and 2 in intercropping. The land share was calculated based on the densities of a species in the intercropping and in the monocropping or on the basis of the number of rows or plant arrangements¹.

The LER is defined as the sum of partial LERs (relative yields) per species (pLER1 and pLER2):

$${{{{{\rm{LER}}}}}} = {{{{{{\rm{pLER}}}}}}}_{1}+{{{{{{\rm{pLER}}}}}}}_{2} = \frac{{{{{{{\rm{Y}}}}}}}_{1}}{{{{{{{\rm{M}}}}}}}_{1}}+\frac{{{{{{{\rm{Y}}}}}}}_{2}}{{{{{{{\rm{M}}}}}}}_{2}}$$

(4)

where Y₁ and Y₂ are the yields (per unit of the total area of the intercropping) of species 1 and 2 in intercropping, and M₁ and M₂ are the yields of species 1 and 2 in monocropping (same as above)¹.

Experimental design for peanut/maize intercropping under pot conditions

We conducted intercropping and monocropping experiments in pots. A typical iron-deficient calcareous sandy soil was collected from 0 to 20-cm-depth in Lihua village, Daxing district of Beijing, China (116°15′ E, 39°35′ N). Soil properties before fertilizer application were determined according to protocols from the previous literature⁶ and listed in Supplementary Table 6. Each pot contained 8 kg of soil amended with basal fertilizers [composition (mg kg⁻¹ soil): N 100 (as Ca(NO₃)₂·4H₂O), P 150 (as KH₂PO₄), K 100 (as KCl), Mg 50 (as MgSO₄·7H₂O) and Zn 5 (as ZnSO₄·7H₂O)].

Peanut (Arachis hypogaea L. cv. Luhua14) and maize (Zea mays L. cv. Zhengdan958) were used. Three cropping treatments were set up. Six peanut plants or three maize plants were grown in a pot for the monocropping treatments (monocropping peanut and monocropping maize). Three peanut plants with three maize plants were grown in a pot for the intercropping treatments. The plants were grown under natural light conditions with 28–33 °C air temperature, 400–450 mmol m⁻² s⁻¹ light intensity and 70–75% relative humidity. During the experiment, soil water content was kept at approximately 80% of the field capacity. Plant samples of peanut were harvested to assess the nutrition status at 46, 53, 63, and 73 days post sowing (dps). At the same time, pots were destroyed to acquire complete soil. Bulk soil was removed by gently shaking. The roots of intercropping maize and intercropping peanut were carefully separated from each other when shaking. Rhizosphere soil was collected for microbiome and soil properties analysis by removing the soil closely adhered to the roots using a sterilized fine brush (Supplementary Fig. 2). The peanut plants and rhizosphere soils from the same pots were mixed into one replicate. Each treatment had four replicates at each time point.

Design for soil sterilization experiment in pot conditions

To assess the role of the rhizosphere microbiome in intercropping peanut iron nutrition improvement, monocropping peanut and peanut/maize intercropping were grown in both normal and sterilized soil. Soil was prepared as described above. For soil sterilization treatment, the soil was γ-irradiated with a maximum dose of 20 kilograys (24 h, ⁶⁰Co-γ, Beijing Atomic High-tech Jinhui Radiation Technology Application Co. Ltd.). Plant species, growth conditions, and crop management were the same as in the experimental design for peanut/maize intercropping in pot conditions. At 73 dps, plant samples of peanut were harvested to assess the iron nutrition status, and rhizosphere soil was collected to measure available iron concentration. The peanut plants and rhizosphere soil from the same pot were mixed and treated as one replicate. Each treatment had three or four replicates.

Measuring iron-related indicators in plants and soil

SPAD value

Chlorophyll levels in fresh young leaves (top fully emerged leaves) from peanut was measured using a chlorophyll meter (Konica–Minolta, Osaka, Japan). Each leaf was measured in four replications and averaged while avoiding the main leaf veins.

Ferric-chelate reductase (FCR) activity of peanut plants

According to previous literature⁴⁴, the fresh peanut plants were washed and then the roots were immersed in saturated CaSO₄ solution for 30 min. After washing, peanut plants were transferred to a 100 Ml reagent consisting of 0.4 Mm 2,2′- bipyridine and 0.1 Mm Fe(III)-EDTA, for a 2-h incubation, protected from light. The absorbance of the solution at 523 nm (A₅₂₃) was measured. Ferric-chelate reductase activity was calculated with the formula (5) below and water was used as a control:

$${{{{{\rm{Ferric}}}}}} - {{{{{\rm{chelate}}}}}}\,{{{{{\rm{reductase}}}}}}\,{{{{{\rm{capacity}}}}}}({{{{{\rm{nmol}}}}}}\,{{{{{{\rm{g}}}}}}}^{-1}\,{{{{{\rm{FW}}}}}}\,2{{{{{\rm{h}}}}}}) \\ = {{{{{{\rm{A}}}}}}}_{523}\times 100/({{{{{\rm{FW}}}}}}\times 8650)\times {10}^{9}$$

(5)

where A₅₂₃ is the absorbance of the solution under 523 nm, FW is the fresh weight of the root, and 8650 is the molar absorptivity of Fe(II)-2,2′- bipyridine (L mol⁻¹ cm⁻¹). 100 is the volume of plant incubation solution.

Deoxymugineic acid (DMA) secretion capacity of maize plants

DMA is only phytosiderphore secreted by maize, so we used phytosiderphore secretion-determination method⁴⁵ to determine the DMA secretion capacity. The fresh maize plants were washed and then the roots were immersed in saturated CaSO₄ solution for 30 min. After washing, maize plants were transformed to 200 mL water for a 4-h incubation, protected from light. Before incubation, 10 mg L⁻¹ micropur (Katadyn Products Inc. Wallisellen, Switzerland) was added to prevent microbial degradation of phytosiderphores. 10 mL above-incubated solution was added 0.5 mL of 1 mM FeCl₃ and shaken for 15 min, then 1 mL sodium acetate buffer (pH 7.0) was added and the solution was shaken for a further 10 min. To reduce Fe(III) to Fe(II), the solutions were filtered into 0.25 ml of 6 M HCl and then 0.5 mL 80 g L⁻¹ hydroxylamine hydrochloride was added. The solution was incubated at 55 °C for 30 min. Then, the solution is mixed with 0.25 mL 2.5 g L⁻¹ ferrozine and 1 mL 2.0 M sodium acetate buffer (pH 4.7). After 5-min incubation, the absorbance of the solution at 562 nm was measured. DMA secretion activity was calculated with the formula (6) below by using water as a control:

$$ {{{{{\rm{DMA}}}}}}\,{{{{{\rm{secretion}}}}}}\,{{{{{\rm{capacity}}}}}}({{{{{\rm{nmol}}}}}}\,{{{{{{\rm{g}}}}}}}^{-1}\,{{{{{\rm{FW}}}}}}\,4{{{{{\rm{h}}}}}}) = {{{{{{\rm{A}}}}}}}_{562}\times 1.35\times 200/\\ ({{{{{\rm{FW}}}}}}\times 1/0.000045)\times {10}^{9}$$

(6)

where A₅₆₂ is the absorbance of the solution under 562 nm. FW is the fresh weight of the root. 1/0.000045 is the molar absorptivity (L mol⁻¹ cm⁻¹). 200 is the volume of plant incubation solution. 1.35 is the volume ratio of the final and original reaction solution.

