Iron limitation-induced endophytic Ammoniphilus assemblage promotes root apoplastic iron remobilization by attenuation of salicylic acid pathways

Zhu, Lin; Zhang, Aijing; Guo, Jiansheng; Liu, Huabing; Xie, Yue; Lu, Xiaomin; Yan, Congsheng; Wang, Jianfei; Zhou, Cheng

doi:10.1038/s41522-025-00765-z

Download PDF

Article
Open access
Published: 01 July 2025

Iron limitation-induced endophytic Ammoniphilus assemblage promotes root apoplastic iron remobilization by attenuation of salicylic acid pathways

Lin Zhu^1,2,
Aijing Zhang¹,
Jiansheng Guo³,
Huabing Liu¹,
Yue Xie ORCID: orcid.org/0000-0002-8481-7186¹,
Xiaomin Lu¹,
Congsheng Yan⁴,
Jianfei Wang¹ &
…
Cheng Zhou ORCID: orcid.org/0000-0002-8746-8663⁵

npj Biofilms and Microbiomes volume 11, Article number: 125 (2025) Cite this article

1205 Accesses
Metrics details

Subjects

Abstract

Plants establish symbiotic associations with root-colonizing microbes to adapt to adverse conditions. However, how root-associated microbiota interacted with their hosts to improve plant growth under nutrient deficient conditions remains poorly understood. In this study, we explored an interaction between tomato plants and root-associated microbiota under iron (Fe) limitation, mediated by bacterial secretion of glutamine. 16S rRNA gene sequencing revealed that Fe-limited conditions altered the composition of root-associated microbiomes, resulting in the enrichment of Ammoniphilus sp. This taxon was isolated and shown to alleviate Fe deficiency symptoms. Moreover, Fe deficiency triggered salicylic acid (SA)-induced hydrogen peroxide (H₂O₂) burst, thereby inhibiting the exudation of Fe-mobilizing phenolics from the roots. However, bacterial secretion of Gln greatly attenuated the SA-induced H₂O₂ production in the roots, thereby enhancing bacterial colonization and promoting apoplastic Fe remobilization. Collectively, these results underscored a microbial strategy for orchestrating plant SA pathways to facilitate the reutilization of root apoplastic Fe.

Evaluation of Pseudomonas sp. for its multifarious plant growth promoting potential and its ability to alleviate biotic and abiotic stress in tomato (Solanum lycopersicum) plants

Article Open access 01 December 2020

Functional mutants of Azospirillum brasilense elicit beneficial physiological and metabolic responses in Zea mays contributing to increased host iron assimilation

Article Open access 06 January 2021

Spatial IMA1 regulation restricts root iron acquisition on MAMP perception

Article 10 January 2024

Introduction

Iron (Fe) deficiency represents a significant challenge for plant growth in calcareous soils, characterized by the substantial presence of free calcium carbonate (CaCO₃) and a pH level exceeding 7.0¹. While Fe is the second most abundant metal element in the Earth’s crust, its bioavailability is restricted under alkaline conditions, resulting in impaired growth and chlorosis in plants^2,3. Fe is an essential element critical for various physiological and biochemical processes in plants, encompassing vital functions such as respiration and DNA synthesis. Disruption of these processes due to Fe deficiency adversely impacts plant growth and development^4,5,6. Concurrently, plants establish symbiotic relationships with beneficial microbes to adapt to Fe-limited environments^7,8,9. However, the roles of root-associated microbiota in mediating plant Fe homeostasis under Fe-limited conditions remain poorly understood.

In natural environments, plants have evolved two highly effective strategies, commonly known as Strategy I and Strategy II, to acquire Fe from external conditions. Strategy I, also referred to as the reduction strategy, is predominantly employed by dicot and non-graminaceous monocot plants¹⁰. This sophisticated strategy involves a series of coordinated steps. Initially, the plasma membrane H⁺-ATPase AHA2 actively pumps protons into the rhizosphere, creating an acidic microenvironment. This acidification facilitates the release of Fe³⁺ from Fe hydroxides¹¹. Subsequently, ferric chelate reductases (FROs) localized on the plasma membrane play a crucial role in reducing ferric (Fe³⁺) ions to their more accessible ferrous (Fe²⁺) form. Finally, the iron-regulated transporter protein 1 (IRT1) efficiently transports the Fe²⁺ ions into root epidermal cells, ensuring their uptake¹². In contrast, graminaceous plants employ Strategy II, also known as the chelation strategy¹³. These plants employ a unique approach by releasing specific compounds called phytosiderophores (PS), which include mugineic acids, into the rhizosphere. The phytosiderophore-Fe³⁺ chelate complex is then transported into root cells via specialized transporter proteins known as yellow stripe (YS) or yellow stripe-like (YSL) proteins^14,15. It’s worth noting that, compared with Strategy II plants, high pH conditions in the rhizosphere serve as a strong buffer for the protons released by Strategy I plants¹⁶. This buffering effect inhibits the activity of FROs in Strategy I plants^17,18,19. Therefore, Strategy I plants exhibit a greater sensitivity to alkaline conditions in comparison to Strategy II plants.

Although Fe remobilization can be challenging, plants have developed adaptive mechanisms to reuse stored Fe under Fe-limited conditions^20,21,22. Apoplastic Fe serves as a vital Fe source for plants during Fe deficiency²³. The regulation of root apoplastic Fe reutilization involves various phytohormones and signaling molecules such as abscisic acid, nitric oxide and putrescine^20,21,22. Furthermore, many researches on red clover and rice have demonstrated that root-secreted phenolic compounds can mobilize apoplastic Fe for improving plant Fe nutrition^23,24,25. However, Fe deficiency can stimulate the production of hydrogen peroxide (H₂O₂) in roots, leading to increased lignin formation and a consequent reduction in the secretion of Fe-mobilizing phenolic compounds^24,26,27,28. Under Fe deficient conditions, wild-type Arabidopsis thaliana plants exhibit more severe chlorosis and less root apoplastic Fe reutilization compared with the salicylic acid (SA) synthesis mutant pad4²⁹. It is well established that SA plays a role in regulating plant defense by inducing H₂O₂ production^30,31,32. Moreover, elevated H₂O₂ levels have been reported to promote lignin synthesis in plants^33,34,35,36. Lignin, an essential byproduct of phenolics metabolism, is an amorphous heteropolymer formed through the oxidative coupling of p-hydroxycinnamyl alcohols in an H₂O₂-dependent reaction^24,37. Elevated pH levels can significantly enhance lignin biosynthesis in roots, requiring substantial phenolic substrates and consequently decreasing phenolic levels within the roots. Furthermore, high pH conditions lead to reduced phenolics secretion, thereby limiting the mobilization of root apoplastic Fe²⁴. Therefore, the SA-induced H₂O₂ production in the Fe-deficient plants may enhance root lignin formation, thereby reducing the secretion of Fe-mobilizing phenolic compounds.

The contribution of root-associated microbiomes is critical in ensuring an adequate supply of Fe to their hosts in natural conditions. Both the structure and function of these microbiomes are impacted by soil factors and root exudates^7,8,9. Importantly, root-secreted coumarins can shape the rhizosphere microbial community and engage in interactions with the rhizosphere microbiota to improve plant Fe nutrition under Fe-limited conditions⁹. Hence, understanding the mechanisms governing host-microbiota interactions is essential for engineering functional microbiota to improve plant growth. Despite the importance of root apoplastic Fe reutilization in plant adaptation to Fe-limited conditions, there is a paucity of knowledge regarding how the assembly of root endophytic microbiota (endo-microbiota) mediates this process.

This study aimed to elucidate how root endo-microbiota steered adaptive responses of host plants to Fe deficiency. Our hypotheses were as follows: (I) the endo-microbiota induced root metabolic changes that conferred improved plant adaptation to Fe-limited conditions, and (II) interference of the microbiota with the SA pathways played a central role in this process. To achieve these objectives, we conducted a comprehensive analysis of changes in root-associated microbiomes in response to Fe limitation. We further isolated several key taxa from the microbiomes and investigated their effects on plant growth, along with the underlying mechanisms of interactions with the hosts under Fe-limited conditions.

Results

Root-associated microbiota was important for plant growth under Fe-limited conditions

To explore the roles of root-associated microbiomes in mediating plant growth under Fe-limited conditions, we conducted an experiment using tomato plants grown in soils with different Fe availability (Fig. 1a). Initially, agricultural soils sourced from tomato fields had an acidic pH of 6.2, ensuring sufficient Fe levels. To simulate Fe-limited environments, CaCO₃ was introduced into the acidic soils, elevating the soil pH to a range of 7.5–8.0, which led to substantial reduction in Fe availability (Supplementary Fig. 1a). While plants grown in both natural and sterilized acidic soils exhibited similar phenotypic characteristics, shoot fresh weight (SFW) and total chlorophyll content were decreased progressively with increasing soil pH. Remarkably, plants grown in the natural soils outperformed those in the sterilized soils under Fe-limited conditions (Fig. 1b–d).

