Recent insights into iron homeostasis and their application in graminaceous crops

Abstract

Higher plants utilize various mechanisms to maintain iron homeostasis. To acquire sparingly soluble iron from the rhizosphere, graminaceous plants synthesize natural iron (III) chelators known as mugineic acid family phytosiderophores (MAs). Recent research has uncovered various genes involved in iron uptake and translocation, as well as factors regulating the expression of these genes, especially in rice. Manipulation of these molecular components is used to produce transgenic crops with enhanced tolerance to iron deficiency, or with a high seed iron content. Since iron homeostasis is closely linked to that of other mineral elements, an understanding of this phenomenon will serve as the basis for the production of crops with low concentrations of toxic metals and transgenic plants for phytoremediation.

1. Introduction

Iron is essential for most living organisms, including plants. Despite its abundance in the Earth’s crust, iron is sparingly soluble under aerobic conditions, especially in high pH and calcareous soils, which account for about 30% of the world’s cultivated soils. Thus, iron deficiency is a widespread agricultural problem that hinders plant growth and lowers crop yields.^1,2) Because plants are the primary food source for humans, the nutritional state of plants is of central importance to human health.³⁾ To acquire enough iron while avoiding toxicity, plants tightly control their uptake, utilization, and storage of iron in response to its availability in the environment.

Since the discovery of the iron solubilizing capacity of root washings from iron-deficient rice and oat plants by Takagi,⁴⁾ the mugineic acid family phytosiderophores (MAs) have been identified as natural iron (III) [Fe(III)] chelators synthesized in the roots of graminaceous plants. Römheld and Marschner⁵⁾ named this iron-acquisition mechanism, which is specific to graminaceous plants, ‘Strategy II’, in comparison with Fe²⁺ transport system after reduction of Fe(III) or ‘Strategy I’ in non-graminaceous plants. The genes involved in the MAs-dependent system have been identified over the past two decades. More recently, iron translocation within plants at the molecular level has begun to be understood. Also, regulatory components of the genes involved in iron uptake and translocation are being identified. In this review, we introduce recent discoveries related to these molecular components in graminaceous plants, and their application to produce transgenic plants with increased tolerance to iron deficiency, improved food quality and those applicable for remediation of heavy metal-polluted soils. Research on these subjects for rice has recently made great advance due to the completion of the rice genome sequencing project and outstanding importance of rice as a staple food crop. Recent advances in iron nutrition in non-graminaceous plants have been reviewed elsewhere.^6–8)

2. Genes involved in iron uptake

The synthesis and secretion of MAs are specific to graminaceous plants and are strongly enhanced under low iron conditions. The identification of molecular components involved in the biosynthesis of MAs has been examined on the basis of extensive physiological studies focused on the identification of MAs and their biosynthetic pathways (reviewed by Refs. 2, 9, 10). To date, nine types of MAs have been identified, and their biosynthetic pathways beginning with methionine as well as the corresponding genes that encode each enzyme in the biosynthetic step have been largely established (Fig. 1 ).^2,9–11) Four sequential enzymatic reactions convert three molecules of L-methionine to 2′-deoximugineic acid (DMA). All graminaceous species examined thus far possess the ability to synthesize DMA, which is further hydroxylated to form other MAs in some species, including barley and rye. The methionine cycle works vigorously to meet the increased demand for methionine in the synthesis of MAs.

Figure 1.

The biosynthetic pathway of MAs in graminaceous plants. SAMS, S-adenosylmethionine synthetase; NAS, nicotianamine synthase; NAAT, nicotianamine aminotransferase; DMAS, deoxymugineic acid synthase; IDS2, iron-deficiency-specific clone no. 2; IDS3, iron-deficiency-specific clone no. 3; NA, nicotianamine; DMA, 2′-deoxymugineic acid; MA, mugineic acid; HMA, 3-hydroxymugineic acid; epiHDMA, 3-epihydroxy-2′-deoxymugineic acid; epiHMA, 3-epihydroxymugineic acid. To date, four other MAs have been identified.

