- Email: [email protected]
A decrease in phytic acid content substantially affects the distribution of mineral elements within rice seeds
Plant Science 238 (2015) 170–177
Contents lists available at ScienceDirect
Plant Science journal homepage: www.elsevier.com/locate/plantsci
A decrease in phytic acid content substantially affects the distribution of mineral elements within rice seeds Hiroaki Sakai a , Toru Iwai b , Chie Matsubara b , Yuto Usui b , Masaki Okamura b , Osamu Yatou c , Yasuko Terada d , Naohiro Aoki b , Sho Nishida b , Kaoru T. Yoshida b,∗ a
Faculty of Agriculture, The University of Tokyo, Bunkyo-ku, Tokyo, Japan Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo-ku, Tokyo, Japan c Crop Development Division, NARO Agricultural Research Center, Inada, Joetsu, Niigata, Japan d Super Photon Ring-8, Japan Synchrotron Radiation Research Institute, Sayo-cho, Sayo-gun, Hyogo, Japan b
a r t i c l e
i n f o
Article history: Received 1 April 2015 Received in revised form 30 May 2015 Accepted 6 June 2015 Available online 14 June 2015 Keywords: Low phytic acid grains Mineral elements Mutant Oryza sativa L. Phosphorus Seed storage substances
a b s t r a c t Phytic acid (myo-inositol hexakisphosphate; InsP6 ) is the storage compound of phosphorus and many mineral elements in seeds. To determine the role of InsP6 in the accumulation and distribution of mineral elements in seeds, we performed fine mappings of mineral elements through synchrotron-based X-ray microfluorescence analysis using developing seeds from two independent low phytic acid (lpa) mutants of rice (Oryza sativa L.). The reduced InsP6 in lpa seeds did not affect the translocation of mineral elements from vegetative organs into seeds, because the total amounts of phosphorus and the other mineral elements in lpa seeds were identical to those in the wild type (WT). However, the reduced InsP6 caused large changes in mineral localization within lpa seeds. Phosphorus and potassium in the aleurone layer of lpa greatly decreased and diffused into the endosperm. Zinc and copper, which were broadly distributed from the aleurone layer to the inner endosperm in the WT, were localized in the narrower space around the aleurone layer in lpa mutants. We also confirmed that similar distribution changes occurred in transgenic rice with the lpa phenotype. Using these results, we discussed the role of InsP6 in the dynamic accumulation and distribution patterns of mineral elements during seed development. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Phytic acid, myo-inositol 1,2,3,4,5,6-hexakisphosphate (InsP6 ), is a hexaphosphoric ester of myo-inositol. InsP6 serves as a phosphorus (P) storage substance in plants. In plant seeds, about 75 ± 10% of total P is accumulated as InsP6 [1], and InsP6 is deposited primarily as mixed salts of various mineral cations such as magnesium (Mg), potassium (K), calcium (Ca), manganese (Mn), iron (Fe), and zinc (Zn) [2,3] because InsP6 has six negatively charged phosphate groups and can chelate strongly with the cations to form insoluble salts; i.e., phytate.
Abbreviations: DAF, days after flowering; EMS, ethyl methanesulfonate; ICP–OES, inductively coupled plasma optical–emission spectrometry; InsP6 , myoinositol 1,2,3,4,5,6-hexakisphosphate; lpa, low phytic acid; -XRF, micro X-ray fluorescence; NT, non-transformant; P, phosphorus; Pi, inorganic phosphate; WT, wild type. ∗ Corresponding author. Tel.: +81 3 5841 8086; fax: +81 3 5841 1306. E-mail address: [email protected] (K.T. Yoshida). http://dx.doi.org/10.1016/j.plantsci.2015.06.006 0168-9452/© 2015 Elsevier Ireland Ltd. All rights reserved.
Cereal grains are utilized not only as staple food items for humans but also as feeds for livestock. However, P in the form of InsP6 is not readily available to monogastric non-ruminant animals, such as humans, pigs, and poultry, because they lack phytase, the enzyme that degrades InsP6 [4]. To provide the optimal level of P for animal growth, feeds have traditionally been supplemented with inorganic phosphate (Pi) [5]. Consequently, non-ruminants excrete large amounts of unabsorbed P derived from InsP6 and surplus Pi in their manure. This leads to the accumulation of P in soil and water, and subsequently to the eutrophication of water areas [5,6]. In addition, non-ruminants do not easily absorb metal elements from seeds because they bind to InsP6 and form phytate [7]. To decrease the environmental loading and to improve the utilization of P and mineral cations, the development of cultivars that contain low levels of phytate in their seeds is a major breeding objective [5]. Several low phytic acid (lpa) mutants, whose seed phytate is reduced by 50–95%, have been isolated in barley [8,9], rice [10,11], wheat [12], maize [13–15], and soybean [16]. For transgenic plants, Shi et al. [17] succeeded in producing transgenic plants with a lowphytate and high-Pi phenotype in maize and soybean through the gene silencing of InsP6 transporter. Kuwano et al. also generated
H. Sakai et al. / Plant Science 238 (2015) 170–177
transgenic rice, in which seed phytate was reduced by 68% and freely available Pi was concomitantly increased through the antisense repression of 1D-myo-inositol 3-phosphate synthase gene [18], which catalyzes the first step in phytic acid biosynthesis and inositol metabolism. Nutrition studies of livestock and human beings have confirmed that low-phytate seeds improve the availability of both P and metals such as Fe and Ca, and reduce P excretion in manure [19,20]. However, the distribution of these metal elements in lowphytate seed is not fully understood. To compare the content and distribution of metal elements in seeds between lpa and wild-type (WT) plants, inductively coupled plasma optical–emission spectrometry (ICP–OES) analyses were conducted using milled rice [21,22]. Although no marked effects on the amount of metal cations were observed, there was a trend for the concentrations of several cations to increase (e.g., K and Mg) in milled lpa seeds compared with milled WT seeds. These results suggest that the lpa phenotype affects the distribution of mineral cations in a seed. However, the precise locations of elements are not clear in studies that used milled grains. Synchrotron-based micro X-ray fluorescence (-XRF) imaging at the Super Photon Ring-8 (SPring-8) facility is non-destructive and can provide elemental mapping images with high resolution; i.e., subcellular localizations of elements, as shown in our previous study [23]. Using -XRF imaging analysis, we have strongly suggested that not all storage mineral cations in a rice seed existed as phytate, and that Zn and copper (Cu) accumulated as a storage form other than phytate at least in the endosperm, because a significant amount of Zn and Cu did not colocalize with P in the endosperm tissues [23]. In contrast, K, Ca, and Fe were colocalized with P, and might primarily accumulate as phytate in the aleurone layer throughout seed development. In this study, we conducted -XRF analysis on the cross sections of rice seeds of the WT, two independent lpa mutants, and one lpa transformant, to determine the differences in the dynamic changes in the distribution of minerals between WT and low-phytate seeds, and to estimate the role of InsP6 in the metal distribution. The nutritional value of the lowphytate seed in relation to the distribution pattern and the storage form of the elements was then considered.
2. Materials and methods 2.1. Plant materials We identified two lpa mutants from an M2 population of rice (Oryza sativa L. var japonica, cv. Koshihikari) mutagenized with ethyl methanesulfonate (EMS), the seeds of which had a high Pi content, because a high Pi content is an indicator of low InsP6 content [5]. Self-pollinated M3 seeds of each putative mutant were planted to confirm the lpa phenotype in the subsequent generation. The two mutants were crossed with indica rice (cv. Kasalath) for rough mapping. The F2 populations segregated into the wild and lpa types at a ratio of 3:1, indicating that each mutant phenotype is caused by a single recessive mutation. Genomic DNA was isolated individually from these mutants and analyzed using PCR-based genetic markers to identify the molecular markers linked to the lpa phenotypes. The primers used for the molecular markers were as follows: a derived cleaved amplified polymorphic sequence (dCAPS) corresponded to marker 3-1, 5 -TCTAATACCATTTTTCAACGAAGC3 and 5 -TAATCAAGTTATTTGCTGCG-3 ; the other markers are shown at http://rgp.dna.affrc.go.jp/E/publicdata/caps/index.html. LPA2623 was mapped roughly to the 5.8 cM region located between the C63223 and C60710 markers on the long arm of chromosome 2 and LPA3847 to the 2.4 cM region located between the 3-1 and R2247 markers on the short arm of chromosome 3 (Fig. S1). An allelism test also revealed that the loci of the two mutants,
171
designated lpa2623 and lpa3847, were distinct from each other. OsPGK1 and OsMRP13/OsABCC13 genes, which were responsible for low-phytate phenotype [24,25], were located in the mapped LPA2623 and LPA3847 regions, respectively (Fig. S1). The mutant and WT plants were grown in pots in a greenhouse under natural light conditions in summer. Only superior caryopses were used for all experiments, because the growth of inferior caryopses was retarded [26]. To produce transgenic lpa rice, a gene consisting of 18kDa oleosin promoter and rice myo-inositol 3-phosphate synthase (RINO1) [27] cDNA in the antisense orientation has been introduced into japonica rice (cv. Kitaake). The detailed methods of transformation, selection of a stable transformant, and growth conditions were described previously [18].