**Fig. 1: Effects of root-associated microbiota on plant growth under Fe-limited conditions.**

Furthermore, plants grown in the natural soils exhibited higher shoot and root soluble Fe content compared to those in the sterilized soils under Fe-limited conditions (Fig. 1e, f). However, no discernible differences in total shoot and root Fe content were observed between plants grown in both the natural and sterilized acidic soils (Supplementary Fig. 1b, c). Apoplasts serve as crucial repositories of Fe in plants, and the reutilization of root apoplastic Fe is closely linked to soluble Fe levels in Fe-deficient roots^20,21,22. We noted significantly lower levels of apoplastic Fe in plants grown in the natural soils compared to those in the sterilized soils under Fe-limited conditions (Fig. 1g). Therefore, these findings strongly suggested that the root-associated microbiomes played a pivotal role in enhancing the remobilization of root apoplastic Fe under Fe-limited conditions.

Effects of Fe-limited conditions on root-associated bacterial communities

We utilized 16S rRNA gene amplicon sequencing to scrutinize root-associated bacterial communities (Fig. 2a). At the phylum level, bacterial communities in both bulk and rhizosphere soil samples were predominantly composed of Proteobacteria, with other taxa such as Actinobacteria, Bacteroidetes, and Firmicutes. Notably, the elevated soil pH levels resulted in a decline in relative abundance of Proteobacteria within root samples, while Firmicutes became the dominant phylum, with Paenibacillaceae and Bacillaceae as the prevailing families within Firmicutes (Fig. 2b, c). Furthermore, alpha diversity, as assessed by the Shannon and Chao1 indices, exhibited significantly higher values in the acidic soils compared to the Fe-limited calcareous soils across roots, bulk, and rhizosphere soil samples (Fig. 2d, e). Principal coordinates analysis (PCoA) analysis revealed significant variations in the bacterial community compositions among various samples. Additionally, PERMANOVA analysis indicated that soil pH exerted the primary influence on the divergence among these samples (Fig. 2f).

**Fig. 2: Comparison of root-associated bacterial communities from normal (acid) and Fe-limited (alkaline) conditions by 16S rRNA gene sequence analysis.**

To further assess the impacts of Fe-limited conditions on the bacterial community compositions, we identified OTUs that were specifically enriched in the rhizosphere and roots of plants cultivated under normal (acid) and Fe-limited (alkaline) conditions (Fig. 3a–c). In the Fe-limited treatments, there was a significant increase in the number of rhizosphere-enriched OTUs compared to normal treatments, particularly with Proteobacteria showing a pronounced enrichment. Conversely, the number of root-enriched OTUs showed a modest increase, with Firmicutes being particularly enriched in the Fe-limited conditions (Fig. 3d). Moreover, Venn diagram depicting shared OTUs revealed a core group of six OTUs (OTU5, OTU7, OTU8, OTU13, OTU87, and OTU119) that were specifically enriched in the roots across different treatments. In contrast, no OTUs specifically enriched in the rhizosphere samples were shared among the different treatments (Fig. 3e, f).

**Fig. 3: Analysis of OTUs specially enriched in different compartments.**

Endophytic microbiota linked with the reutilization of root apoplastic Fe

We screened which endo-bacterial OTUs were correlated with Fe-limited (pH 7.5 and pH 8.0) treatments and the reutilizing efficiency of root apoplastic Fe (rREU^apoFe). A total of 377 and 303 endo-bacterial OTUs significantly differed between the normal (pH 6.2) and Fe-limited (pH 7.5 and pH 8.0) treatments, respectively (Fig. 4a, b). Among these, 130 OTUs were significantly enriched, while 86 OTUs were depleted in the Fe-limited treatments (>2-fold increase in abundance) compared with normal treatments, respectively (Fig. 4c, d). Spearman’s rank correlation analysis identified 106 OTUs that showed positive correlations with rREU^apoFe (FDR-adjusted p value < 0.05) (Fig. 4e; Supplementary Table 1). Among those responsive taxa, a particularly noteworthy pattern emerged in OTU5, classified as Ammoniphilus sp. The average relative abundance of Ammoniphilus OTU5 was notably higher in the pH 7.5 treatments, averaging 12.5% of the total bacterial community, but decreased to 6.2% in the pH 8.0 treatments. In contrast, this OTU represented only 1.2% of the total bacterial community in the normal treatments (Fig. 4f). Furthermore, Ammoniphilus OTU5 exhibited a strong correlation with rREU^apoFe (R = 0.84) (Fig. 4g). As mentioned earlier, Ammoniphilus OTU5 was specifically enriched in the roots, prompting us to conduct further investigations into the function of this taxon.

**Fig. 4: The importance of root endophytic microbiota in the promotion of root apoplastic Fe reutilization.**

Endophytic Ammoniphilus isolates enhanced plant adaptation to Fe-limited conditions

To investigate the roles of Ammoniphilus sp. in mediated plant adaptation to Fe-limited conditions, we obtained a total of 286 bacterial isolates from tomato roots. Among these isolates, 59 were identified as Ammoniphilus sp. isolates. To avoid potential duplicates, we excluded isolates that showed 100% identity in their 16S rRNA gene sequences, resulting in 19 distinct Ammoniphilus isolates. These isolates were phylogenetically categorized into five major groups (Fig. 5a). From each group, we selected one isolate for further assays, named AS12, AS15, AS27, AS63, and AS72. These isolates shared nucleotide identities of 98.15%, 98.94%, 100%, 97.84% and 98.68%, respectively, with OTU5 that was identified in the sequencing analysis. We chose these five isolates to assess their impact on plant growth under Fe-limited conditions using gnotobiotic and hydroponic systems.

**Fig. 5: Evaluation of the roles of *Ammoniphilus* sp. in the promotion of root apoplastic Fe reutilization.**

To evaluate the capacity of distinct Ammoniphilus groups to improve plant growth under highly Fe-limited calcareous conditions (pH 8.0), we conducted experiments involving the cultivation of tomato plants with different bacterial strains. In gnotobiotic systems, various Ammoniphilus isolates were assessed for their ability to rescue Fe-limiting plant growth (Fig. 5b). We observed that two Ammoniphilus isolates increased total chlorophyll content and growth-rescuing activity. However, the remaining isolates did not significantly enhance plant adaptation to Fe-limited conditions (Fig. 5c–e). Moreover, shoot and root soluble Fe content was significantly higher in the inoculated plants compared with the control plants under Fe-limited conditions (Fig. 5f, g). Conversely, root apoplastic Fe was remarkably less in the inoculated plants than the control plants (Fig. 5h). To further examine the effects of two Ammoniphilus sp. isolates (AS27 and AS63) on plant Fe deficiency tolerance, hydroponic systems were employed (Fig. 5i). The relieved leaf chlorosis in the inoculated plants corresponded with the measured chlorophyll content (Fig. 5j, k). Consistently, shoot soluble Fe content was significantly higher in the inoculated plants compared with the control plants under Fe-limited conditions (Fig. 5l). Furthermore, qPCR analysis indicated that AS27 achieved more efficient colonization than the other isolates (Supplementary Fig. 2). Considering its superior ability to enhance plant Fe deficiency resistance, root colonization, and sequence similarity with OTU5, we selected the isolate AS27 as a representative strain of Ammoniphilus sp. for subsequent experiments.

Ammoniphilus sp. attenuated the SA-mediated pathways in Fe-deficient roots

To unravel the mechanisms underlying the increased tolerance of plants to Fe deficiency induced by Ammoniphilus sp. AS27, we conducted a comparative RNA-Seq analysis of root transcriptomes. PCA analysis revealed distinct alterations in the transcriptomes in response to -Fe and AS27 treatments (Fig. 6a). Among the identified differentially expressed genes (DEGs), 327 and 195 genes were transcriptionally induced, while 80 and 89 genes were repressed after 36 and 72 h of −Fe treatment, respectively. Additionally, a total of 1,679 DEGs (847 up-regulated and 832 down-regulated) and 557 DEGs (470 up-regulated and 87 down-regulated) were discerned between the control (heat-killed AS27) and AS27-inoculated roots following 36 and 72 h of −Fe treatment, respectively (Fig. 6b, c). In the Fe-deficient roots, 116 up-regulated and 26 down-regulated DEGs were shared between +Fe and −Fe treatments (Supplementary Fig. 3a,b). AS27 was found to regulate 53 up-regulated and 47 down-regulated DEGs shared between the control and AS27-inoculated roots under −Fe treatment (Supplementary Fig. 3c, d). KEGG enrichment analysis indicated a notable enrichment of shared DEGs in multiple pathways, such as glutathione metabolism, plant-pathogen interaction, and photosynthesis, all of which were influenced by −Fe treatments (Fig. 6d). However, several genes associated with the plant-pathogen interaction pathway, activated by Fe deficiency, were significantly suppressed by AS27 (Fig. 6e).