Most of the biosynthetic genes have been cloned first from barley, and then from rice, maize and wheat.^9,10) The spatial expression patterns of the DMA biosynthetic genes in rice have been investigated by histochemical observation of promoter-β-glucuronidase (GUS) transgenic rice lines. Rice nicotianamine synthase genes OsNAS1, OsNAS2, nicotianamine aminotransferase gene OsNAAT1, and deoxymugineic acid synthase gene OsDMAS1 show similar expression patterns in roots and leaves, with expression observed mainly in the phloem tissues of roots and leaves in the presence of adequate iron and strong induction throughout all root and leaf tissues in response to iron deficiency.^12–14) These results strongly suggest that DMA is synthesized in all root cells under iron-deficient conditions. A rice knockout mutant of OsNAAT1 exhibited severe growth retardation under aerobic conditions and a complete defect in DMA secretion, indicating that OsNAAT1 encodes the sole functional enzyme possessing NAAT activity in rice.¹⁵⁾

In contrast to the biosynthetic pathway of MAs, the molecular components involved in the secretion of MAs remain unclear. Vesicular transport and the subsequent diurnal secretion of MAs have been suggested.^16–19)

MAs secreted to the rhizosphere solubilize Fe(III), and the resulting Fe(III)–MAs complexes are taken up by roots. The gene encoding the Fe(III)–MAs transporter in this iron uptake step, Yellow Stripe 1 (YS1), was first isolated in maize.²⁰⁾ Disruption of YS1 leads to leaf chlorosis due to a defect in Fe(III)–MAs uptake.²¹⁾ Subsequently, a barley homolog of YS1 (HvYS1) was isolated that was shown to transport Fe(III)–MAs.²²⁾ HvYS1 is induced in iron deficiency and is localized to root epidermal cells, suggesting its role in uptake. Maize YS1, but not barley HvYS1, was shown to be localized at the distal side in root epidermal cells.^22,23)

To identify the rice transporter gene for Fe(III)–DMA uptake, we analyzed the expression of 18 YS1-like (YSL) genes in rice (OsYSL1-18) using micro-dissected root tissues. Among these genes, OsYSL15 was strongly up-regulated in the root epidermis under iron-deficient conditions.²⁴⁾ OsYSL15 promoter-GUS analysis also revealed strong expression in the epidermal and exodermal cells of iron-deficient roots, where DMA synthesis and the uptake of Fe(III)–DMA complexes is thought to occur.²⁴⁾ Furthermore, OsYSL15 transports Fe(III)–DMA in yeast and Xenopus oocytes.²⁴⁾ Two insertional mutants of OsYSL15 have been shown to exhibit chlorotic phenotypes in response to iron deficiency.²⁵⁾ These results strongly indicate that OsYSL15 is the rice counterpart of the YS1/YSL genes for Fe(III)–DMA uptake (Fig. 2 ). HvYS1 and OsYSL15 expression fluctuates daily,^23,24,26) possibly so that iron uptake can be coordinated with the diurnal secretion of MAs.¹⁶⁾ In contrast, the expression of maize YS1 does not show clear daily fluctuations.²³⁾

Figure 2.

A simplified scheme of the uptake and translocation of iron in rice. Ovals represent transporters. Iron flow is depicted in red arrows. Xylem loading and phloem unloading of iron would require efflux-type transporters, which are little characterized.

In non-graminaceous plants, iron uptake from the rhizosphere is mediated by the induction of Fe(III)-chelate reductase and subsequent transport of Fe²⁺ ions across the root plasma membrane (Strategy I).⁵⁾ Eide et al.²⁷⁾ isolated the Arabidopsis IRT1 gene, which is the dominant ferrous transporter in this uptake process.²⁸⁾ Various transporter genes homologous to IRT1 have been found in plants, animals, protists, and fungi; they are known collectively as the zinc-regulated transporter, iron-regulated transporter-like protein (ZIP) family.²⁹⁾ Rice, in spite of being a Strategy II plant, possesses homologs of the Arabidopsis IRT1 gene, OsIRT1 and OsIRT2, the ferrous transport capacity of which was demonstrated by functional complementation in yeast.^30,31) OsIRT1 expression is strongly induced in iron-deficient roots, and OsIRT2 is expressed similarly, but at lower levels. Promoter-GUS analysis indicated that OsIRT1 is expressed mainly in the epidermis, exodermis, and inner layer of the cortex in deficient roots, as well as in the companion cells of shoots. Moreover, analysis using the positron-emitting tracer imaging system (PETIS) revealed that rice is able to take up both Fe(III)–DMA and Fe²⁺. Thus, rice plants possess a system other than the DMA-based Strategy II for iron uptake (Fig. 2). Such a system to take up Fe²⁺ seems reasonable because rice is commonly grown under submerged conditions in which the dominant form of soil iron is Fe²⁺.