Active iron in young leaves

Fresh young leaves from peanut were washed with distilled water and collected to assess HCl-extractable iron (‘active iron’) according to the previous literature⁶. 1 g of young leaves was taken and mixed with 10 mL of 1 M hydrochloric acid. The supernatant was shaken for 5 h at 150 r.p.m. and then filtered with filter paper. The iron concentration was measured by inductively coupled plasma-optical emission spectrometry (ICP-OES) using a 7300DV system (Perkin Elmer, Waltham, USA). Active iron in young leaves was calculated with the formula (7) below and deionized water was used as a control:

$${{{{{\rm{Active}}}}}}\,{{{{{\rm{iron}}}}}}\,{{{{{\rm{in}}}}}}\,{{{{{\rm{young}}}}}}\,{{{{{\rm{leaves}}}}}}(\mu {{{{{\rm{g}}}}}}\,{{{{{{\rm{g}}}}}}}^{-1}) = ({{{{{{\rm{CONCN}}}}}}}_{{{{{{\rm{Fe}}}}}}}-{{{{{{\rm{CONCN}}}}}}}_{{{{{{\rm{control}}}}}}})\times 10/{{{{{\rm{FW}}}}}}$$

(7)

where CONCN_Fe and CONCN_control are the Fe concentration (μg mL⁻¹) of filtrate of sample and control, respectively, and FW is the fresh weight of the young leaves. Multiplication with 10 is applied to account for the volume of hydrochloric acid.

Available iron in the rhizosphere

10 g of air-dried rhizoplane soil was weighed and added to 20 mL of diethylenetriaminepentaacetic acid (DTPA) extracting agent, containing 0.005 M DTPA, 0.01 M CaCl₂ and 0.1 M triethanolamine. Then, the mixture was shaken for 2 h at 180 r.p.m. at 25 °C. Subsequently, the solution was filtered and the iron concentration in the filtrate was measured by ICP-OES. Available iron in the rhizosphere was calculated with the formula (8) below and deionized water was used as a control:

$${{{{{\rm{Available}}}}}}\,{{{{{\rm{iron}}}}}}\,{{{{{\rm{in}}}}}}\,{{{{{\rm{the}}}}}}\,{{{{{\rm{rhizosphere}}}}}}(\mu {{{{{\rm{g}}}}}}\,{{{{{{\rm{g}}}}}}}^{-1}) = ({{{{{{\rm{CONCN}}}}}}}_{{{{{{\rm{Fe}}}}}}}-{{{{{{\rm{CONCN}}}}}}}_{{{{{{\rm{control}}}}}}})\times 20/{{{{{\rm{weight}}}}}}$$

(8)

where CONCN_Fe and CONCN_control are Fe concentration (μg mL⁻¹) of the soil sample and control, weight is the weight of the soil sample. Multiplication with 20 is applied to account for the volume of DTPA-extracting agents.

Soil DNA extraction, 16S rRNA amplicon sequencing

16S rRNA amplicon sequencing was performed to examine the impact of intercropping on rhizosphere microbiome compositions. As with any microbiome sequencing study, it is important to consider potential contamination sources. We can identify two such potential sources in our rhizosphere microbiome sample collection scheme, which are (a) incomplete root separation and (b) soil mixing during plant harvest. For (a), it could be that peanut and maize roots cannot fully reliably be distinguished and separated when collecting them. We can refute this possibility because the roots of peanuts and maize can be clearly distinguished based on their color: roots of maize are white, while roots of peanut are dark brown (Supplementary Fig. 25). This distinct difference allowed us the completely and reliably separate the roots of the two species. For (b), there is the possibility that loose soil particles originally associated with the root of one of the species are transferred to the soil collected from the other species during the harvesting process. To prevent this source of contamination, we first vigorously shook the roots of the collected plant to get rid of all loose soil particles that are typically associated with the sandy soil. Subsequently, we only collected the rhizosphere soil that was tightly attached around the peanut or maize roots. These two points ensure to reduce the contamination risk to a minimum and that the convergence we observed stems from the biological interactions between the fine roots of the two plant species. DNA was isolated from 500 mg of soil sample using the FastDNA® Spin Kit for Soil (MP Biomedicals, Solon, USA) according to the manufacturer’s instructions. The hypervariable region V3–V4 of the bacterial 16S rRNA gene was amplified with primer 338F (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) by an ABI GeneAmp® 9700 PCR thermocycler (Applied Biosystems, Foster City, USA). PCR conditions were as follows: 3 min at 95 °C, then 27 cycles of each 30 s at 95 °C, 30 s at 55 °C and 30 s at 72 °C. After these cycles, there was a 10-min elongation time at 72 °C. Purified amplicons were pooled in equimolar and paired-end sequenced on an Illumina MiSeq PE300 platform (Illumina, San Diego, USA) according to the standard protocols by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China).

The upstream analysis progress of the amplicon sequence was processed using QIIME2 v2021.8⁴⁶. Using the DADA2 workflow in QIIME2, barcodes, PCR primers and low-quality sequences were trimmed, and the remaining reads were subsequently denoised and merged to generate the amplicon sequence variants (ASV) feature table. A taxonomy native Bayes classifier trained on the V3–V4 region (338F and 806R) of the reference 16S sequence (SILVA release 138) was used to assign taxonomic identities to the representative sequences. Only ASV that were at least annotated to the phylum level were retained. Features identified as chloroplast or mitochondrial were removed. Features with a total abundance (summed across all samples) of less than ten or present in less than six samples (10% samples) were filtered. We normalized the sequencing depth to 17,063 counts per sample using scaling with ranked subsampling (SRS)⁴⁷, preserving the original community structure by minimizing subsampling errors. Representative sequences were filtered based on the filtered ASV table, then used to generate an approximately-maximum-likelihood phylogenetic tree using FastTree 2.1.8 in QIIME2 with default parameters.

PCoA analysis and Bray-Curtis distance calculation were performed by R ‘phyloseq’ package (version 1.34.0) based on the relative abundance of ASV. Differential microbiota between groups was identified by linear discriminant analysis (LDA) and LDA effect size (LEfSe) with an LDA threshold of 3 and p-value threshold of 0.05. LefSe analysis was performed on the Huttenhower lab Galaxy server (http://huttenhower.sph.harvard.edu/galaxy). The format conversion from ASV table to table for Lefse was performed by R ‘amplicon’ package (version 1.14.2). The correlation between rhizobacteria genus biomark’s relative abundance and soil plant iron nutrition was also based on relative abundance and calculated using non-parametric Spearman’s rank correlation (presented as Spearman rho) using R ‘phyloseq’ package (version 2.2.5).

Isolation of siderophore-secreting rhizobacteria

We isolated siderophore-secreting-rhizobacteria from the rhizosphere of intercropping peanut to assess the role they play in intercropping peanut iron nutrition improvement. The isolation method was modified according to protocols from the previous literature⁴⁸ (Fig. 4a) and details are described below. Briefly, 1 g of the soil sample from intercropping peanut rhizosphere was moderately agitated in 20 mL of sterile 0.9% NaCl solution for 10 min first. Then the microbial suspension (serially diluted) was spread on agar plates with 1% (v/v) chrome azurol S (CAS) indicator. CAS indicator contained 1 mM CAS, 0.1 mM FeCl₃; 4 mM hexadecyltrimethylammonium bromide (HDTMA). Other components of the CAS agar medium consisted of (mg L⁻¹): sucrose, 2000; casein acid hydrolysate, 3; CaCl₂, 222; MgSO₄, 241; agar, 1800 and 0.05% (v/v) 0.1 mM phosphate buffer solution (24.27 g L⁻¹ Na₂HPO₄·12H₂O; 5.905 g L⁻¹ NaH₂PO₄·2H₂O, 0.75 g L⁻¹ KH₂PO₄, 2.5 g L⁻¹NH₄Cl, 1.25 g L⁻¹ NaCl, pH = 6.8). The pH of the medium was adjusted to 7.0. The ternary complex CAS-Fe(III)-HTDMA forms a blue dye. Siderophores remove Fe(III) from the dye by chelating Fe(III), which induces a color change from blue to orange (in agar plate) or red (in solution)⁴⁵. The plates were incubated at 30 °C for 48 h in the dark. The microbial colonies with or without an orange halo were picked. The isolates were then re-streaked on LB plates for colony purification. The final collection consisted of 324 siderophore-secreting rhizobacteria isolates. All purified isolates were cultured in 1 mL liquid LB medium at 30 °C with shaking (rotary shaker at 180 r.p.m.) before freezing and storing at −80 °C in 15% glycerol. Requests for strains should be addressed to zuoym@cau.edu.cn.