**Fig. 6: Effects of *Ammoniphilus* sp. on root transcriptomic profiles.**

Venn diagram analysis highlighted 12 core genes shared among different treatments (Fig. 6f). We then examined the commonly shared genes with known functions related to Fe uptake and defense responses (Fig. 6g). The expression of IRT1, encoding an iron-regulated transporter, and FRO1, encoding a ferric chelate reductase, was markedly induced by Fe deficiency. The core gene list revealed an abundance of GO terms linked to defense responses such as PR1, PR4, and PR4-like genes. The transcription of these genes was markedly up-regulated under Fe deficiency, while the inoculation with AS27 inhibited their transcription. Additionally, a total of 290 DEGs (168 up-regulated and 122 down-regulated) and 859 DEGs (459 up-regulated and 400 down-regulated) were identified between +Fe and –Fe+AS27 treatments after 36 and 72 h of treatment, respectively (Supplementary Fig. 4a). Compared with +Fe treatments, the inoculation with AS27 greatly weakened the expression of several Fe acquisition- and defense-related genes in the Fe-deficient roots (Supplementary Fig. 4b). Moreover, a total of 154 DEGs (69 up-regulated and 85 down-regulated) were shared between +Fe and –Fe+AS27 treatments, respectively (Supplementary Fig. 4c, d). KEGG enrichment analysis revealed most of the DEGs were enriched into amino acid transport and metabolism, MAPK signaling pathway, and ascorbate and aldarate metabolism (Supplementary Fig. 4e).

It is well documented that PR1 serves as a marker for the SA-dependent plant defense responses³⁸. In line with the data of RNA-Seq, qPCR analysis revealed that the inoculation with AS27 remarkably inhibited the expression of PR1 in the Fe-deficient roots (Fig. 6h). In addition, the levels of free SA were notably lower in the AS27-inoculated roots compared with the control roots under -Fe treatment (Fig. 6i). Additionally, AS27 did not grow on minimal media with SA as the sole carbon source (Fig. 6j). To further assess whether SA directly inhibited Ammoniphilus growth in vitro, AS27 growth was evaluated in minimal media containing glucose with 0.5 mM SA. The results indicated that SA did not exhibit inhibitory effects on bacterial growth (Fig. 6j).

Bacterial Gln suppressed the Fe deficiency-induced activation of plant SA pathways

We next analyzed metabolic changes of both the control (heat-killed AS27) and AS27-inoculated roots under Fe deficiency (Fig. 7a). A total of 196 and 119 metabolites were significantly altered in the roots following 36 and 72 h of AS27 treatments, respectively (Fig. 7b, c). Additionally, a total of 24 metabolites were enriched, while 33 metabolites were depleted, shared between the control and AS27-inoculated plants under −Fe treatments (Fig. 7d). Among the shared enriched metabolites, one amino acid (Gln) and 23 secondary metabolites were categorized into different chemical classes, including alcohols, alkaloids, and fatty acids (Fig. 7e, Supplementary Table 2). HPLC analysis showed that the levels of Gln were notably higher in the AS27-inoculated roots compared with the controls under −Fe treatments (Supplementary Fig. 5). Moreover, Gln was also highly abundant in bacterial culture of AS27 (Fig. 7f).

**Fig. 7: Effects of *Ammoniphilus* sp. on root metabolic profiles.**

Gln can serve as a negative regulator mediating the SA signaling pathways in plants³⁹. Given the increased levels of Gln in the AS27-inoculated roots, we further verified whether: (1) bacterial secretion of Gln enhanced plant resistance to Fe deficiency, and (2) bacterial Gln dampened the activation of SA pathways triggered by Fe deficiency. Our findings revealed that ΔglnA bacteria, lacking the ability to produce Gln, did not improve the plant’s ability to tolerate Fe deficiency. However, the augmented capability of plants to withstand Fe deficiency was significantly recovered in the glnA gene complemented strain (CΔglnA). Compared with WT bacteria, ΔglnA bacteria exhibited modestly delayed growth in vitro (Supplementary Fig. 6). Furthermore, the addition of 0.5 mM Gln into the growth medium markedly alleviated leaf chlorosis caused by Fe deficiency (Fig. 7g). In accordance to the phenotype observed, Gln treatment significantly increased the content of shoot and root soluble Fe (Supplementary Fig. 7a, b), but a reverse trend was observed for root apoplastic Fe (Supplementary Fig. 7c). Moreover, treatment with either WT bacteria (AS27) or 0.5 mM Gln significantly reduced the levels of SA and the expression of the SA marker gene PR1 in the Fe-deficient roots, but this effect was not observed in ΔglnA bacteria (Fig. 7h, i). Additionally, treatment with 3 μM SA exacerbated chlorotic symptoms in the Fe-deficient plants, and fully abolished the AS27-induced plant Fe deficiency resistance (Fig. 7g–i). Conversely, chlorotic symptoms were markedly alleviated in plants overexpressing NahG, a salicylate hydroxylase that degrades SA (Fig. 7g).

We next investigated whether bacterial Gln played a role in promoting their colonization in the Fe-deficient roots. To characterize the bacteria-plant interaction, we inoculated AS27 and ΔglnA bacteria onto plants grown in hydroponic systems. Fe deficiency promoted root colonization of AS27, whereas ΔglnA bacteria failed to colonize the Fe-deficient roots as efficiently as AS27. Interestingly, the addition of exogenous Gln not only enhanced root colonization of AS27, but also significantly restored ΔglnA bacterial colonization (Fig. 7j). We further hypothesized that bacterial Gln was necessary to attenuate the activation of the SA pathway, and therefore, the colonization of Gln-deficient bacteria could be restored on the plants with compromised SA-mediated plant immunity. Supporting this hypothesis, a visible increase in root colonization of ΔglnA bacteria was observed in the NahG transgenic plants exposed to either Fe-sufficient or Fe-limited conditions (Fig. 7j). Based on these results, we concluded that SA restricted the growth of ΔglnA bacteria in the Fe-deficient roots, and bacterial Gln effectively counteracted this restriction.

Bacterial Gln enhanced root apoplastic Fe reutilization by inhibition of H₂O₂ burst

The apoplastic Fe content in the roots under Fe deficiency was reduced by 28%, and this reduction was further intensified by inoculation with AS27 (Fig. 8a). Furthermore, the levels of root soluble Fe was notably higher in the inoculated plants compared with the control (heat-killed AS27) plants under Fe deficiency (Fig. 8b). Since phenolics secreted by Fe-deficient roots facilitated the remobilization of apoplastic Fe in dicotyledons²³, we monitored phenolic secretion in the growth medium. The secretion of phenolics was increased by 45% on the third day of −Fe treatment. However, in −Fe+AS27 treatment, it was increased by more than approximately two-fold (Fig. 8c).

**Fig. 8: Effects of SA-induced root H₂O₂ burst on root apoplastic Fe remobilization and bacterial colonization.**

The accumulation of SA was negatively related to root soluble Fe levels under Fe deficiency²⁹. Furthermore, the significant utilization of phenolics as substrates for lignin formation in Fe-deficient roots resulted in reduced availability of Fe-mobilizing phenolics²⁴. Based on these observations, it was postulated that Ammoniphilus sp. weakened Fe deficiency-induced lignin formation by secreting Gln to inhibit the SA pathways. Fe deficiency significantly induced lignin deposition in the roots, as indicated by intensified lignin staining (Fig. 8d). Correspondingly, there was a notable increase in the content of lignin in the roots experiencing Fe deficiency (Fig. 8e). However, the secretion of phenolics was significantly decreased in the Fe-deficient roots (Fig. 8c). In contrast, both the AS27-inoculated plants and NahG transgenic plants exhibited the opposite trend. However, the ΔglnA mutant failed to inhibit lignin formation and promote phenolic secretion in the Fe-deficient roots. Moreover, the AS27-induced effects were fully abolished in the Fe-deficient roots when 3 μM SA was applied exogenously (Fig. 8d–f). Additionally, the Gln-treated plants exhibited weaker lignin formation and stronger phenolic secretion compared with the control plants (Fig. 8d–f).