3. Genes involved in iron translocation

The translocation of iron and other minerals inside the plant body involves a sequence of processes that require various metal chelators and transporters. The chemical properties of iron, including poor solubility and high reactivity, compel plants to use suitable chelating molecules inside their bodies.^1,32) Physiological and molecular studies have indicated that one of the principal chelators inside the plant body is nicotianamine (NA),^32,33) which is a biosynthetic precursor of MAs (Fig. 1) synthesized by both graminaceous and non-graminaceous plants. Among the 18 OsYSL genes identified, OsYSL2 expression is strongly induced in iron-deficient leaves.³⁴⁾ Electrophysiological analyses of Xenopus oocytes showed that OsYSL2 transports Fe(II)–NA and manganese (II) [Mn(II)]–NA, but not Fe(III)–MAs. OsYSL2 promoter-GUS analysis revealed that OsYSL2 is expressed in root companion cells and the phloem cells of leaves and leaf sheaths, where OsNAS1-3 are also expressed.^12,34) OsYSL2 knockdown rice lines produced using RNA interference (RNAi) accumulated less iron and manganese in shoots and seeds.³⁵⁾ These results indicate that OsYSL2 is responsible for the long-distance transport of NA-chelated iron and manganese.

In addition to the central role of NA, various lines of evidence suggest that MAs play a role in the internal distribution of iron in graminaceous plants. Endogenous MAs were detected in the shoots of barley and rice, and the amount of MAs increased dramatically under iron deficiency.³⁶⁾ DMA was also detected in rice phloem³⁷⁾ and xylem sap.³⁸⁾ The genes involved in DMA biosynthesis in rice, OsNAS1-3, OsNAAT1, and OsDMAS1, are co-expressed in phloem companion cells in roots and leaves.^12–14) Furthermore, the OsYSL15 transporter gene is also expressed in the phloem companion cells of roots and leaves, as well as in reproductive organs.²⁴⁾

Recently, we reported another Fe(III)–DMA transporter gene, OsYSL18.³⁹⁾ OsYSL18 is expressed in restricted plant parts, including the phloem parenchyma and companion cells at the base of every leaf sheath, suggesting its role in phloem iron transport. Based on physiological studies using tracer elements, the basal part of the shoot was designated as a discrimination center (DC), where phloem and xylem structures are associated and minerals and metabolites accumulate prior to translocation to other plant parts.⁴⁰⁾ Recently, we also reported that iron-52 (⁵²Fe) supplied as ⁵²Fe(III)–DMA to the roots of barley is translocated mainly via phloem to the youngest leaves, whereas it is translocated mostly via xylem to older leaves, suggesting the importance of iron transfer from xylem to phloem in the DC and/or roots.⁴¹⁾ OsYSL18 expression was specific to the region close to the DC. Therefore, phloem loading of Fe(III)–DMA via OsYSL18 might be important for iron distribution to leaf sheaths, especially where xylem iron translocation is not fully developed.

Organic acids are also common metal chelators inside the plant body. In particular, Fe(III)–citrate has long been believed to be the dominant form of iron in xylem sap.^42,43) Since xylem is apoplastic, efflux-type transporters for metals and/or metal-chelator complexes should be needed for long-distance transport in addition to influx-type transporters, including YSLs. FRD3, an Arabidopsis transporter of the multidrug and toxin efflux family, has been shown to be involved in citrate efflux into xylem, which is needed for efficient iron transport within the plant.⁴⁴⁾ A FRD3-like gene in rice, OsFRDL1, is expressed in root pericycle cells adjacent to the protoxylem and metaxylem.⁴⁵⁾ Recently, Yokosho et al.⁴⁶⁾ reported the function of OsFRDL1 as a citrate effluxer required for efficient iron translocation. Molecular components known to be involved in iron translocation in rice is summarized in Fig. 2.

We recently developed a highly sensitive quantification method for NA and DMA using liquid chromatography/electrospray ionization time-of-flight mass spectrometry (LC/ESI-TOF-MS).^38,47) Using this method, comparable amounts of NA, DMA, and iron were detected in rice xylem sap.³⁸⁾ Time-course measurement during iron deficiency treatment suggested that Fe(III)–DMA is increasingly important for xylem iron translocation during deficiency, whereas Fe(II)–NA might be the predominant chemical form of iron in xylem sap under iron-sufficient conditions.