Measuring the siderophore-secreting capacity of rhizobacteria

To quantify the siderophore-secreting capacity of each rhizobacterium isolate under iron-limited conditions, the rhizobacteria isolates were revived by transferring 50-μL volumes of their freezer stocks into 5 mL liquid LB medium and cultured 18 h at 30 °C with shaking (rotary shaker at 180 r.p.m.). Then, the 50-μL volumes of revived cultures were transferred into 5 mL liquid standard succinic medium (SSM) without added iron (6 g L⁻¹ K₂HPO₄, 3 g L⁻¹ KH₂PO₄, 1 g L⁻¹ (NH₄)₂SO₄, 0.2 g L⁻¹ MgSO₄·7H₂O, 4 g L⁻¹ succinic acid, pH = 7.0). After 48-h incubation at 30 °C, the bacterial cultures were diluted to OD₆₀₀ = 0.8. Then, cell-free supernatant from diluted bacterial culture was harvested by centrifugation (12,096 × g, 10 min at 4 °C) and filtration (using a 0.22 µm filter). Siderophore concentration in the supernatant was assayed using a modified version of the universal chemical CAS assay. Briefly, CAS assay solution was prepared as follows: a 21.9 mg quantity of HDTMA was dissolved in 25 mL water while constantly stirring over low heat. 1.5 mL of 1 mM FeCl₃·6H₂O (in 10 mM HC1) was mixed with 7.5 mL of 2 mM CAS. This solution was slowly added to the HDTMA solution while stirring, and the mixture volume was made up to 100 mL with deionized water. Cell-free supernatant or deionized water as a control reference was mixed with an equal volume of CAS assay solution in a 96-well plate and incubated for four h at room temperature (Supplementary Fig. 8). Siderophores removed the Fe(III) from pigmented Fe(III)-CAS-HDTMA complex, inducing a color change from blue to red in the CAS medium solution, lowering the OD₆₃₀ measurements. The OD₆₃₀ of the cell-free supernatants (A) and deionized water controls (Ar) was then measured using a plate reader (SpectraMax M5, Molecular Devices, Sunnyvale, USA). Siderophore-secreting capacity can be quantified using the following formula: Su = (Ar−A) ÷ Ar. This formula quantifies the relative reduction in the OD₆₃₀ measurement induced by iron-chelating compounds that retrieve Fe(III) from the pigmented Fe(III)-CAS-HDTMA complex. The Su-value exhibits exponential instead of linear changes when the siderophore is excessive (Su > 0.73). Thus, the supernatant for the CAS assay reaction solution should be gradient diluted by 1/2 CAS assay solution to ensure Su < 0.73. Desferrioxamine B (DFOB), a purchasable siderophore, was used to establish a calibration curve to estimate the concentration of siderophores in the supernatants of the rhizobacteria. Serial dilutions (1, 3, 6, 12, and 24 µM) of the DFOB standard were mixed with CAS assay solution and incubated for four h at room temperature before OD₆₃₀ determination. The Su-value of 24 µM DFOB was 0.71, and the calibration curve equation was: siderophore equivalent = 32.99 × Su + 27.46 (R² = 0.999, p < 0.001). Results are expressed as µM equivalents of DFOB. To calculate the iron-binding capacity of bacterial siderophores, one mmol of DFOB is considered to bind one mmol Fe(III).

Tree construction and phylogenetic analysis of rhizobacterial siderophore-secreting isolates

We sequenced the 16S rRNA genes of the siderophore-secreting rhizobacterial isolates to confirm their genus-group and evolutionary relationships. According to the manufacturer’s protocol, the genomic DNA was extracted using Wizard® Genomic DNA Purification Kit (Promega, Charbonnières, France). Purified genomic DNA was quantified by a TBS-380 fluorometer (Turner BioSystems Inc., Sunnyvale, USA). High-quality DNA (OD_260/280 = 1.8~2.0, >20 μg) was used to do further research. The 16S rRNA gene was amplified using 27 F (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1492 R (5′-TACGGYTACCTTGTTACGACTT-3′) primers by the polymerase chain reaction (PCR). The PCR reactions (25 μL) contained one μL bacterial DNA, 12.5 μL master mix, one μL each of the forward and reverse primers, and 9.5 μL deionized water. The PCR was run as follows: initial: denaturation at 95 °C for 5 min, 30 cycles of denaturation at 94 °C for 30 s, annealing at 58 °C for 30 s, extension at 72 °C for 1 min 30 s, and the final extension at 72 °C for 10 min. The PCR products were sequenced by Beijing Genomics Institution Co., Ltd (Beijing, China).

The 16S rRNA gene sequences were trimmed at both ends to obtain maximum overlap using the software DNAStar (version 7.1.0, DNASTAR Inc., Madison, USA). The 16S rRNA gene sequences and homologous sequence similarity were identified using the NCBI database (https://www.ncbi.nlm.nih.gov/). We constructed an approximately-maximum-likelihood phylogenetic tree using FastTree 2.1.8 in QIIME2 with default parameters. Trees were visualized using ggtree v2.4.1 in R v4.2.2.

Total genomic DNA extraction and sequencing of rhizobacteria isolate

The genome of Pseudomonas 1502IPR-01 isolate, which had the highest siderophore-secreting capacity in all rhizobacteria isolates, was sequenced using a combination of PacBio RS II Single Molecule Real Time (SMRT) and Illumina sequencing platforms. The Illumina data were used to evaluate the complexity of the genome.

For Illumina sequencing, at least 1 μg genomic DNA was used in sequencing library construction. DNA samples were sheared into 400–500 bp fragments using a M220 Focused Acoustic Shearer (Covaris, Woburn, USA) following the manufacturer’s protocol. Illumina sequencing libraries were prepared from the sheared fragments using the NEXTflex™ Rapid DNA-Seq Kit (Bioo Scientific, Austin, USA). Briefly speaking, 5’ prime ends were first end-repaired and phosphorylated. Next, the 3’ ends were A-tailed and ligated to sequencing adapters. The third step was to enrich the adapters-ligated products using PCR. The prepared libraries were then used for paired-end Illumina sequencing (2 × 150 bp) on an Illumina HiSeq X Ten machine (Illumina, San Diego, USA).

For Pacific Biosciences sequencing, an aliquot of 15 μg DNA was spun in a Covaris g-TUBE (Covaris, Woburn, USA) at 4355 × g for 60 s using an Eppendorf 5424 centrifuge (Eppendorf, NY, USA). DNA fragments were then purified, end-repaired, and ligated with SMRTbell sequencing adapters following the manufacturer’s recommendations (Pacific Biosciences, Menlo Park, USA). Following the manufacturer’s recommendations, the resulting sequencing library was purified three times using 0.45 × volumes of Agencourt AMPure XP beads (Beckman Coulter Genomics, Danvers, USA). Next, a~10 kb insert library was prepared and sequenced on one SMRT cell using standard methods.

