An underground tale: contribution of microbial activity to plant iron acquisition via ecological processes

Abstract

Background

Iron (Fe) deficiency in crops is a worldwide agricultural problem. Plants have evolved several strategies to enhance Fe acquisition, but increasing evidence has shown that the intrinsic plant-based strategies alone are insufficient to avoid Fe deficiency in Fe-limited soils. Soil micro-organisms also play a critical role in plant Fe acquisition; however, the mechanisms behind their promotion of Fe acquisition remain largely unknown.

Scope

This review focuses on the possible mechanisms underlying the promotion of plant Fe acquisition by soil micro-organisms.

Conclusions

Fe-deficiency-induced root exudates alter the microbial community in the rhizosphere by modifying the physicochemical properties of soil, and/or by their antimicrobial and/or growth-promoting effects. The altered microbial community may in turn benefit plant Fe acquisition via production of siderophores and protons, both of which improve Fe bioavailability in soil, and via hormone generation that triggers the enhancement of Fe uptake capacity in plants. In addition, symbiotic interactions between micro-organisms and host plants could also enhance plant Fe acquisition, possibly including: rhizobium nodulation enhancing plant Fe uptake capacity and mycorrhizal fungal infection enhancing root length and the nutrient acquisition area of the root system, as well as increasing the production of Fe³⁺ chelators and protons.

INTRODUCTION

Iron (Fe) is an essential nutrient for plants, and it serves as a cofactor for a wide variety of cellular processes, such as oxygen transport, cellular respiration, chlorophyll biosynthesis, thylakoid biogenesis and chloroplast development (Nishio et al., 1985; Kobayashi and Nishizawa, 2012). However, the Fe bioavailability in well-aerated soils is often severely limited, particularly in calcareous soils, which occupy 30 % of the Earth's surface. Hence, Fe-deficiency-induced chlorosis is a serious problem leading to yield loss and reduced quality in agricultural production (Kim and Guerinot, 2007; Zheng, 2010). Fe deficiency in plants is also closely related to the prevalence of Fe-deficiency-induced anaemia in humans (Murgia et al., 2012). Plants have evolved at least two mechanisms favouring efficient acquisition of Fe. Strategy I, which occurs in non-graminaceous plants, relies on acidification of the rhizosphere to increase the solubility of ferric Fe compounds through proton extrusion, trans-plasma membrane electron transfer to reduce Fe to its more soluble ferrous form via ferric chelate reductase (FRO2) and transport of Fe into root cells by iron-regulated transporter 1 (IRT1) (Marschner et al., 1995; Santi and Schmidt, 2009; Ivanov et al., 2012; Kobayashi and Nishizawa, 2012). Strategy II, which is utilized by the Gramineae, relies on extrusion of mugineic acid family phytosiderophores (MAs) via efflux transporter of MAs (e.g. TOM1) to solubilize Fe in the rhizosphere, and subsequent transport of the Fe(III)–phytosiderophore complex across the plasma membrane of the root epidermal cell via yellow stripe1 transporter 1 (YS1) (Curie et al., 2001, 2009; Nozoye et al., 2011; Kobayashi and Nishizawa, 2012).

These strategies have been thought to ensure normal growth for many so-called ‘Fe-efficient’ plants under Fe-limited conditions. In the last decade, however, several lines of evidence have shown that these strategies alone are insufficient to prevent plants from suffering Fe deficiency in Fe-limited soils. For example, sunflower (Helianthus annuus) plants grown in sterile soil show severe chlorosis, poor growth and lower tissue Fe levels than those grown in non-sterile soil (Masalha et al., 2000). Similarly, Fe acquisition and growth of rape and red clover plants also significantly decrease when the plants are grown in sterile soils, but normal growth can be quickly restored by adding Fe–EDDHA to the sterile soil or spraying Fe–EDTA on the leaves (Rroco et al., 2003; Jin et al., 2006). These results provide evidence that soil microbial activity probably plays a critical role in plant Fe acquisition. Further supporting this, de Santiago et al. (2009, 2011, 2013) observed that Fe acquisition in wheat, white lupin and cucumber plants increases when their calcareous media are inoculated with the soil fungus Trichoderma asperellum strain T34. How soil micro-organisms promote plant Fe acquisition is still largely unknown. Nevertheless, researchers have made great efforts to uncover this interesting and important underground mechanism in recent decades and have obtained many valuable clues. Based on these clues, this present prospective review discusses the possible mechanisms for soil micro-organism promotion of plant Fe acquisition.

