Phytate exudation by the roots of Pteris vittata can dissolve colloidal FePO4

Phosphorus (P) is limiting nutrient in many soils, and P availability may often depend on iron (Fe) speciation. Colloidal iron phosphate (FePO4coll) is potentially present in soils, and we tested the hypothesis that phytate exudation by Pteris vittata might dissolve FePO4coll by growing the plant in nutrient solution to which FePO4coll was added. The omission of P and Fe increased phytate exudation by P. vittata from 434 to 2136 mg kg−1 as the FePO4coll concentration increased from 0 to 300 mM. The total P in P. vittata tissue increased from 2880 to 8280 mg kg−1, and the corresponding increases in the trichloroacetic acid (TCA) extractable P fractions were inorganic P (860–5100 mg kg−1), soluble organic P (250–870 mg kg−1), and insoluble organic P (160–2030 mg kg−1). That is, FePO4-solubilizing activity was positive correlated with TP, TCA P fractions in P. vittata, TP in growth media, and root exudates. This study shows that phytate exudation dissolved FePO4coll due to the chelation effect of phytic acid on Fe; however, the wider question of whether phytic acid excretion was prompted by deprivation of P, Fe, or both remains to be answered.


Introduction
Phosphorus (P) is an important nutrient for plant growth (Meyer et al. (Missong et al.2018)) and world food production (Fresne et al. (Fresne et al.2021)), playing a major role in plant metabolic processes (He et al. (He2020)). Since the 1960s, the extensive use of P-fertilizers (Lin et al. (Liu et al.2016)), and many other P products has caused excessive P levels to more frequently disturb water bodies and aquatic systems (Lei et al. (Lessl and Ma2013)) and cause many environmental problems such as eutrophication (Liu et al. (Liu et al.2017)) and also threat to human health (Xu et al. (Zhang et al.2007)).
In soil solutions, the colloids carrying P govern the mobility and availability of P in the aqueous phase of soil (Gottselig et al. (Gottselig2017)); however, the availability of P also depends on the P concentration and its speciation (Montalvo et al. (Moradi2020)). The colloids (diameter of 1−1000 nm) are rich in Al/Fe oxyhydroxides (Wang et al. (Wang et al.2021); Eltohamy et al. (Eltohamy et al.2021); Wang et al. (Xu et al.2020)) and strongly sorb P in solution decreasing its availability (Montalvo et al. (Moradi2020)), accelerate P mobility (Moradi et al. (Nielsen et al.2013)), and transport P to the aquatic environment (Vendelboe et al. (Violante and Caporale2015)). Also, high concentrations of Fe/Al oxyhydroxides in the Ah soil horizon may promote P sorption in the colloidal phase and thereby promote P loss via leaching (Missong et al. (Montalvo et al.2015)). The association of colloidal P with Fe oxyhydroxides showed the important role of Fe in catchment areas, the P mobility may be increased by the Fe rich colloids, but decreased the P availability (Baken et al. (Baken et al.2016)). Jiang et al. ((Jones1998)) claimed that the majority of P bind to Fe oxides, and after the dissolution of Fe oxides, this P may be released and available for plants and microbes, while the previous study of Baken et al. ((Baken et al.2014)) showed that Fe rich colloids can decrease the P bioavailability to algae and promote eutrophication of natural water. Thus, it is very important to reuse the FePO 4coll from soil solution before going to water bodies. P deficiency affects a plant's metabolism processes, causing the synthesis and release of organic acids into the rhizosphere (Ryan et al. 2001;Wang et al. (Wang et al.2016); Chen et al. (Chen et al.2017)). Different plants produce different root exudates which promote the uptake of nutrients (Han et al. (Han et al.2017)) and P utilization (Wang and Lambers (Wang et al.2013)). For example, P deficiency induces citric acid exudation by white lupin (Lupinus albus) in response to P deficits (Cheng et al. (Cheng et al.2011)), and oats (Avena sativa) appear to do likewise (Wang et al. (Wang et al.2020)). In contrast, Chinese brake fern (Pteris vittata) may respond by releasing phytic acid (Fu et al. (Fu et al.2017)). In the rhizosphere, organic acids may act as metal chelators and affect the dissolution and release of P from mineral phases (Jones (Lambers and Plaxton2015)).
Chinese brake fern (Pteris vittata) is a widely distributed plant in Asia, Europe, Africa, and Australia, can be grow in d i f f e r e n t e n vi r o nm en t s ( W an e t al . ( W a ng a n d Lambers2019)), with a high yield of about 36 ton per hectare (Song et al. (Subbarao et al.1997)). Besides, Pteris vittata (P. vittata) is also native to soils (tropical) of low nutrient status that contain mostly unavailable organic and inorganic P (Liu et al. (Mathews et al.2010)). This association suggests that P. vittata may access the P and Fe from insoluble phases such as those described in the preceding paragraph and/or mineralize organic P (Fu et al. (Fu et al.2017)). Unlike the more typical organic acids (citrate and oxalate), phytic acid has been detected in root exudates of P. vittata (Tu et al. (Tu et al.2011)) where it may be the main source of P storage, as it is in cereal grains and their products (Thavarajah et al. (Trela2010)). Phytic acid forms strong chelates with Fe ions at a wide pH range (De Stefano et al. (De Stefano et al.2003); Trela (Tu et al.2004)) and consequently also adsorbs to Fe/Al soil minerals (Chen and Arai (Chen and Arai2019)).
Colloidal P and Fe are frequently associated (Niyungeko et al. 2018;Fresne et al. (Fresne et al.2021), and affect the uptake of P from colloids (Montalvo et al. (Moradi2020); Zhang et al. (Ryan et al. 2001)). In addition, Sega et al. ((Song et al.2019)) claimed that organic acid exudation may dissolve nanoFePO 4 and release P for plant uptake; however, they did not identify which organic acids may be involved. Moreover, these earlier studies did not mention the potential role of phytate exudation in dissolution of FePO 4coll , possibly because phytic acid contains a high constituent concentration of P. There is however evidence that phytate exudation from P. vittata releases arsenic (an analog to P) from contaminated soil (Tu et al. (Tu et al.2011)); however, the mechanism remains obscure due to the complexities of working in a soil system. To avoid these complexities in our study of P. vittata and the effect of phytate exudation on FePO 4coll dissolution, we used a hydroponic system.