Genome assembly

The complete genome sequence of Pseudomonas sp. 1502IPR-01 was assembled using both the PacBio and Illumina reads. The original image data was transferred into sequence data via base calling, defined as raw data, and saved as a FASTQ file. Those FASTQ files contain read sequences, and quality information is included. A statistic of quality information was applied for quality trimming, by which the low-quality data can be removed to form clean data. The reads were then assembled into a contig using a hierarchical genome assembly process (HGAP) and canu, a software that performs scalable and accurate long-read assembly via adaptive k-mer weighting and repeat separation. The last circular step was checked and finished manually, generating a complete genome with seamless chromosomes and plasmids. Finally, error correction of the PacBio assembly results was performed using the Illumina reads using Pilon. All of the above analyses were performed using I-Sanger Cloud Platform (www.i-sanger.com) from Shanghai Majorbio.

Phylogenetic tree construction of the 1502IPR-01 and representatives of Pseudomonas spp

To confirm the exact evolutionary relationships of 1502IPR-01 and representatives of Pseudomonas spp., a multi-locus sequence analysis-based phylogenetic tree was generated using a previously established method⁴⁹. Specifically, TBLASTN+ (v. 2.10.0) was used to identify and extract sequences for 100 housekeeping genes used in Davis et al.⁴⁹. All 100 housekeeping genes of 1502IPR-01 and 23 completely sequenced Pseudomonas spp. were aligned using MAFFT (v.7.427). Phylogenetic tree construction was done using IQ-TREE2. Trees were visualized using ggtree v2.4.1 in R v4.2.2. Information of used sequenced Pseudomonas spp. strains were listed in Supplementary Table 7.

Pyoverdine biosynthesis and transport gene cluster analysis

The pyoverdine biosynthesis and transport genes locus in 1502IPR-01 genome were identified using TBLASTN+ (v. 2.10.0) with the protein sequences from Pseudomonas aeruginosa PAO1 used as queries against the predicted protein sequences from 1502IPR-01 genome as the databases. Sequence comparisons were conducted using Smith-Waterman algorithm by SnapGene software (version 6.0.2, GSL Biotech LLC, Boston, USA) to obtain DNA sequence identity. Visualization was performed by Adobe Illustrator CC, 2020 (Adobe, San Jose, USA).

Pyoverdine production and purification

For pyoverdine production, Pseudomonas sp. 1502IPR-01 was cultured in an above-mentioned SSM medium. After culturing at 30 °C at 180 r.p.m for 48 h, the culture media (10 L) was isolated from the cell materials by centrifugation at 12,096 × g for 15 min and filtration with filter paper. The crude culture extracts were adjusted to pH 5.5 by the addition of HCl and adsorbed by XAD-4, a polymeric adsorbent resin. After removal of supernatant, the resin was then filtered, and the pyoverdine-containing fraction was subsequently eluted with MeOH-H₂O 1:1 (v/v). The MeOH-H₂O 1:1 (v/v) elute was evaporated to dryness in a vacuum. Then the residues were dissolved in 20 mL H₂O and subjected to the Shimadzu semi-preparative HPLC (Shimadzu, Tokyo, Japan) at 20 °C with a flow rate of 3.5 mL min⁻¹. The injection volume was 100 μL. Samples were chromatographed using a solvent system of 0.1% trifluoroacetic acid (TFA) in H₂O (solvent A) and MeCN-MeOH 4:1 (v/v, solvent B). A linear binary gradient-0–15 min 11.0–12.5% B, 15–20 min 12.5–13.0% B, 20–25 min 13.0–100.0% B-was used. The eluate was monitored at 380 nm. The faction containing pyoverdine was collected at 13–14.5 min and then adjusted to pH 7.0 with 0.1% ammonium hydroxide before evaporation. The salt in the fraction generated during the neutralization of TFA was removed by a Spectra/Por Dialysis membrane with a molecular weight cut-off (MWCO): 500–1000 Da (Spectrum Lab., Rancho Dominguez, USA). Then, the purity > 99.5% of isolated pyoverdine was confirmed by HPLC.

Structural identification of pyoverdine

Mass spectrometry

LC-HRMS spectra were obtained on an Agilent 6530 Q-TOF mass spectrometer coupled to an Agilent 1260 HPLC (Agilent Technologies GmbH, Waldbronn, Germany); Thermo Orbitrap MS (Orbitrap Fusion Lumos, Thermo Fisher Scientific, San Jose, USA); ESI-QQQ-MSMS (Xevo TQ-S micro, Waters, Milford, USA).

NMR data were obtained using (operating at 500.13 MHz for ¹H-NMR and 125.75 MHz for ¹³C-NMR) (Bruker Biospin GmbH, Karlsruhe, Germany). All spectra were measured at 303 K. The residual solvent signals were used for referencing spectra in the ¹H and ¹³C dimensions.

UV/Vis

The UV spectra were recorded by a SPD-M20A Shimadzu photodiode array detector (PDA, detection 190–800 nm) (Shimadzu, Kyoto, Japan) connected to a Shimadzu HPLC system (LC-20AR, Shimadzu, Japan) during semi-preparative HPLC separations in MeCN-MeOH-H₂O containing 0.1% TFA in H₂O.

Semi-preparative HPLC

Semi-preparative HPLC separations were conducted on a Shimadzu HPLC system (LC-20AR, Shimadzu, Japan) with an Atlantis Prep OBD T3 column (5 μm, 150 × 19 mm, Waters, USA).

Chemicals

Ultrapure water was acquired from Watsons Ltd. (Hong Kong, China). Acetonitrile (HPLC grade) and TFA (HPLC grade) used for UHPLC/Q-TOF-MS were purchased from Merck (Darmstadt, Germany) and Roe Scientific Inc. (Newark, USA), respectively. Sodium carbonate (HPLC grade) was purchased from Sigma-Aldrich (St. Louis, USA). The dialysis membrane was purchased from Spectrum Laboratories, Inc. (Rancho Dominquez, USA).

Pyoverdine (Supplementary Fig. 11a), obtained as a yellowish-green crystalline powder, showed the typical UV/Vis spectra of pyoverdines: 387 nm at pH 7.0 and a split band of 365 and 407 nm at pH 3.0 and 10.0. The molecular formula of C₅₇H₈₄N₁₆O₂₅ was determined by the pseudo-molecular ion peak at m/z 1393.5876 [M + H]⁺ (calculated for C₅₇H₈₅N₁₆O₂₅⁺, 1393.5872) in positive HR-ESI-MS (Supplementary Fig. 11b). Total hydrolysis of pyoverdine indicated a peptide chain containing the three types of amino acids: serine (Ser), lysine (Lys) and 5-N-formyl-5-N-hydroxyornithine (FoOHOrn) (Supplementary Fig. 12).