In plants, SA often triggers the production of H₂O₂, which antagonizes microbial colonization^30,40,41. Moreover, elevated H₂O₂ levels have been shown to reduce internal iron availability in the Fe-deficient plants²⁶. Therefore, we hypothesized that inhibiting SA-induced H₂O₂ burst was responsible for Ammoniphilus sp. AS27 to achieve efficient colonization and promote root apoplastic Fe reutilization. To confirm this, we initially used the fluorescent probe H₂DCF DA to visualize H₂O₂ production in the roots. The Fe-deficient roots accumulated higher levels of H₂O₂ compared with the Fe-sufficient roots, as indicated by intense green fluorescence. However, the inoculation with AS27 significantly reduced H₂O₂ production in the Fe-deficient roots compared with the ΔglnA-inoculated plants (Fig. 8g). The H₂O₂ burst in the Fe-deficient roots was also repressed by 0.5 mM Gln, similar to those observed in the NahG transgenic plants. However, treatment with 3 μM SA greatly reduced bacterial colonization and root apoplastic Fe reutilization in the AS27-inoculated roots, accompanied by a substantial accumulation of H₂O₂ under Fe deficiency (Fig. 8g–j). Additionally, we observed a significant recovery of bacterial colonization by ΔglnA in the Fe-deficient roots upon treatment with 1 mM glutathione (GSH), a reactive oxygen species (ROS) scavenger (Fig. 8h). Furthermore, treatment with GSH alone markedly inhibited lignin formation and enhanced the reutilization of root apoplastic Fe in the ΔglnA-inoculated roots exposed to Fe deficiency (Fig. 8d, j). Therefore, our data indicated that bacterial Gln was essential for inhibiting the SA-induced H₂O₂ burst, thereby promoting bacterial colonization and the remobilization of root apoplastic Fe.

Discussion

The root-associated microbiota provides great benefits for their hosts, particularly under nutrient-limited conditions^9,42,43,44. Gaining insights into the mechanisms that underlie these beneficial services is paramount for enhancing plant health and fitness. Given the sessile nature of plants, they possess the ability to selectively recruit specific members from root microbiomes to adapt to challenging environments⁴⁵. However, the precise contributions of the recruited microbiota to plant stress resistance, as well as the underlying mechanisms, remain poorly elucidated. In this context, we identified one Ammoniphilus species within root microbiomes that played a pivotal role in promoting the reutilization of root apoplastic Fe, potentially through the modulation of the SA pathway. Our results indicated that Fe deficiency triggered the SA-induced H₂O₂ burst in the roots, thereby enhancing lignin biosynthesis and inhibiting the secretion of Fe-mobilizing phenolics. Furthermore, the production of Gln by root-inhabiting Ammoniphilus sp. emerged as a critical factor in attenuating the activation of plant SA pathways by Fe-limited conditions. These findings underscore the crucial roles of root endophytic microbiota in remobilizing root apoplastic Fe.

Soil-borne microbes are integral in shaping plant growth, health, and performance^46,47. In this study, soil sterilization increased the sensitivity of plants to Fe-limited conditions in comparison to those grown in natural soils. This highlighted the importance of root-associated microbiota in facilitating plant growth under Fe-limited conditions. Furthermore, the enhanced reutilization of root apoplastic Fe in plants cultivated in natural soils strongly suggested that the microbiota played a pivotal role in regulating root apoplastic Fe mobilization. An in-depth analysis of the root-associated bacterial community compositions unveiled significant differences among all the samples, primarily attributed to the variations in soil pH. In the Fe-limited soils, there was a discernible decrease in the relative abundance of Proteobacteria, while Firmicutes became dominant in the root microbiota. Family-level analysis further showed that the abundance of families including Paenibacillaceae and Bacillaceae was enhanced in the roots under Fe-limited conditions. However, alkaline soil promoted specific enrichment of the family Rhodocyclaceae in the roots across different grass species⁴⁸. This indicates that different plant species establish unique associations with specific microbial taxa to adapt to Fe-limited conditions. Further investigation into the endo-microbiota that thrived under Fe-limited conditions led to the identification of specific OTUs closely linked to the roots. Fe-limited treatments resulted in an increase in their abundance, particularly belonging to the family Paenibacillaceae, indicating their potential role in assisting plants in adaption to Fe-limited conditions. Several OTUs that significantly differed between the normal and Fe-limited treatments showed positive correlations with rREU^apoFe. Notably, we identified Ammoniphilus sp. as one such OTU, exhibiting the strongest correlation with rREU^apoFe and the highest abundance in the roots under Fe-limited conditions. The presence of Ammoniphilus sp. and other responsive taxa positively correlated to root apoplastic Fe reutilization strongly suggested their potential role as mediators of plant adaptive responses to Fe deficiency.

Recent research has unveiled a fascinating interplay between plant-derived phenolics, such as coumarins, and root microbiome during Fe starvation^7,8,9. Root-secreted coumarins are triggered under Fe deficiency and facilitate an interaction between the host and root microbiota, enhancing host Fe nutrition⁹. This symbiotic relationship underscores the vital role of root microbiota in the plant’s adaptive response to Fe-limited environments. Voges et al.⁸ have proposed a mechanism wherein Fe deficiency induces the secretion of coumarins from the roots, which not only act as chelators to mobilize Fe³⁺ but also generate ROS such as H₂O₂. Importantly, ROS exert selective pressure on the composition of root microbiome. Within this intricate relationship, the plant’s immune system emerges as a pivotal mediator of microbial colonization^49,50,51. Plant immune reactions are recognized as important barriers to microbial colonization, hindering bacterial overgrowth and influencing the composition of the root microbiome. SA, a key signaling molecule in the regulation of plant defense responses, has been shown to shape the root microbiome⁵². Conversely, root-associated microbes can also impact plant immune responses, creating a feedback loop in the plant-microbe interaction^53,54,55. In this study, the abundance of a specific taxon, Ammoniphilus, in the roots was significantly increased under Fe-limited conditions. The inoculation with Ammoniphilus sp. reduced the accumulation of SA in the Fe-deficient roots, thereby inhibiting local immune responses and facilitating root colonization. However, SA treatment significantly impeded bacterial colonization in the Fe-deficient roots. Conversely, the bacterial mutant ΔglnA, with reduced capacity to colonize the Fe-deficient roots, exhibited modest overgrowth in the NahG transgenic plants expressing an SA hydroxylase. These findings indicated the critical role of the plant immune system in modulating bacterial colonization under Fe-limited conditions.

It has been previously been indicated that Arabidopsis plants deficient in SA biosynthesis exhibited increased resistance to Fe deficiency. Moreover, these mutants displayed enhanced reutilization of root apoplastic Fe compared with WT plants²⁹. Root apoplastic Fe can be reclaimed and transported to above-ground tissues to alleviate chlorosis resulting from Fe deficiency^20,21,22. In our study, the inoculation with Ammoniphilus sp. significantly promoted the reutilization of root apoplastic Fe under Fe-limited conditions. However, the application of exogenous SA inhibited root apoplastic Fe remobilization in the inoculated plants, rendering them more sensitive to Fe deficiency. Of particular interest was the mechanism by which Ammoniphilus sp. suppressed the accumulation of endogenous SA in the Fe-deficient roots. Metabolic analyses revealed that a total of 24 shared metabolites were significantly enriched in the roots of inoculated plants. Among these metabolites, Gln was one of the most abundant amino acids present in both the culture of Ammoniphilus sp. and the roots. Perturbed glutamine homeostasis in plants has been shown to mediate the SA signaling pathways, as Gln deficiency enhances the SA-mediated defense responses³⁹. We discovered that exogenous Gln could alleviate Fe deficiency-induced chlorosis and root apoplastic Fe remobilization. Furthermore, the SA levels were markedly reduced in the Fe-deficient roots treated with exogenous Gln, similar to the results observed in Ammoniphilus-inoculated plants. However, these effects were not observed in the ΔglnA-treated plants. Hence, our findings suggested that bacterial secretion of Gln suppressed the Fe deficiency-induced increases in root SA accumulation, thereby promoting bacterial colonization and enhancing host adaptation to Fe-limited conditions.

One mechanism of action of SA in plants is its capacity to stimulate the production of H₂O₂, which contributes to plant resistance against biotic stress^30,31,32. In this study, the accumulation of SA in the Fe-deficient roots coincided with an increase in H₂O₂ levels. The application of exogenous SA further elevated the H₂O₂ levels in the Fe-deficient roots. However, both the Ammoniphilus-inoculated and NahG transgenic plants exhibited significantly lower levels of H₂O₂ compared with the control plants. Previous work by Chen et al.²⁴ has indicated that elevated H₂O₂ levels can enhance the phenylpropane metabolism by increasing lignin synthesis and reducing phenolic secretion, thereby diminishing the efficiency of apoplastic iron mobilization. The apoplastic Fe pool plays a crucial role as a source of Fe under Fe-limited conditions. Root apoplastic Fe can be mobilized by phenolic compounds released by Fe deficient roots^20,23. Lignin, a major component of the apoplasts, is closely related to the mobilization of apoplastic Fe under Fe-limited conditions. Lignin synthesis is a H₂O₂-dependent process that requires significant amounts of phenolic substrates, leading to a reduction in phenolics within the roots²⁴. In this study, the inoculation with Ammoniphilus sp. inhibited Fe deficiency-induced lignin synthesis and promoted the secretion of phenolics from the roots. However, the exogenous application of SA, which increased root H₂O₂ levels, counteracted the effects induced by Ammoniphilus sp. Furthermore, the removal of cellular H₂O₂ was shown to enhance phenolic secretion and inhibit lignin synthesis. Therefore, we concluded that the inoculation with Ammoniphilus sp. suppressed H₂O₂-induced lignin synthesis and increased phenolic secretion, which contributed to the enhanced remobilization of root apoplastic Fe.