Materials and methods
Plant conditioning P. vittata plants (height 5-10 cm) were purchased from Guangdong province, and seedlings of pea (Pisum sativum) and lettuce (Lactuca sativa) from a local nursery in Hangzhou, Zhejiang Province, China. The P. vittata, pea, and lettuce were first conditioned in 0.2-strength aerated Hoagland (Table S1) solution at pH 6.5. The solution was buffered with 1 mM KOH-MES (2-(N-morpholino) ethane sulfonic acid) (Mathews et al. (Meyer et al.2020)). Water losses due to evapotranspiration were replaced daily with Milli-Q water and replaced every 2 weeks. The plants were raised in a growth chamber at~70% relative humidity, 25°C day/night temperature, and a 16 h light and 8 h darkness (Wan et al. (Wan2020)). When the new fronds and leaves were emerging, the conditioned plants were used in experiments 2.2-2.4.
During conditioning, the plants were raised in a growth chamber at~70% relative humidity, 25°C day/night temperature, with 16 h of light per day (Wan et al. (Wan2020)).

Phosphorus accumulation experiment
There were three plant species P. vittata, pea, and lettuce. Each pot contained three conditioned plants of the one species and contained 500 mL of half-strength of aerated Hoagland nutrient solution. The composition of the solution was modified to contain 0, 50, 100, and 150 mg P. L −1 as potassium dihydrogen phosphate (KH 2 PO 4 ). Solutions were topped up daily with pure water to replace loss due to evapotranspiration, and solutions were replaced twice a week. Plants were harvested after 15 days, and the roots were separated from the leaves and dried at 80°C.
Subsamples of dried leaves/fronds roots were ground and digested in concentrated nitric acid (HNO 3 ). Diluted digests were analyzed for P using an inductively coupled plasmaatomic emission spectrometer (ICP-OES) according to the manufacturer's instructions (iCAP 6000 series, Thermo Fisher scientific, USA).
Organic acid excretion by P. vittata, pea, and lettuce Three conditioned P. vittata, pea and lettuce plants (the "Plant conditioning" section) were transferred to separate pots containing 500 mL of P and Fe free, half strength Hoagland nutrient solution, buffered at pH 6.0 with 0.5 M MES buffer for 3 days (Fayiga et al. (Fayiga et al.2008)). The organic acid concentrations were determined as described in section S 2.2.