It has been well documented that mass spectrometry is a powerful method for structural elucidation of pyoverdines⁵⁰. Herein, we obtained the following characteristic fragment ions of pyoverdine by using an orbitrap higher-energy collisional dissociation (HCD) approach with varying collision energies. An ion at m/z 204.08, the most characteristic ion for the pyoverdine chromophore (A1’-2, Supplementary Fig. 11c), was clearly recognized in the orbitrap MS (Supplementary Fig. 11d). Nearby, another abundant ion at m/z 270.09, corresponding to the chromophore connected with a carboxylic amide side chain (A1’-1) (Supplementary Fig. 11c, d), could be explained as the fragment ion of the opening of the tetrahydropyrimidine ring of the chromophore by retro-Diels-Alder (RDA) reaction. Then, the side chain ketoglutaric acid (kgl) of pyoverdine was evidenced by the loss of [CO₂ + H₂O] from the A1 (m/z 445.14) and B1 (m/z 473.13) ions yielding the two corresponding A1’ (m/z 383.14) and B1’ (m/z 411.13) ions (Supplementary Fig. 11e)⁵⁰. Hence, the structure-fragmentation pathways from B1 to A1’-2 were proposed as shown in Supplementary Fig. 11c. Furthermore, the series of B-ions from B8 (m/z 1265.49) to B1 (m/z 473.13) allowed to determine of the amino acid sequence of the peptide chain as Ser–Ser–FoOHOrn–Ser–Ser–(Lys–FoOHOrn–Lys–Ser) (Supplementary Figs. 12–14). The series of B’-ions (lossing of H₂O + CO₂ from the kgl chain of the B-ions, Supplementary Figs. 12–14) form B8’ (m/z 1203.49) to B1’ (m/z 411.13) corroborated the peptide sequence. Taken together, the primary structure of pyoverdine was deduced as kgl-chr-Ser–Ser–FoOHOrn–Ser–Ser–(Lys–FoOHOrn–Lys–Ser), which is identical to the pyoverdine from Pseudomonas spp. CFML 95-275⁵¹.

The structure of pyoverdine deduced from MS data was further confirmed by the NMR data. On the basis of a combination of 1D (¹H- and ¹³C-NMR) and 2D (COSY, NOESY, HSQC, and HMBC) NMR experiments (Supplementary Figs. 15–20), the ¹H- and ¹³C-data of pyoverdine were assigned in Supplementary Tables 4 and 5. In addition, the key COSY and HMBC correlations are shown in Supplementary Fig. 21. The results showed that ¹H-NMR data of pyoverdine was almost identical to that of the reported pyoverdine⁵¹. While the ¹³C-NMR chemical shift values of pyoverdine shifted 1–2 ppm, which could be caused by the different acquisition temperatures or machines. The pyoverdine molecule consists of three parts: (i) a dihydroquinoline-type chromophore (chr): (1S)-5-amino-2,-3-dihydro-8,9-dihydroxy-1H-pyrimido-[1,2-a]quinolone-1-carboxylic acid, (ii) a peptide comprising 9 amino acids: Ser-Ser-FoOHOrn-Ser-Ser-(Lys-FoOHOrn-Lys-Ser) (Ser: serine; FoOHOrn: 5-N-formyl-5-N-hydroxyornithine; Lys: lysine), and (iii) a side chain (α-ketoglutaric acid, kgl) bound to the nitrogen atom at position C-3 of the chromophore.

The capacity of pyoverdine of Pseudomonas sp. 1502IPR-01 to solubilize iron from Fe(OH)₃

Fe(OH)₃ is the main insoluble form of iron in calcareous soil. Therefore, we determined the capacity of the pyoverdine of Pseudomonas sp. 1502IPR-01 to solubilize iron from insoluble Fe(OH)₃ stocks according to previous paper⁵². For this purpose, 0, 0.025, 0.2, 0.5, 1 μmol equivalent of purified pyoverdine (calculated based on the DFOB calibration curve) was added into 2 mL 0.5 M sodium acetate solution. Then, all solutions were added to 0.5 mL 5 mM Fe(OH)₃ (Fe(OH)₃ solution was prepared by neutralizing FeCl₃ with NaOH to pH 6.5–7.0) suspensions to a total volume of 5 mL. The above solutions were incubated at 55 °C for five hours, and the supernatant was collected after centrifugation at 1000 × g at 25 °C for 10 min. Finally, the iron content in the supernatant was determined by ICP-OES using a 7300DV system (Perkin Elmer, Waltham, USA).

Functional verification of Pseudomonas sp. 1502IPR-01 and its siderophore

We identified the effect of Pseudomonas sp. 1502IPR-01 and its siderophore on peanut iron nutrition improvement in both greenhouse and field conditions. The normal and sterilized iron-deficient calcareous sandy soil was prepared as described above for pot experiment. Peanut (Arachis hypogaea L. cv. Luhua14) and maize (Zea mays L. cv. Zhengdan958) were used. Four treatment groups were set up: (1) monocropping peanut grown in normal soil; (2) monocropping peanut grown in sterilized soil; (3) peanut/maize intercropping grown in normal soil; (4) peanut/maize intercropping grown in sterilized soil. Six monocropping peanut plants were grown in a pot for the monocropping peanut treatment, and three peanut plants with three maize plants were grown in a pot for the intercropping treatment. Each pot contained 8 kg of soil. The rest of the greenhouse culture conditions remained identical as described above. At anthesis (50 dps), pod bearing (64 dps), and fruit-swelling stage (75 dps), 600 mL of 10⁸ CFU mL⁻¹ 1502IPR-01 strain solution (1 × 10¹⁰ CFU plant⁻¹) or 600 mL of 20 μM purified pyoverdine (2 μmol plant⁻¹) solution were irrigated into the rhizosphere as 1502IPR-01 and pyoverdine treatment. The 600 mL of water was irrigated into the rhizosphere as control. It is worth noting that we applied 20 μM pyoverdine solution to treat the rhizosphere of peanut according to previous literature²¹ and our own research experience⁶. The peanut plants in the same pot were mixed as a replicate. Each treatment had three or four replicates. At 84 dps, SPAD values and active iron concentration in young leaves were measured. Rhizosphere soil was collected to measure the available iron concentration.

The field experiment was conducted in Xichang village (116°10′ E, 39°39′ N), Fangshan district of Beijing, and Ligu village (115°9′ E, 36°7′ N) in Puyang city, Henan province, China in 2020. In Beijing, the peanut cultivar was Luhua14. The average temperature from May to September was 25.6 °C, and the annual precipitation was 603 mm. The type of soil in Beijing is calcic cambisol. The soil property was pH 7.9, total N 0.038%, available P (Olsen-P) 5.12 mg·kg⁻¹, available K (NH₄OAc-K) 60.32 mg·kg⁻¹, available Fe 3.35 mg·kg⁻¹. In Henan, the peanut cultivar was Yuhua37. The average temperature from May to September was 25.4 °C, and the annual precipitation was 604.1 mm. The type of soil in Henan is fluvo-aquic soil. The soil property was pH 7.7, total N 0.041%, available P (Olsen-P) 7.42 mg·kg⁻¹, available K (NH₄OAc-K) 70.2 mg·kg⁻¹, available Fe 3.74 mg·kg⁻¹. 900 kg hm⁻² compound fertilizer (N-P-K: 19-19-19) was applied before sowing, and field management was according to local practices. Four treatments were set up: (1) 150 mL water was irrigated in the rhizosphere, and 16 mL water was sprayed on young leaves for one peanut plant as a control treatment; (2) 150 mL 10⁸ CFU mL⁻¹ 1502IPR-01 strain solution was irrigated in rhizosphere, and 16 mL water was sprayed on young leaves for one peanut plant as 1502IPR-01 treatment; (3) 150 mL 48 μM purified pyoverdine (7.2 μmol plant⁻¹) was irrigated into rhizosphere, and 16 mL water was sprayed on young leaves for one peanut plant as pyoverdine treatment; (4) 150 mL water was irrigated in the rhizosphere and 16 mL 48 μM EDTA-FeNa solution was sprayed on young leaves for one peanut plant as EDTA-Fe treatment. All spraying solutions contained 0.1% Triton X-100 as surfactant. Each treatment consisted of six field plot replications in Beijing and eight field plot replications in Puyang. Each field plot replication consisted of 16 holes, and two peanut plants were grown in one hole. The plant spacing was 30 cm, and row spacing was 60 cm. At anthesis (53 dps in Beijing and Puyang), pod bearing (65 dps in Beijing and 68 dps in Puyang), and fruit-swelling (79 dps in Beijing and 84 dps in Puyang) stage, the respective treatments were repeated. Two weeks after the third treatment, SPAD values and active iron concentration in young leaves were measured. Rhizosphere soil was collected to measure the available iron concentration. Then, peanut plants were harvested to measure yield.

