The Phytosiderophore Efflux Transporter TOM2 Is Involved in Metal Transport in Rice*

Background: Phytosiderophores are important metal chelators for graminaceous plants. Results: Repression of TOM2 causes severe growth defects. Conclusion: TOM2 is a crucial efflux transporter of phytosiderophores. Significance: TOM2 may enhance crop yield and nutritional quality. Iron is an essential metal element for all living organisms. Graminaceous plants produce and secrete mugineic acid family phytosiderophores from their roots to acquire iron in the soil. Phytosiderophores chelate and solubilize insoluble iron hydroxide in the soil. Subsequently, plants take up iron-phytosiderophore complexes through specific transporters on the root cell membrane. Phytosiderophores are also thought to be important for the internal transport of various transition metals, including iron. In this study, we analyzed TOM2 and TOM3, rice homologs of transporter of mugineic acid family phytosiderophores 1 (TOM1), a crucial efflux transporter directly involved in phytosiderophore secretion into the soil. Transgenic rice analysis using promoter-β-glucuronidase revealed that TOM2 was expressed in tissues involved in metal translocation, whereas TOM3 was expressed only in restricted parts of the plant. Strong TOM2 expression was observed in developing tissues during seed maturation and germination, whereas TOM3 expression was weak during seed maturation. Transgenic rice in which TOM2 expression was repressed by RNA interference showed growth defects compared with non-transformants and TOM3-repressed rice. Xenopus laevis oocytes expressing TOM2 released 14C-labeled deoxymugineic acid, the initial phytosiderophore compound in the biosynthetic pathway in rice. In onion epidermal and rice root cells, the TOM2-GFP fusion protein localized to the cell membrane, indicating that the TOM2 protein is a transporter for phytosiderophore efflux to the cell exterior. Our results indicate that TOM2 is involved in the internal transport of deoxymugineic acid, which is required for normal plant growth.

Iron is an essential metal element for all living organisms. Graminaceous plants produce and secrete mugineic acid family phytosiderophores from their roots to acquire iron in the soil. Phytosiderophores chelate and solubilize insoluble iron hydroxide in the soil. Subsequently, plants take up iron-phytosiderophore complexes through specific transporters on the root cell membrane. Phytosiderophores are also thought to be important for the internal transport of various transition metals, including iron. In this study, we analyzed TOM2 and TOM3, rice homologs of transporter of mugineic acid family phytosiderophores 1 (TOM1), a crucial efflux transporter directly involved in phytosiderophore secretion into the soil. Transgenic rice analysis using promoter-␤-glucuronidase revealed that TOM2 was expressed in tissues involved in metal translocation, whereas TOM3 was expressed only in restricted parts of the plant. Strong TOM2 expression was observed in developing tissues during seed maturation and germination, whereas TOM3 expression was weak during seed maturation. Transgenic rice in which TOM2 expression was repressed by RNA interference showed growth defects compared with non-transformants and TOM3-repressed rice. Xenopus laevis oocytes expressing TOM2 released 14 C-labeled deoxymugineic acid, the initial phytosiderophore compound in the biosynthetic pathway in rice. In onion epidermal and rice root cells, the TOM2-GFP fusion protein localized to the cell membrane, indicating that the TOM2 protein is a transporter for phytosiderophore efflux to the cell exterior. Our results indicate that TOM2 is involved in the internal transport of deoxymugineic acid, which is required for normal plant growth.
Iron is essential for all living organisms, including humans and plants. In plants, iron is necessary for growth and is required for various cellular activities ranging from photosynthesis to respiration. Under aerobic conditions, iron is oxidized to iron hydroxide (Fe(OH) 3 ), which is poorly soluble in water. Therefore, although mineral soils contain 6% iron by weight, most of it is not available to plants. This phenomenon is exacerbated in high-pH soils, including calcareous soils, constituting a major problem for crop production. Moreover, iron deficiency leads to leaf chlorosis, poor yields, and decreased nutritional quality. Iron uptake in humans is ultimately dependent on the iron that plants take up from the soil. Therefore, increasing the efficiency of iron uptake in food plants could have a dramatic, positive impact on both crop productivity and human health.