2.2. Seed Pi measurement To analyze Pi levels, crushed powdery whole seed was extracted with 12.5% (w/v) trichloroacetic acid containing 25 mM MgCl2 . The Pi concentration was colorimetrically determined as described by Chen et al. [28]. The detailed method was described previously [29].
2.3. Ion chromatography analysis Mature rice seeds were dehusked, and dried for 48 h at 60 ◦ C. The dried seed was weighed and crushed with a hammer. The crushed samples were ground with a mortar and pestle and then homogenized in 2.4% (w/v) HCl. The HCl extracts were diluted 40-fold with deionized water and then subjected to ion chromatography. The details of the extraction and ion chromatography method were described previously [29]. More than three independent seeds were used in the experiment.
2.4. Quantitative analysis of starch Seed starch contents were measured according to the method of Okamura et al. [30]. Two to three dehusked mature seeds were dried for 48 h at 60 ◦ C and then ground into powders using a MultiBeads Shocker (Yasui Kikai, Osaka, Japan). The powdered samples were extracted twice with 80% (v/v) ethanol for 20 min at 80 ◦ C, with mixing every 5 min. After centrifugation at 11,000 × g for 5 min, the precipitates were resuspended in distilled water, and boiled for 2 h with mixing every 30 min. The enzymatic method using an F-kit for starch (J.K. International, Tokyo, Japan) was used for the assay. The starch concentration was determined spectrophotometrically at 340 nm. The analysis was repeated at least three times using seeds derived from independent plants.
2.5. ICP–OES analysis Mature seeds were dehusked and dried for 48 h at 60 ◦ C. Six dried seeds were weighed and crushed together using a MultiBeads Shocker (Yasui Kikai), and then digested with 2 mL of 0.08 N HNO3 for 30 min at 80 ◦ C. Subsequently, samples were heated to 120 ◦ C for at least 1 h to evaporate water. Then, 1 mL of H2 O2 was added and the samples were heated sequentially for 30 min at 80 ◦ C, 1 h at 120 ◦ C, and overnight at 80 ◦ C. The samples were then dissolved in 1 mL of 0.08 N HNO3 . Elemental concentrations in the acid-digested samples were analyzed by inductively coupled plasma optical-emission spectrometry (ICP–OES; model SPS3500, Hitachi High-Tech Science Co., Tokyo, Japan) in accordance with the manufacturer’s instructions. The experiment was repeated three times using three independent seed samples.
172
H. Sakai et al. / Plant Science 238 (2015) 170–177
2.6. -XRF analysis Developing seeds of the WT and two mutants at 10, 20, and 30 days after flowering (DAF), and mature seeds (about 50 DAF) of non-transgenic and transgenic plants were collected for elemental mapping by synchrotron-based -XRF analysis using the BL37XU beamline of the SPring-8 facility in Hyogo, Japan. The dehusked seeds were vacuum-freeze dried for 2 days, and 200-m cross sections of the center of seeds were subjected to -XRF analysis. The detailed methods of the -XRF analysis were described previously [23], but here the beam size was 0.6 m in the vertical direction × 0.8 m in the horizontal direction at the sample position when we analyzed the developing seeds. The scanning step size was set to 50 m to obtain elemental maps of cross-sections of whole seeds, or 1 m for elemental maps of the lateral side of seeds. Three independent seeds were used for whole cross-section mappings, and one representative of the three seeds was used for lateral-side line mappings.
3. Results 3.1. Morphological traits of low phytic acid mutants We screened an M2 population of rice (cv. Koshihikari) mutagenized with EMS for increased Pi levels using a molybdate staining assay, because the grains of lpa mutants generally accumulate high levels of Pi instead of InsP6 [5]. We identified two single-gene recessive mutants that had greatly increased Pi levels in seeds, lpa2623 and lpa3847. The allelism test and the rough mapping revealed that these two lpa mutants were non-allelic (Fig. S1). Although there was no significant difference in the germination rate, seedling growth, plant height, and heading date in the two lpa mutants and the WT, the traits of the seeds were significantly different. The weights of dried mature seeds of the mutants were significantly lighter than those of the WT (Fig. 1A). The reduction in seed weight in comparison with the WT was 11% for lpa3847 and 33% for lpa2623. The reduction rates were significantly different between lpa2623 and lpa3847. 3.2. Biochemical components in the seeds of low phytic acid mutants The reduced seed weight of lpa mutants compared to the WT (Fig. 1A) indicated the possibility that the starch content was lower
Fig. 1. Phenotypes of seeds obtained from the wild type (WT) and two lpa mutants. (A) Dry weight of mature seeds. Dehusked seeds were dried for 48 h at 60 ◦ C. Each value represents the mean ± SE (n = 20). Columns with different letters are significantly different (Tukey’s test, P < 0.01). (B) Comparison of starch contents of mature seeds. Values represent the relative fold change of each content per seed compared with the value of WT. Each value represents the mean ± SE (n = 4 for left panel, 3 for right panel). Columns with different letters are significantly different (t-test, P < 0.05 for left panel, P < 0.01 for right panel).
in the mutant seeds, because starch is the main component of seeds. The starch measurements revealed that the starch content of seeds decreased significantly in both lpa mutants in comparison with the WT (Fig. 1B). The reduction rates were 9.22% for lpa3847 and 41.9% for lpa2623. Thus, the reduction rate of seed weight was correlated with the reduction rate of starch content. The InsP6 and Pi content of mature seeds of the WT and lpa mutants was determined by ion-chromatographic analysis and a colorimetric molybdate assay, respectively (Fig. 2A and B). The seed InsP6 content decreased significantly in both lpa mutants in comparison with the WT. The reduction rate of the InsP6 content of lpa2623 (74.7%) was higher than that of lpa3847 (66.1%) (Fig. 2A). From the chromatographic analysis, the accumulation of any other inositol phosphates in mature seeds of both lpa mutants was not detected, as in the seeds of the WT (data not shown). In contrast to InsP6 , the Pi content of both mutants increased by 16.1-fold for lpa3847 and 7.67-fold for lpa2623 (Fig. 2B). The total P content of seeds of the WT, lpa3847, and lpa2623 was 31.4, 30.6, and 28.8 g/seed, respectively (Fig. 2C). These values were not significantly different. To confirm if the decrease of InsP6 in the mutant seeds affected the quantity of mineral cations in seeds, the K, Fe, Zn, and Cu contents of mature seeds were measured by ICP–OES (Fig. 3). The contents of all four metal elements in the seeds of lpa2623 and lpa3847 were identical to those of the WT. The translocation of these elements from vegetative organs into developing seeds was therefore unaffected by the change in the storage form of phosphorus. 3.3. Mineral localizations in the developing seeds of low phytic acid mutants During seed development, P and metal elements were actively translocated from vegetative organs into seeds until 14 DAF, and the translocation of most elements was completed in the period from 21 to 28 DAF [23]. Therefore, in this study, seeds at 10, 20, and 30 DAF were subjected to elemental mapping by -XRF analysis with a 50-m step size. We determined the distribution patterns of P, K, Fe, Zn, and Cu in the cross sections of the center of developing rice seeds (Fig. 4). In the WT seeds, as in a previous study using Kitaake rice seeds [23], P, K, and Fe were mainly localized in the aleurone layer (Fig. 4). P, K, and Fe levels clearly increased in the aleurone layer from 10 to 20 DAF. P and K were most abundant in the lateral to ventral sides of the aleurone layer at 20 and 30 DAF in the WT. At any stage of seed development, Fe in the WT seed was mainly localized in the dorsal side of the aleurone layer adjacent to vascular bundles, which are the gates of translocation from vegetative organs (Fig. 4, Fig. S2). In comparison with P, K, and Fe, Zn and Cu were broadly distributed around the aleurone layer of the WT seed and the signals were also detected in the inner endosperm during seed development (Fig. 4). The accumulation of Zn around the aleurone layer of the WT progressed substantially from 10 to 20 DAF. Cu was most abundant around the lateral side of the aleurone layer in the WT seeds at any stage of seed development, and actively spread into the endosperm from 10 to 20 DAF. In both lpa mutants, the signals of P and K in the aleurone layer were weaker than in the WT at all stages of seed development (Fig. 4). This means that large amounts of P and K diffused into the inner endosperm in both lpa mutants, because the amounts of P and K in lpa seeds were identical to those in the WT seeds (Figs. 2C and 3). The accumulation and distribution patterns of Fe in the seeds of both lpa mutants were similar to those in the seeds of the WT (Fig. 4). However, the region where Fe was abundant in both lpa mutants displayed a tendency to expand from the dorsal side to
H. Sakai et al. / Plant Science 238 (2015) 170–177
173
Fig. 2. InsP6 (A), Pi (B), and total phosphorus (C) levels in mature seeds obtained from the wild type (WT) and two lpa mutants. (A) InsP6 contents were determined by ion-chromatography (n 3). (B) Pi contents were measured by colorimetric assay using molybdate staining (n = 5). (C) Total phosphorus contents were measured by ICP–OES analysis (n = 3). Each value represents the mean ± SE. Columns with the same letters are not significantly different (Tukey’s test, P < 0.01 for (A and B), P < 0.05 for (C)).