Effect of P. vittata on FePO 4coll dissolution
The experiment was conducted in Fe-and P-free 0.2-strength Hoagland media. Five conditioned P. vittata plants (the "Plant conditioning" section) were grown in 500 mL per pot of sterile, Fe and P free, MES (0.5 M) buffered half strength Hoagland solution to investigate the role of phytic acid excretion on the dissolution of FePO 4coll . The FePO 4coll was prepared and characterized as described in the supplementary material (S2.2, Fig. 1). The hydroponic solutions contained solid phase FePO 4coll at mass/volume concentrations of (i) 0 mM FePO 4coll (control), (ii) 100 mM FePO 4coll , (iii) 200 mM FePO 4 c o l l , and (iv) 300 mM FePO 4 c o l l . Chloramphenicol at 30 mg L −1 was added to minimize microbial activity. Evapotranspiration losses were replaced with pure water daily, and the solution was replaced every 2 weeks. Analysis for P and Fe, in the growth media and P. vittata, was conducted as in section S 2.2. The trichloroacetic acid (TCA)  Error bars indicate the SE (n =3), values followed by the same letter are not significantly different (P<0.05) P fractions in P. vittata samples were analyzed as explained in section S 2.3. In addition, the FePO 4 -solubilizing activity in root exudates of P. vittata and phytase activity in fresh roots of P. vittata were determined according to Subbarao et al. ((Thavarajah et al.2009)) and Richardson et al. ((Sega2020)), explained in section S 2.4.
Effect of oxalic acid, citric acid, and phytic acid on FePO 4coll dissolution Solid FePO 4coll was prepared and characterized after Liu et al. (2017b) as detailed in supplementary material (S 2.2). The in vitro experiment was performed in 10 mM NaClO 4 . The 100 μM FePO 4coll was mixed with phytic acid, oxalic and citric acid at organic acid: FePO 4coll molar ratios of 1:1, 2:1, 3:1, 10:1, 30:1, and 60:1 in a centrifuge tube with final volume 50 mL. The pH was adjusted to 6.0 after the addition of organic acids using 0.1 L NaOH and HCl, and the redox potential of the solution were maintained at 67.2 mV. To minimize microbial activity, chloramphenicol at 30 mg L −1 was also added (Subbarao et al. (Thavarajah et al.2009)). Samples were shaken at 200 rpm for 24 h at 28°C then centrifuged at 5000 rpm for 10 min. The supernatant was analyzed for P and Fe using ICP-MS calibrated and operated as per the manufacturer's instructions (NexION300X, PerkinElmer, USA).

Statistical analysis
Results were compared using analysis of variance and a 5% significance level with the least significant difference (Tukey). Linear regression modeling was used to predict the effect of the Fe-P-solubilizing activity and TCA fractions in fern, TP in root exudates and growth media and effect of TP in P. vittata, growth media and root exudates and Fe in growth media on biomass of P. vittata by using the SPSS software (Version 20.0. SPSS Inc, USA), while the origin 2020 (Origin Lab Corporation, USA) was used for graphical illustration. Visual Minteq 3.1 was used to predict the phytate, and FePO 4coll speciation by keeping the pH 6.0 and zero ionic strength and allowing Fe phytate phases to precipitate.