Our promoter-GUS analysis of rice genes involved in iron acquisition and translocation also revealed that the majority of these genes are expressed in flowers and maturing seeds grown in iron-replete soil, as well as in germinating seeds (Fig. 3 ).^24,34,39,48) OsYSL2, OsYSL15, and OsYSL18 are also co-expressed in flowers and the vascular bundles of developing seeds.^24,34,39) Thus, both Fe(II)–NA and Fe(III)–DMA may be involved in seed-iron loading. In addition, OsYSL18 is strongly expressed in pollen and pollen tubes, suggesting its role in pollen function and fertilization by transporting Fe(III)–DMA into pollen cells.³⁹⁾ Notably, OsYSL15 knockdown seedlings showed a severe arrest in germination and early growth that was rescued by high iron supplementation,²⁴⁾ demonstrating the role of OsYSL15 in iron homeostasis during the early stages of growth, facilitating internal Fe(III)–DMA translocation and/or Fe(III)–DMA uptake from the outer surface of seedlings. Recently, synchrotron-based X-ray microfluorescence imaging of iron, zinc, manganese, and copper in rice seeds during germination was carried out at the Super Photon ring-8 GeV (Spring-8) facility.⁴⁹⁾ The distribution of iron in the endosperm and embryo changed during germination, and it differed from that of zinc, manganese, and copper, which may be determined by various expressed metal transporters as suggested by microarray analysis.⁴⁹⁾

Figure 3.

Localization of transporter gene expression in germinating seeds, as observed by histochemical staining of GUS activity in the OsYSL2 (A), OsIRT1 (B) or OsYSL15 (C) promoter-GUS transgenic rice lines in fully mature seeds (0 d) and seeds 1–3 days after sowing.

4. Genes involved in the regulation of iron deficiency responses

Under low iron availability, graminaceous plants induce various genes, many of which are involved in iron acquisition and utilization.^{2,9,10,50,51)} We identified the novel iron deficiency-responsive cis-acting elements IDE1 and IDE2,⁵²⁾ which are the first identified elements related to micronutrient deficiencies in plants. IDE1 and IDE2 synergistically induce Fe-deficiency-responsive expression in tobacco roots, as well as in rice roots and leaves.^52,53) Sequences similar to IDE1 or IDE2 have been found in various Fe deficiency-inducible promoters in barley, rice, tobacco, and Arabidopsis.^50,52,54) This suggests that gene regulatory mechanisms involving IDEs are not only conserved among graminaceous (Strategy II) plants, but are also functional in non-graminaceous (Strategy I) species.

The introduction of graminaceous iron deficiency-responsive promoters into rice, tobacco, and Arabidopsis has revealed complicated patterns of compatibility.^52,55–57) While the barley IDS3 promoter confers iron deficiency responses in all three species, the promoters of HvNAS1 and OsNAS1 are responsive in rice and tobacco, but not in Arabidopsis. The tissue specificities conferred by these promoters are still more complicated, and are possibly dependent on combinations of promoter elements and the presence and/or activation of relevant transcription factors in each tissue. In roots, iron deficiency-induced gene expression is proposed to be mediated by shoot-derived long-distance and local iron signals.^58,59) Each iron deficiency-induced gene appears to respond to one or both of these signals.^57,60)

In a recent search for transcription factors that interact with IDEs, we successfully identified two rice transcription factors, IDEF1 (IDE-binding factor 1) and IDEF2, which bind specifically to IDE1 and IDE2, respectively.^61,62) IDEF1 and IDEF2 belong to uncharacterized branches of the plant-specific transcription factor families ABI3/VP1 and NAC, respectively, and they possess novel properties of sequence recognition. IDEF1 recognizes the CATGC sequence within IDE1, whereas IDEF2 predominantly recognizes CA[A/C]G[T/C][T/C/A][T/C/A] within IDE2 as its core binding site. Northern blotting and histochemical analyses of promoter-GUS transgenic rice lines revealed that IDEF1 and IDEF2 are constitutively expressed during both vegetative and reproductive growth.^61–64)