Fig. 3. The K, Fe, Zn, and Cu contents of mature seeds from the wild type (WT) and two lpa mutants. The mineral contents were measured by ICP–OES analysis. Each value represents the mean ± SE of three replicates. Columns with the same letters are not significantly different (Tukey’s test, P < 0.05).
the lateral and ventral direction in the aleurone layer after 10 DAF (Fig. 4). This tendency was more pronounced in lpa2623 than in lpa3847. The Zn and Cu signals, which were broadly diffused from the aleurone layer to the endosperm in WT seeds, were
concentrated around the aleurone layer in both lpa mutants (Fig. 4). Zn in the lpa seeds increased around the aleurone layer from 10 to 20 DAF and was concentrated as seed development proceeded. Although the progress of Cu accumulation in developing seeds differed slightly between lpa2623 and lpa3847 seeds, the concentration of Cu around the aleurone layer was prominent in both lpa mutants at 30 DAF. The accumulation of Cu in the aleurone layer was slightly more apparent in lpa2623 than in lpa3847. To compare the precise distributions of Zn and Cu around the aleurone layer in the WT and lpa mutants, the step size of the -XRF analysis was changed from 50 to 1 m. We analyzed the lateral sides of the cross sections of 30-DAF seeds of the WT and lpa mutants (Fig. 5). From a distribution map of P in a seed, we estimated the area of the aleurone layer because the concentrated localization of the P in the aleurone layer was observed in both WT and lpa mutants (Fig. 4). As an example, the distribution map of P of the WT seed is shown in Fig. 5A. We determined the maximum detection limit for the normalized X-ray fluorescence intensity in the elemental map of P as the fluorescence intensities of most of the points in the aleurone layer exceed the limit. Using this limit, the aleurone layer was displayed as a white color in the distribution map of P, as seen in Fig. 5B. From the determination of the boundaries of the aleurone layer, the estimated widths of the aleurone layers of the WT, lpa2623, and lpa3847 were 67, 65, and 77 m, respectively. These values did not deviate from the general width of the aleurone layer of a rice seed. We then drew a line from the seed coat to the inner endosperm (red arrow in Fig. 5B). We performed a line mapping, which defined a series of fluorescence intensities that were obtained in one direction. To compare the line mapping of elements in the WT and in two lpa mutants, the widths of the aleurone layers of the WT and mutants were adjusted. The normalized fluorescence intensities of P, Zn, and Cu on the line from the seed coat to the endosperm in the lateral side of the seed are displayed in Fig. 5C. The intensities of P in the aleurone layers of both lpa mutants were lower than in the WT, as shown in Fig. 4. The intensities of Zn and Cu in both lpa mutants were higher than in the WT in both the aleurone layer and the endosperm adjacent to the aleurone layer (subaleurone layer) (Fig. 5C). The distribution of Cu was substantially different in the lpa mutants. In the WT seed, the area with the most abundant Cu was in the subaleurone layer. In contrast, Cu was abundant in the aleurone layer, and in the boundaries between the aleurone layer and the endosperm of the lpa2623 and the lpa3847 seeds. This tendency was more pronounced in lpa2623 than in lpa3847.
174
H. Sakai et al. / Plant Science 238 (2015) 170–177
Fig. 4. Elemental maps of P, K, Fe, Zn, and Cu in the cross sections of developing seeds at 10, 20, and 30 days after flowering (DAF). Normalized X-ray fluorescence intensities are scaled from red (maximum) to blue (minimum) for individual elements. Normalized intensities exceeding maximum limits of scale bars are displayed as white. To generate the maps, a step size of 50 m was used. Dorsal side is at the top in all figures. Maps representative of three independent seeds are shown. Bars = 300 m. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
3.4. Mineral localization in the seeds of low-phytic-acid transgenic rice We have generated transgenic rice plants carrying a rice 1Dmyo-inositol-3-phosphate synthase (RINO1) antisense gene, which codes the enzyme that catalyzes the first step in InsP6 biosynthesis and inositol metabolism, driven by an 18-kDa oleosin promoter, which promotes expression in the embryo and aleurone layer [18]. The reduction rate of InsP6 of the lpa transformant was 68% compared to that of non-transformant (NT) (cv. Kitaake). We conducted -XRF analysis on the cross sections of the center of mature seeds of the transgenic lpa rice and NT rice. The concentrations of P and four metal elements (K, Fe, Zn, and Cu) in the seed of the transformant were similar to those of the NT (Fig. S3). The changes in the distribution patterns of these five mineral elements in the lpa transformant (Fig. 6) were consistent with those observed in lpa
mutants at 30 DAF (Fig. 4). In the lpa transformant, the P and K signals in the aleurone layer were weaker than those of the NT. The abundant region of Fe in the seed of the transformant expanded to the lateral and ventral directions in the aleurone layer in comparison with the NT. The Zn and Cu signals were concentrated around the aleurone layer in the lpa transformant. 4. Discussion The fine -XRF imaging analysis revealed that P, K, and Fe in the WT (cv. Koshihikari) accumulated mainly in the aleurone layer during seed development, and Zn and Cu were broadly distributed from the aleurone layer to the inner endosperm. These distribution patterns of minerals agreed with those in our previous study using another cultivar, Kitaake [23]. Ogawa et al. investigated the patterns of mineral accumulation within the aleurone layers during
H. Sakai et al. / Plant Science 238 (2015) 170–177
175
Fig. 6. Elemental maps of P, K, Fe, Zn, and Cu in cross sections of mature seeds from non-transgenic rice (NT) and transgenic lpa rice (RINO1). Normalized X-ray fluorescence intensities are scaled from red (maximum) to blue (minimum) for individual elements. To generate the maps, a step size of 50 m was used. Dorsal side is at the top in all figures. Maps representative of three independent seeds are shown. RINO1: transgenic rice, in which antisense RINO1 gene was expressed under the control of the 18-kDa oleosin promoter. Bars = 300 m. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Elemental distributions of P, Zn, and Cu around the aleurone layer on the lateral side of a seed cross section at 30 days after flowering (DAF). (A and B) Elemental maps of P in a wild-type (WT) seed. Maximum detection limits of X-ray fluorescence intensity are 120 (A) and 22 (B). Normalized X-ray fluorescence intensities are scaled from red (maximum) to blue (minimum). Normalized intensities exceeding maximum limits of the scale bars are displayed as white. To generate the map, a step size of 1 m was used. Dorsal side is at the top in the figures. Bars = 20 m. Coat: seed coat, Aleu: aleurone layer, Endo: Endosperm. (C) Elemental line maps of P, Zn, and Cu from seed coat to endosperm in the lateral side of seed at 30 DAF. Two vertical lines in each figure represent both sides of the boundaries of the aleurone layer. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
rice seed development using an electron microprobe X-ray analyzer and reported that the accumulation of K in aleurone cells was delayed relative to that of P and Mg and was not observed until 19 DAF [31]. In our studies, however, K accumulation was not delayed and we detected K signals in aleurone cells at 10 DAF [23]. This inconsistency is probably due to differences in the sensitivity of the methods used. Since synchrotron-based -XRF analysis is a very sensitive method compared to other methods, we could detect a trace accumulation of K at the earlier stage. Using -XRF imaging analysis, we further revealed the dynamic changes in the mineral distributions of two independent lpa mutants and an lpa transformant in this study. The P content and the levels of four metal elements (K, Fe, Zn, and Cu) in the seeds of both lpa2623 and lpa3847 mutants, with 77.4 and 66.1% reduction in seed InsP6 , respectively, were the same as in the WT seeds (Figs. 2C and 3). Liu et al. also showed that there was no major difference in the total contents of P, K, Mg, Ca, Mn, Fe, and Zn between the seeds of the WT and an lpa1-1 mutant with a 45% reduction in seed phytate [32]. These results indicate that the storage form of phosphorus does not affect the translocation of minerals from vegetative organs into developing seeds, and that the changes in the distribution patterns of minerals observed in lpa seeds were not due to the changes in the amounts of minerals. The chromosome rough mapping of the lpa2623 and lpa3847 mutants revealed that the mutations occurred at two different loci, which were predicted to be in a different region from the RINO1 gene