MICROBIAL COMMUNITY STRUCTURE IN THE RHIZOSPHERE DEPENDS ON THE FE STATUS OF PLANTS

Soils have highly distinct microbial communities, which are thought to result from many different selection factors, including the texture, nutrient and organic matter content and pH of the soil (Berg and Smalla, 2009; Liang et al., 2012). Environmental factors such as climate, soil aeration and moisture content could also affect the make up of microbial communities in soils (Gelsomino et al., 1999; Carelli et al., 2000; Marschner et al., 2004). The rhizosphere is regarded as a component of the soil system, but it can be markedly different from the bulk soil in terms of the above-mentioned factors, because most of these factors vary due to the activity of plant roots, such as the consumption of nutrients and exudation of protons and organic compounds.

Iron serves as a cofactor for a wide variety of cellular processes in plants, and deficiency of Fe in plants affects root behaviour, particularly the exudation of organic compounds, which are referred to as root exudates (Jin et al., 2006, 2007; Carvalhais et al., 2011). In addition to the exudation of phytosiderophores by Strategy II plants, Fe deficiency often strongly induces the roots of Strategy I plants to release reductants and chelators including phenolic compounds, flavins and organic acids (Curie and Briat, 2003; Hell and Stephan, 2003; Jin et al., 2006, 2007, 2010; Kobayashi and Nishizawa, 2012). As a consequence, Fe deficiency in plants can be predicted to alter the chemical and physical properties of rhizospheric soils, in turn affecting which micro-organisms are found in the rhizosphere and their relative abundance. Indeed, several studies have provided evidence for this. Using PCR–DGGE (polymerase chain reaction–denaturing gradient gel electrophoresis) of rDNA, a culture-independent method, Yang and Crowley (2000) demonstrated that the structure of the microbial community in the barley rhizosphere varies with the plant's Fe nutritional status. In the rhizosphere of a transgenic tobacco plant overaccumulating Fe, Robin et al. (2006, 2007) observed a decrease in Fe availability and shifts in the structure of the microbial community. Recently, we grew red clover plants in an Fe-limited calcareous soil and observed a clearly different PCR–DGGE profile of rhizospheric microbial 16S rDNA, as compared with that from plants grown in the same soil but with foliar application of 100 µm Fe–EDTA solution (Fig. 1).

Fig. 1.

Effect of Fe status on the microbial community in soil. (A) Microbial community 16S rDNA fingerprints of bacteria from rhizosphere soil of Fe-stressed and Fe-sufficient red clover as determined by PCR–DGGE. (B) Line image profiles generated by image analysis of gels shown in A. For the Fe-sufficient treatment, the leaves of red clover plants were sprayed with 100 µM Fe–EDTA solution every other day, while for the Fe-stressed treatment, the leaves were sprayed with deionized water.

The above PCR–DGGE studies demonstrate that the structure of microbial communities in the rhizospheres of Fe-deficient plants can be quite different from that of Fe-sufficient plants, but the exact mechanism underlying this difference remains an open question. To clarify whether the alteration of microbial communities is associated with root exudates, we analysed root exudates from Fe-deficient red clover plants and found that phenolics represent the major component (Jin et al., 2007). Plant phenolic compounds have both antimicrobial and growth-promoting effects (Blum et al., 2000; Dakora and Phillips, 2002; Jin et al., 2006; Medina et al., 2006; Cicerale et al., 2012; Sugiyama and Yazaki, 2012). When a calcareous soil solution was incubated on agar plates containing phenolic root exudates from Fe-deficient red clover, only a few microbial species thrived while growth of the rest was inhibited (Fig. 2; Jin et al., 2006, 2008). Similarly, Masaoka et al. (1997) observed that phenolic exudates collected from Fe-deficient alfalfa plants inhibit the growth of Fusarium oxysporum f. sp. phaseoli, but promote the growth of Rhizobium meliloti. Further, we recently found that the siderophore secretion profile of micro-organisms from soil containing added phenolic root exudates is similar to that found in the rhizosphere soil of Fe-deficient plants (Jin et al., 2010). These studies attest to a possible relationship between Fe-deficiency-induced exudation of phenolics and changes in the composition of the microbial community in the rhizosphere. However, more direct evidence is still necessary to confirm such a relationship. In addition, as mentioned previously, Fe deficiency also induces exudation of phytosiderophores, flavins or organic acids, depending on the plant species (Kobayashi and Nishizawa, 2012), but to date there is no information on whether and how these root exudates link to changes in rhizosphere microbial communities.

Fig. 2.

Effect of the phenolics from Fe-deficient root exudates on the growth of soil micro-organisms. A 100 µL aliquot of soil solution (10⁻⁴ dilution) was spread on broth peptone agar plates with or without phenolics (0·6 µmol mL⁻¹), and then incubated at 30 °C for 2 d. The phenolics were collected from root exudates of Fe-deficient red clover.