The characterization of transgenic rice plants with altered IDEF1 or IDEF2 expression has revealed the physiological functions of these transcription factors in iron homeostasis. Transgenic rice lines with introduced IDEF1 under the control of the iron deficiency-inducible IDS2 promoter exhibited a slower progression of leaf chlorosis in iron-free hydroponic culture and improved early growth when germinated in calcareous soil.⁶¹⁾ Conversely, IDEF1 knockdown lines generated by RNAi exhibited hypersensitivity in iron-free hydroponic culture.⁶³⁾ During the early stages of iron deficiency, most known iron uptake/utilization-related genes, including OsIRO2, OsIRT1, OsYSL15, OsYSL2, OsNAS1, OsNAS2, OsNAS3, and OsDMAS1, are positively regulated by IDEF1 (Fig. 4 ).⁶³⁾ In the subsequent stages of iron deficiency, however, the IDEF1-mediated regulation of these iron uptake/utilization-related genes became less obvious. In turn, the expression of several iron deficiency-induced genes encoding late embryogenesis abundant (LEA) proteins, including Osem gene, was increasingly regulated by IDEF1 (Fig. 4).⁶³⁾ These results and in silico cis-distribution analysis using microarray data suggest that IDEF1 has a dual function in iron deficiency responses; namely, (i) the coordinated transactivation of iron utilization-related genes via CATGC-containing IDE1-like elements, especially at the early stage, and (ii) the transactivation of seed maturation-related genes via RY elements, especially during the subsequent stages of iron deficiency.^63,65)

Figure 4.

Model of the gene regulatory network mediated by IDEF1, IDEF2, and OsIRO2 in response to iron deficiency. Dotted lines indicate possible regulation. Ovals indicate transcription factors. All of the indicated genes except IDEF1 and IDEF2 are transcriptionally induced in response to iron deficiency.

IDEF2 also regulates iron homeostasis by inducing another subset of deficiency-responsive genes.⁶²⁾ IDEF2 knockdown lines generated by RNAi and IDEF2 dysfunction lines created using Chimeric REpressor Gene-Silencing Technology (CRES-T) exhibit aberrant iron distribution between the roots and shoots, and are defective in the induction of many iron deficiency-responsive genes, including OsYSL2.⁶²⁾ The regulatory patterns of these IDEF2-dependent genes are largely unaltered between iron sufficiency and early or subsequent deficiency.⁶⁴⁾ The gene regulatory pathways mediated by IDEF1 and IDEF2 are partially, but not predominantly, overlapped.^62,63) Promoter-GUS analysis revealed that IDEF1 and IDEF2 are expressed in many plant parts, including root vascular bundles, lateral roots, and leaf blades, as well as in reproductive tissues, including pollen, suggesting their widespread roles.⁶⁴⁾

To clarify the molecular mechanisms that regulate iron acquisition, we also characterized iron deficiency-induced transcription factors. Microarray analyses revealed that the basic helix-loop-helix (bHLH) transcription factor gene IRO2 is strongly upregulated by iron deficiency in shoots and roots in barley and rice.⁶⁶⁾ The expression of rice OsIRO2 (AK073385) is also reportedly induced by gibberellin in the basal region of rice leaf sheaths.⁶⁷⁾ The core sequence for OsIRO2 binding was determined to be CACGTGG.⁶⁶⁾ We produced transgenic rice plants with enhanced or repressed OsIRO2 expression by introducing the 35S-OsIRO2 cassette or using RNAi.⁶⁸⁾ In iron-deficient hydroponic culture, OsIRO2-overexpressing lines showed enhanced DMA secretion and slightly improved growth compared to non-transformed plants, whereas OsIRO2-repressed lines showed reduced DMA secretion and hypersensitivity to iron deficiency. Microarray and Northern blot analyses revealed that the expression of OsIRO2 is positively related to various iron deficiency-induced genes in roots, including those responsible for DMA biosynthesis (OsNAS1, OsNAS2, OsNAAT1, OsDMAS1, and various genes involved in the methionine cycle) and Fe(III)–DMA uptake (OsYSL15) (Fig. 4). OsIRO2 also affects the expression of some iron deficiency-inducible transcription factors that possess OsIRO2-binding core sequences in their promoter regions.⁶⁸⁾ Importantly, OsIRO2 itself possesses multiple IDEF1-binding core sequences in its promoter region and is positively regulated by IDEF1.⁶¹⁾ Based on these results, a sequential link in the iron-deficiency response involving IDEF1, IDEF2, OsIRO2, and its downstream iron deficiency-inducible transcription factors is proposed (Fig. 4).^61–63,68)

5. Other mineral elements

Iron transporters and chelators often interact with a wide range of metals other than iron. Thus, iron homeostasis is closely linked to that of other metals, such as zinc, copper, manganese, nickel, and cadmium. Because many of these are essential plant nutrients and all essential and non-essential metals are toxic in excess, the complicated relationships in mineral element translocation are a matter of concern for humans and in agriculture. In particular, zinc deficiency constitutes a major problem in both plant and human nutrition, and cadmium toxicity is a serious problem in human health.