176
H. Sakai et al. / Plant Science 238 (2015) 170–177
locus (Fig. S1). Therefore, the similar changes in the distribution patterns of minerals observed in the two mutants and the transformant might be caused by the low-phytate phenotype. However, we could not exclude the possibility that the changes in the mineral distribution were due in part to the reduction in the weights of lpa seeds (Fig. 1) and that the reduction in starch, which accumulates in the endosperm in rice, affected the mineral distribution. The accumulations of P and K in the aleurone layer of both lpa mutants were considerably less than those in the WT at all stages of seed development (Fig. 4). This tendency was also observed in the seeds of the lpa transformant. In a previous study, we concluded that translocated Pi was immediately synthesized into InsP6 and then accumulated in the aleurone layer cells of rice seeds [23]. In this study, because the amounts of P and K were identical to those in the WT (Figs. 2C and 3), and because P and K strong signals were not detected in the endosperm, P and K likely diffuse uniformly and thinly in the endosperm of lpa mutants. After translocation from vegetative organs, Pi that was not used for InsP6 biosynthesis in lpa mutants might immediately diffuse into the endosperm (Fig. S2). Also, K that was not captured by the reduced InsP6 might immediately diffuse into the endosperm. A decrease in the P and K concentrations in the bran of lpa mutants was also observed in rice and barley using milling products [21,33]. However, Ren et al. reported that concentrations of K, Ca, Zn, Mg, Fe, and Cu increased in milled rice (endosperm) of an lpa mutant, even though the distribution of P was not altered [22]. All of these studies used milled products, and therefore the precise distribution pattern of elements could not be analyzed. In this study, for the first time, the precise positions of the elements in lpa seeds could be determined using -XRF analysis. In contrast to K, the distribution pattern of Fe in seeds of both the lpa mutants and the lpa transformant were similar to those of the WT and NT, except that the region of Fe abundance in the aleurone layer of low-phytate seeds had a tendency to expand to the lateral and ventral directions (Figs. 4 and 6). The difference in the change in the distribution of Fe and K in lpa seeds (i.e., a minor change for Fe and a large change for K) may indicate a difference in the affinity of Fe and K for InsP6 . In the aleurone layer of the WT seeds, Fe was abundant in the dorsal side where vascular bundles existed; i.e., the translocation sites (Fig. S2)[34], while K was abundant in the lateral to ventral side of the aleurone layer, which is distant from the translocation sites. This suggests that Fe is captured by InsP6 immediately after its translocation into a seed. In contrast, in the aleurone layer K likely moves from the dorsal to the lateral and ventral sides before being captured by InsP6 (Fig. S2) [34]. Even in a low-phytate seed, the reduced InsP6 may preferentially bind to Fe, and K, having a lower affinity for InsP6 than Fe, then diffuses into the endosperm from the aleurone layer without forming a phytate salt (Fig. S2). The increased accumulations of Fe on the lateral and ventral sides of the aleurone layer of lpa mutants and the lpa transformant compared to the WT and NT occur because the reduced amount of InsP6 on the dorsal side of the aleurone layer could not capture all of the translocated Fe. In both the lpa mutants and the lpa transformant, the amount of Zn and Cu in the aleurone and subaleurone layers increased compared to the WT and NT (Figs. 4–6). Iwai et al. suggested that Zn is loosely bound and Cu is scarcely bound to InsP6 in rice WT seeds, and that large amounts of Zn and Cu may exist in a storage form other than phytate in the endosperm [23]. Iwai et al. also reported that Zn and Cu moved from the aleurone layer to the endosperm as seed development progressed [23]. During seed development, the aleurone cells are gradually filled with globoids and oil bodies, and therefore, metal elements that do not form phytate salt will be gradually extruded from the aleurone layer to the endosperm. A reduction in globoid size was observed in the aleurone layer of lpa seeds in rice, barley, and corn [32,35,36]. Because the estimated
width of the aleurone layers of the two lpa mutants were similar to that of WT, it is plausible that Zn and Cu in the lpa seeds were not moved out into the endosperm, but remained in the spaces between small globoids in the aleurone layer. As a result, the Zn and Cu signals in the aleurone layer were considerably higher in lpa seeds than in WT seeds (Figs. 4 and 5). A provocative question is why K that did not form phytate in low-phytate seeds did not remain in the aleurone layer as did Zn and Cu, but diffused uniformly into the endosperm. It is suggested that Zn in grains is mainly bound to S; i.e., it forms a metal thiolate complex with peptides and/or proteins [37,38], which is stored around the aleurone layer. Divalent and trivalent cations, such as Zn, can form a bridge between the negative charges of thiolate ligands. However, K cannot form a bridge between thiolate ligands because it is a monovalent cation; therefore, free K ions may diffuse promptly and uniformly into the endosperm of low-phytate seeds. Lpa2623 displayed a more severe phenotype than lpa3847, including greater changes in the distribution patterns of Fe and Cu. The lower concentration of InsP6 and/or starch in lpa2623 might result in greater changes in the distribution patterns of Fe and Cu in comparison with the lpa3847. Rose et al. examined the genetic variation of P partitioning in rice plants and showed that the partitioning of P between grain and straw at maturity (P harvest index) was highly correlated with the harvest index [39]. In comparison, in our study, the mineral contents, including P, of lpa seeds were not affected by the reduction in grain yields, which was due mainly to the reduction in starch content. This suggests that carbon loading and mineral loading into a seed are controlled by different regulatory mechanisms and that mineral nutrient contents might be improved without affecting grain yield. Reducing the phytate content of grains could enhance mineral absorption from mixed diets [40]. We found that reducing the InsP6 content of rice seeds did not simply result in all mineral elements diffusing into the endosperm without forming phytate. Fe might be bound to InsP6 even in low-phytate seeds, and hence low-phytate rice did not greatly increase Fe utilization. In addition, there was a dynamic change in the distribution patterns of many mineral elements in low-phytate seeds. Therefore, the mineral content may differ between the WT and mutants whether brown or white rice (milled rice) is used. Attention should be paid to both the storage form and the change in the amount of mineral elements in individual parts of the low-phytate seed. In conclusion, to our knowledge, this is the first study to identify the precise dynamics of mineral elements in low-phytate seeds during seed development by utilizing synchrotron-based -XRF fluorescence imaging. We showed differences in the affinity of each element for InsP6 , and the role of InsP6 in the accumulation pattern of metal elements. To understand the dynamics of mineral elements in seeds more precisely, the chemical forms of P and metal elements during seed development should be determined.
Acknowledgments The ICP–OES analysis was supported by Japan Advanced Plant Science Network. The -XRF analysis was performed at the BL37XU in the SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2010B1695 and No. 2013B1209). The authors are grateful to Dr. M. Takahashi, Ms. H. Aihara, Mr. N. Morishita, Mr. Y. Tagashira, Mr. G. Hayakawa, and the members of Laboratory of Plant Nutrition and Physiology, Utsunomiya University for the technical assistance for the -XRF analysis. We are grateful to Ms. T. Shimizu for her help in rough mapping of lpa2623 mutant.
