The rice variety Tianyou 998 used in this study is more sensitive to Fe²⁺ compared to other rice varieties and forms a larger number of iron plaques (Fu et al., 2012). The rice seeds were cleaned by rinsing with distilled water to remove impurities, and then disinfected with a 10% hydrogen peroxide (H₂O₂) solution for 10 minutes. Clean culture dishes were chosen and washed with distilled water. Filter paper was placed at the bottom of each dish, and nutrient solution was added to saturate the filter paper. The sterilized rice seeds were placed on filter paper, and the dishes were incubated at 30 °C for germination.

Healthy and uniform rice seedlings were chosen from the culture dishes. These seedlings were placed in preprepared plastic foam trays and slightly immersed in 1/4 strength complete nutrient solution. The nutrient solution was replaced every 3 d, and the rice plant growth was monitored. When the rice plants reached the three-leaf stage, they were transplanted into PVC pipes containing quartz sand and cultivated for 21 d. The nutrient solution was replaced every 3 d during this period. After 21 d, the rice roots were meticulously extracted and cleaned for subsequent analyzes.

The composition of the complete nutrient solution used in this experiment was as follows: NH₄NO₃ 0.429 mmol/L, Ca(NO₃)₂·4H₂O 1 mmol/L, MgSO₄·7H₂O 1.667 mmol/L, KH₂PO₄ 1 mmol/L, K₂SO₄ 0.16 mmol/L, Fe(III)-EDTA 50 μmol/L, MnSO₄ 9.1 μmol/L, ZnSO₄ 0.15 μmol/L, CuSO₄ 0.16 μmol/L, (NH₄)₄MoO₂₄·4H₂O 0.52 μmol/L, and H₃BO₃ 19 μmol/L. The stock solution was prepared with the aforementioned formula (Fu et al., 2016).

To investigate the effects of the Fe²⁺ and PO₄ ^3- concentrations on the formation of rice arbuscular mycorrhizal iron plaques, an iron-phosphorus ratio experiment was devised. Four different FeCl₂·4H₂O concentrations were selected, with three parallel experiments for each concentration: (1) the Fe0.4 treatment: 0.4 mmol/L FeCl₂·4H₂O, (2) the Fe1.2 treatment: 1.2 mmol/L FeCl₂·4H₂O, (3) the Fe2.0 treatment: 2.0 mmol/L FeCl₂·4H₂O, and (4) the Fe3.2 treatment: 3.2 mmol/L FeCl₂·4H₂O. All rice experiments were carried out with a phosphorus concentration of 0.04 mmol/L. The rice plants were cultured for 2 d, after which varying iron concentrations were introduced while maintaining a constant phosphorus concentration.

Similarly, to explore the effects of PO₄ ^3- concentrations on the formation of rice arbuscular mycorrhizal iron plaques, another iron-phosphorus ratio experiment was conducted. These experiments included four PO₄ ^3- concentrations: (1) the P0.1 treatment: 0.1 mmol/L KH₂PO₄, (2) the P0.033 treatment: 0.033 mmol/L KH₂PO₄, (3) the P0.02 treatment: 0.02 mmol/L KH₂PO₄, and (4) the P0.01 treatment: 0.01 mmol/L KH₂PO₄. As with the Fe experiments, after 2 d of cultivation with the corresponding KH₂PO₄ concentration, the same iron concentration (0.1 mmol/L Fe) was introduced while keeping the phosphorus concentration constant. The rice plants were cultured for an additional 2 d.

Tweezers are used to pick out the quartz sand one by one to delicately remove the rice roots. The rice roots were delicately removed, washed with distilled water, placed on a flat surface, and scanned with a root scanner to record the data.

Standard solutions with known DCB-Fe concentrations were prepared. These solutions had concentrations of 1 μg/mL, 2 μg/mL, 3 μg/mL, 4 μg/mL, and 5 μg/mL. The absorbance values of these solutions were utilized to generate a standard curve.

After 24 h of treatment, the fresh rice roots (0.2 g) were extracted for enzyme measurements. The superoxide dismutase (SOD) activity was measured with the method of Jiang et al., reported (Jiang et al., 2023), and one unit of activity inhibited the reduction of nitroblue tetrazolium (NBT) by 50%. The peroxidase (POD) and catalase (CAT) activities were determined with the guaiacol method and potassium permanganate titration method, respectively (Thounaojam et al., 2012).

The main root length, total root length, root surface area, and root volume were measured with a root scanner. The rice root vitality was assessed with the amount of α-naphthylamine oxidation over 24 h (Zhang et al., 2021).

The rice roots, labeled according to their concentrations, were washed with deionized water. Subsequently, they were dried at 75°C for 4 h in an oven and analyzed with a field emission scanning electron microscope (FESEM) (Hitachi SU801, Japan). A small amount of powder was placed on conductive tape, and the sample surface was observed after platinum/carbon coating. The acceleration voltage was 5 kV, the current was 5 μA, and the working distance (WD) was 7.9 nm.