Pot experiment with pyoverdine null mutant

To further strengthen the point that pyoverdine is produced in soil and involved in peanut iron nutrition, we conducted additional experiments with the laboratory strain P. aeruginosa PAO1 and its defined isogenic pyoverdine null mutant PAO1 ΔpvdDpchEF. PAO1 ΔpvdDpchEF is a double knockout mutant of the primary siderophore pyoverdine (through the deletion of pvdD) and the secondary siderophore pyochelin (through the deletion of pchE and pchF) in P. aeruginosa PAO1. We used this double knockout mutant because pyochelin is upregulated when pyoverdine is knocked out and can compensate for the effect of pyoverdine. To assess the role of pyoverdine, we thus needed to use the double knockout mutant. The normal iron-deficient calcareous sandy soil was prepared as described above for the pot experiment with each pot containing 8 kg of soil. Six monocropping peanut plants (Arachis hypogaea L. cv. Luhua14) were grown in a pot for the monocropping peanut treatment. At anthesis (50 dps), pod bearing (64 dps), and fruit-swelling stage (75 dps), 600 mL of 10⁸ CFU mL⁻¹ (1 × 10¹⁰ CFU plant⁻¹), solutions of Pseudomonas sp. 1502IPR-01, P. aeruginosa PAO1 wildtype, and the siderophore-deficient mutant PAO1 ΔpvdDpchEF were irrigated into the rhizoplane. 600 mL of water was irrigated into the rhizosphere as control. The rest of the greenhouse culture conditions remained identical to the ones described above. The peanut plants growing in the same pot were mixed and treated as one replicate. Each treatment had three or four replicates. At 84 dps, SPAD values and active iron concentration in young leaves were measured. Rhizosphere soil was collected to measure the available iron concentration.

Data analysis

Data sets were tested for normal distribution with the Shapiro-Wilk test and for heterogeneity of variance using Levene’s test. For data sets that met the conditions of normal distribution and homogeneity of variance, parametric Student’s t-test and ANOVA with LSD as post-hoc test were used. For data sets that did not meet these conditions, the BoxCox algorithm was used for data transformation to meet the assumptions of parametric tests. If these conditions were still not met after transformation, non-parametric Wilcoxon tests, Kruskal–Wallis tests with Dunnett T3 test (for one-way designs) or Scheirer-Ray-Hare test (for two-way designs) were used. The BH algorithm was used to correct p-values in multiple comparisons. All the above tests, as well as PERMANOVA and correlation analyses, were analyzed in R (version 4.2.2; https://www.r-project.org) using the ‘stats’ (version 4.2.2), ‘car’ (version 3.1-1), ‘forecast’ (version 8.2.0), ‘ggpubr’ (version 0.4.0), ‘stats’ (version 4.2.2), ‘agricolae’ (version 1.3-5), ‘ggpubr’ (version 0.4.0), ‘stats’ (version 4.2.2), ‘PMCMRplus’ (version 1.9.6), ‘rcompanion’ (version 2.4.21), ‘vegan’ (version 2.5-7), and ‘psych’ (version 2.2.5) packages. For data sets that did not meet normal distribution, correlations were calculated using non-parametric Spearman’s rank correlation (presented as Spearman rho). For data sets that met normal distribution, correlations were calculated using parametric Pearson correlation (presented as Pearson’s r). The heatmap was generated using ‘ComplexHeatmap’ (version 2.15.1) package in R (version 4.2.2) based on the relative abundance of the genus. Then, plots including conceptual diagrams were assembled and aesthetically perfected by Adobe Illustrator CC, 2020 (Adobe, San Jose, USA). All tests were two-sided. The China administrative border map used in the figures was downloaded from a public platform: resource and environment center data cloud platform (http://www.resdc.cn/).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Raw data of 16S rRNA amplicon sequencing used in this study are deposited in the NCBI database under BioProject PRJNA788265. 16S rRNA gene sequences of 46 isolated rhizobacteria are deposited on GenBank with accession OL824679 - OL824724. The Pseudomonas sp. 1502IPR-01 reference strain genome sequence, determined for this study, is deposited at the NCBI database under BioProject PRJNA788188. The raw data generated in this study are available in the Figshare repository under accession code (https://doi.org/10.6084/m9.figshare.24648306)⁵³ and also are provided in the Source Data file/Supplementary Data with this paper. Source data are provided with this paper.

Code availability

The analysis code that supports the findings of this study is available on GitHub (https://github.com/wnq13579/Peanut-maize-intercropping-microbiome) and deposited in Zenodo with https://doi.org/10.5281/zenodo.10212602⁵⁴ (https://zenodo.org/records/10212602).