In conclusion, our study has shed light on the roles of root endo-microbiota in facilitating plant adaptation to iron-limited conditions. The recruited Ammoniphilus sp. effectively increased phenolic secretion and enhanced root apoplastic Fe remobilization under Fe-limited conditions. Moreover, bacterial secretion of Gln inhibited the SA-induced H₂O₂ burst triggered by Fe deficiency, thereby promoting bacterial colonization within the roots. Overall, this research has provided valuable insights into the intricate interactions between host plants and root-associated microbiota, offering significant potential for improving plant growth in Fe-limited soils.

Methods

Plant materials and growth conditions

Seeds of tomato (Solanum lycopersicum L.) were surface-sterilized and then placed in the dark at 25 °C. After two days, the seedlings were transplanted to plastic pots containing non-sterile (natural) and sterile soil. The soil was sterilized by γ-irradiation (Xiyue Radiation Technology Co., Ltd., Nanjing, China) to eliminate the effects of resident soil community. The soil used was collected from a vegetable crop field and classified as yellow soil, containing 35.52 g kg⁻¹ of organic matter, 76.68 mg kg⁻¹ of inorganic nitrogen, 65.07 mg kg⁻¹ of available phosphorus, 185.25 mg kg⁻¹ of available potassium, EC of 0.34 mS cm⁻¹, and pH of 6.2. The top 20 cm of soil layer was dried, ground, and sieved through a 3 mm mesh. For preparing calcareous soils, the pH of soils supplemented with 12.6 g kg⁻¹ or 18.5 g kg⁻¹ CaCO₃ was approximately 7.5 or 8.0, respectively. The plants were grown in a growth chamber at 25 °C, with a photoperiod of 16 h light and 8 h dark, 70% relative humidity, and a photosynthetic photon flux density of 200 μmol m⁻² s⁻¹.

Soil sampling, bacterial isolation, and DNA sequencing

Tomato plants were cultivated in pots containing soils with different pH levels. After three weeks of culture, samples of bulk and rhizosphere soils and root tissues were collected. Bulk soil samples were obtained from 0 to 10 cm soil layer and then passed through a 2 mm mesh. Rhizosphere soil was obtained by washing the soil closely adhering to the roots⁵⁶. The remaining roots were sterilized using 1% sodium hypochlorite for 5 min, followed by rinsing with a sterile saline solution. Moreover, to maximize the isolation of culturable bacterial strains, roots (1.0 g) harvested from tomato plants cultivated in calcareous soils were firstly homogenized in 20 mL of sterile saline solution. Subsequently, 100 μL of diluted root homogenate was applied to diverse kinds of culture media, including TSA, LB, King’B and NA. In total, 54 samples were obtained, including bulk soil, rhizosphere soil, and roots (3 compartments × 3 treatments × 6 repetitions).

Total DNA was extracted from soil and root samples using the Power Soil DNA isolation kit (Mobio Laboratories, Carlsbad, CA, USA). The V4 high variable region of bacterial 16S rRNA was amplified using a pair of specific primers, 515 F and 806 R, followed by sequencing on an Illumina Hiseq2500 platform. Quality filtering was performed for generating high-quality clean reads based on the Trimmomatic (V0.33) quality control process⁵⁷. Clean reads were generated using FLASH (V1.2.11) based on the overlap between the paired-end reads. Removal of barcodes and primers resulted in the generation of effective Clean Tags. These Clean Tags were then analyzed using USEARCH software (V10), and sequences with a similarity of ≥ 97% were grouped into the same operational taxonomic unit (OTU)⁵⁸. Taxonomic assignment for each representative sequence was performed using the SILVA database, with a confidence threshold of ≥ 0.5. Following the elimination of OTUs and Tags annotated as chloroplasts or mitochondria (16S amplicons), along with those unannotated at the kingdom level, an OTU table containing the information about OTU taxonomy was generated for the analysis of bacterial communities.

Gnotobiotic and hydroponic experiments

Highly alkaline (Fe-limited) soils with a pH of 8.0 were prepared using the procedure mentioned earlier. After two days of germination, tomato seedlings were transplanted into glass bottles containing sterile soil, with or without bacterial inoculum. Bacterial suspensions were collected and introduced into the soil at a final density of 5 × 107 CFU g⁻¹ soil. Heat-killed bacteria were obtained by incubation of bacterial suspensions at 99 °C for 30 min. For root light-shielding experiments, light-protected glass bottles were covered with opaque black films, allowing only the shoots to be exposed to light. These glass bottles were then vertically placed in a growth chamber. After 10 days of culture, chlorophyll content and shoot fresh weight were measured. Plant growth rescue activity was calculated using the formula: \(\% \,{\rm{growth\; rescue}}=({{SFW}}_{{inoculated\; on\; FeL}}-{{SFW}}_{{axenic\; on\; FeL})}/({{SFW}}_{{axenic\; on\; FeS}}-{{SFW}}_{{axenic\; on\; FeL}})\times 100 \%\) (FeS: Fe-sufficient conditions; FeL: Fe-limited conditions).

Hydroponic systems were employed to assess the microbe-induced Fe deficiency tolerance in plants. In brief, tomato seedlings were grown in aerated half-strength (1/2) Hoagland nutrient solutions (pH of 6.2) for two weeks. After that, plants were transferred to 1/2 Hoagland nutrient solutions containing either 0 (pH 8.0, adjusted with 5 mM HEPES) or 50 μM Fe (III)-EDTA (pH 6.2, adjusted with 5 mM MES) Fe (III)-EDTA¹. For bacterial inoculation, bacterial suspensions were introduced into the nutrient solutions at a final concentration of 5 × 10⁷ CFU mL⁻¹. The nutrient solutions were replaced every other day. A total of 15 plants per treatment were harvested for further assays.

RNA sequencing analysis

Two-week-old tomato plants were transferred to 1/2 Hoagland nutrient solution with either pH 6.2 (adjusted with 5 mM MES) or pH of 8.0 with (adjusted with 5 mM HEPES), with or without bacterial inoculum at a final concentration of 5 × 10⁷ CFU mL⁻¹. These plants were incubated for 36 and 72 h, respectively, in the same growth chamber mentioned earlier. For each treatment, we collected three replicates, with each replicate consisting of five plants. Total RNA was extracted for constructing RNA-seq libraries using the Novaseq 6000 platform (Illumina, CA, United States). Raw data were processed to eliminate adapters and low-quality reads using Trimmomatic v0.38. High-quality reads were then aligned to the tomato genome annotation ITAG4.1⁵⁹. Gene Ontology (GO) Term Finder was applied to identify GO terms that annotated a list of enriched genes with FDR-adjusted p value < 0.05. KEGG enrichment analysis of genes significantly correlated with each gene cluster at FDR-adjusted p value < 0.05 was performed using clusterProfiler⁶⁰. Additionally, qRT-PCR reactions were conducted with SlActin serving as an internal control⁶¹. The primers used for qPCR analysis were listed in Supplementary Table 3.

H₂O₂ and lignin staining

The accumulation of H₂O₂ was detected using 2’,7’-dichlorodihydrofluorescein diacetate (H₂DCFDA) staining²⁶. Root tissues were immersed in a buffer solution containing 20 µM H₂DCFDA (Sigma-Aldrich) and 20 µM phosphate (pH 6.0), followed by a 20 min dark incubation. These samples were then rinsed with distilled water to remove excess dye. The roots were visualized using a confocal microscope (Zeiss LSM 310 META) with excitation at 488 nm. To visualize lignin in the roots, a lignin staining solution was prepared by dissolving phloroglucinol in anhydrous ethanol, and adding an equal volume of concentrated hydrochloric acid²⁷. Root tissues were submerged in the staining solution and incubated at 25 °C for 5 minutes.

Assays of SA-catabolizing bacteria and construction of NahG transgenic plants

To assess the ability of bacteria to catabolize SA, bacterial isolates were cultured on minimal salts media ((NH₄)₂SO₄ 0.5 g, KH₂PO₄ 1.0 g, Na₂HPO₄·2H₂O 3.5 g, MgCl₂·6H₂O 0.1 g, Ca(NO₃)₂·4H₂O 0.05 g, FeSO₄·7H₂O 0.001 g per L) supplemented with 0.5 mM SA or 20 g L⁻¹ glucose. Additionally, the coding sequence of the NahG gene encoding salicylate hydroxylase was initially obtained by searching the NCBI database. Subsequently, the coding sequence of NahG was synthesized and inserted into pCAMBIA1300. The resulting plasmid was then introduced into Agrobacterium tumefaciens GV3101 through freeze-thaw transformation. Tomato plants were then genetically transformed using a leaf-disc method⁶². T2 transgenic lines were used for further experiments. The primers used in this study were listed in Supplementary Table 3.