To acquire iron, higher plants have two strategies for the uptake of oxidized Fe(III) from the rhizosphere (1). All higher plants, except graminaceous plants, take up iron using ferric chelate reductases to reduce ferric iron to Fe(II), which is then absorbed by ferrous iron transporters (2)(3)(4). Alternatively, graminaceous plants, including important staple crops such as rice, wheat, and barley, secrete natural iron chelators called mugineic acid family phytosiderophores (MAs) 2 from their roots (5). These MAs have six coordination sites (three -COOH, two -NH, and one -OH) that bind to iron and are thought to form octahedral Fe(III) complexes that are soluble (6). The secreted MAs chelate and solubilize gelatinous Fe(OH) 3 in the soil, forming Fe(III)-MA complexes that are absorbed into root cells through the Fe(III)-MA transporters called yellow stripe 1 (YS1)/YS1-like (YSL) transporters, which localize to the root cell membrane (7,8). The biosynthetic pathway for MAs in graminaceous plants has been elucidated (9,10). S-adenosylmethionine, the precursor of MAs, is converted to 2Ј-deoxymugineic acid (DMA) via four sequential steps catalyzed by S-adenosylmethionine synthetase, nicotianamine * This work was supported by a grant-in-aid for scientific research on priority areas from the Ministry of Education, Culture, Sports, Science, and Technology, Japan (to N. K. N. and N. U.) and by a grant-in-aid for Young Scientists (B) grant number from JSPS KAKENHI (to T. N. synthase, nicotianamine aminotransferase, and deoxymugineic acid synthase (9 -13). Although rice and maize secrete DMA, other species, including barley and rye, further hydroxylate DMA to other MAs. The production and secretion of MAs increase markedly in response to iron deficiency. The secretion of MAs in barley follows a distinct diurnal rhythm, with a peak just after sunrise or initial illumination (14,15). MAs are thought to be specific to graminaceous plants. On the other hand, nicotinamide (NA), an intermediate in MAs biosynthesis and a structural analog of MAs, is produced in all plants examined so far, including non-graminaceous plants such as Arabidopsis, tomato, and tobacco (16 -20).
In the plant body, free Fe 2ϩ is toxic. Therefore, iron transport requires complex formation with some component. MAs and NA are also essential for the translocation of metal nutrients inside the body of the plant. MAs and NA are able to chelate not only iron but also various transition metals such as zinc, manganese, and copper (16,17,(21)(22)(23). These metals are micronutrients and are important for plant development because they are necessary for the activities and structures of various proteins. Because MAs have been identified in the xylem and phloem of rice and barley, they are also suggested to play an important role in the long-distance transport of metals in graminaceous plants (23)(24)(25)(26). NA has been suggested to play an essential role in metal translocation and accumulation in developing seeds on the basis of an analysis of the tomato chloronerva mutant (19) and NA-deficient transgenic tobacco (Nicotiana tabacum) plants (20). In addition, MAs and NA have been suggested to be involved in metal transport during rice seed germination on the basis of results from physiological analyses using YSL knockdown rice lines and expression analyses using promoter-␤-glucuronidase (GUS) and microarrays (8,(27)(28)(29).