References

Li, C. et al. Syndromes of production in intercropping impact yield gains. Nat. Plants 6, 653–660 (2020).
Article PubMed Google Scholar
Brooker, R. W. et al. Facilitation and biodiversity–ecosystem function relationships in crop production systems and their role in sustainable farming. J. Ecol. 109, 2054–2067 (2021).
Article Google Scholar
Zuo, Y. & Zhang, F. Iron and zinc biofortification strategies in dicot plants by intercropping with gramineous species. A review. Agron. Sustain. Dev. 29, 63–71 (2009).
Article CAS Google Scholar
National Bureau of Statistics of China. China Statistical Yearbook (in Chinese) (China Statistics Press, 2021).
Dai, J. et al. From leguminosae/gramineae intercropping systems to see benefits of intercropping on iron nutrition. Front. Plant Sci. 10, 605 (2019).
Article PubMed PubMed Central Google Scholar
Xiong, H. et al. Molecular evidence for phytosiderophore-induced improvement of iron nutrition of peanut intercropped with maize in calcareous soil. Plant Cell Environ. 36, 1888–1902 (2013).
Article PubMed CAS Google Scholar
Curie, C. & Mari, S. New routes for plant iron mining. N. Phytol. 214, 521–525 (2017).
Article CAS Google Scholar
Riaz, N. & Guerinot, M. L. All together now: regulation of the iron deficiency response. J. Exp. Bot. 72, 2045–2055 (2021).
Article PubMed PubMed Central CAS Google Scholar
Kobayashi, T., Nozoye, T. & Nishizawa, N. K. Iron transport and its regulation in plants. Free Radic. Biol. Med. 133, 11–20 (2019).
Article PubMed CAS Google Scholar
Suzuki, M. et al. Development of a mugineic acid family phytosiderophore analog as an iron fertilizer. Nat. Commun. 12, 1558 (2021).
Article ADS PubMed PubMed Central CAS Google Scholar
Lurthy, T., Pivato, B., Lemanceau, P. & Mazurier, S. Importance of the rhizosphere microbiota in iron biofortification of plants. Front. Plant Sci. 12, 744445 (2021).
Article PubMed PubMed Central Google Scholar
Zhang, J. et al. NRT1.1B is associated with root microbiota composition and nitrogen use in field-grown rice. Nat. Biotechnol. 37, 676–684 (2019).
Article PubMed CAS Google Scholar
Voges, M., Bai, Y., Schulze-Lefert, P. & Sattely, E. S. Plant-derived coumarins shape the composition of an Arabidopsis synthetic root microbiome. Proc. Natl Acad. Sci. USA 116, 12558–12565 (2019).
Article ADS PubMed PubMed Central Google Scholar
Cadot, S. et al. Specific and conserved patterns of microbiota-structuring by maize benzoxazinoids in the field. Microbiome 9, 103 (2021).
Article PubMed PubMed Central CAS Google Scholar
Harbort, C. J. et al. Root-secreted coumarins and the microbiota interact to improve iron nutrition in Arabidopsis. Cell Host Microbe 28, 825–837 (2020).
Article PubMed PubMed Central CAS Google Scholar
Rajniak, J. et al. Biosynthesis of redox-active metabolites in response to iron deficiency in plants. Nat. Chem. Biol. 14, 442–450 (2018).
Article PubMed PubMed Central CAS Google Scholar
Ghosh, S. K., Bera, T. & Chakrabarty, A. M. Microbial siderophore - A boon to agricultural sciences. Biol. Control 144, 104214 (2020).
Article CAS Google Scholar
Sorrenti, G., Masiello, C. A. & Toselli, M. Biochar interferes with kiwifruit Fe-nutrition in calcareous soil. Geoderma 272, 10–19 (2016).
Article ADS CAS Google Scholar
Maitra, S. et al. Intercropping—a low input agricultural strategy for food and environmental security. Agronomy 11, 343 (2021).
Article CAS Google Scholar
Shirley, M., Avoscan, L., Bernaud, E., Vansuyt, G. & Lemanceau, P. Comparison of iron acquisition from Fe-pyoverdine by strategy I and strategy II plants. Botany 89, 731–735 (2011).
Article CAS Google Scholar
Vansuyt, G., Robin, A., Briat, J. F., Curie, C. & Lemanceau, P. Iron acquisition from Fe-pyoverdine by Arabidopsis thaliana. Mol. Plant-Microbe Interact. 20, 441–447 (2007).
Article PubMed CAS Google Scholar
Soares, E. V. Perspective on the biotechnological production of bacterial siderophores and their use. Appl. Microbiol. Biotechnol. 106, 3985–4004 (2022).
Article PubMed CAS Google Scholar
Gu, S. et al. Competition for iron drives phytopathogen control by natural rhizosphere microbiomes. Nat. Microbiol. 5, 1002–1010 (2020).
Article PubMed PubMed Central CAS Google Scholar
Zuo, Y. M., Zhang, F. S., Li, X. L. & Cao, Y. P. Studies on the improvement in iron nutrition of peanut by intercropping with maize on a calcareous soil. Plant Soil 220, 13–25 (2000).
Article CAS Google Scholar
Li, L. et al. Root distribution and interactions between intercropped species. Oecologia 147, 280–290 (2006).
Article ADS PubMed Google Scholar
Gao, Y. et al. Distribution of roots and root length density in a maize/soybean strip intercropping system. Agric. Water Manag. 98, 199–212 (2010).
Article Google Scholar
Li, Q. S. et al. Belowground interactions impact the soil bacterial community, soil fertility, and crop yield in maize/peanut intercropping systems. Int. J. Mol. Sci. 19, 622 (2018).
Article PubMed PubMed Central CAS Google Scholar
Figueiredo, A. R. T., Özkaya, Ö., Kümmerli, R. & Kramer, J. Siderophores drive invasion dynamics in bacterial communities through their dual role as public good versus public bad. Ecol. Lett. 25, 138–150 (2022).
Article PubMed Google Scholar
Kümmerli, R. Iron acquisition strategies in pseudomonads: mechanisms, ecology, and evolution. Biometals 36, 777–797 (2023).
Article PubMed Google Scholar
Ahmed, E. & Holmstrom, S. J. Siderophores in environmental research: roles and applications. Micro. Biotechnol. 7, 196–208 (2014).
Article CAS Google Scholar
Shen, H. et al. AhDMT1, a Fe²⁺ transporter, is involved in improving iron nutrition and N₂ fixation in nodules of peanut intercropped with maize in calcareous soils. Planta 239, 1065–1077 (2014).
Article PubMed CAS Google Scholar
Dong, Q. et al. Maize and peanut intercropping improves the nitrogen accumulation and yield per plant of maize by promoting the secretion of flavonoids and abundance of Bradyrhizobium in rhizosphere. Front. Plant Sci. 13, 957336 (2022).
Article PubMed PubMed Central Google Scholar
Martínez-Hidalgo, P. & Hirsch, A. M. The nodule microbiome: N₂-fixing rhizobia do not live alone. Phytobiomes J. 1, 70–82 (2017).
Article Google Scholar
Li, B. et al. Root exudates drive interspecific facilitation by enhancing nodulation and N₂ fixation. Proc. Natl Acad. Sci. USA 113, 6496–6501 (2016).
Article ADS PubMed PubMed Central CAS Google Scholar
Hu, H. Y. et al. Nitrogen fixation and crop productivity enhancements co‐driven by intercrop root exudates and key rhizosphere bacteria. J. Appl. Ecol. 58, 2243–2255 (2021).
Article CAS Google Scholar
Vora, S. M., Joshi, P., Belwalkar, M. & Archana, G. Root exudates influence chemotaxis and colonization of diverse plant growth promoting rhizobacteria in the pigeon pea – maize intercropping system. Rhizosphere 18, 100331 (2021).
Article Google Scholar
Wang, Z. et al. Forms of nitrogen inputs regulate the intensity of soil acidification. Glob. Change Biol. 29, 4044–4055 (2023).
Article CAS Google Scholar
Weng, Z. H., Li, G., Sale, P. W. G. & Tang, C. X. Application of calcium nitrate with phosphorus promotes rhizosphere alkalization in acid subsoil. Eur. J. Soil Sci. 73, e13153 (2022).
Article CAS Google Scholar
Ferreira, C. M. H. et al. Calcareous soil interactions of the iron(III) chelates of DPH and Azotochelin and its application on amending iron chlorosis in soybean (Glycine max). Sci. Total Environ. 647, 1586–1593 (2019).
Article ADS PubMed CAS Google Scholar
Havlin, J. L., Beaton, J. D., Tisdale, S. L. & Nelson, W. L. Soil Fertility and Fertilizers: An Introduction to Nutrient Management 7th edn (Pearson College Div, 2004).
Federal Ministry of Labour and Social Affairs (BMAS). Technical Rules for Biological Agents: Classification of Prokaryotes (Bacteria and Archaea) into Risk Groups (TRBA 466)-2015 (2015).
Cornu, J. Y. et al. Contrasting effects of pyoverdine on the phytoextraction of Cu and Cd in a calcareous soil. Chemosphere 103, 212–219 (2014).
Article ADS PubMed CAS Google Scholar
Verma, S., Chakdar, H., Kumar, M., Varma, A. & Saxena, A. K. Microorganisms as a sustainable alternative to traditional biofortification of iron and zinc: status and prospect to combat hidden hunger. J. Soil Sci. Plant Nutr. 21, 1700–1717 (2021).
Article CAS Google Scholar
Li, C. J., Zhu, X. P. & Zhang, F. S. Role of shoot in regulation of iron deficiency responses in cucumber and bean plants. J. Plant Nutr. 23, 1809–1818 (2000).
Article CAS Google Scholar
Reichman, S. M. & Parker, D. R. Critical evaluation of three indirect assays for quantifying phytosiderophores released by the roots of Poaceae. Eur. J. Soil Sci. 58, 844–853 (2007).
Article Google Scholar
Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37, 852–857 (2019).
Article PubMed PubMed Central CAS Google Scholar
Beule, L. & Karlovsky, P. Improved normalization of species count data in ecology by scaling with ranked subsampling (SRS): application to microbial communities. PeerJ 8, e9593 (2020).
Article PubMed PubMed Central Google Scholar
Schwyn, B. & Neilands, J. B. Universal chemical-assay for the detection and determination of siderophores. Anal. Biochem. 160, 47–56 (1987).
Article PubMed CAS Google Scholar
Davis, E. W., Okrent, R. A., Manning, V. A. & Trippe, K. M. Unexpected distribution of the 4-formylaminooxyvinylglycine (FVG) biosynthetic pathway in Pseudomonas and beyond. PLoS ONE 16, e0247348 (2021).
Article PubMed PubMed Central CAS Google Scholar
Budzikiewicz, H., Schafer, M., Fernandez, D. U., Matthijs, S. & Cornelis, P. Characterization of the chromophores of pyoverdins and related siderophores by electrospray tandem mass spectrometry. Biometals 20, 135–144 (2007).
Article PubMed CAS Google Scholar
Sultana, R. et al. A pyoverdin from Pseudomonas sp. CFML 95-275. Z. Naturforsch. C. 55, 857–865 (2000).
Article PubMed CAS Google Scholar
Jin, C. W., Li, G. X., Yu, X. H. & Zheng, S. J. Plant Fe status affects the composition of siderophore-secreting microbes in the rhizosphere. Ann. Bot. 105, 835–841 (2010).
Article PubMed PubMed Central CAS Google Scholar
Wang, N. Q. et al. Microbiome convergence enables siderophore-secreting-rhizobacteria to improve iron nutrition and yield of peanut intercropped with maize. Figshare. https://doi.org/10.6084/m9.figshare.24648306 (2023).
Wang, N. Q. et al. Microbiome convergence enables siderophore-secreting-rhizobacteria to improve iron nutrition and yield of peanut intercropped with maize. wnq13579/Peanut-maize-intercropping-microbiome. Zenodo. https://doi.org/10.5281/zenodo.10212602 (2023).