Measurement of Fe content

To measure total Fe content, both shoot and roots were harvested and rinsed with 1.5 mM CaCl₂, and then digested with a mixture of HNO₃/HClO₄ (4:1, v/v). Soluble Fe in shoots and roots was extracted as previously described by Cassin et al.,⁶³. For analyzing soil available Fe, air-dried soil samples were sieved and extracted using a DTPA solution. The extract was then filtered for measuring the Fe concentration⁶⁴. Total and soluble Fe concentrations were measured using inductively coupled plasma atomic emission spectroscopy (ICP-AES). Additionally, apoplastic Fe content was measured according to the method reported by Jin et al.,²³. Root samples were immersed in 10 mL of 0.5 mM CaSO₄ for 15 min and then treated with a solution containing 1.5 mM 2.2-bipyridyl and 12.5 mM Na₂S₂O₄. Absorbance of the resulting solutions, which contained the Fe²⁺-bipyridyl complex, was recorded at 520 nm. Furthermore, the reutilizing efficiency of root apoplastic Fe (ApoFe) was calculated using the formula: \({{\rm{rREU}}}^{{ApoFe}}=({{ApoFe}}_{{sterile}}-{{ApoFe}}_{{non}-{sterile}})/{{ApoFe}}_{{sterile}}{\rm{x}}100 \%\).

Determination of chlorophyll, SA, phenolic compounds, and lignin content

Leaf chlorophyll was extracted using 90% acetone, and total chlorophyll content was calculated based on the formula: 20.21 × A645 + 8.02 × A663⁶⁵. Free SA content was measured using HPLC (Agilent 1260 Infinity II)⁶⁶. Briefly, plant tissues (200 mg) were ground and then extracted with methanol. After evaporating the methanol, the resulting precipitate was dissolved and digested with β-glucosidase to remove glycosylated SA. The resulting samples were then treated with 10% TCA and centrifuged at 12,000 × g for10min. The supernatant was extracted with a solution of consisting of ethyl acetate, cyclohexane, and isopropanol in a 100:99:1 (v/v) ratio. The organic phase was dried and analyzed using HPLC. Phenolics in the solutions were collected as previosuly reported by Jin et al.,²³. The concentration of total phenolic compounds was measured and expressed as molar equivalents of gallic acid. Root lignin content was quantified as previously described by Hatfield & Fukushima⁶⁷, with minor modifications. Approximately 1.0 g of fresh root tissues was used for lignin extraction, followed by determining the absorbance of the supernatant at 280 nm.

Metabolomic analysis and quantification of Gln content

Fresh root samples (100 mg) harvested from different treatments were ground and then extracted with 1.0 mL of cold water/methanol (1:3, v/v). After centrifugation at 13,500 × g and 4 °C for 10 min, the resulting samples were concentrated under vacuum and dissolved in methanol. After centrifuging again, the supernatant was filtered for metabolome analysis using liquid chromatograph-mass spectrometry (LC-MS) analysis. Furthermore, MetaboAnalyst v.4.0 software was applied to screen differentially abundant metabolites based on the criteria: |log₂(fold changes)| ≥ 1.0, FDR-adjusted p value with < 0.05, and variable importance in projection (VIP) > 1⁶⁸. Six plants were pooled as one replicate, and a total of five replicates were conducted. To detect the metabolites in bacterial culture, bacterial strains were inoculated into the minimal salts media supplemented with root homogenate (50 g L⁻¹), and cultured at 28 °C for 16 h. Then, bacterial cultures were centrifuged at 4 °C and 8000 rpm for 15 minutes. The resulting supernatant was concentrated and then dissolved with 2-chlorophenylalanine solution (4 mg L⁻¹) prepared in 80% methanol for LC-MS analysis. In addition, the content of Gln in the roots was extracted and measured as described previously⁶⁹.

Quantitative analysis of bacterial colonization and genetic manipulation

For qPCR analysis of bacterial colonization, a standard curve was initially established. Ten-fold serial dilutions of bacterial cells were performed, resulting in concentrations ranging from 10⁻¹ to 10⁻⁸ CFU mL⁻¹, and total DNA was then extracted. qPCR assays were performed using the AceQ qPCR SYBR Green Master Mix Kit (Vazyme Biotech, China). Additionally, 100 μL of each dilution was plated on LB agar plates. After 16 h of incubation, colony numbers were enumerated, providing colony counts for each dilution. The standard curve was then constructed by plotting Ct values against colony numbers. To quantify bacterial cells within roots, total DNA was extracted from root tissues. Then, the estimation of bacterial cell CFUs within the roots was conducted based on the constructed standard curve.

Genomic DNA from Ammoniphilus sp. AS27 was used as a template for PCR amplification of the homologous arm sequences flanking the glnA gene, and a neomycin resistance gene (Neo) was also amplified. The resulting sequences were then ligated into the pMD-18T for generating the recombinant plasmids pMD-up, pMD-down, and pMD-Neo, respectively. The upstream of glnA from pMD-up was digested and ligated into the pBluescript (pBlue-up). Then, the Neo gene from pMD-Neo was digested and ligated into the pBlue-up for generating the pMD-up::Neo. Finally, the downstream of glnA from the pMD-down was digested and ligated into the pBlue-up::Neo to generate the pBlue-glnA::Neo. The resulting vectors were transformed into Ammoniphilus sp. for the construction of ΔglnA mutant (ΔglnA). In addition, the growth of ΔglnA mutant was detected in NA liquid medium. Bacterial strains were incubated at 28 °C and 200 rpm for 24 h. The culture with an initial OD600 of 0.1 was then transferred into fresh liquid medium and incubated for 16 h. Bacterial samples were collected every 2 h, and bacterial concentration was determined by measuring the absorbance at 600 nm. Moreover, to complement the glnA disruption in ΔglnA mutant, the glnA gene, along with its promoter was amplified and cloned into the vector pUBC19. The resulting vector was then transformed into the ΔglnA mutant to obtain the glnA gene complemented strain (CΔglnA). For confocal microscopy analysis of bacterial colonization, green fluorescent strains were generated by transforming bacterial cells with the pBE2R-EGFP plasmid using a Bio-Rad Gene Pulser electroporator (Bio-Rad, Madrid, Spain) set at 2.4 kV, 200 Ω, and 25 μF with 0.1 cm cuvettes. The primers used in this study were listed in Supplementary Table 3.

Statistical analysis

Statistical analyses were carried out using the IBM SPSS 22.0 (IBM Corporation, New York, USA) and R software (Version 4.0.1). A significance level of p < 0.05 was applied for all statistical tests. In R software, Spearman’s rank correlation coefficients were performed for evaluating the relationship between the OTU abundance and rREU^apoFe at FDR-adjusted p value < 0.05⁷⁰. Manhattan plots were generated using the “ggplot2” package⁷¹. To examine the differences in the bacterial communities, PCoA analysis was performed using a Bray-Curtis dissimilarity matrix. PERMANOVA, conducted using the R vegan package, was used to assess the effect of soil pH on the bacterial community. MEGA11.0 was used for phylogenetic analysis of different bacterial isolates⁷².

Reporting summary

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

Data availability

All the sequencing dataset acquired in this study have been deposited in the China National GeneBank Nucleotide Sequence Archive (CNSA) database (accession No. CRA011449 and CRA011782). The data supporting the findings of this study are available in supplementary files or can be obtained from the corresponding authors upon request.