Deionized water-washed rice roots were carefully placed into 1.5 mL conical tubes. Subsequently, 0.5 mL of a fixative solution, which contained a 2.5% glutaraldehyde solution, was added to each tube. These tubes were then subjected to centrifugation for 1 hour to create compact aggregates. Following centrifugation, the fixative solution was carefully poured into the centrifuge tubes, ensuring complete submersion of the samples. The prepared samples were subsequently cold-fixed at a temperature of 4°C within a refrigerator for a duration of 12 h.

For subsequent stages of processing, the fixed samples were embedded in epoxy resin. To maintain a low temperature during this process, the samples were initially placed in foam boxes accompanied by ice packs. To facilitate the observation process, the prepared samples were sectioned. For this purpose, the samples were embedded in epoxy resin and allowed to polymerize in darkness for a period of 48 h. Subsequently, sectioning was executed with an ultramicrotome equipped with a diamond knife (Leica EM·UC6). The resulting sections, with approximate thicknesses of 50 nm, were then positioned onto porous carbon-coated copper grids for observation.

The data were subjected to analysis of variance, and post hoc comparisons were carried out with Duncan’s multiple range test at P<0.05. The statistical software program SPSS version 13.0 (SPSS Inc., Chicago, IL, USA) was used.

To learn how Fe²⁺ and PO₄ ^3- affected rice root growth and hence the formation of iron plaques on the root surfaces, different Fe²⁺ and PO₄ ^3- concentrations were applied during rice growth. The results demonstrated that the primary root length of rice increased from 10.80±0.12 to 12.52±0.30 cm, total root length increased from 126.51±2.69 to 136.77±1.46 cm/plant, root surface area increased from 14.19±0.11 to 18.20±0.16 cm²/plant, and root volume increased from 0.175±0.003 to 0.192 ±0.002 cm³/plant within the Fe²⁺ concentration range of 0.4 to 3.2 mmol/L ( Table 1 ). We found that the taproot, total root, root surface area and root volume of rice exhibited increasing trends as the concentration of Fe²⁺ in the nutrient solution was increased from 0.4 to 3.2 mmol/L. This suggested that the presence of Fe²⁺ in the environment promoted rice root growth. The taproot length, total root length, and root surface area of the rice exhibited significant increases when the concentration of PO₄ ^3- in the nutrient solution was raised from 0.01 to 0.1 mmol/L. The increases in the fresh root weight and the root volume, however, were not pronounced, possibly due to the reductions in root diameter with PO₄ ^3- deficiencies. This was consistent with previous findings indicating that a PO₄ ^3- deficiency promoted plant root growth, increases in the root hair length and/or density and a decrease in the root diameter, ultimately resulting in a greater surface area for soil probing (Gahoonia et al., 2001; Deng et al., 2020).

The extraction of iron from iron plaques on the rice root surface of DCB was employed to quantitatively compare the extent of iron plaque formation (Zhang et al., 2023). As depicted in Figure 1 , an increase in the concentration of extrinsic Fe²⁺ in the nutrient solution resulted in a significant elevation of the extracted DCB-Fe concentration (P>0.05). These findings suggested that augmenting the Fe²⁺ concentration promoted iron plaque formation on the root surface. When controlling the Fe²⁺ concentration at 0.3 mmol/L, only the PO₄ ^3- concentration in the nutrient solution was changed. Figure 1B shows that an increase in the PO₄ ^3- concentration resulted in a decrease in the DCB-Fe concentration on the root surface. The increased PO₄ ^3- concentration in the nutrient solution significantly hampered the formation of iron plaques on the root surfaces.

The vitality of the rice root was evaluated based on the oxidation of α-naphthylamine with varying concentrations of Fe²⁺ and PO₄ ^3-. The vitality of the rice roots gradually improved with increasing concentrations of Fe²⁺ in the nutrient solution ( Figure 2A ). When the Fe²⁺ concentration reached 3.2 mmol/L, there was only a slight enhancement in the rice root vitality, suggesting that a higher concentration of Fe²⁺ did not exert a significant promotive effect on vitality. Cultivating the rice roots in a PO₄ ^3–enriched nutrient solution provided an initial increase in the root vitality with increasing PO₄ ^3- concentration. The PO₄ ^3- deficiency negatively impacted the root vitality of the rice.

We quantified the activities of the root SOD, POD and CAT enzymes in rice roots exposed to varying concentrations of Fe²⁺ and PO₄ ^3-. The findings demonstrated a 36.7 % increase in the SOD enzyme activity, a 32.4% increase in the POD enzyme activity, and a 33.6% increase in the CAT enzyme activity with increases in the Fe²⁺ concentration from 0.4 mmol/L to 3.2 mmol/L, indicating that higher Fe²⁺ levels promoted root vitality and enhanced enzymatic function ( Figures 2B-D ). When the concentration of PO₄ ^3- was increased from 0.01 mmol/L to 0.02 mmol/L, the activities of the SOD, POD and CAT enzymes decreased. However, the activities of these three enzymes exhibited increases with further increases in the concentration of PO₄ ^3- from 0.02 mmol/L to 0.1 mmol/L ( Figures 2B-D ), suggesting that both moderate increases in PO₄ ^3- concentration and its deficiency can positively impact root vitality and enzyme activity.