In all plant species, NA is thought to play crucial roles in divalent metal translocation.³²⁾ In addition, enhanced NA production in rice, Arabidopsis, and tobacco confers tolerance to excess amounts of various metals, especially nickel,^69,70) indicating its role in protection from metal toxicity.

In graminaceous plants, zinc is thought to be taken up as free zinc (II) [Zn(II)] ions, while the Zn(II)–MAs complex has been proposed as a possible form for uptake.^71,72) Suzuki et al.⁷³⁾ showed that barley induces the expression of MAs biosynthetic genes not only under iron deficiency, but also under zinc deficiency, leading to increased MAs secretion. Moreover, analysis by the PETIS confirmed that more Zn(II)–MAs than Zn²⁺ were absorbed by zinc-deficient barley roots. In contrast to barley, rice does not increase DMA secretion under zinc-deficient conditions.⁷⁴⁾ However, Zn(II)–DMA is translocated to zinc-deficient rice leaves at a higher rate than Zn²⁺, suggesting that zinc is also transported internally in an DMA-chelated form.⁷⁴⁾ In contrast, Meda et al.⁷⁵⁾ reported that chelate formation of cadmium (II) [Cd(II)]–DMA is much weaker than that of Fe(III)–DMA and Zn(II)–DMA, and that DMA does not function in cadmium uptake. DMA-mediated iron acquisition was suggested to be advantageous for improving iron uptake in the presence of cadmium.

ZIP transporters in various organisms transport not only Fe²⁺, but also other divalent cations, including Zn²⁺, Mn²⁺, and Cd²⁺.⁶⁾ Among the rice ZIP members, OsZIP4 expression is strongly induced under zinc deficiency.⁷⁶⁾ Along with its Zn²⁺-transporting capacity, OsZIP4 expression in vascular tissues, leaf mesophyll cells, and apical meristems suggests its predominant role in zinc translocation and utilization in rice. Overexpression of OsZIP4 under the control of the constitutive 35S promoter leads to zinc accumulation in roots and severe zinc deficiency in shoots, suggesting that OsZIP4 needs to be strictly regulated.⁷⁷⁾

The barley ortholog of IRT1, HvIRT1, was reported to transport Fe²⁺, Mn²⁺, Zn²⁺, and Cd²⁺.⁷⁸⁾ Both iron and manganese deficiencies induce HvIRT1 expression, and increased expression of HvIRT1 in a manganese-efficient genotype with a higher Mn²⁺ uptake rate suggests the contribution of HvIRT1 to manganese uptake in barley.⁷⁸⁾ On the other hand, OsIRT1 and OsIRT2 do not complement yeast growth defective in Zn²⁺, Mn²⁺, or Cu²⁺ uptake.³¹⁾ However, these transporters enhance yeast sensitivity to Cd²⁺,⁷⁹⁾ suggesting their ability to transport Cd²⁺, and thus may constitute a substantial influx route of cadmium into rice grains.

Recently, we performed a microarray analysis of rice following cadmium treatment.⁸⁰⁾ Many genes responded to short- and/or long-term cadmium exposure. Genes involved in DMA biosynthesis were induced in the roots following long-term cadmium treatment, suggesting induced iron deficiency by cadmium stress. Similarly, a microarray analysis of rice plants treated with excess zinc revealed the induction of DMA biosynthetic genes.⁸¹⁾ However, induced expression of OsIRO2, but not OsIRT1, in the presence of excess zinc, suggested the presence of a pathway that induces OsIRO2-mediated DMA biosynthesis in zinc-overtreated plants, which is not simply caused by iron deficiency.

6. Production of transgenic crops with improved nutritional traits

Iron deficiency tolerance is divergent among graminaceous plants and is thought to be dependent on the amount and types of MAs that they secrete. Rice, sorghum, and maize secrete only small amounts of DMA among the MAs, and thus are susceptible to low iron availability. We have long aimed to produce transgenic crops, especially rice, with enhanced tolerance to iron deficiency (reviewed by Ref. 10). This was first achieved by introducing a genome fragment containing barley HvNAAT-A and HvNAAT-B into rice.⁸²⁾ These transgenic lines secreted an increased amount of MAs and exhibited substantial tolerance to calcareous soils. Subsequently, various rice lines carrying one or more barley genes for MAs biosynthesis have been produced, some of which have been subjected to field experiments in calcareous soil under paddy conditions (Fig. 5 ).⁸³⁾ Two rice lines, one with a genome fragment containing barley NA synthase gene HvNAS1 and another with a barley genome fragment of mugineic acid synthase gene IDS3, exhibited tolerance to low iron availability. Thus, a transgenic approach to increasing the tolerance of rice to low iron availability is practical for improving agricultural productivity in calcareous soils.