References

Terés, J. et al. Soil carbonate drives local adaptation in Arabidopsis thaliana. Plant Cell Environ. 42, 2384–2398 (2019).
Article PubMed PubMed Central Google Scholar
Clemens, S. & Weber, M. The essential role of coumarin secretion for Fe acquisition from alkaline soil. Plant Signal Behav. 11, e1114197 (2016).
Article PubMed Google Scholar
Palmer, C. M., Hindt, M. N., Schmidt, H., Clemens, S. & Guerinot, M. L. MYB10 and MYB72 are required for growth under iron-limiting conditions. PLoS Genet. 9, e1003953 (2013).
Article PubMed PubMed Central Google Scholar
Vélez-Bermúdez, I. C. & Schmidt, W. How plants recalibrate cellular iron homeostasis. Plant Cell Physiol. 36, 154–162 (2022).
Article PubMed Google Scholar
Walker, E. L. & Connolly, E. L. Time to pump iron: iron-deficiency-signaling mechanisms of higher plants. Curr. Opin. Plant Biol. 11, 530–535 (2008).
Article CAS PubMed Google Scholar
Riaz, N. & Guerinot, M. L. All together now: regulation of the iron deficiency response. J. Exp. Bot. 72, 2045–2055 (2021).
Article CAS PubMed PubMed Central Google Scholar
Stringlis, I. A. et al. MYB72-dependent coumarin exudation shapes root microbiome assembly to promote plant health. Proc. Natl. Acad. Sci. USA 115, E5213–E5222 (2018).
Article CAS PubMed PubMed Central Google Scholar
Voges, M. J. E. E. E., 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 PubMed PubMed Central 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 CAS PubMed PubMed Central Google Scholar
Marschner, H. & Römheld, V. Strategies of plants for acquisition of iron. Plant Soil 165, 261–274 (1994).
Article CAS Google Scholar
Santi, S. & Schmidt, W. Dissecting iron deficiency-induced proton extrusion in Arabidopsis roots. N. Phytol. 183, 1072–1084 (2009).
Article CAS Google Scholar
Martín-Barranco, A., Spielmann, J., Dubeaux, G., Vert, G. & Zelazny, E. Dynamic control of the high-affinity iron uptake complex in root epidermal cells. Plant Physiol. 184, 1236–1250 (2020).
Article PubMed PubMed Central Google Scholar
Kobayashi, T., Itai, R. N. & Nishizawa, N. K. Iron deficiency responses in rice roots. Rice 7, 27 (2014).
Article PubMed PubMed Central Google Scholar
Murata, Y. et al. A specific transporter for iron (III)-phytosiderophore in barley roots. Plant J. 46, 563–572 (2006).
Article CAS PubMed Google Scholar
Inoue, H. et al. Rice OsYSL15 is an iron-regulated iron (III)-deoxymugineic acid transporter expressed in the roots and is essential for iron uptake in early growth of the seedlings. J. Biol. Chem. 284, 3470–3479 (2009).
Article CAS PubMed Google Scholar
Ohwaki, Y. & Sugahara, K. Active extrusion of protons and exudation of carboxylic acids in response to iron deficiency by roots of chickpea (Cicer arietinum L). Plant Soil 189, 49–55 (1997).
Article CAS Google Scholar
Alcantara, E., Romera, F. J., Canete, M. & de la Guardia, M. D. Effects of bicarbonate and iron supply on Fe(III) reducing capacity of roots and leaf chlorosis of the susceptible peach rootstock “Nemaguard. J. Plant Nutr. 23, 1607–1617 (2000).
Article CAS Google Scholar
Lucena, C. et al. Bicarbonate blocks the expression of several genes involved in the physiological responses to Fe deficiency of Strategy I plants. Funct. Plant Biol. 34, 1002–1009 (2007).
Article CAS PubMed Google Scholar
Hsieh, E. J. & Waters, B. M. Alkaline stress and iron deficiency regulate iron uptake and riboflavin synthesis gene expression differently in root and leaf tissue: implications for iron deficiency chlorosis. J. Exp. Bot. 67, 5671–5685 (2016).
Article CAS PubMed PubMed Central Google Scholar
Lei, G. J. et al. Abscisic acid alleviates iron deficiency by promoting root iron reutilization and transport from root to shoot in Arabidopsis. Plant Cell Environ. 37, 852–863 (2014).
Article CAS PubMed Google Scholar
Zhu, X. F., Wang, B., Song, W. F., Zheng, S. J. & Shen, R. F. Putrescine alleviates iron deficiency via NO-dependent reutilization of root cell-wall Fe in Arabidopsis. Plant Physiol. 170, 558–567 (2016).
Article CAS PubMed Google Scholar
Graziano, M. & Lamattina, L. Nitric oxide accumulation is required for molecular and physiological responses to iron deficiency in tomato roots. Plant J. 52, 949–960 (2007).
Article CAS PubMed Google Scholar
Jin, C. W. et al. Iron deficiency-induced secretion of phenolics facilitates the reutilization of root apoplastic iron in red clover. Plant Physiol. 144, 278–285 (2007).
Article CAS PubMed PubMed Central Google Scholar
Chen, H., Zhang, Q., Cai, H., Zhou, W. & Xu, F. H₂O₂ mediates nitrate-induced iron chlorosis by regulating iron homeostasis in rice. Plant Cell Environ. 41, 767–781 (2018).
Article CAS PubMed Google Scholar
Tsai, H. H. & Schmidt, W. Mobilization of iron by plant-borne coumarins. Trends Plant Sci. 22, 538–548 (2017).
Article CAS PubMed Google Scholar
Ramírez, L., Bartoli, C. G. & Lamattina, L. Glutathione and ascorbic acid protect Arabidopsis plants against detrimental effects of iron deficiency. J. Exp. Bot. 64, 3169–3178 (2013).
Article PubMed Google Scholar
Carrió-Seguí, À. et al. The altered expression of microRNA408 influences the Arabidopsis response to iron deficiency. Front Plant Sci. 10, 324 (2019).
Article PubMed PubMed Central Google Scholar
Donnini, S., Dell’Orto, M. & Zocchi, G. Oxidative stress responses and root lignification induced by Fe deficiency conditions in pear and quince genotypes. Tree Physiol. 31, 102–113 (2021).
Article Google Scholar
Shen, C. et al. Involvement of endogenous salicylic acid in iron-deficiency responses in Arabidopsis. J. Exp. Bot. 67, 4179–4193 (2016).
Article CAS PubMed Google Scholar
Harfouche, A. L., Rugini, E., Mencarelli, F., Botondi, R. & Muleo, R. Salicylic acid induces H₂O₂ production and endochitinase gene expression but not ethylene biosynthesis in Castanea sativa in vitro model system. J. Plant Physiol. 165, 734–744 (2008).
Article CAS PubMed Google Scholar
Pasqualini, S. et al. Salicylic acid modulates ozone-induced hypersensitive cell death in tobacco plants. Physiol. Plant. 115, 204–212 (2002).
Article CAS PubMed Google Scholar
Chen, Z. et al. Induction, modification, and transduction of the salicylic acid signal in plant defense responses. Proc. Natl. Acad. Sci. USA 92, 4134–4137 (1995).
Article CAS PubMed PubMed Central Google Scholar
Su, G., An, Z., Zhang, W. & Liu, Y. Light promotes the synthesis of lignin through the production of H₂O₂ mediated by diamine oxidases in soybean hypocotyls. J. Plant Physiol. 162, 1297–1303 (2005).
Article CAS PubMed Google Scholar
Liu, W. et al. Drought-induced ABA, H₂O₂ and JA positively regulate CmCAD genes and lignin synthesis in melon stems. BMC Plant Biol. 21, 83 (2021).
Article PubMed PubMed Central Google Scholar
Tobimatsu, Y. & Schuetz, M. Lignin polymerization: how do plants manage the chemistry so well?. Curr. Opin. Biotechnol. 56, 75–81 (2019).
Article CAS PubMed Google Scholar
Wu, G. et al. Activation of host defense mechanisms by elevated production of H₂O₂ in Transgenic Plants. Plant Physiol. 115, 427–435 (1997).
Article CAS PubMed PubMed Central Google Scholar
Vanholme, R. et al. Metabolic engineering of novel lignin in biomass crops. N. Phytol. 96, 978–1000 (2012).
Article Google Scholar
Molinari, S., Fanelli, E. & Leonetti, P. Expression of tomato salicylic acid (SA)-responsive pathogenesis-related genes in Mi-1-mediated and SA-induced resistance to root-knot nematodes. Mol. Plant Pathol. 15, 255–264 (2014).
Article CAS PubMed Google Scholar
Liu, G. et al. Amino acid homeostasis modulates salicylic acid-associated redox status and defense responses in Arabidopsis. Plant Cell 22, 3845–3863 (2010).
Article CAS PubMed PubMed Central Google Scholar
Gerna, D., Roach, T., Mitter, B., Stöggl, W. & Kranner, I. Hydrogen peroxide metabolism in interkingdom interaction between bacteria and wheat seeds and seedlings. Mol. Plant Microbe Interact. 33, 336–348 (2020).
Article CAS PubMed Google Scholar
Morcillo, R. J. et al. Rhizobacterium-derived diacetyl modulates plant immunity in a phosphate-dependent manner. EMBO J. 39, e102602 (2020).
Article CAS PubMed Google Scholar
Wang, C. et al. Functional assembly of root-associated microbial consortia improves nutrient efficiency and yield in soybean. J. Integr. Plant Biol. 63, 1021–1035 (2021).
Article CAS PubMed Google Scholar
Castrillo, G. et al. Root microbiota drive direct integration of phosphate stress and immunity. Nature 543, 513–518 (2017).
Article CAS PubMed PubMed Central Google Scholar
Pii, Y. et al. The interaction between iron nutrition, plant species and soil type shapes the rhizosphere microbiome. Plant Physiol. Biochem. 99, 39–48 (2016).
Article CAS PubMed Google Scholar
Custódio, V. et al. Sculpting the soil microbiota. Plant J. 109, 508–522 (2022).
Article PubMed Google Scholar
Pieterse, C. M. J., de Jonge, R. & Berendsen, R. L. The soil-borne supremacy. Trends Plant Sci. 21, 171–173 (2016).
Article CAS PubMed Google Scholar
Bakker, P. A. H. M. et al. The soil-borne identity and microbiome-assisted agriculture: Looking back to the future. Mol. Plant 13, 1394–1401 (2020).
Article CAS PubMed Google Scholar
Lopes, L. D., Hao, J. & Schachtman, D. P. Alkaline soil pH affects bulk soil, rhizosphere and root endosphere microbiomes of plants growing in a Sandhills ecosystem. FEMS Microbiol. Ecol. 97, fiab028 (2021).
Article CAS PubMed Google Scholar
Tang, J. et al. Plant immunity suppression via PHR1-RALF-FERONIA shapes the root microbiome to alleviate phosphate starvation. EMBO J. 41, e109102 (2022).
Article CAS PubMed PubMed Central Google Scholar
Song, S., Liu, Y., Wang, N. R. & Haney, C. H. Mechanisms in plant-microbiome interactions: lessons from model systems. Curr. Opin. Plant Biol. 62, 102003 (2021).
Article CAS PubMed Google Scholar
Teixeira, P. J. P., Colaianni, N. R., Fitzpatrick, C. R. & Dangl, J. L. Beyond pathogens: microbiota interactions with the plant immune system. Curr. Opin. Microbiol. 49, 7–17 (2019).
Article CAS PubMed Google Scholar
Lebeis, S. L. et al. Salicylic acid modulates colonization of the root microbiome by specific bacterial taxa. Science 349, 860–864 (2015).
Article CAS PubMed Google Scholar
Cheng, Y. T., Zhang, L. & He, S. Y. Plant-microbe interactions facing environmental challenge. Cell Host Microbe 26, 183–192 (2019).
Article CAS PubMed PubMed Central Google Scholar
Syed, A. B., Rahman, S. F., Singh, E., Pieterse, C. M. J. & Schenk, P. M. Emerging microbial biocontrol strategies for plant pathogens. Plant Sci. 267, 102–111 (2018).
Article Google Scholar
Bai, B. et al. The root microbiome: Community assembly and its contributions to plant fitness. J. Integr. Plant Biol. 64, 230–243 (2022).
Article PubMed Google Scholar
Li, Z. et al. A simplified synthetic community rescues Astragalus mongholicus from root rot disease by activating plant-induced systemic resistance. Microbiome 9, 217 (2021).
Article CAS PubMed PubMed Central Google Scholar
Caporaso, J. G. et al. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7, 335–336 (2010).
Article CAS PubMed PubMed Central Google Scholar
Edgar, R. C. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996–998 (2013).
Article CAS PubMed Google Scholar
Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009).
Article CAS PubMed PubMed Central Google Scholar
Mao, X. Z., Cai, T., Olyarchuk, J. G. & Wei, L. P. Automated genome annotation and pathway identification using the KEGG Orthology (KO) as a controlled vocabulary. Bioinformatics 2, 3787–3793 (2005).
Article Google Scholar
Huang, J. et al. A simplified synthetic rhizosphere bacterial community steers plant oxylipin pathways for preventing foliar phytopathogens. Plant. Physiol. Biochem. 202, 107941 (2023).
Article CAS PubMed Google Scholar
Barriuso, J. et al. Transgenic tomato plants alter quorum sensing in plant growth-promoting rhizobacteria. Plant Biotechnol. J. 6, 442–452 (2008).
Article CAS PubMed Google Scholar
Cassin, G., Mari, S., Curie, C., Briat, J. F. & Czernic, P. Increased sensitivity to iron deficiency in Arabidopsis thaliana over accumulating nicotianamine. J. Exp. Bot. 60, 1249–1259 (2009).
Article CAS PubMed PubMed Central Google Scholar
Lindsay, W. L. & Norvell, W. A. Development of a DTPA soil test for zinc, iron, manganese and copper. Soil Sci. Soc. Am. J. 42, 421–428 (1978).
Article CAS Google Scholar
Porra, R. J. The chequered history of the development and use of simultaneous equations for the accurate determination of chlorophylls a and b. Photosynth. Res. 73, 149–156 (2022).
Article Google Scholar
Tanabe, N. et al. The basic helix-loop-helix transcription factor, bHLH11 functions in the iron-uptake system in Arabidopsis thaliana. J. Plant Res. 132, 93–105 (2019).
Article CAS PubMed Google Scholar
Hatfield, R. & Fukushima, R. S. Can lignin be accurately measured?. Crop Sci. 45, 832–839 (2005).
Article CAS Google Scholar
Xiao, B. et al. Flooding-induced rhizosphere Clostridium assemblage prevents root-to-shoot cadmium translocation in rice by promoting the formation of root apoplastic barriers. J. Hazard. Mater. 439, 129619 (2022).
Article CAS PubMed Google Scholar
Tabuchi, M. et al. Severe reduction in growth rate and grain filling of rice mutants lacking OsGS1;1, a cytosolic glutamine synthetase1;1. Plant J. 42, 641–651 (2005).
Article CAS PubMed Google Scholar
Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Methodol. 57, 289–300 (1995).
Article Google Scholar
Wang, W. et al. Long-term nitrogen application decreases the abundance and copy number of predatory myxobacteria and alters the myxobacterial community structure in the soil. Sci. Total Environ. 708, 135114 (2020).
Article CAS PubMed Google Scholar
Kumar, S., Stecher, G. & Tamura, K. MEGA7: Molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870–1874 (2016).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