The previously mentioned sterilization experiments and PCR–DGGE analyses provide evidence that soil micro-organisms play an important role in plant Fe acquisition, and micro-organism community structure in the rhizosphere changes along with the Fe status of plants, prompting the question of whether the plant-induced rhizosphere behaviour is beneficial for the plant itself. The ratio of siderophore-producing micro-organisms isolated from soil solution incubated with phenolic root exudates of Fe-deficient red clover is much higher than that isolated from phenolic-free control soil solution (Jin et al., 2006, 2008). Furthermore, micro-organisms from the rhizosphere of Fe-deficient plants have a greater siderophore secretion capacity than those from Fe-sufficient plants (Jin et al., 2010). Due to their strong Fe-chelating ability, secretion of siderophores is expected to increase Fe solubility in rhizosphere soil (Jin et al., 2008). The detailed functions of siderophore compounds will be discussed below. Together, these results suggest that the plant Fe-deficiency-induced changes in microbial community structure in the rhizosphere should favour the Fe uptake of the plant itself. Nevertheless, more direct evidence is also necessary to confirm this.

MECHANISMS OF MICROBIAL ACTIVITY IN IMPROVING THE FE NUTRITION OF PLANTS

Siderophore production

Siderophores are low molecular weight organic chelators with very high and specific affinity for Fe(III). In aerobic calcareous soil, many micro-organisms synthesize and release siderophores to overcome low Fe availability resulting from the minimal solubility of Fe hydroxides (Schalk et al., 2011). Siderophores increase Fe solubility by chelation to form siderophore–Fe complexes (Lemanceau et al., 2009; Saha et al., 2012). There are almost 500 compounds identified as siderophores, which can be classified as hydroxamate, catecholate and/or hydoxycarboxylic acid siderophores based on their ligand architecture (Boukhalfa and Crumbliss, 2002; Bonnefoy and Holmes, 2012). Figure 3 shows some representative siderophore structures with hydroxamic acid and catechol moieties. Because of their solubilizing effect on Fe hydroxides, the production of siderophores in the rhizosphere has been proposed to be the key microbial activity that benefits plant Fe acquisition (Masalha et al. 2000; Yehuda et al., 2000; Jin et al., 2006, 2010; Desai and Archana, 2011; Hayat et al., 2012). In a soil cultivation study, Carrillo-Castañeda et al. (2005) observed that Fe acquisition of common bean plants was increased by siderophore-producing micro-organisms. Furthermore, using hydroponic culture, various microbial Fe–siderophore chelates have been demonstrated to serve as good sources of Fe for plants, for example Fe–ferrioxamine for oat (Crowley et al., 1991), Fe–pyoverdine for arabidopsis (Vansuyt et al., 2007), Fe–erobactin for soybean and oat (Chen et al., 1998, 2000), and Fe–rhizoferrin for tomato, barley and corn (Yehuda et al., 2000). In addition, the siderophore secreted by a Pseudomonas isolated from a soil solution incubated with phenolic root exudates of Fe-deficient red clover plants can dissolve Fe from an insoluble Fe source, and the dissolved Fe can be easily utilized by plants (Jin et al., 2010). Interestingly, the Fe levels in plants supplied with Fe in the form of Fe–siderophore are much higher than those in plants supplied with the same Fe concentration in the form of Fe–EDTA (Jin et al., 2010). A similar phenomenon was observed by Vansuyt et al. (2007), wherein the Fe concentration in arabidopsis plants fed with Fe–pyoverdine was higher than in those fed with Fe–EDTA. These data indicate that Fe–siderophores may be incorporated into the roots of Strategy I plants more efficiently than other Fe sources. The redox potential E(mV) of siderophore–Fe³⁺ chelates is generally low, within the range of –330 to –750 mV (Boukhalfa and Crumbliss, 2002). In plants, the redox potential of most biological reductants is higher than this; for example, E = –320 mV for NADPH, which is suggested to be the electron donor for Fe(III) reduction by ferric chelate reductase in Strategy I plants (Robinson et al., 1999). Bienfait et al. (1983) demonstrated that there is little reduction of Fe(III)–ferrioxamine B and the Fe(III)–aerobactin complexes. Therefore, it seems that Fe(III)–siderophore chelates are probably acquired via a reduction-independent pathway in Strategy I plants. In Strategy II plants, although Fe acquisition does not depend on a reduction mechanism, the YS1 transporter works efficiently for Fe(III)–phytosiderophore but not for other Fe(III) chelates such as microbial Fe(III)–siderophores (Römheld and Marschner, 1986; Curie et al., 2001). Accordingly, acquisition of Fe(III)–siderophore chelates by the roots of Strategy II plants may also‡ be achieved via an as yet unidentified plasmalemma pathway. It is possible that endocytosis is involved in Fe–siderophore incorporation into root cells (Lemanceau et al., 2009).

Fig. 3.

Representative siderophore structures with hydroxamic acid and catechol moieties (Boukhalfa and Crumbliss, 2002). Ferrioxamine E, ferrioxamine B, rhodotorulic acid and alcaligin are shown for the hydroxamic acid moiety, and entrobactin represents the catechol Fe(III)-chelating moiety.