H. Sakai et al. / Plant Science 238 (2015) 170–177
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.plantsci.2015.06. 006 References [1] V. Raboy, myo-Inositol-1,2,3,4,5,6-hexakisphosphate, Phytochemistry 64 (2003) 1033–1043. [2] J.N.A. Lott, J.S. Greenwood, G.D. Batten, Mechanisms and regulation of mineral nutrient storage during seed development, in: J. Kigel, G. Galili (Eds.), Seed Development and Germination, Marcel Dekker, New York, 1995, pp. 215–235. [3] L. Bohn, L. Josefsen, A.S. Meyer, S.K. Rasmussen, Quantitative analysis of phytate globoids isolated from wheat bran and characterization of their sequential dephosphorylation by wheat phytase, J. Agric. Food Chem. 55 (2007) 7547–7552. [4] P.B. Holm, K.N. Kristiansen, H.B. Pedersen, Transgenic approaches in commonly consumed cereals to improve iron and zinc content and bioavailability, J. Nutr. 132 (2002) 514S–516S. [5] V. Raboy, Approaches and challenges to engineering seed phytate and total phosphorus, Plant Sci. 177 (2009) 281–296. [6] L. Bohn, A.S. Meyer, S.K. Rasmussen, Phytate: impact on environment and human nutrition. A challenge for molecular breeding, J. Zhejiang Uni. Sci. B 9 (2008) 165–191. [7] J.N.A. Lott, Accumulation of seed reserves of phosphorus and other minerals, in: D.R. Murray (Ed.), Seed Physiology, Development, vol. 1, Academic Press, Sydney, 1984, pp. 139–166. [8] S.R. Larson, K.A. Young, A. Cook, T.K. Blake, V. Raboy, Linkage mapping of two mutations that reduce phytic acid content of barley grain, Theor. Appl. Genet. 97 (1998) 141–146. [9] S.K. Rasmussen, F. Hatzack, Identification of two low-phytate barley (Hordeum vulgare L.) grain mutants by TLC and genetic analysis, Hereditas 129 (1998) 107–112. [10] S.R. Larson, J.N. Rutger, K.A. Young, V. Raboy, Isolation and genetic mapping of a non-lethal rice (Oryza sativa L.) low phytic acid 1 mutation, Crop Sci. 40 (2000) 1397–1405. [11] Q.L. Liu, X.H. Xu, X.L. Ren, H.W. Fu, D.X. Wu, Q.Y. Shu, Generation and characterization of low phytic acid germplasm in rice (Oryza saiva L.), Theor. Appl. Genet. 114 (2007) 803–814. [12] M. Guttieri, S.J. Barker, L. Perez-Grau, Identification and characterization of a low phytic acid wheat, Crop Sci. 44 (2004) 418–424. [13] V. Raboy, P.F. Gerbasi, K.A. Young, S.D. Stoneberg, S.G. Pickett, A.T. Bauman, P.P.N. Murthy, W.F. Sheridan, D.S. Ertl, Origin and seed phenotype of maize low phytic acid 1-1 and low phytic acid 2-1, Plant Physiol. 124 (2000) 355–368. [14] J. Shi, H. Wang, Y. Wu, J. Hazebroek, R.B. Meeley, D.S. Ertl, The maize low-phytic acid mutant lpa2 is caused by mutation in an inositol phosphate kinase gene, Plant Physiol. 131 (2003) 507–515. [15] J. Shi, H. Wang, J. Hazebroek, D.S. Ertl, T. Harp, The maize low-phytic acid 3 encodes a myo-inositol kinase that plays a role in phytic acid biosynthesis in developing seeds, Plant J. 42 (2005) 708–719. [16] W.D. Hitz, T.J. Carlson, P.S. Kerr, S.A. Sebastian, Biochemical and molecular characterization of a mutation that confers a decreased raffinosaccharide and phytic acid phenotype on soybean seeds, Plant Physiol. 128 (2002) 650–660. [17] J. Shi, H. Wang, K. Schellin, B. Li, M. Faller, J.M. Stoop, R.B. Meeley, D.S. Ertl, J.P. Ranch, K. Glassman, Embryo-specific silencing of a transporter reduces phytic acid content of maize and soybean seeds, Nat. Biotechnol. 25 (2007) 930–937. [18] M. Kuwano, T. Mimura, F. Takaiwa, K.T. Yoshida, Generation of stable ‘low phytic acid’ transgenic rice through antisense repression of the 1D-myo-inositol 3phosphate synthase gene (RINO1) using the 18-kDa oleosin promoter, Plant Biotechnol. J. 7 (2009) 96–105. [19] C. Mendoza, F.E. Viteri, B. Lonnerdal, K.A. Young, V. Raboy, K.H. Brown, Effect of genetically modified, low-phytic acid maize on absorption of iron from tortillas, Am. J. Clin. Nutr. 68 (1998) 1123–1127.
177
[20] T.L. Veum, D.R. Ledoux, D.W. Bollinger, V. Raboy, A. Cook, Low-phytic acid barley improves calcium and phosphorus utilization and growth performance in growing pigs, J. Anim. Sci. 80 (2002) 2663–2670. [21] R.J. Bryant, J.A. Dorsch, K.L. Peterson, J.N. Rutger, V. Raboy, Phosphorus and mineral concentrations in whole grain and milled low phytic acid (lpa) 1-1 rice, Cereal Chem. 82 (2005) 517–522. [22] X.L. Ren, Q.L. Liu, H.W. Fu, D.X. Wu, Q.Y. Shu, Density alteration of nutrient elements in rice grains of a low phytate mutant, Food Chem. 102 (2007) 1400–1406. [23] T. Iwai, M. Takahashi, K. Oda, Y. Terada, K.T. Yoshida, Dynamic changes in the distribution of minerals in relation to phytic acid accumulation during rice seed development, Plant Physiol. 160 (2012) 2007–2014. [24] S.I. Kim, C.B. Andaya, S.S. Goyal, T.H. Tai, The rice OsLpa1 gene encodes a novel protein involved in phytic acid metabolism, Theor. Appl. Genet. 117 (2008) 769–779. [25] X.H. Xu, H.J. Zhao, Q.L. Liu, T. Frank, K.H. Engel, Mutations of multi-drug resistance-associated protein ABC transporter gene 5 result in reduction of phytic acid in rice seeds, Theor. Appl. Genet. 119 (2009) 75–83. [26] M. Kuwano, F. Takaiwa, K.T. Yoshida, Differential effects of a transgene to confer low phytic acid in caryopses located at different positions in rice panicles, Plant Cell Physiol. 50 (2009) 1387–1392. [27] K.T. Yoshida, T. Wada, H. Koyama, R. Mizobuchi-Fukuoka, S. Naito, Temporal and spatial patterns of accumulation of the transcript of myo-inositol-1phosphate synthase and phytin-containing particles during seed development in rice, Plant Physiol. 119 (1999) 65–72. [28] P.S. Chen, T.Y. Toribara, H. Warner, Microdetermination of phosphorus, Anal. Chem. 28 (1956) 1756–1758. [29] M. Kuwano, A. Ohyama, Y. Tanaka, T. Mimura, F. Takaiwa, K.T. Yoshida, Molecular breeding for transgenic rice with low-phytic-acid phenotype through manipulating myo-inositol 3-phosphate synthase gene, Mol. Breed. 18 (2006) 263–272. [30] M. Okamura, T. Hirose, Y. Hashida, T. Yamagishi, R. Ohsugi, N. Aoki, Starch reduction in rice stems due to a lack of OsAGPL1 or OsAPL3 decreases grain yield under low irradiance during ripening and modifies plant architecture, Funct. Plant Biol. 40 (2013) 1137–1146. [31] M. Ogawa, K. Tanaka, Z. Kasai, Accumulation of phosphorus, magnesium and potassium in developing rice grains: followed by electron microprobe X-ray analysis focusing on the aleurone layer, Plant Cell Physiol. 20 (1979) 19–27. [32] J.C. Liu, I. Ockenden, M. Truax, J.N.A. Lott, Phytic acid-phosphorus and other nutritionally important mineral nutrient elements in grains of wild-type and low phytic acid (lpa1-1) rice, Seed Sci. Res. 14 (2004) 109–116. [33] K. Liu, K.L. Peterson, V. Raboy, Comparison of the phosphorus and mineral concentrations in bran and abraded kernel fractions of a normal barley (Hordeum vulgare) cultivar versus four low phytic acid isolines, J. Agric. Food Chem. 55 (2007) 4453–4460. [34] K. Hoshikawa, Anthesis, fertilization and development of caryopsis, in: T. Matsuo, S. Shimizu, K. Hoshikawa, E. Maeda, K. Yamazaki (Eds.), Science of the Rice Plant, vol. 1, Morphology, Nobunkyo, Tokyo, 1990, pp. 273–306. [35] I. Ockenden, J.A. Dorsch, M.M. Reid, L. Lin, L.K. Grant, V. Raboy, J.N.A. Lott, Characterization of the storage of phosphorus, inositol phosphate and cations in grain tissues of four barley (Hordeum vulgare L.) low phytic acid genotypes, Plant Sci. 167 (2004) 1131–1142. [36] L. Lin, I. Ockenden, J.N.A. Lott, The concentrations and distribution of phytic acid-phosphorus and other mineral nutrients in wild-type and low phytic acid 1-1 (lpa 1-1) corn (Zea mays L.) grains and grain parts, Can. J. Bot. 83 (2005) 131–141. [37] E. Lombi, K.G. Scheckel, J. Pallon, A.M. Carey, Y.G. Zhu, A.A. Meharg, Speciation and distribution of arsenic and localization of nutrients in rice grains, New Phytol. 184 (2009) 193–201. [38] D.P. Persson, T.H. Hansen, K.H. Laursen, J.K. Schjoerring, S. Husted, Simultaneous iron, zinc, sulfur and phosphorus speciation analysis of barley grain tissues using SEC–ICP–MS and IP–ICP–MS, Metallomics 1 (2009) 418–426. [39] T.J. Rose, J. Pariasca-Tanaka, M.T. Rose, Y. Fukuta, M. Wissuwa, Genotypic variation in grain phosphorus concentration, and opportunities to improve P-use efficiency in rice, Field Crop. Res. 119 (2010) 154–160. [40] V. Raboy, Seed phosphorus and the development of low-phytate crops, in: B.L. Turner, A.E. Richardson, E.J. Mullaney (Eds.), Inositol Phosphates: Linking Agriculture and the Environment, CAB International, Oxfordshire, 2006, pp. 111–132.