Recently, we identified transporter of MAs (TOM1) as an efflux transporter of DMA in rice, barley, and maize (40,41). Expression of TOM1, HvTOM1, and ZmTOM1 was strongly induced in iron-deficient roots, whereas, in rice and barley, TOM1 and HvTOM1 showed a diurnal pattern in their expression. In transgenic rice with overexpressed or repressed expression of TOM1, the level of TOM1 expression correlates highly with the level of phytosiderophore secretion from the roots. Moreover, overexpression of TOM1 and HvTOM1 resulted in increased tolerance to iron deficiency. These results suggest that TOM1 is the main transporter for MA secretion under iron deficiency. TOM1, HvTOM1, and ZmTOM1 are members of the major facilitator superfamily, which is widely distributed among plants (42). The efflux transporter of NA (ENA1) in rice also belongs to this family (40). Zinc-induced facilitator 1 (ZIF1), the ortholog of TOM1 in Arabidopsis, has been reported to be an efflux transporter of NA localized in the vacuolar membrane and hypothesized to transport NA from the cytoplasm into vacuoles (43,44). Rice contains five homologs of TOM1. Two are located in tandem with TOM1 on chromosome 11 (Os11g0135000, TOM2 and Os11g0135900, TOM3) and three others are located in tandem on chromosome 12 (Os12g0132500, Os12g0132800, and Os12g0133100), similar to TOM genes on chromosome 11. However, their functions have not yet been identified. In this study, we analyzed the function of TOM2 and TOM3 in rice to advance our understanding of the role of MAs and their efflux transporters in graminaceous plants. We found that TOM2 has the ability to transport phytosiderophores to the cell exterior and is necessary for normal plant growth.
GUS Activity Assay-GUS activity in the roots and shoots of transgenic plants was determined using a histochemical assay (34). GUS activity in reproductive organs and germinating seeds was also determined histochemically according to methods described previously (27).
Plant Growth Conditions and Elemental Analysis-Rice plants were grown hydroponically. Seeds were surface-sterilized with a 2.5% sodium hypochlorite solution and then germinated for 1 week. After germination, the seedlings were transferred to a 20-liter plastic container containing a nutrient solution of the following composition: 0.7 mM K 2 SO 4 , 0. For chlorophyll analysis, plants were grown hydroponically as described above. On day 12 after transition to the iron-deficient medium, the youngest and oldest leaves at that time were analyzed for chlorophyll content using a Soil & Plant Analyzer Development (SPAD-5) chlorophyll meter (Konica Minolta, Tokyo, Japan). Plant length and weight were measured 2 weeks after transition to the iron-deficient medium. Three to five plants for each line were used for each analysis, the same experiments were replicated three times, and the reproducibility of the experiments was confirmed. The concentrations of iron, zinc, copper, and manganese were determined using inductively coupled plasma MS (15). All experiments were performed in triplicate.
Construction of Oocyte Expression Vectors and Efflux Experiments-The TOM2 ORF of pTOM2 was excised from pTOM2 with EcoI and XhoI and cloned into a plasmid constructed previously (50). Then capped cRNA derived from these plasmids was injected into Xenopus laevis oocytes (50). Assays of efflux activity using X. laevis oocytes were performed according to a method described previously (40).
Construction of the GFP Fusion Gene for the Transient Expression and Observation of TOM2-synthetic GFP Localization-An attL/attR substrate recombination reaction between pTOM2 and pDEST35S-synthetic GFP (51) generated an expression vector containing the Cauliflower mosaic virus 35 S promoter-TOM2-synthetic GFP gene fusion sequence ( 35 S::TOM2-GFP). Onion epidermal cells were transformed with 35 S::TOM2-GFP using the Biolistic PDS-1000/He particle delivery system (Bio-Rad), and synthetic GFP fluorescence was observed (31).

Results
We recently isolated TOM1 (Os11g0134900) as an efflux transporter of DMA in rice (40). To characterize the TOM family in rice, we carried out a homology search of TOM1 using a rice genome database (RAP-DB) and found five homologous genes of TOM1: TOM2 (Os11g0135000), TOM3 (Os11g0135900), Os12g0132500, Os12g0132800, and Os12g0133100 (Fig. 1A). TOM1, TOM2, and TOM3 are located on chromosome 11 in tandem (Fig. 1B), whereas, similarly, Os12g0132500, Os12g0132800, and Os12g0133100 are located on chromosome 12 in tandem. We compared their genomic sequences and found that the TOM1, TOM2, and TOM3 genomic sequences are quite similar to those of Os12g0132500, Os12g0132800, and Os12g0133100, respectively (Fig. 1C). The coding sequence of each gene predicted by the genomic sequence was also quite similar. We attempted to clone Os12g0132500, Os12g0132800, and Os12g0133100 but were unsuccessful in cloning their cDNA. Therefore, we focused on analyzing the functions of TOM2 and TOM3 in this study.