Generation of protons

Theoretically, the solubility of Fe decreases up to 1000-fold for each unit increase in pH (Guerinot and Yi, 1994). Hence, acidification of the rhizosphere can have an enormous impact on Fe solubility in the vicinity of the roots. The two most frequently used nitrogen fertilizers in crop production are urea and ammonium. Upon the application of urea to soil, it is rapidly hydrolysed to ammonia by urease enzymes (Singh et al., 2013). In well-aerated soils, ammonia and ammonium are converted to nitrate via nitrification, which is accompanied by release of protons (H⁺) (Van Miegroet and Cole, 1984). It should be noted that uptake of nitrogen as ammonium in plants also releases protons (Xu et al., 2012). Therefore, both nitrification and ammonium uptake can lead to soil acidification. However, the key process causing soil acidification due to ammonium-based fertilizers is nitrification, which is catalysed by ammonia-oxidizing prokaryotes (Jetten et al., 1997). Accordingly, in calcareous soils, the micro-organisms associated with nitrification can be expected to increase Fe solubility via H⁺ generation, and thereby facilitate plant Fe acquisition. Supporting this conclusion, Malhi et al. (1998) found in a 27 year field experiment that both soil acidification and the DTPA-extractable Fe concentration increase significantly with the rate of ammonium nitrate fertilization. Nevertheless, a contribution of plant ammonium uptake to the increase of extractable Fe in soil cannot be excluded.

In calcareous soils, many phosphate-solubilizing bacteria (PSB) also excrete H⁺. For instance, both Penicillium bilaji and Penicillium cf. fuscum significantly lower the soil pH in the presence of ammonium (Asea et al., 1988), and similar results are observed for Penicillium aurantiogriseum and Penicillium simplicissimum (Illmer and Schinner, 1995; Illmer et al., 1995; Ahuja et al., 2007). Thus, in calcareous soils, the presence of H⁺-excreting PSB may also help to increase Fe solubility. However, little information is available regarding the effect of H⁺-excreting PSB on plant Fe uptake. Besides PSB, mycorrhizae also excrete H⁺. The role of mycorrhizae in plant Fe acquisition will be discussed below.

Production of hormonal compounds

Enhancing Fe-deficiency-inducible responses can be predicted to increase plant acquisition of Fe from Fe-limited soils. Indeed, Zhang et al. (2009) reported that the soil bacterium Bacillus subtilis GB03 could enhance Fe acquisition of arabidopsis plants by activating the Fe-deficiency-inducible responses, suggesting that soil micro-organisms could regulate plant Fe acquisition via a signalling process. In the last decade, plant physiologists have made a great effort to uncover the signals responsible for triggering Fe deficiency responses in plant roots, and several hormonal compounds have been identified as signalling elements (Hindt and Guerinot, 2012; Ivanov et al., 2012; Kobayashi and Nishizawa, 2012). These phytohormones include auxin (Jin et al., 2008; Chen et al., 2010), nitric oxide (NO) (Graziano and Lamattina, 2007), ethylene (Garcia et al., 2011), cytokinin (Séguéla et al., 2008) and brassinosteroids (Wang et al., 2012). Among these, auxin, NO and ethylene are particularly interesting with regard to revealing potential interactions between soil micro-organisms and Fe uptake of plants since these compounds can be generated by soil micro-organisms.

Production of auxin-like compounds

Indole-3-acetic acid (IAA) is the most active auxin in plants, and it is known to stimulate both rapid (e.g. increases in cell elongation) and long-term (e.g. cell division and differentiation) responses (Zhao et al., 2010). The majority of micro-organisms isolated from rhizosphere soil can produce IAA (Patten et al., 2012). In addition to IAA, micro-organisms such as Paenibacillus polymyxa, Azospirilla and Klebsiella pneumoniae also release other auxin-like compounds, such as indole-3-butyric acid (IBA), indole-3-ethanol (TOL), indole-3-carboxylic acid and indole-3-aldehyde (Hayat et al., 2010). Some examples of soil micro-organisms known to produce auxin compounds are listed in Table 1. The ratio of auxin-producing micro-organisms in soil solution incubated with phenolic root exudates of Fe-deficient red clover is higher than that in phenolic-free control soil solution (Jin et al., 2006, 2008), suggesting that Fe deficiency in plants may lead to beneficial effects on the growth of auxin-producing micro-organisms in the rhizosphere. Nevertheless, this needs to be confirmed by in situ comparisons between rhizosphere soils of Fe-sufficient and Fe-deficient plants.

Table 1.