Role of phytate exudation in FePO 4coll dissolution
In P. vittata, TP, TCA fractions, and Fe contents also increased with increasing FePO 4coll concentration. The total P concentration increased from 2880 to 8280 mg kg −1 and the TCA fractions were inorganic P (860-5100 mg kg −1 ), soluble organic P (250-870 mg kg −1 ), and insoluble organic P (160-2030 mg kg −1 ) (Fig. 3). There was also a dose response relationship between the external concentration of FePO 4coll and the internal concentrations of Fe, e.g., with 300 mM FePO 4coll the plants contained the highest Fe concentration (P <0.05, 3560 mg kg −1 fresh weight) (Fig. 3). The total P in root exudates also increased as the treatments concentration increases (Fig. 3).  5 Simple linear regression between Fe-P-solubilizing activity and TP in P. vittata, growth media, root exudates, insoluble, soluble and inorganic P in P. vittata and between biomass of P. vittata and TP in P. vittata, growth media, root exudates, and Fe in growth media Biomass Fe, P content in growth media, phytase, and Fe-P-solubilizing activity The biomass of P. vittata also increased as the FePO 4coll concentration increased from 0 to 300 mM (Fig. 4). The concentrations of P and Fe in the hydroponic media followed dose response relationship with the FePO 4coll addition. The P and Fe were higher (80.65 and 51.06 mM) (P <0.05) in 300 mM followed by 200 mM (72.32 and 36.1 mM) while 100 mM have less P and Fe concentration (Fig. 4). In root exudates, the phytase and Fe-P-solubilizing activity increased as the concentration of FePO 4coll increased from 0 to 300 mM. The Fig. 6 Effects of organic acids (phytic acid, oxalic acid, and citric acid) on P and Fe dissolution from FePO 4coll (s) at a molar ratio of 1:1, 2:1, 3:1, 10:1, 30:1, and 60: 1.
Error bars indicate the SE (n =3), values followed by the same letter are not significantly different (P<0.05) phytase activity was greater in 300 mM FePO 4coll (8.3 h −1 g −1 root FW), followed by 200 and 100 mM, while the control ha the least phytase activity (3.2 h −1 g −1 root FW). The same trend was observed in the Fe-P-solubilizing activity which also increased from 0.86 to 6.85 (μg P mL −1 ) as FePO 4 c o l l concentration increased from 0 to 300 mM (Fig. 4).

Solubilizing effects of organic acids on FePO4 coll in vitro
The dissolution of FePO 4coll increased as the organic acids: FePO 4coll molar ratio increased (Fig. 6). In the case of phytic acid, a molar ratio 1:1 (phytic acid: FePO 4coll ) released 32.5 μM P and 3.12 μM Fe, while the same molar ratios of citric and oxalic acids released 14.65 and 16.3 μM P and 4.5 and 5.4 μM Fe. At the higher molar ratio of 60:1 (organic acid: FePO 4coll ), the phytic, citric acid, and oxalic acid released 69.8, 45.2, and 54.3 μM P respectively and the corresponding Fe concentrations were 69.8, 36.82, and 25.6 μM (Fig. 5). Equilibrium speciation modeling (Visual MINTEQ) predicted that at lower ratios of (phytic acid: FePO 4coll ) more P and less Fe would be released into solution as a consequence of the predominance of the insoluble tetraFe-phytate complex, while at higher molar ratios, such as 60:1, the more soluble monoFe phytate complex would predominate (Table 1).
Phytic acid exudation P. vittata released phytic acid under P and Fe limited conditions (FePO 4coll , 0−300 mM) (Fig. 2C). Phytic acid is a major source of P in plants, and it is not usually secreted under environmental stress, e.g., Arabidopsis thaliana, Triticum aestivum, and Brassica species release malic and citric acids. Similarly, Oryza sativa releases only citric acid, and P. esinoformis releases oxalic and citric acids (Fu et al. (Fu et al.2017)). Plants commonly release organic acids because they release P through the desorption of P from metal oxide  (Jones (Lambers and Plaxton2015)). The rhizosphere pH depends upon the HCO 3 − , OH − , and H + (secreted by roots) which is determined by the complement of ions taken up by roots (Hinsinger et al. (Hinsinger et al.2003)). There is also a possibility that P. vittata may release phytate due to Fe stress by analogy with the secretion of phytosiderophores (Ahmed and Holmström (Ahmed and Holmström2014)) that complex Fe (Li et al. (Lin et al.2019)), facilitating transport to the roots (Ahmed and Holmström (Ahmed and Holmström2014)). That is, our study does not discriminate between deprivation of P or of Fe as the cause of phytate release by the roots.