Promoter-GUS Analysis of TOM2 and TOM3-To assess the tissue-specific localization of TOM2 and TOM3, we generated transgenic plants in which the GUS gene uidA was driven by the TOM2 or TOM3 promoter (Fig. 2). Under iron-sufficient conditions, TOM2 and TOM3 expression was not observed in roots (Fig. 2, A, C, and E-G). TOM2 expression was induced in roots by iron deficiency (Fig. 2, B and H-K) and observed in areas near where the lateral roots emerged (Fig. 2B). In cross-section, TOM2 expression was observed in the cortex and central cylinder, both of which are involved in metal translocation, but not in the exodermis, which is involved in metal uptake from the soil, or in the endodermis (Fig. 2, H-J). In the vertical section, TOM2 expression occurred in root radicles, where cell division is active (Fig. 2K). Under iron-deficient conditions, TOM3 expression was only occasionally detected near the root tips (Fig. 2D).
TOM2 expression was also observed in the basal part of the shoots, which is suggested to be important for iron transfer from the xylem to phloem for translocation to the youngest leaf (52). Strong expression was detected near the junction between roots and shoots (Fig. 2, L and M). The vertical section (Fig. 2N) and cross-section (Fig. 2O) showed that TOM2 was expressed in the vascular bundle and root radicles of the basal part of the shoots. In the cross-section of leaf sheaths 5 cm above the basal part of the shoots, TOM2 expression was observed in large vascular bundles, the outer layer, and veins between large vascular bundles (Fig. 2P). In leaf blades, TOM2 expression was observed in phloem cells in the large vascular bundles (Fig. 2Q).  The expression patterns of TOM2 in the shoots were not altered by iron status. The expression of TOM3 was not observed in shoots (data not shown).
TOM2 and TOM3 expression was also observed in developing and germinating seeds (Fig. 2, R and S). During seed development, nutrients are transported from leaves through vascular bundles to the embryo and endosperm (53). During germination, these nutrients, accumulated in the endosperm, are transported through the epithelium to the embryo (54). In developing seeds, TOM2 expression was observed in the dorsal vascular bundles, epithelium, and scutellum of the embryo, whereas TOM3 expression was observed in the anther and aleurone layer (Fig. 2R). In germinating seeds, TOM2 expression was detected in the epithelium, scutellum, and dorsal vascular bundles, whereas TOM3 expression was observed in the aleurone layer (Fig. 2S).
Functions of TOM2 and TOM3 in Rice-To examine the function of TOM2 and TOM3 in rice plants, transgenic plants in which their expression was repressed using RNAi (RNAi plants) were generated. Quantitative real-time PCR of the RNAi plant roots confirmed that the expression of TOM2 and TOM3 was repressed in TOM2 RNAi and TOM3 RNAi plants, respectively (Fig. 3, D and E). In normal soil culture, TOM2 RNAi plants showed growth defects compared with non-transformants (NTs), whereas TOM3 RNAi plants showed no significant differences in vegetative growth (Fig. 3, A-C and F-H).  1 and 2, from lines #12, #19, and #21). Values represent the means of three replicates. Error bars represent mean Ϯ S.D. NOVEMBER 13, 2015 • VOLUME 290 • NUMBER 46

JOURNAL OF BIOLOGICAL CHEMISTRY 27693
Tiller number (Fig. 3A) and seed fertilities of TOM2 RNAi plants decreased compared with TOM3 RNAi and NT plants (Fig. 3, B and C). Dry weight of the shoots at harvest and yields of TOM2 RNAi plants were dramatically lower than those of TOM3 RNAi and NT plants, whereas shoot length did not differ significantly differ (Fig. 3, F-H). Metal concentrations in the seeds of TOM2 RNAi, TOM3 RNAi, and NT rice were measured (Fig. 3I). Manganese concentrations were higher in TOM2 RNAi seeds compared with NT and TOM3 RNAi seeds. Iron, zinc, and copper concentrations did not differ significantly in TOM2 RNAi, TOM3 RNAi, and NT plants.