Contents lists available at ScienceDirect
Plant Science journal homepage: www.elsevier.com/locate/plantsci
A decrease in phytic acid content substantially affects the distribution of mineral elements within rice seeds Hiroaki Sakai a , Toru Iwai b , Chie Matsubara b , Yuto Usui b , Masaki Okamura b , Osamu Yatou c , Yasuko Terada d , Naohiro Aoki b , Sho Nishida b , Kaoru T. Yoshida b,∗ a
Faculty of Agriculture, The University of Tokyo, Bunkyo-ku, Tokyo, Japan Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo-ku, Tokyo, Japan c Crop Development Division, NARO Agricultural Research Center, Inada, Joetsu, Niigata, Japan d Super Photon Ring-8, Japan Synchrotron Radiation Research Institute, Sayo-cho, Sayo-gun, Hyogo, Japan b
a r t i c l e
i n f o
Article history: Received 1 April 2015 Received in revised form 30 May 2015 Accepted 6 June 2015 Available online 14 June 2015 Keywords: Low phytic acid grains Mineral elements Mutant Oryza sativa L. Phosphorus Seed storage substances
a b s t r a c t Phytic acid (myo-inositol hexakisphosphate; InsP6 ) is the storage compound of phosphorus and many mineral elements in seeds. To determine the role of InsP6 in the accumulation and distribution of mineral elements in seeds, we performed fine mappings of mineral elements through synchrotron-based X-ray microfluorescence analysis using developing seeds from two independent low phytic acid (lpa) mutants of rice (Oryza sativa L.). The reduced InsP6 in lpa seeds did not affect the translocation of mineral elements from vegetative organs into seeds, because the total amounts of phosphorus and the other mineral elements in lpa seeds were identical to those in the wild type (WT). However, the reduced InsP6 caused large changes in mineral localization within lpa seeds. Phosphorus and potassium in the aleurone layer of lpa greatly decreased and diffused into the endosperm. Zinc and copper, which were broadly distributed from the aleurone layer to the inner endosperm in the WT, were localized in the narrower space around the aleurone layer in lpa mutants. We also confirmed that similar distribution changes occurred in transgenic rice with the lpa phenotype. Using these results, we discussed the role of InsP6 in the dynamic accumulation and distribution patterns of mineral elements during seed development. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Phytic acid, myo-inositol 1,2,3,4,5,6-hexakisphosphate (InsP6 ), is a hexaphosphoric ester of myo-inositol. InsP6 serves as a phosphorus (P) storage substance in plants. In plant seeds, about 75 ± 10% of total P is accumulated as InsP6 [1], and InsP6 is deposited primarily as mixed salts of various mineral cations such as magnesium (Mg), potassium (K), calcium (Ca), manganese (Mn), iron (Fe), and zinc (Zn) [2,3] because InsP6 has six negatively charged phosphate groups and can chelate strongly with the cations to form insoluble salts; i.e., phytate.
Abbreviations: DAF, days after flowering; EMS, ethyl methanesulfonate; ICP–OES, inductively coupled plasma optical–emission spectrometry; InsP6 , myoinositol 1,2,3,4,5,6-hexakisphosphate; lpa, low phytic acid; -XRF, micro X-ray fluorescence; NT, non-transformant; P, phosphorus; Pi, inorganic phosphate; WT, wild type. ∗ Corresponding author. Tel.: +81 3 5841 8086; fax: +81 3 5841 1306. E-mail address: [email protected] (K.T. Yoshida). http://dx.doi.org/10.1016/j.plantsci.2015.06.006 0168-9452/© 2015 Elsevier Ireland Ltd. All rights reserved.
Cereal grains are utilized not only as staple food items for humans but also as feeds for livestock. However, P in the form of InsP6 is not readily available to monogastric non-ruminant animals, such as humans, pigs, and poultry, because they lack phytase, the enzyme that degrades InsP6 [4]. To provide the optimal level of P for animal growth, feeds have traditionally been supplemented with inorganic phosphate (Pi) [5]. Consequently, non-ruminants excrete large amounts of unabsorbed P derived from InsP6 and surplus Pi in their manure. This leads to the accumulation of P in soil and water, and subsequently to the eutrophication of water areas [5,6]. In addition, non-ruminants do not easily absorb metal elements from seeds because they bind to InsP6 and form phytate [7]. To decrease the environmental loading and to improve the utilization of P and mineral cations, the development of cultivars that contain low levels of phytate in their seeds is a major breeding objective [5]. Several low phytic acid (lpa) mutants, whose seed phytate is reduced by 50–95%, have been isolated in barley [8,9], rice [10,11], wheat [12], maize [13–15], and soybean [16]. For transgenic plants, Shi et al. [17] succeeded in producing transgenic plants with a lowphytate and high-Pi phenotype in maize and soybean through the gene silencing of InsP6 transporter. Kuwano et al. also generated
H. Sakai et al. / Plant Science 238 (2015) 170–177
transgenic rice, in which seed phytate was reduced by 68% and freely available Pi was concomitantly increased through the antisense repression of 1D-myo-inositol 3-phosphate synthase gene [18], which catalyzes the first step in phytic acid biosynthesis and inositol metabolism. Nutrition studies of livestock and human beings have confirmed that low-phytate seeds improve the availability of both P and metals such as Fe and Ca, and reduce P excretion in manure [19,20]. However, the distribution of these metal elements in lowphytate seed is not fully understood. To compare the content and distribution of metal elements in seeds between lpa and wild-type (WT) plants, inductively coupled plasma optical–emission spectrometry (ICP–OES) analyses were conducted using milled rice [21,22]. Although no marked effects on the amount of metal cations were observed, there was a trend for the concentrations of several cations to increase (e.g., K and Mg) in milled lpa seeds compared with milled WT seeds. These results suggest that the lpa phenotype affects the distribution of mineral cations in a seed. However, the precise locations of elements are not clear in studies that used milled grains. Synchrotron-based micro X-ray fluorescence (-XRF) imaging at the Super Photon Ring-8 (SPring-8) facility is non-destructive and can provide elemental mapping images with high resolution; i.e., subcellular localizations of elements, as shown in our previous study [23]. Using -XRF imaging analysis, we have strongly suggested that not all storage mineral cations in a rice seed existed as phytate, and that Zn and copper (Cu) accumulated as a storage form other than phytate at least in the endosperm, because a significant amount of Zn and Cu did not colocalize with P in the endosperm tissues [23]. In contrast, K, Ca, and Fe were colocalized with P, and might primarily accumulate as phytate in the aleurone layer throughout seed development. In this study, we conducted -XRF analysis on the cross sections of rice seeds of the WT, two independent lpa mutants, and one lpa transformant, to determine the differences in the dynamic changes in the distribution of minerals between WT and low-phytate seeds, and to estimate the role of InsP6 in the metal distribution. The nutritional value of the lowphytate seed in relation to the distribution pattern and the storage form of the elements was then considered.
