Findings

Iron (Fe) is an essential micronutrient for all higher organisms. Plants require Fe for several cellular processes, including respiration, chlorophyll biosynthesis, and photosynthetic electron transport (Marschner 1995). The molecular mechanism of Fe transport in rice has been well documented (Bashir et al. 2006; Bashir and Nishizawa 2006; Bashir et al. 2010; Bashir et al. 2011b; Bashir et al. 2013a; Ishimaru et al. 2012; Kobayashi and Nishizawa 2012). Once inside a plant, Fe enters root cells and is transported to the shoot and seeds. Fe performs vital roles in subcellular organelles such as chloroplasts and mitochondria, and defects in mitochondrial Fe homeostasis significantly affect plant growth (Bashir et al. 2011a; Bashir et al. 2011c; Ishimaru et al. 2009). As excess Fe in the cytoplasm may be toxic, it is either stored as ferritin in chloroplasts or is diverted to the vacuole. Knockout mutants for the rice vacuolar metal transporters OsVIT1 and OsVIT2 were recently reported to accumulate increased amounts of Fe in their seeds (Zhang et al. 2012). This accumulation was mainly observed in the embryo (Zhang et al. 2012), which is removed during milling. In this short report, we describe the characterization of a mutant in which the expression of OsVIT2 was knocked down and we show that changes in the expression of OsVIT2 and mitochondrial iron transporter (MIT) affect seed Fe localization in brown rice as well as in polished rice seeds.

We characterized a T-DNA line (An et al. 2003a; Jeong et al. 2006) in which the T-DNA was integrated ~500 bp upstream of the start codon of OsVIT2 (Os09g0396900), as shown in Figure 1a. (For details see Additional file 1 “Methods”). Genomic polymerase chain reaction (PCR) using primers specific for the T-DNA integration site confirmed the homozygous status of the plants (Figure 1b), while primers specific for exon 3 were used to check the quality of the DNA (Figure 1c). Quantitative PCR analysis confirmed that the expression of OsVIT2 was significantly downregulated in the osvit2 line compared to wild-type (WT) plants (Figure 1d). Note that similar to indica rice, the sequence of OsVIT2 is not complete in RAP-DB (http://rapdb.dna.affrc.go.jp/) and the sequence of OsVIT2 showed 100% similarity to that reported for indica rice (Zhang et al. 2012). Data related to the expression of OsVIT2 were generated through rice global gene expression profile data sets maintained at http://ricexpro.dna.affrc.go.jp (Sato et al. 2011a; Sato et al. 2011b). OsVIT2 expression is upregulated in the presence of excess Fe (Bashir et al. 2011c), and the expression of OsVIT2 could be observed through all developmental stages (Figure 1e–l). In roots, the expression of OsVIT 2 increased slightly from 27 days after transplantation (DAT) to 76 DAT and was not regulated diurnally (Figure 1e). A similar trend was observed in the stems (Figure 1f). In leaves, the expression of OsVIT2 increased slightly at noon from 27 to 76 DAT and at 125 DAT, while at midnight, it decreased slightly from 27 to 76 DAT and then increased at 127 DAT (Figure 1g). OsVIT2 expression did not change significantly during anther development (Figure 1h). In the embryo, OsVIT2 expression increased slightly from 1 to 3, 5, and 7 days after fertilization (DAF; Figure 1i). In ovary, OsVIT2 expression first increased from 7 to 10 DAF, then decreased from 10 to 14 and 28 DAF, and then increased again at 42 DAF (Figure 1j). In endosperm, the expression of OsVIT2 first increased from 7 to 10 DAF and then decreased from 10, 14, 28, and 42 DAF (Figure 1k). In addition, OsVIT2 expression remained largely unchanged during pistil development (Figure 1l). These results clearly support the earlier results of Zhang et al. (2012) showing that OsVIT2 plays a critical role in transporting Fe and zinc (Zn) from leaves to seeds.