Next, plants were grown in iron-deficient hydroponic culture medium to confirm the phenotype of TOM2 RNAi (Fig. 4,  A-D). In agreement with the phenotype seen under soil culture, both the shoot and root weights of TOM2 RNAi were significantly lower than in NTs (Fig. 4, B and D), although the shoot and root lengths did not differ significantly (Fig. 4, A and C). In the shoots, iron requirement is particularly high in newly developing leaves, where various iron-containing proteins, used for various metabolic processes such as photosynthesis, are newly biosynthesized. As a consequence, symptoms of iron deficiency, namely chlorosis (a decrease in chlorophyll content), first appear in the youngest leaves. In addition, iron absorbed from barley roots is transmitted to the youngest leaf and the other leaves mainly via the phloem and xylem, respectively (52). Therefore, the chlorophyll contents (in SPAD units) and metal concentration of the oldest leaves and the youngest leaves at the sampling time were measured (Figs. 4 and 5). The SPAD values of the oldest leaves of TOM2 RNAi plants were lower than in NT plants (Fig. 4E), whereas the SPAD values at the sampling time of the youngest leaves of TOM2 RNAi plants were slightly lower than in the NTs (Fig. 4F). Iron concentrations in the youngest leaves and roots were similar between TOM2 RNAi plants and NT, whereas iron concentrations in the fifth leaves from the youngest leaves and leaf sheath of TOM2 RNAi lines tended to be higher than in the NTs (Fig. 5). Zinc and copper  concentrations in the youngest leaves, leaf sheath, and roots of TOM2 RNAi lines were higher than in NTs. Moreover, manganese concentrations were significantly higher in the youngest leaves, fifth-youngest leaves, and leaf sheaths of TOM2RNAi lines compared with NTs.
We further analyzed the change in expression of genes involved in iron homeostasis (Fig. 6). The expression levels of TOM1 and OsIRO2, encoding an iron deficiency-induced transcription factor regulating iron homeostasis (55), were slightly higher in lines 2 and 4 of TOM2 RNAi rice compared with NT plants. In contrast, the expression levels of TOM3 and OsIRT1, encoding a ferrous iron transporter (56), and OsYSL2 did not differ significantly between TOM2 RNAi rice and NTs. The expression level of OsFer1 and OsFer2, encoding iron storage proteins, ferritins (57) (quantified collectively as OsFer), in TOM2 RNAi rice decreased compared with the NTs in iron-sufficient roots, whereas their expression was similar in TOM2 RNAi rice and NT plants in iron-deficient roots.
DMA Efflux Activity of TOM2-To examine the DMA transport activity of TOM2, 14 C-labeled DMA was synthesized from S-adenosylmethionine as described previously (40). We injected capped TOM2 cRNA, TOM1 cRNA (positive control), or water (negative control) into X. laevis oocytes and incubated them for 2 days. Then the oocytes were loaded with 14 C-labeled DMA and monitored for the release of 14 C into the medium (Fig. 7A). Oocytes expressing TOM2 exhibited a higher rate of DMA efflux than control oocytes (Fig. 7A). TOM2 efflux activity was similar to that of TOM1. Subcellular localization of the TOM2 protein was investigated by fusing the coding sequence of TOM2 with GFP. In onion epidermal cells transiently expressing the TOM2-GFP fusion protein, fluorescence was localized to the cell membrane (Fig. 7, B-D). In addition, cell membrane localization of the TOM2-GFP fusion protein was confirmed in rice roots stably expressing TOM2-GFP (Fig. 7, E-G), collectively indicating that TOM2 encodes a transporter of DMA efflux to the cell exterior.