Role of phytate in FePO 4coll dissolution
The secreted phytate in root exudates may dissolve FePO 4coll in the following ways (1) lower down pH or forming the insoluble complex with metals at the mineral surface which further change the mineral electric surface potential negative and affect the P binding to mineral and (2) by the effect of phosphorylated inositol rings on Fe (Wang and Lambers (Wang et al.2013); Lambers and Plaxton (Lei et al.2020);Violante and Caporale (Wan et al.2010)). At low molar concentration of phytic acid and Fe, water insoluble, tetraferric phytate complex formed, but at higher phytate concentrations, this was replaced by the water soluble monoferric phytate complex (Trela (Tu et al.2004); Nielsen et al. (Prokůpková et al.1996)).
In our experiment, the uptake of P (Fig. 3) shows that the released phytic acid dissolved the FePO 4coll , which is consistent with the results of in vitro chelating experiments at 1:1 molar ratio (phytic acid: FePO 4coll ) ( Table 2 and Fig. 6). The data showed that Fe was precipitated by phytic acid as Fephytate at a low concentration of phytic acid (Table 2). These data were confirmed by visual MINTEQ software (Table 1). At low concentration of phytic acid, the strong chelation may be due to the insoluble Fe-phytate, and as concentration increased, the Fe-phytate became water soluble. The Fe uptake may be because of plant-secreted siderophores phytosiderophores that make complex with Fe, then transported to plant across the root plasma membrane (Ahmed and Holmström (Ahmed and Holmström2014)). The transported Fe-siderophore complex reduced to Fe 2+ in the plant cell membrane and then uptake by cells. However, there is a need for more investigation for P. vittata and phytosiderophores' role in FePO 4coll dissolution. Our results agreed with Sega et al. ((Song et al.2019)) who claimed that root exudation may dissolute the FePO 4 nanoparticles which release P and Fe. However, they did not mention which organic acid could dissolve FePO 4 . However, in addition the molar ratio (P/Fe) FePO 4coll (product) was of 1:0.5, but according to Fig. 4, the molar ratio of P/Fe in solution was 1:2.6, 1:2.03, and 1:1.5 for treatments that initially contained 100, 200, and 300 mM of FePO 4coll . That is, at the highest FePO 4 concentration, the high P/Fe ratio in solution at the end of the experiment ratio showed that the phytate exudation dissolved the FePO 4coll .
Role of phytase and Fe-P-solubilizing activity on FePO 4coll The phytase enzyme also plays the dominant role in P and Fe released in our experiment. Under P starvation, plant cells promote the activity of phytases (Tu et al. (Vendelboe et al.2012)), and Fe adsorption from Fe-phytate is improved by adding phytase enzymes to oat-based beverages (Zhang et al. (Zhang et al.2021)). The phytase secretion in our experiment is independent of microbial activity under limiting P and Fe conditions, which is consistent with our finding of high concentrations of Fe and P in growth media and also in P. vittata tissue (Fig. 4). We also claimed that Fe-Psolubilizing activity in root exudates of P. vittata played an important role in FePO 4coll dissolution as evidenced by positive correlations between increased Fe-P-solubilizing activity and increased uptake of Fe and P by P. vittata (Fig. 5), which extends the findings of Subbarao et al. ((Thavarajah et al.2009)). This may explain the apparently conflicted environmental adaptation of excreting a P-rich substance (phytate) in response to a P-limitation; however, our results do not exclude the possibility that this response was elicited by an Felimitation, or a joint P and Fe limitation. Nonetheless, several studies have shown that P. vittata exudes phytic acid under stress caused by root exposure to minerals such as phosphate rock and FeAsO 4 (Fu et al. (Fu et al.2017); Liu et al. (Liu et al.2017)). Moreover, the hypothetical cost/benefit ration of such a response would be multiplied in soil where the released phytate would likely be constrained to the rhizosphere rather than diluted in a hydroponic medium.

Conclusion
The combined deprivation of P and Fe caused P. vittata to release phytic, which efficiently dissolved FePO 4coll releasing P and Fe that were taken up by the plant. However, whether deprivation of one or both of P and Fe is sufficient to cause phytate exudation remains unknown. The phytase activity in root exudates may have facilitated the uptake of the dissolved Fe and P, since FePO 4coll increased both plant biomass and the internal concentrations of P and Fe. Fe-P-solubilizing activity is not unique to the root exudates of P. vittata; however, the release of phytic acid as a stress response merits further investigation.