2. Materials and methods 2.1. Plant materials We identified two lpa mutants from an M2 population of rice (Oryza sativa L. var japonica, cv. Koshihikari) mutagenized with ethyl methanesulfonate (EMS), the seeds of which had a high Pi content, because a high Pi content is an indicator of low InsP6 content [5]. Self-pollinated M3 seeds of each putative mutant were planted to confirm the lpa phenotype in the subsequent generation. The two mutants were crossed with indica rice (cv. Kasalath) for rough mapping. The F2 populations segregated into the wild and lpa types at a ratio of 3:1, indicating that each mutant phenotype is caused by a single recessive mutation. Genomic DNA was isolated individually from these mutants and analyzed using PCR-based genetic markers to identify the molecular markers linked to the lpa phenotypes. The primers used for the molecular markers were as follows: a derived cleaved amplified polymorphic sequence (dCAPS) corresponded to marker 3-1, 5 -TCTAATACCATTTTTCAACGAAGC3 and 5 -TAATCAAGTTATTTGCTGCG-3 ; the other markers are shown at http://rgp.dna.affrc.go.jp/E/publicdata/caps/index.html. LPA2623 was mapped roughly to the 5.8 cM region located between the C63223 and C60710 markers on the long arm of chromosome 2 and LPA3847 to the 2.4 cM region located between the 3-1 and R2247 markers on the short arm of chromosome 3 (Fig. S1). An allelism test also revealed that the loci of the two mutants,
171
designated lpa2623 and lpa3847, were distinct from each other. OsPGK1 and OsMRP13/OsABCC13 genes, which were responsible for low-phytate phenotype [24,25], were located in the mapped LPA2623 and LPA3847 regions, respectively (Fig. S1). The mutant and WT plants were grown in pots in a greenhouse under natural light conditions in summer. Only superior caryopses were used for all experiments, because the growth of inferior caryopses was retarded [26]. To produce transgenic lpa rice, a gene consisting of 18kDa oleosin promoter and rice myo-inositol 3-phosphate synthase (RINO1) [27] cDNA in the antisense orientation has been introduced into japonica rice (cv. Kitaake). The detailed methods of transformation, selection of a stable transformant, and growth conditions were described previously [18].
2.2. Seed Pi measurement To analyze Pi levels, crushed powdery whole seed was extracted with 12.5% (w/v) trichloroacetic acid containing 25 mM MgCl2 . The Pi concentration was colorimetrically determined as described by Chen et al. [28]. The detailed method was described previously [29].
2.3. Ion chromatography analysis Mature rice seeds were dehusked, and dried for 48 h at 60 ◦ C. The dried seed was weighed and crushed with a hammer. The crushed samples were ground with a mortar and pestle and then homogenized in 2.4% (w/v) HCl. The HCl extracts were diluted 40-fold with deionized water and then subjected to ion chromatography. The details of the extraction and ion chromatography method were described previously [29]. More than three independent seeds were used in the experiment.
2.4. Quantitative analysis of starch Seed starch contents were measured according to the method of Okamura et al. [30]. Two to three dehusked mature seeds were dried for 48 h at 60 ◦ C and then ground into powders using a MultiBeads Shocker (Yasui Kikai, Osaka, Japan). The powdered samples were extracted twice with 80% (v/v) ethanol for 20 min at 80 ◦ C, with mixing every 5 min. After centrifugation at 11,000 × g for 5 min, the precipitates were resuspended in distilled water, and boiled for 2 h with mixing every 30 min. The enzymatic method using an F-kit for starch (J.K. International, Tokyo, Japan) was used for the assay. The starch concentration was determined spectrophotometrically at 340 nm. The analysis was repeated at least three times using seeds derived from independent plants.
2.5. ICP–OES analysis Mature seeds were dehusked and dried for 48 h at 60 ◦ C. Six dried seeds were weighed and crushed together using a MultiBeads Shocker (Yasui Kikai), and then digested with 2 mL of 0.08 N HNO3 for 30 min at 80 ◦ C. Subsequently, samples were heated to 120 ◦ C for at least 1 h to evaporate water. Then, 1 mL of H2 O2 was added and the samples were heated sequentially for 30 min at 80 ◦ C, 1 h at 120 ◦ C, and overnight at 80 ◦ C. The samples were then dissolved in 1 mL of 0.08 N HNO3 . Elemental concentrations in the acid-digested samples were analyzed by inductively coupled plasma optical-emission spectrometry (ICP–OES; model SPS3500, Hitachi High-Tech Science Co., Tokyo, Japan) in accordance with the manufacturer’s instructions. The experiment was repeated three times using three independent seed samples.
172
H. Sakai et al. / Plant Science 238 (2015) 170–177
2.6. -XRF analysis Developing seeds of the WT and two mutants at 10, 20, and 30 days after flowering (DAF), and mature seeds (about 50 DAF) of non-transgenic and transgenic plants were collected for elemental mapping by synchrotron-based -XRF analysis using the BL37XU beamline of the SPring-8 facility in Hyogo, Japan. The dehusked seeds were vacuum-freeze dried for 2 days, and 200-m cross sections of the center of seeds were subjected to -XRF analysis. The detailed methods of the -XRF analysis were described previously [23], but here the beam size was 0.6 m in the vertical direction × 0.8 m in the horizontal direction at the sample position when we analyzed the developing seeds. The scanning step size was set to 50 m to obtain elemental maps of cross-sections of whole seeds, or 1 m for elemental maps of the lateral side of seeds. Three independent seeds were used for whole cross-section mappings, and one representative of the three seeds was used for lateral-side line mappings.
3. Results 3.1. Morphological traits of low phytic acid mutants We screened an M2 population of rice (cv. Koshihikari) mutagenized with EMS for increased Pi levels using a molybdate staining assay, because the grains of lpa mutants generally accumulate high levels of Pi instead of InsP6 [5]. We identified two single-gene recessive mutants that had greatly increased Pi levels in seeds, lpa2623 and lpa3847. The allelism test and the rough mapping revealed that these two lpa mutants were non-allelic (Fig. S1). Although there was no significant difference in the germination rate, seedling growth, plant height, and heading date in the two lpa mutants and the WT, the traits of the seeds were significantly different. The weights of dried mature seeds of the mutants were significantly lighter than those of the WT (Fig. 1A). The reduction in seed weight in comparison with the WT was 11% for lpa3847 and 33% for lpa2623. The reduction rates were significantly different between lpa2623 and lpa3847. 3.2. Biochemical components in the seeds of low phytic acid mutants The reduced seed weight of lpa mutants compared to the WT (Fig. 1A) indicated the possibility that the starch content was lower
Fig. 1. Phenotypes of seeds obtained from the wild type (WT) and two lpa mutants. (A) Dry weight of mature seeds. Dehusked seeds were dried for 48 h at 60 ◦ C. Each value represents the mean ± SE (n = 20). Columns with different letters are significantly different (Tukey’s test, P < 0.01). (B) Comparison of starch contents of mature seeds. Values represent the relative fold change of each content per seed compared with the value of WT. Each value represents the mean ± SE (n = 4 for left panel, 3 for right panel). Columns with different letters are significantly different (t-test, P < 0.05 for left panel, P < 0.01 for right panel).