Discussion
The TOM Gene Family May Be Duplicated in the Rice Genome-The rice genome contains five genes homologous to TOM1 (Fig. 1A). Of these, TOM1, TOM2, and TOM3 are located in tandem on chromosome 11, whereas the three other genes (Os12g0132500, Os12g0132800, and Os12g0133000) are located on chromosome 12, also in tandem (Fig. 1B). The gene structure on chromosome 12 is quite similar to TOM1-3 on chromosome 11. Genomic sequences of the predicted coding regions of TOM1 and Os12g0132500 share 98% identity (Fig.  1C), whereas TOM2 and Os12g0132800 and TOM3 and Os12g0133000 show 94% and 97% identity, respectively. We could not find a cDNA clone with the predicted coding sequence of Os12g0132500. Therefore, whether Os12g0132500 is transcribed in rice is uncertain. TOM2 has several rice Tos17 insertion mutants. We confirmed that Tos17 was definitely inserted into the TOM2 genomic region, but the expression level of TOM2 in these mutants, as detected by Northern blot analysis, was not altered compared with the wild type. In addition, TOM2 Tos17 mutants showed no phenotypic changes as seen in TOM2 RNAi plants in this study. We could not distinguish the expression between TOM2 and Os12g0132800 in TOM2 Tos17 mutants. Os12g0132800 expression may complement TOM2, and Os12g0132500, Os12g0132800, and Os12g0133000 on chromosome 12 may be duplicates of the TOM family on chromosome 11. These observations suggest that DMA secretion and transport by the TOM family are important for plant nutrition. The function of the TOM family may be conserved by duplication of the genomic fragment in rice.
TOM2 Is Involved in the Internal Transport of Metals by DMA Efflux Transport-MAs are able to chelate iron and other metals, such as zinc, copper, and manganese (21,22). MAs are abundant in rice roots and shoots, and the amount of DMA in the xylem sap increases under iron deficiency, suggesting that MAs play important roles in internal metal transport (23-25). Moreover, chemical speciation of iron-binding ligands revealed DMA as a dominant chelator in rice phloem sap (26). These results indicate that MAs play important roles both in iron acquisition from the soil and in internal metal transport inside the plant body. In this study, promoter-GUS analysis showed that TOM2 is expressed in the vascular tissues in roots, shoots, and seeds (Fig. 2). TOM2 expression was very high in the vascular tissues of the basal part of the shoots, which connects the roots and shoots (Fig. 2, L-O). Using a positron-emitting tracer imaging system, iron was shown to first accumulate at the basal part of the shoots after absorption from the roots and was then distributed throughout the plant body (52). Therefore, TOM2 and DMA might be involved in iron distribution mediated in this part of the plant. We clarified that TOM2 has efflux activity using [ 14 C]DMA in X. laevis oocytes (Fig. 7A). In onion epidermal cells and rice roots, TOM2 localized to the cell membrane (Fig. 7, B-G). These results demonstrate that TOM2 is an efflux transporter of DMA from the cytosol to the cell exterior. Our results suggest that TOM2 is involved in DMA-mediated metal transport inside the plant body and supports the importance of DMA as a chelating molecule facilitating metal transport in the plant body in addition to its well known function in iron uptake from the soil.
TOM2 is Crucial for Normal Plant Growth-We observed growth defects in TOM2 RNAi rice but not in TOM3 RNAi rice (Figs. 3 and 4). In TOM2 RNAi plants, the concentrations of iron and other metals, such as zinc, copper, and manganese, tended to be higher than in the NTs in the youngest leaves, fifth-youngest leaves, the leaf sheath, roots, and seeds (Figs. 3 and 5). These results suggest that metal translocation was aberrant in TOM2 RNAi plants. Previously, phenolics efflux zero 1 and 2 (PEZ1 and PEZ2), the efflux transporters of protocatechuic acid, have been reported to contribute to the long-distance transport of iron through the solubilization of precipitated apoplasmic iron in the root xylem (58,59), suggesting that some transporters are necessary to mobilize metals accumulated in the apoplasm to transport them to sinks in plants.