Figure 5.

Iron deficiency tolerance of transgenic rice plants in a calcareous paddy field. (A) Photograph of the entire field at 35 days after transplanting. The rice lines tested are those transformed with barley genome fragment(s) containing (i) IDS3, (ii) HvNAS1, and (iii) HvNAS1, HvNAAT-A and -B, as well as non-transformants. (B) Visual comparison between IDS3-transformed line (right) and non-transformants (left) at 44 days after transplanting, as illustrated in the red box in (A).

Another effective approach to improving iron deficiency tolerance was accomplished by introducing ferric-chelate reductase activity in rice.⁸⁴⁾ Although rice is able to take up Fe²⁺ via the OsIRT1 transporter, its ferric-chelate reductase activity is low and not induced in response to iron deficiency.³¹⁾ Therefore, we introduced a reconstructed and mutagenized derivative of the yeast ferric reductase gene FRE1, designated refre1/372,⁸⁵⁾ which has been artificially evolved to possess enhanced enzymatic activity at a high pH for preferential function in calcareous soils. We used the OsIRT1 promoter to drive the expression of refre1/372, thereby enabling the expression of ferric-chelate reductase and ferrous transporter in the same cells under the same conditions.⁸⁴⁾ Transgenic rice plants with the introduced OsIRT1 promoter-refre1/372 construct successfully induced ferric-chelate reductase expression and activity in iron-deficient roots, leading to increased iron uptake compared to the controls, as revealed by the PETIS. The transformants exhibited enhanced tolerance to low iron availability in both hydroponic culture and calcareous soil. When grown in calcareous soil until harvest, the transformants yielded 7.9 times more grain than the vector controls, demonstrating that creating a complete Strategy I system in rice by enhancing ferric-chelate reductase activity is effective in improving iron-deficiency tolerance.

As mentioned above, the manipulation of transcription factors regulating iron homeostasis could be another potent method for enhancing deficiency tolerance. We recently found that the overexpression of OsIRO2 was much more effective than IDEF1 induction in conferring long-term iron deficiency tolerance to rice plants grown in calcareous soils (unpublished results). When grown in a calcareous soil until maturation, 35S promoter-OsIRO2 transformants grew better than non-transformants, resulting in higher grain yields.⁸⁶⁾

Understanding iron homeostasis also paves the way for fortifying rice grains with iron and zinc, which would greatly contribute to human nutrition. Field trials revealed that a rice line carrying a barley IDS3 genome fragment was not only tolerant to low iron availability in calcareous soils, but was also capable of accumulating more zinc and iron in its grains in both calcareous and Andosol paddy fields.^83,87) Because hydroxylated MAs are more stable under mildly acidic conditions,⁸⁸⁾ MAs synthesized by IDS3 would be favorable for the internal translocation of iron and zinc. Furthermore, the overexpression of barley HvNAS1 under the control of either the 35S or OsActin1 promoter resulted in three- and two-fold increases in the iron and zinc concentrations, respectively, in polished T₁ grains as compared to non-transformants.⁸⁹⁾ Similarly, activation-tagged OsNAS3 overexpression rice lines showed three- and two-fold increases in iron and zinc, respectively, in their grains.⁷⁰⁾ In these transgenic rice lines, the amounts of NA and DMA were increased.^70,89) These results indicate that enhanced NA and DMA production results in the efficient translocation of iron and zinc to rice grains. Further, Lee et al.⁷⁰⁾ showed that increased iron became incorporated into a low molecular mass of possible Fe–NA cluster, which was effective in alleviating anemic symptoms in mice. More recently, we produced transgenic rice lines with iron concentrations that were increased up to 4.4-fold in their polished seeds by expressing OsYSL2 under the control of a sucrose transporter (OsSUT1) promoter.³⁵⁾ The overexpression of OsIRT1 in rice also led to slight increases in iron and zinc accumulation in grains.⁹⁰⁾