Figure 1
figure 1

Characterization of the OsVIT2 knockdown mutant ( Osvit2 ) and expression of OsVIT2 during different developmental stages of rice. a: Gene structure and positions of the primers specific for the T-DNA insertion. b: Bands corresponding to the primers are shown in red. c: Bands corresponding to the primers are shown black. d: Quantitative reverse-transcription PCR analysis showing the expression of Osvit2 in WT and knockdown plants. A vertical bar followed by an asterisk indicates a significant difference from WT according to the Tukey–Kramer test (p < 0.05; WT, n = 4; osvit2, n = 8). e: Roots at 27 and 76 DAT. f: Stems at 27 and 76 DAT. g: Leaf blades at 27, 76, and 125 DAT. h: The anther size was 0.3–0.6, 0.7–1.0, 1.2–1.5, and 1.6–2.0 mm. i: Embryos 1, 3, 5, and 7 days after flowering. j: Ovaries 7, 10, 14, 28, and 42 days after flowering. k: Endosperm, 7, 10, 14, 28, and 42 days after flowering. l: The pistil size was 5–10, 10–14, and 14–18 mm (n = 3). Error bars represent the SD.

Full size image

We grew WT and osvit2 plants under different Fe concentrations. Under Fe-deficient conditions, no difference was observed in shoot length, root length, or chlorophyll content (SPAD value; Figure 2a, d, and g), while at 100 and 500 μΜ Fe-EDTA, shoot growth was significantly retarded compared to WT plants (Figure 2b and c), whereas the root length and SPAD value remained unchanged (Figure 2e, f, h, and i). We also measured the concentrations of Fe, Zn, manganese (Mn), and copper (Cu) in the roots and shoots of WT and osvit2 plants. No difference was observed between WT and osvit2 shoots for all of these metals when the plants were grown in the absence of Fe (Figure 3a, g, m, and s). When plants were grown in the presence of 100 μΜ Fe-EDTA, the concentrations of Fe, Zn, and Mn decreased significantly in osvit2 compared to WT plants. The concentration of Cu also decreased. When plants were grown under Fe excess conditions, the concentrations of Fe, Mn, and Cu decreased significantly in the shoots of osvit2 plants. The roots of osvit2 plants accumulated more Fe and Zn when grown under Fe-deficient conditions (Figure 3d and j), while the metal concentration in osvit2 roots was comparable to that in WT plants following growth with 100 or 500 μM Fe-EDTA (Figure 3e, f, k, l, p–r, and v–x). Pearl staining analysis showed that osvit2 plants accumulated more Fe in their embryos compared to WT plants (Figure 4a and b). Mutants for AtVIT1 (Kim et al. 2006) and OsVIT1 and OsVIT2 (Zhang et al. 2012) have been reported to have disturbed Fe accumulation in seeds.

Figure 2
figure 2

Morphological characteristics of osvit2 plants grown at different Fe concentrations. a–c: Shoot length. d–f: Root length. g–i: SPAD value. a, d, and g: –Fe (0 μM Fe-EDTA). b, e, and h: +Fe (100 μM Fe-EDTA). c, f, and i: ++Fe (500 μM Fe-EDTA). Vertical bars followed by an asterisk indicate a significant difference from the WT according to the Tukey–Kramer test (p < 0.05; n = 4).

Full size image
Figure 3
figure 3

Metal concentration of osvit2 plants grown at different Fe concentrations. a–c: Shoot Fe concentration. d–f: Fe concentration. g–i: Shoot Zn concentration. j–l: Root Zn concentration. m–o: Shoot Mn concentration. p–r: Root Mn concentration. s–u: Shoot Cu concentration. v–x: Root Cu concentration. a, d, g, j, m, p, s, and v: –Fe (0 μM Fe-EDTA). b, e, h, k, n, q, t, and w: +Fe (100 μM Fe-EDTA). c, f, i, l, o, r, u, and x: ++Fe (500 μM Fe-EDTA). Vertical bars followed by an asterisk indicate a significant difference from the WT according to the Tukey–Kramer test (p < 0.05; n = 4).