Members of the YSL family in rice are also involved in the internal transport of metals. OsYSL15, which imports Fe(III)-DMA, is strongly expressed in iron-deficient roots, which suggests that OsYSL15 functions in the absorption of Fe(III)-DMA from the soil (8). In addition, the expression of OsYSL15 has been observed in seeds, and its repression by RNAi led to defects in seed germination, suggesting that OsYSL15 is also involved in the internal transport of iron. OsYSL2, which imports Fe(II)-NA and Mn(II)-NA, is strongly expressed in iron-deficient leaves (36). Enhancement of OsYSL2 expression driven by a sucrose transporter promoter increased the iron concentration in polished seeds up to 4.4-fold compared with the NTs (60). These results suggest that OsYSL2 is important for phloem transport of metals into seeds. In addition, OsYSL16 and OsYSL18, Fe(III)-DMA transporters, are specifically expressed in the xylem or phloem and reproductive organs (29,38,61). These results suggest that the expression of specialized transporters in restricted areas is necessary to maintain metal homeostasis throughout the plant body. Promoter-GUS analysis in this study indicated that TOM2 is expressed in the vascular tissues of shoots and seeds under both iron-sufficient and iron-deficient conditions (Fig. 2, L-S). In addition, under irondeficient conditions, TOM2 expression was high in the vascular tissues of roots (Fig. 2, B and H-K). These expression patterns are consistent with an analysis of transcriptomes by laser microdissection microarray in which TOM2 was expressed mainly in the vascular bundles and induced by iron deficiency in the cortex (62). TOM2 expression was very high in the root primordium during the vegetative stage (Fig. 2, B and K) and in the embryo during the reproductive stage and germination (Fig.  2R). These expression patterns suggest that TOM2 is involved in the mobilization of metals through the secretion of DMA into the vascular bundles for phloem and/or xylem metal loading to supply sufficient metals for plant differentiation and growth. In TOM2 RNAi rice, the expression level of OsFer (summation of OsFer1 and OsFer2) was lower than in the NTs, and the expression of other iron homeostasis-related genes, such as TOM1 and OsIRO2, tended to be induced (Fig. 6). OsFer1 and OsFer2 encode the iron storage protein ferritin, which releases iron under iron deficiency (57). The expression levels of OsFer1 and OsFer2 are down-regulated under iron deficiency and up-regulated under iron excess conditions (57). Decreases in OsFer expression and the slight induction of TOM1 and OsIRO2 in TOM2 RNAi rice suggest that TOM2 RNAi plants are physiologically deficient in iron compared with NT plants. TOM2 may play a crucial role in metal mobilization and utilization in the apoplasm through DMA secretion to solubilize metal nutrients to facilitate their transport to sinks in the plant body.
In conclusion, we show that the TOM family has important roles both in the acquisition of iron from the soil and in metal translocation in graminaceous plants through the efflux transport of DMA. TOM1 expression increases dramatically under iron-deficient conditions and may be involved mainly in the secretion of phytosiderophores into the soil from root cells (40). In contrast, TOM2 may function primarily in the translocation of metals inside the plant body under normal growth conditions. The function of TOM3 has not yet been determined, but is thought to also have a specific role in the transport of phytosiderophores because of sequence similarity to TOM1 and TOM2. The TOM family has important roles in the maintenance of metal homeostasis through the secretion of DMA into the soil and also by translocation of metals to sink areas in plants. In addition, analogy between MAs and NA suggests that the TOM family is also involved in NA-mediated metal translocation in all higher plants. Recent advances have shown that specialized plant membrane transporters can be used to enhance the yields of staple crops and increase the nutrient content and resistance to key stresses such as iron deficiency, which could expand the amount of available arable land (63). Characterization of the TOM family is a significant step in advancing these efforts.
Author Contributions-T. N. designed, performed, and analyzed the experiments and wrote the paper. S. N. provided technical assistance in the experiments shown in Fig. 3. T. K. and H. N. provided assistance for writing the paper. Y. S. and N. U. provided technical assistance for the experiments shown in Fig. 7. N. K. N. conceived and coordinated the study and wrote the paper with T. N. All authors reviewed the results and approved the final version of the manuscript.