To investigate the morphologies of rice iron plaques treated with different Fe²⁺ and PO₄ ^3- nutrient solutions, the rice roots were dried and observed with SEM. The results revealed that upon treatment with an Fe²⁺ concentration of 0.4 mmol/L, the rice root surfaces exhibited a distribution of small iron oxide particles measuring approximately 1 μm, along with larger particles measuring approximately 6 μm when observed at a magnification of 2000 times ( Figure 3A ; Supplementary Figure 1A ). When the concentration of Fe²⁺ was increased to 1.2 mmol/L, a significant increase in the number of particles on the surfaces of the rice roots was observed, as shown in Figure 3B ; Supplementary Figure 1B . Moreover, these particles exhibited a predominant morphology with larger particles measuring approximately 2 μm, along with larger iron oxide particles measuring approximately 10 μm. As the concentration of Fe²⁺ was increased to 2.0 mmol/L, distinct particles were observed on the rice root surfaces, with sizes of 13 μm (large), 5 μm (medium) and less than 1 μm (small) ( Figure 3C ; Supplementary Figure 1C ). This observation suggested that the elevated Fe²⁺ levels facilitated the dissolution and subsequent recrystallization processes of the iron plaque particles present on the root surfaces. The concentration of Fe²⁺ was raised to 3.2 mmol/L, and the particles on the root surfaces vanished to give a homogeneous iron plaque. The surface was uniformly covered by an iron plaque with a thickness of approximately 1 μm ( Figure 3D ; Supplementary Figure 1D ).

The morphological changes occurring during the formation of iron plaque on the rice root surface when using a PO₄ ^3- nutrient solution exhibited an opposite trend to that observed with Fe²⁺ treatment. Following cultivation with 0.01 mmol/L PO₄ ^3-, there was a relatively low occurrence of larger particles measuring 3 μm on the surfaces of the rice roots ( Figure 4A ; Supplementary Figure 2A ). The particle sizes on the rice root surfaces decreased to less than 1 μm as the concentration of PO₄ ^3- was increased to 0.033 mmol/L, and the quantity decreased ( Figures 4B, C ; Supplementary Figure 2C ). When the concentration of PO₄ ^3- reached 0.1 mmol/L, the rice root surfaces exhibited smooth textures devoid of any discernible particles ( Figure 4D ; Supplementary Figure 2D ), implying a gradual reduction and eventual disappearance of the iron oxide particles with increasing PO₄ ^3-.

To enhance the observation of iron surface responses to variations in the Fe²⁺ and PO₄ ^3- concentrations within the nutrient solution, HRTEM was employed for examination after preparing ultrathin sections of the roots ( Figures 5 , 6 ). When the Fe²⁺ concentration in the nutrient solution reached 0.4 mmol/L, a colloidal iron plaque with a thickness of approximately 10 μm was observed on the root surface, featuring tiny particles measuring approximately 200 μm at its center, as shown in Figure 5A . As the concentration of Fe²⁺ was further increased to 1.2 mmol/L, a surface layer with a thickness of only 2 μm consisting of aggregated small particles covered the root surface ( Figure 5B ). Upon closer examination, it was revealed that this layer consisted of nanoparticles ranging from one to several tens of particles, each measuring approximately 200 nm in diameter ( Figure 5C ). When the concentration of Fe²⁺ reached 2.0 mmol/L, the thickness of the particle layer increased to approximately 5 μm ( Figure 5D ). Finally, upon reaching a concentration of 3.2 mmol/L, there were no further increases in the thickness of the particle layer ( Figure 5E ). At this juncture, a distinct 3 μm and block-like hematite morphology emerged within the central region of the layer, which was enveloped by a colloidal coating ( Figure 5F ).

When maintaining a constant Fe²⁺ concentration and adjusting only the PO₄ ^3- concentration in the nutrient solution, the results were consistent with the SEM observations ( Figure 6 ). Upon addition of PO₄ ^3- at a concentration of 0.01 mmol/L, a surface layer composed of small particles and with a thickness of 3 μm formed, which resembled the morphology observed upon adding Fe²⁺ at a concentration of 1.2 mmol/L ( Figure 6A ). However, when the concentration of PO₄ ^3- in the nutrient solution was further increased to 0.02 mmol/L, it was observed that the thickness of this layer decreased to approximately 500 nm ( Supplementary Figures 3 , 4 ). Additionally, small particle aggregates emerged on the outer sides of the root surfaces. This observation suggested that a decrease in the PO₄ ^3- concentration made the iron plaque on the root surface more susceptible to detachment, as shown in Figure 6B . When the concentration of PO₄ ^3- reached 0.033 mmol/L, there is a monolayer consisting of nanoparticles with a diameter of approximately 200 nm was observed on the iron plaque located on the root surface ( Figure 6C ). Upon reaching the maximum concentration of 0.1 mmol/L, the iron plaque was transformed into a colloidal form with a thickness less than 100 nm, predominantly composed of iron oxide ( Figure 6D ).