Download references

Acknowledgements

This research was financially supported by the National Natural Science Foundation of China (grant no. 32372810 to Y.-M. Z., 32102469 to N.-Q. W., and 42325704 to Z. W.), the National Key Research and Development Program of China (grant no. 2022YFD1901500/2022YFD1901501, 2023YFD1700203 to Y.-M. Z. and 2021YFD1900100 to Z. W.), and the Swiss National Science Foundation SNSF (grant no. 31003A_182499 to R. K.). We thank Dr. Prof. Yang Bai from the Chinese Academy of Sciences, Dr. Prof. Chuihua Kong, and Dr. Prof. Ling Xu from China Agricultural University for valuable discussions and suggestions.

Author information

These authors contributed equally: Nanqi Wang, Tianqi Wang, Yu Chen.

Authors and Affiliations

College of Resources and Environmental Sciences, State Key Laboratory of Nutrient Use and Management (SKL-NUM), National Academy of Agriculture Green Development, China Agricultural University, 100193, Beijing, China
Nanqi Wang, Tianqi Wang, Qiaofang Lu, Kunguang Wang, Zhechao Dou, Zhiguang Chi, Wei Qiu, Jing Dai, Lei Niu, Jianyu Cui, Fusuo Zhang & Yuanmei Zuo
Jiangsu Key Laboratory for the Research and Utilization of Plant Resources, Jiangsu Province Engineering Research Center of Eco-cultivation and High-value Utilization of Chinese Medicinal Materials, Institute of Botany, Jiangsu Province and Chinese Academy of Sciences, 210014, Nanjing, Jiangsu, China
Yu Chen
Department of Plant Pathology, The Key Laboratory of Plant Immunity, Nanjing Agricultural University, Nanjing, 210095, China
Ming Wang
Jiangsu provincial key lab for solid organic waste utilization, Key lab of organic-based fertilizers of China, Jiangsu Collaborative Innovation Center for Solid Organic Wastes, Educational Ministry Engineering Center of Resource-saving fertilizers, Nanjing Agricultural University, Nanjing, 210095, China
Zhong Wei
Department of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland
Rolf Kümmerli

Authors

Nanqi Wang
View author publications
You can also search for this author in PubMed Google Scholar
Tianqi Wang
View author publications
You can also search for this author in PubMed Google Scholar
Yu Chen
View author publications
You can also search for this author in PubMed Google Scholar
Ming Wang
View author publications
You can also search for this author in PubMed Google Scholar
Qiaofang Lu
View author publications
You can also search for this author in PubMed Google Scholar
Kunguang Wang
View author publications
You can also search for this author in PubMed Google Scholar
Zhechao Dou
View author publications
You can also search for this author in PubMed Google Scholar
Zhiguang Chi
View author publications
You can also search for this author in PubMed Google Scholar
Wei Qiu
View author publications
You can also search for this author in PubMed Google Scholar
Jing Dai
View author publications
You can also search for this author in PubMed Google Scholar
Lei Niu
View author publications
You can also search for this author in PubMed Google Scholar
Jianyu Cui
View author publications
You can also search for this author in PubMed Google Scholar
Zhong Wei
View author publications
You can also search for this author in PubMed Google Scholar
Fusuo Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Rolf Kümmerli
View author publications
You can also search for this author in PubMed Google Scholar
Yuanmei Zuo
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Y.-M. Z., Z. W., N.-Q. W. and T.-Q. W. designed the research. N.-Q. W. and T.-Q. W. performed and analyzed most of the experiments in the laboratory and field, and Q.-F. L. K.-G. W., Z.-C. D., Z.-G. C., W.-Q., J.-D., L.-N., and J.-Y. C. performed and analyzed most of the experiments in the greenhouse. Y. C. and M. W. provided lab conditions and intellectual input for identifying the siderophore structure and wrote the section “Structural identification of pyoverdine” in methods. R. K. and Z. W. provided intellectual input, and guidance to microbiome data analysis and helped to interpret data. M. W. provided some intellectual input for this manuscript, and guidance to microbiome data analysis and visualization. N.-Q. W., T.-Q. W., F.-S. Z., R. K., Z. W., and Y.-M. Z. wrote and revised the manuscript. All of the authors discussed the results and commented on the manuscript. Y.-M. Z. supervised the study.

Corresponding authors

Correspondence to Ming Wang, Zhong Wei or Yuanmei Zuo.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Pierre Cornelis, and Eiko Kuramae for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Peer Review File

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Supplementary Data 4

Supplementary Data 5

Supplementary Data 6

Supplementary Data 7

Supplementary Data 8

Supplementary Data 9

Supplementary Data 10

Supplementary Data 11

Supplementary Data 12

Supplementary data 13

Description of Additional Supplementary Files

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Wang, N., Wang, T., Chen, Y. et al. Microbiome convergence enables siderophore-secreting-rhizobacteria to improve iron nutrition and yield of peanut intercropped with maize. Nat Commun 15, 839 (2024). https://doi.org/10.1038/s41467-024-45207-0

Download citation

Received: 07 May 2023
Accepted: 17 January 2024
Published: 29 January 2024
DOI: https://doi.org/10.1038/s41467-024-45207-0

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.