Production of auxin-like compounds by soil micro-organisms

Micro-organisms	Auxin compounds	References
Paenibacillus polymyxa	Indole-3-acetic acid; indole-3-butyric acid; indole-3-ethanol; indole-3-carboxylic acid; indole-3-aldehyde	Lebuhn et al. (1997)
Azospirillum brasilense	Indole-3-acetic acid; indole-3-ethanol	El-Khawas and Adachi (1999)
Klebsiella pneumoniae	Indole-3-acetic acid; indole-3-ethanol; indole-3-pyruvic acid; indole-3-acetaldehyde	El-Khawas and Adachi (1999)
Rhizobium leguminosarum	Indole-3-acetic acid	Dazzo et al. (2000); Bhattacharjee et al. (2012)
Azotobacter sp.	Indole-3-acetic acid	Zahir et al. (2000); Ahmad et al. (2005)
Rhizobacteria (unidentified)	Indole-3-acetic acid	Asghar et al. (2002)
Pseudomonas fluorescens	Indole-3-acetic acid	Dey et al. (2004)
Rhizobacteria (unidentified)	Indole-3-acetic acid	Khalid et al. (2004)
Azospirillum brasilenseA3, A4, A7, A10, CDJA; Bacillus circulans P2; Bacillus sp. P3; Bacillus magaterium P5; Bacillus. Sp. Psd7; Streptomyces anthocysnicus; Pseudomonas aeruginosa Psd5; Pseudomonas pieketti Psd6; Pseudomonas fluorescens MTCC103	Indole-3-acetic acid	Thakuria et al. (2004)
Pseudomonas sp.	Indole-3-acetic acid	Roesti et al. (2006)
Soil bacterial isolates	Auxin (unidentified)	Jin et al. (2006)
Rhizobium leguminosarum b. Trifolii ACCC18002	Indole-3-acetic acid	Jin et al. (2006)
Bacillus cereus RC18; Bacillus licheniformis RC08; Bacillus megaterium RC07; Bacillus subtilis RC11; Bacillus. OSU-142; Bacillus M-13; Pseudomonas putida RC06; Paenibacillus polymyxa RC05 and RC14	Indole-3-acetic acid	Çakmakçi et al. (2007)
Mesorhizobium loti MP6	Indole-3-acetic acid	Chandra et al. (2007)
Pseudomonas tolaasii ACC23; Pseudomonas fluorescens ACC9; Alcaligenes sp. ZN4; Mycobacterium sp. ACC14	Indole-3-acetic acid	Dell'Amico et al. (2008)
Bacillus sp; Paenibacillus sp.	Indole-3-acetic acid	Beneduzi et al. (2008)
Enterobacter aerogenes sp. NII-0907; Enterobacter aerogenes sp. NII-0929; Enterobacter cloacae subsp. cloacae sp. NII-0931; Enterobacter asburiae sp. NII-0934	Indole-3-acetic acid	Deepa et al. (2010)
Streptomyces strains C	Indolyl-3-acetic acid	Sadeghi et al. (2012)
Enterobacter; Klebsiella	Indolyl-3-acetic acid	de Santi Ferrara et al. (2013)

Auxin has been demonstrated to be an important chemical signal enhancing Fe-deficiency-inducible responses. Exogenous addition of synthetic auxin, either IAA or α-naphthaleneacetic acid, enhances Fe-deficiency-induced reduction of ferric Fe, expression of FRO2 and IRT1, and development of root hairs and lateral roots to increase the surface area for Fe uptake (Jin et al., 2008; Chen et al., 2010; Wu et al., 2012). Accordingly, production of auxin-like compounds by soil micro-organisms can be considered a beneficial microbial activity for plant Fe uptake under Fe-limited conditions. In support of this, auxins produced by a microbe that was isolated from soil mixed with phenolics secreted from Fe-deficient red clover plants markedly enhance the activity of ferric chelate reductase in roots of Fe-deficient plants (Jin et al., 2006).

Production of NO gas

Nitric oxide is a small, highly reactive and membrane-permeable gaseous molecule. In the last decade, NO has been recognized as a plant hormone due to its signalling functions in numerous cellular and physiological processes (Shapiro, 2005; Palmieri et al., 2008; Baudoin, 2011). In addition to eukaryotic organisms such as plants, soil micro-organisms also produce NO. The processes of nitrification and denitrification have been identified as the most important sources of NO generation in soil micro-organisms (Gasche and Papen, 2002; Kitzler et al., 2006; Kim et al., 2012). The critical factor affecting generation of NO by soil micro-organisms (Davidson et al., 2000; Kim et al., 2012) is the soil water content, since it controls the oxygen content of the soil, which in turn determines whether nitrification or denitrification is the dominant process in the soil. Therefore, in dry, well-aerated soils, the oxidative process of nitrification dominates, and the more oxidized gas, NO, is the most common nitrogen oxide produced by soil micro-organisms (Davidson et al., 2000; Stange et al., 2013). NO production in soils also depends on soil N availability (Kitzler et al., 2006). The mean level of NO can reach 250 ppb in forest soils (Rudolph and Conrad, 1996), whereas it is probably much higher in agricultural soils due to N fertilizer input during crop cultivation. Like auxin, NO acts as an enhancer signal in the regulation of Fe-deficiency-inducible responses. Exogenous addition of the NO donor S-nitrosoglutathione (GNSO) promotes the Fe-deficiency-induced reduction of ferric Fe, expression of FRO2 and IRT1, and development of root hairs and lateral roots (Graziano and Lamattina, 2007; Chen et al., 2010; Jin et al., 2011; Meiser et al., 2011; Zhang et al., 2012). Therefore, it is reasonable to assume that NO generation by soil micro-organisms may enhance Fe acquisition of plants grown in Fe-limited soils. Experimental evidence from soil culture is still necessary to test this, however.