This study was funded by the Natural Science Foundation of Universities in Anhui Province (2023AH020024, gxyq2022053), the University-Level Advanced Discipline (XK-XJGY003), Anhui Provincial Collaborative Innovation Project (202306e08020014), the County-Level Science and Technology Plan Project (2023TPY04) and the Discipline (Major) Leader Development Program (DTR2024033). The funder played no role in study design, data collection, analysis and interpretation of data, or the writing of this manuscript.

Author information

Authors and Affiliations

Key Lab of Bio-Organic Fertilizer Creation, Ministry of Agriculture and Rural Affairs, College of Biomedicine and Health Science, Anhui Science and Technology University, Chuzhou, China
Lin Zhu, Aijing Zhang, Huabing Liu, Yue Xie, Xiaomin Lu & Jianfei Wang
School of Life Science and Technology, Tongji University, Shanghai, China
Lin Zhu
School of Medicine, Zhejiang University, Hangzhou, China
Jiansheng Guo
Institute of Horticulture, Anhui Academy of Agricultural Sciences, Hefei, China
Congsheng Yan
Jiangsu Provincial Key Lab of Solid Organic Waste Utilization, Jiangsu Collaborative Innovation Center of Solid Organic Wastes, Educational Ministry Engineering Center of Resource-Saving Fertilizers, Nanjing Agricultural University, Nanjing, China
Cheng Zhou

Authors

Lin Zhu
View author publications
Search author on:PubMed Google Scholar
Aijing Zhang
View author publications
Search author on:PubMed Google Scholar
Jiansheng Guo
View author publications
Search author on:PubMed Google Scholar
Huabing Liu
View author publications
Search author on:PubMed Google Scholar
Yue Xie
View author publications
Search author on:PubMed Google Scholar
Xiaomin Lu
View author publications
Search author on:PubMed Google Scholar
Congsheng Yan
View author publications
Search author on:PubMed Google Scholar
Jianfei Wang
View author publications
Search author on:PubMed Google Scholar
Cheng Zhou
View author publications
Search author on:PubMed Google Scholar

Contributions

C.Z. conceived the study. C.Z. and J.W. designed the experiments. L.Z., H.L., A.Z., C.Y., and J.G. performed the experimental work. Y.X. and X.L. designed the gnotobiotic experiments. C.Z., J.W., and L.Z. contributed to writing the manuscript. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Jianfei Wang or Cheng Zhou.

Ethics declarations

Competing interests

The authors declare no competing interests.

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

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, 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 licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence 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 licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Zhu, L., Zhang, A., Guo, J. et al. Iron limitation-induced endophytic Ammoniphilus assemblage promotes root apoplastic iron remobilization by attenuation of salicylic acid pathways. npj Biofilms Microbiomes 11, 125 (2025). https://doi.org/10.1038/s41522-025-00765-z

Download citation

Received: 16 October 2023
Accepted: 18 June 2025
Published: 01 July 2025
DOI: https://doi.org/10.1038/s41522-025-00765-z