Full size image
Figure 4
figure 4

Localization of Fe in WT, osvit2 , mit-2 , and mit-2:MIT lines. a: WT DJ. b: osvit2. c: WT DJ. d: mit-2. e: mit-2 complemented with the MIT open reading frame.

Full size image

We previously reported that in mit-2 plants, the expression of OsVIT2 was upregulated (Bashir et al. 2011c), so we assessed whether the concentration of Fe and other metals also changed in mit-2 seeds. mit-2 seeds accumulated less Fe in the embryo compared to WT plants, and in mit-2 lines complemented with MIT, the localization of Fe was comparable to that in WT plants. We further measured the concentration of different metals in mit-2 and osvit2 seeds. The concentrations of Fe, Zn, and Cu increased in osvit2 seeds, while that of Mn decreased (Figure 5). As we previously reported that the expression of OsVIT2 was upregulated in mit-2 plants (Bashir et al. 2011c), we analyzed mit-2 seeds for changes in metal accumulation. The concentration of Fe was higher in leaves harvested from mit-2 plants. A knockout line of osvit2 was previously shown to accumulate an increased amount of Fe in its seeds (Zhang et al. 2012); however, in that report, the authors did not analyze the metal concentration in polished rice. As the embryo is removed during milling, leaving the endosperm as the only edible part, we measured the concentration of Fe and other metals in polished rice seeds (rice endosperm). In polished osvit2 seeds, the concentrations of Fe, Zn, and Cu were significantly elevated compared to WT seeds (Figure 5e–g). Rice is an important agronomical crop and is used as a staple food by approximately half of the world’s population. Rice is poor in nutrients such as Fe, and people who depend on rice as a staple food often suffer from Fe deficiency (Bashir et al. 2013b; Bashir et al. 2010). Fe and Zn deficiencies cause 0.8 million deaths annually, while the number of people suffering from these deficiencies is up to 2 billion (World Health Organization 2003). Thus, breeding rice plants that are capable of accumulating more Fe and Zn in the endosperm is important (Bashir et al. 2012; Bashir et al. 2013b). These results indicate that the knockout/knockdown of osvit2 may be utilized for biofortification programs. In mit-2 seeds, Fe accumulation was significantly lower compared to WT plants (Figure 6a), while the concentration of other metals did not change significantly (Figure 6a–d). Polished mit-2 seeds also accumulated significantly less Fe, while they accumulated more Zn compared to WT seeds (Figure 6e and f); no difference was observed in other metals. OsVIT2 overexpression lines accumulated less Fe in their seeds, and in mit-2, the reduction in Fe accumulation may be caused by increased OsVIT2 expression. Signaling between different subcellular organelles (Vigani et al. 2013) may be responsible for changes in metal localization in rice seeds. These results suggest that subcellular Fe transporters affect seed Fe localization; thus, it may be possible to regulate the expression of these transporters to biofortify rice with Fe and Zn without causing any adverse effects on plant growth and development.

Figure 5
figure 5

Measurement of the metal concentration in WT and osvit2 brown rice and polished seeds. a–d: Brown rice. e and f: Polished rice. a and e: Fe concentration. b and f: Zn concentration. c and g: Cu concentration. d and h: Mn concentration. Vertical bars followed by an asterisk indicate a significant difference from the WT according to the Tukey–Kramer test (p < 0.05; n = 4).

Full size image
Figure 6
figure 6

Measurement of the metal concentration in WT and mit-2 brown rice and polished seeds. a–d: Brown rice. e and f: Polished rice. a and e: Fe concentration. b and f: Zn concentration. c and g: Cu concentration. d and h: Mn concentration. Vertical bars followed by different letters are significantly different from each other according to the Tukey–Kramer test (p < 0.05; n = 4).

Full size image