Production of ethylene gas

Ethylene is also a small, membrane-permeable gaseous signal molecule in plants. Ethylene concentrations as low as 10 nL L⁻¹ can evoke plant responses. Interestingly, ethylene is a common constituent of the soil atmosphere and may accumulate to 10 µL L⁻¹ in soil when conditions favour ethylene production or inhibit ethylene degradation (Smith, 1976). Therefore the ethylene concentration in most soil conditions is high enough to be biologically active for plants. Up to now, all evidence points to micro-organisms as the major source of ethylene in soil (Jäckel et al., 2004), and bacteria, fungi and yeast may all produce ethylene. Accumulation of ethylene in the soil greatly depends on the oxygen level in the soil, with low oxygen concentrations favouring ethylene accumulation (Xu and Inubushi, 2007). Thus, aerobic conditions in Fe-limited soil seem to be disadvantageous for ethylene generation by micro-organisms. However, López-Millán et al. (2000) reported that the oxygen consumption rate in root tips of Fe-deficient plants is about 3-fold more than that of Fe-sufficient plants. Therefore, it is reasonable to assume that there is less oxygen in the rhizosphere soil of Fe-deficient plants; in other words, the growth of Fe-deficient plants may favour ethylene generation in rhizosphere soil. Moreover, a study by Arshad and Frankenberger (1991) revealed that ethylene generation is inhibited in the presence of Fe(III) >10 mg kg⁻¹ soil, which suggests that the low level of available Fe in Fe-limited soils should also favour ethylene production. Like auxin and NO, ethylene acts as an enhancer of Fe-deficiency-inducible responses. Exogenous addition of the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) significantly increases the Fe-deficiency-induced reduction of ferric Fe, expression of FRO2 and IRT1, and development of root hairs (García et al., 2011; Wu et al., 2011). Therefore, generation of ethylene may be another microbial activity that improves plant Fe acquisition in Fe-limited soils.

Micro-organisms enhance plant Fe uptake via symbiotic interactions with plants

Rhizobium–legume symbiosis

Rhizobium nodulation is ubiquitous in leguminous plants. The most important function of nodules is symbiotic N₂ fixation, a process in which Fe-containing proteins play very important roles (Terpolilli et al., 2012). Nodulated legumes thus have an increased need for Fe as compared with non-nodulated plants. In order to meet this increased Fe demand, legumes have developed a mechanism to enhance Fe-deficiency-induced responses in the roots upon nodulation. For example, secretion of both protons and reductants in Fe-deficient roots of peanut plants is significantly increased by rhizobium nodulation (Terry et al., 1988). Soerensen et al. (1988) obtained similar results in soybean plants. Rhizobium nodulation also clearly enhances the activity of ferric chelate reductase in roots of Fe-deficient red clover plants (Deryl and Skorupska, 1992; Jin et al., 2007). In a split-root experiment, Soerensen et al. (1989) found more Fe(III)–EDTA reduction activity on the Bradyrhizobium-infected side of soybean roots than on the non-inoculated side, and the reduction activity on the infected side was greater on the root below the nodule clusters. Furthermore, in a recent study by Slatni et al. (2012), overaccumulation of H⁺-ATPase and IRT1 proteins was observed especially around the cortex cells of nodules from Fe-deficient common bean plants. These results suggest that the positive effects of rhizobium nodulation on Fe-deficiency-induced responses could be locally controlled, probably by a signal derived from nodules.