Nutrient accumulation in grains is achieved by the combined effects of uptake from the soil and translocation from source organs. Uauy et al.⁹¹⁾ reported that the ancestral wild wheat allele encodes a NAC transcription factor (NAM-B1) that accelerates senescence and increases the remobilization of nutrients from leaves to developing grains. Knockdown of the corresponding gene delayed senescence and reduced the wheat grain protein, iron, and zinc contents.^91,92) Although rice counterparts of this gene have not been identified, another rice NAC gene, OsNAC5, is up-regulated in flag leaves during grain filling and might be involved in the remobilization of iron, zinc, and amino acids to grains.⁹³⁾ OsNAC5 expression in leaves is regulated by abscisic acid⁹³⁾ and is induced under iron deficiency in some cases,⁶⁶⁾ but not in others.⁹³⁾

Since the report of the overexpression of ferritin, a common iron storage protein, in rice grains by Goto et al.,⁹⁴⁾ biofortification of rice seeds with iron using ferritin has been attempted in many laboratories worldwide. Wirth et al.⁹⁵⁾ recently reported a greater than six-fold increase in iron concentration in rice endosperm following the combined introduction of the Arabidopsis NAS gene AtNAS1 driven by the 35S promoter and Pvferritin driven by the rice seed storage globulin promoter. Transgenic rice lines produced recently by our group carrying various combinations of iron homeostasis/storage-related genes driven by several tissue-specific promoters have been shown to possess still greater polished seed concentrations of iron and zinc (unpublished data).

7. Future perspectives

Based on the fundamental knowledge accumulated during continuous studies of graminaceous plant responses to iron deficiency and recent advances in genome information, much progress has been made in recent years on iron homeostasis in rice. Many crucial genes for iron translocation and regulation have been identified and characterized. Other recent findings include the induction of glutathione reductase expression and activity in iron-deficient graminaceous plants,⁹⁶⁾ suggesting a close relationship between iron homeostasis and cellular redox status. We identified a rice-specific mitochondrial iron deficiency-induced gene (MIR), whose knockout severely impairs rice growth.⁹⁷⁾ Mechanisms of subcellular compartmentalization for iron and other mineral elements are poorly understood, especially in graminaceous plants. Also, mechanisms of iron deficiency sensing and signaling, as well as the secretory mechanism of MAs and NA into apoplastic spaces, have not been clarified. A comprehensive understanding of the speciation of metal chelates and transporters in each step of metal translocation will become increasingly important for future advances in this field and for the production of crops with improved nutritional traits, including those with low cadmium in the edible parts and high cadmium plants applicable for phytoremediation.

Non-standard abbreviation list

bHLH

basic helix-loop-helix

DMA

2′-deoximugineic acid

DMAS

deoxymugineic acid synthase

GUS

β-glucuronidase

IDE

iron deficiency-responsive element

IDEF

iron deficiency-responsive element-binding factor

MAs

mugineic acid family phytosiderophores

NA

nicotianamine

NAAT

nicotianamine aminotransferase

NAS

nicotianamine synthase

PETIS

positron-emitting tracer imaging system

RNAi

RNA interference

ZIP, zinc-regulated transporter

iron-regulated transporter-like protein

Profile

Naoko K Nishizawa was born in 1945. She earned her bachelor’s degree in Agricultural Chemistry, Faculty of Agriculture at the University of Tokyo in 1968 and her Ph. D. in Agricultural Sciences from the University of Tokyo. She started her career with studies on the ultra-structural analysis of crop plants and performed pioneering work on crop uptake of organic compounds and found that rice plants uptake and utilize organic nitrogen as a nutrient source by the mechanism of heterophagy in 1977. This was the first demonstration of the occurrence of heterophagy mechanism in the plants. She employed as an assistant professor at the University of Tokyo in 1982. She and Dr. Satoshi Mori started to work on iron acquisition mechanism in graminaceous plants in collaboration with Prof. Seiichi Takagi who found mugineic family phytosiderophores, and clarified the biosynthetic pathway of phytosiderophores. This study led to subsequent extensive studies that identified many proteins and genes participated in iron acquisition and their regulatory network, and developed the transgenic rice with enhanced tolerance to iron deficiency in calcareous soil. From 1995 to 1996, she was in the laboratory of plant molecular biology of The Rockefeller University and working on the molecular mechanism of signal transduction pathway in plants. She was promoted to a lecturer in 1996 and to full professor in 1997. She moved to Research Institute for Bioresources and Biotechnology at Ishikawa Prefectural University as full professor in 2009. She is an elected member of Science Council of Japan and was awarded the Japan Prize of Agricultural Sciences in 2010.