The question arises of whether the enhancement of Fe-deficiency-induced responses in nodulated roots merely meets the increased Fe requirement of the nodules, or systemically improves the Fe nutrition of whole plants. Using the radioactive isotope ⁵⁵Fe, Deryl and Skorupska (1992) found that rhizobium nodulation promotes Fe transport from roots to shoots in red clover. In addition, we observed that Fe acquisition and growth of red clover significantly decrease when the plants are grown in sterilized calcareous soil, but are restored by nodulation with Rhizobium leguminosarum bv. trifolii ACCC18002 (Jin et al., 2006). Most recently, Mishra et al. (2011, 2012) demonstrated that nodulation of roots by R. leguminosarum-PR1 elevates the Fe content in pea (Pisum sativum L.) and lentil (Lens culinaris L.) plants. All of these results suggest that rhizobium nodulation can systemically improve the Fe nutrition of plants through enhancing Fe-deficiency-induced responses. However, the mechanism through which rhizobium nodulation enhances Fe-deficiency-induced responses still remains largely unknown. It appears to occur independently of N₂ fixation, as nodules that are null for N₂ fixation still enhance the activity of ferric chelate reductase in roots (Deryl and Skorupska, 1992).

It is interesting to note that leguminous plants secrete more phenolics under Fe-deficient conditions. Several phenolic compounds can stimulate the growth of nodulating rhizobia, and also act as signal molecules to induce rhizobial nod gene expression, both of which enhance rhizobium nodulation (Hassan and Mathesius, 2012). Therefore, the Fe-deficiency-induced secretion of phenolics may favour the growth of nodulating rhizobia in leguminous plants. Indeed, phenolics secreted by Fe-deficient alfalfa significantly stimulate the growth of Rhizobium meliloti (Masaoka et al., 1997). Thus, phenolics secreted by Fe-deficient legumes may improve Fe acquisition of the legumes themselves via a mechanism involving phenolic-induced rhizobium nodulation.

It is also worth noting that the nodulation of legume roots can be improved by inoculation of siderophore-producing rhizobial strains under Fe-deficient conditions (e.g. Tang et al., 1991; Arora et al., 2001). The underlying mechanism may be that siderophore production facilitates the Fe acquisition of nodules, and ultimately promotes the synthesis of Fe-containing proteins, which figure prominently in nodules (Terpolilli et al., 2012). Considering the resulting availability of Fe–siderophores for plant growth and the effect of rhizobium nodulation on Fe-deficiency-induced responses of plant roots, inoculation with siderophore-producing rhizobia can be predicted to enhance Fe nutrition of leguminous plants more than inoculation with rhizobia that do not produce siderophores.

Mycorrhizal symbiosis

The mycorrhizal symbioses formed between plant roots and mycorrhizal fungi are of great interest to ecologists and plant nutritionists because of their potential influence on ecosystem processes, their role in determining plant diversity in natural communities and the ability of mycorrhizae to enhance nutrient acquisition of host roots (Wagg et al., 2011; Smith and Smith, 2012). The majority of plants growing under natural conditions have associated mycorrhizae (Smith and Reed, 1997; Harrison, 2012). It is generally accepted that morphological effects are the most important mechanism by which mycorrhizae increase nutrient acquisition of plants (Clark and Zeto, 2000). For example, in ectomycorrhiza-infected pine seedlings, the total hypha length is >100-fold greater than the total root length (Rousseau et al., 1994; Smith and Reed, 1997), and similar findings have been reported for ectomycorrhiza-infected willow seedlings (Jones et al., 1990). In addition to increases in length, the available surface area for absorbing nutrients is also increased remarkably by the formation of mycorrhizal hyphae (Smith and Reed, 1997). Accordingly, although plants with mycorrhizal roots have access to pools of soil nutrients similar to those of non-mycorrhizal plants, the hyphal network allows considerably larger volumes of soil to be explored. These morphological effects should theoretically enhance plant acquisition of all essential mineral elements, including Fe.

Mycorrhizae also excrete H⁺ and low molecular weight organic chelating compounds, such as citric acid, oxalic acid and siderophores (Li et al., 1991; Winkelmann, 2007; Bharadwaj et al., 2012), which should facilitate mobilization of sparingly available Fe in the rhizosphere soil, and therefore also theoretically promote plant Fe acquisition.

Several studies have investigated the effect of mycorrhizal symbiosis on plant Fe uptake. Some of these studies, however, found that acquisition of Fe was enhanced in mycorrhizae-infected plants (Cress et al., 1986; Raju et al., 1990; Treeby, 1992; Medeiros et al., 1993; Clark and Zeto, 1996; El-Ghandour et al., 1996; Al-Karaki and Clark, 1998; Al-Karaki et al., 1998; Caris et al., 1998; Purakayastha et al., 1998; Pirazzi et al, 1999; Wang et al., 2007; Suri et al., 2011; Amanullah et al., 2012; Labidi et al., 2012), whereas others found it to be decreased (Pacovsky and Fuller, 1988; Kothari et al., 1990, 1991; Clark et al., 1999). This discrepancy is probably due to differences in soil properties and growth environments for plant culture among these studies. For instance, Treeby (1992) and Medeiros et al. (1993, 1994) found that mycorrhizal plants grown at low pH had higher Fe acquisition than those grown at high pH. In addition, Rajue (1990) observed that Glomus macrocarpum (a species of vesicular-arbuscular mycorrhizal fungi)-colonized plants grown at 25 or 30 °C had 10-fold more Fe than those grown at 20 °C. Furthermore, both the plant species and the mycorrhiza species affect the contribution of mycorrhizal symbiosis to plant Fe uptake (Clark and Zeto, 2000). Overall, the majority of the above studies provide evidence that mycorrhizal symbiosis can beneficially affect plant Fe acquisition. However, further investigation is necessary to explore the effects of soil/growth conditions so that the contradictory results can be reconciled.

CONCLUSIONS AND OUTLOOK

Based on the results discussed herein, we have developed a model for how rhizospheric micro-organisms enhance plant Fe acquisition (Fig. 4). When the plant suffers Fe deficiency stress, organic exudates are secreted from roots and accumulate in the rhizosphere. Then, the root exudates selectively alter the microbial community in the rhizosphere by altering the physical–chemical properties of soil, and/or via their antimicrobial and/or growth-beneficial effects. Favoured micro-organisms then produce siderophores and protons, both of which increase Fe bioavailability, and hormones that enhance plant Fe-deficiency-inducible responses. In addition, plant Fe acquisition could also be promoted by symbioses between some micro-organisms and plants, possibly including: rhizobium nodulation (enhancing plant Fe-deficiency-inducible responses) and mycorrhizal symbiosis (increasing root length and nutrient acquisition area of the root system and increasing Fe availability).

Fig. 4.

Proposed model for microbially enhanced plant Fe acquisition. The ‘M’ in the small circle denotes the original microbial community in the rhizosphere. The ‘NM’ in the small circle denotes the newly formed microbial community.

Although studies have indicated that plant root exudates may be responsible for the alterations of microbial community structure in the rhizosphere that occur in response to plant Fe status, the direct evidence for this is limited. This is probably because the components of root exudates and the composition of soil micro-organism communities are both quite complicated. However, recent advances in chemical analysis and molecular biotechnologies may help to elucidate the above relationship. For example, secondary ion mass spectrometry (SIMS) can be used to trace the uptake of ¹³C and ¹⁵N from labelled compounds by cells (Cliff et al., 2002; Herrman et al., 2007; Clode et al., 2009). The SIMS technology has been used successfully to visualize and measure nutrient absorption by roots and rhizosphere micro-organisms at the nicromolar and nanomolar scale (Herrman et al., 2007; Clode et al., 2009). Hence, this technology can be used to study the root exudate-mediated root–microbe interactions between individual root cells and individual microbes in situ. In addition, the recent development of high-throughput tag-encoded FLX amplicon pyrosequencing (Acosta-Martinez et al., 2008; Su et al., 2012) may also facilitate diversity analyses and species identification of micro-organisms inhabiting the rhizospheric soil. This method generates massive amounts of 16S rRNA gene sequences from hitherto uncultured bacterial groups (Uroz, et al., 2010). For a more in-depth discussion of the technologies that could be used to study root–microbe interactions, the reader is referred to a recent review by Marschner et al. (2011). Besides the above technologies, screening plant mutants failing to exude specific organic compounds may help to clarify which kinds of root exudates are responsible for the Fe-deficiency-induced alteration of microbial community structure in the rhizosphere.

Soil microbial activity clearly plays a central role in plant Fe uptake, but the mechanism underlying this process remains largely unknown. It is likely to involve siderophore and proton production, both of which improve Fe bioavailability in the rhizosphere, and hormone generation, which triggers plant development of increased Fe uptake capacity (Fig. 4). Screening micro-organism mutants that fail to produce siderophores or hormones, and analysis of the effects of such mutants on plant Fe uptake in soil cultivation systems, will help to test this directly. In addition, the mechanism of Fe–siderophore acquisition by plants and their subsequent utilization within plants also remains unknown. Iron–siderophores may be directly acquired by root cells via an unidentified component that is independent of the IRT1 transporters in Strategy I plants, and also independent of the YS1 transporter in Strategy II plants. Screening mutant plants by using Fe–siderophores as the sole Fe source may help to uncover the unidentified component(s) used for Fe–siderophore uptake. In addition, the SIMS technology mentioned above may facilitate elucidation of the mechanism of Fe–siderophore uptake by plant roots.

Rhizobial and mycorrhizal symbioses might also improve the Fe nutrition of host plants (Fig. 4). However, the beneficial effects of mycorrhizal symbiosis on plant Fe nutrition may depend on the soil properties and growth environment, as well as plant and mycorrhiza species. Therefore, it is still necessary to determine what soil proprieties, growth environments, plant species and mycorrhiza species co-operate to improve plant Fe nutrition, information that may help to promote crop production in calcareous soils.