in the mutant seeds, because starch is the main component of seeds. The starch measurements revealed that the starch content of seeds decreased significantly in both lpa mutants in comparison with the WT (Fig. 1B). The reduction rates were 9.22% for lpa3847 and 41.9% for lpa2623. Thus, the reduction rate of seed weight was correlated with the reduction rate of starch content. The InsP6 and Pi content of mature seeds of the WT and lpa mutants was determined by ion-chromatographic analysis and a colorimetric molybdate assay, respectively (Fig. 2A and B). The seed InsP6 content decreased significantly in both lpa mutants in comparison with the WT. The reduction rate of the InsP6 content of lpa2623 (74.7%) was higher than that of lpa3847 (66.1%) (Fig. 2A). From the chromatographic analysis, the accumulation of any other inositol phosphates in mature seeds of both lpa mutants was not detected, as in the seeds of the WT (data not shown). In contrast to InsP6 , the Pi content of both mutants increased by 16.1-fold for lpa3847 and 7.67-fold for lpa2623 (Fig. 2B). The total P content of seeds of the WT, lpa3847, and lpa2623 was 31.4, 30.6, and 28.8 g/seed, respectively (Fig. 2C). These values were not significantly different. To confirm if the decrease of InsP6 in the mutant seeds affected the quantity of mineral cations in seeds, the K, Fe, Zn, and Cu contents of mature seeds were measured by ICP–OES (Fig. 3). The contents of all four metal elements in the seeds of lpa2623 and lpa3847 were identical to those of the WT. The translocation of these elements from vegetative organs into developing seeds was therefore unaffected by the change in the storage form of phosphorus. 3.3. Mineral localizations in the developing seeds of low phytic acid mutants During seed development, P and metal elements were actively translocated from vegetative organs into seeds until 14 DAF, and the translocation of most elements was completed in the period from 21 to 28 DAF [23]. Therefore, in this study, seeds at 10, 20, and 30 DAF were subjected to elemental mapping by -XRF analysis with a 50-m step size. We determined the distribution patterns of P, K, Fe, Zn, and Cu in the cross sections of the center of developing rice seeds (Fig. 4). In the WT seeds, as in a previous study using Kitaake rice seeds [23], P, K, and Fe were mainly localized in the aleurone layer (Fig. 4). P, K, and Fe levels clearly increased in the aleurone layer from 10 to 20 DAF. P and K were most abundant in the lateral to ventral sides of the aleurone layer at 20 and 30 DAF in the WT. At any stage of seed development, Fe in the WT seed was mainly localized in the dorsal side of the aleurone layer adjacent to vascular bundles, which are the gates of translocation from vegetative organs (Fig. 4, Fig. S2). In comparison with P, K, and Fe, Zn and Cu were broadly distributed around the aleurone layer of the WT seed and the signals were also detected in the inner endosperm during seed development (Fig. 4). The accumulation of Zn around the aleurone layer of the WT progressed substantially from 10 to 20 DAF. Cu was most abundant around the lateral side of the aleurone layer in the WT seeds at any stage of seed development, and actively spread into the endosperm from 10 to 20 DAF. In both lpa mutants, the signals of P and K in the aleurone layer were weaker than in the WT at all stages of seed development (Fig. 4). This means that large amounts of P and K diffused into the inner endosperm in both lpa mutants, because the amounts of P and K in lpa seeds were identical to those in the WT seeds (Figs. 2C and 3). The accumulation and distribution patterns of Fe in the seeds of both lpa mutants were similar to those in the seeds of the WT (Fig. 4). However, the region where Fe was abundant in both lpa mutants displayed a tendency to expand from the dorsal side to
H. Sakai et al. / Plant Science 238 (2015) 170–177
173
Fig. 2. InsP6 (A), Pi (B), and total phosphorus (C) levels in mature seeds obtained from the wild type (WT) and two lpa mutants. (A) InsP6 contents were determined by ion-chromatography (n 3). (B) Pi contents were measured by colorimetric assay using molybdate staining (n = 5). (C) Total phosphorus contents were measured by ICP–OES analysis (n = 3). Each value represents the mean ± SE. Columns with the same letters are not significantly different (Tukey’s test, P < 0.01 for (A and B), P < 0.05 for (C)).
Fig. 3. The K, Fe, Zn, and Cu contents of mature seeds from the wild type (WT) and two lpa mutants. The mineral contents were measured by ICP–OES analysis. Each value represents the mean ± SE of three replicates. Columns with the same letters are not significantly different (Tukey’s test, P < 0.05).
the lateral and ventral direction in the aleurone layer after 10 DAF (Fig. 4). This tendency was more pronounced in lpa2623 than in lpa3847. The Zn and Cu signals, which were broadly diffused from the aleurone layer to the endosperm in WT seeds, were
concentrated around the aleurone layer in both lpa mutants (Fig. 4). Zn in the lpa seeds increased around the aleurone layer from 10 to 20 DAF and was concentrated as seed development proceeded. Although the progress of Cu accumulation in developing seeds differed slightly between lpa2623 and lpa3847 seeds, the concentration of Cu around the aleurone layer was prominent in both lpa mutants at 30 DAF. The accumulation of Cu in the aleurone layer was slightly more apparent in lpa2623 than in lpa3847. To compare the precise distributions of Zn and Cu around the aleurone layer in the WT and lpa mutants, the step size of the -XRF analysis was changed from 50 to 1 m. We analyzed the lateral sides of the cross sections of 30-DAF seeds of the WT and lpa mutants (Fig. 5). From a distribution map of P in a seed, we estimated the area of the aleurone layer because the concentrated localization of the P in the aleurone layer was observed in both WT and lpa mutants (Fig. 4). As an example, the distribution map of P of the WT seed is shown in Fig. 5A. We determined the maximum detection limit for the normalized X-ray fluorescence intensity in the elemental map of P as the fluorescence intensities of most of the points in the aleurone layer exceed the limit. Using this limit, the aleurone layer was displayed as a white color in the distribution map of P, as seen in Fig. 5B. From the determination of the boundaries of the aleurone layer, the estimated widths of the aleurone layers of the WT, lpa2623, and lpa3847 were 67, 65, and 77 m, respectively. These values did not deviate from the general width of the aleurone layer of a rice seed. We then drew a line from the seed coat to the inner endosperm (red arrow in Fig. 5B). We performed a line mapping, which defined a series of fluorescence intensities that were obtained in one direction. To compare the line mapping of elements in the WT and in two lpa mutants, the widths of the aleurone layers of the WT and mutants were adjusted. The normalized fluorescence intensities of P, Zn, and Cu on the line from the seed coat to the endosperm in the lateral side of the seed are displayed in Fig. 5C. The intensities of P in the aleurone layers of both lpa mutants were lower than in the WT, as shown in Fig. 4. The intensities of Zn and Cu in both lpa mutants were higher than in the WT in both the aleurone layer and the endosperm adjacent to the aleurone layer (subaleurone layer) (Fig. 5C). The distribution of Cu was substantially different in the lpa mutants. In the WT seed, the area with the most abundant Cu was in the subaleurone layer. In contrast, Cu was abundant in the aleurone layer, and in the boundaries between the aleurone layer and the endosperm of the lpa2623 and the lpa3847 seeds. This tendency was more pronounced in lpa2623 than in lpa3847.
174
H. Sakai et al. / Plant Science 238 (2015) 170–177
Fig. 4. Elemental maps of P, K, Fe, Zn, and Cu in the cross sections of developing seeds at 10, 20, and 30 days after flowering (DAF). Normalized X-ray fluorescence intensities are scaled from red (maximum) to blue (minimum) for individual elements. Normalized intensities exceeding maximum limits of scale bars are displayed as white. To generate the maps, a step size of 50 m was used. Dorsal side is at the top in all figures. Maps representative of three independent seeds are shown. Bars = 300 m. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
3.4. Mineral localization in the seeds of low-phytic-acid transgenic rice We have generated transgenic rice plants carrying a rice 1Dmyo-inositol-3-phosphate synthase (RINO1) antisense gene, which codes the enzyme that catalyzes the first step in InsP6 biosynthesis and inositol metabolism, driven by an 18-kDa oleosin promoter, which promotes expression in the embryo and aleurone layer [18]. The reduction rate of InsP6 of the lpa transformant was 68% compared to that of non-transformant (NT) (cv. Kitaake). We conducted -XRF analysis on the cross sections of the center of mature seeds of the transgenic lpa rice and NT rice. The concentrations of P and four metal elements (K, Fe, Zn, and Cu) in the seed of the transformant were similar to those of the NT (Fig. S3). The changes in the distribution patterns of these five mineral elements in the lpa transformant (Fig. 6) were consistent with those observed in lpa
mutants at 30 DAF (Fig. 4). In the lpa transformant, the P and K signals in the aleurone layer were weaker than those of the NT. The abundant region of Fe in the seed of the transformant expanded to the lateral and ventral directions in the aleurone layer in comparison with the NT. The Zn and Cu signals were concentrated around the aleurone layer in the lpa transformant. 4. Discussion The fine -XRF imaging analysis revealed that P, K, and Fe in the WT (cv. Koshihikari) accumulated mainly in the aleurone layer during seed development, and Zn and Cu were broadly distributed from the aleurone layer to the inner endosperm. These distribution patterns of minerals agreed with those in our previous study using another cultivar, Kitaake [23]. Ogawa et al. investigated the patterns of mineral accumulation within the aleurone layers during
H. Sakai et al. / Plant Science 238 (2015) 170–177
175
Fig. 6. Elemental maps of P, K, Fe, Zn, and Cu in cross sections of mature seeds from non-transgenic rice (NT) and transgenic lpa rice (RINO1). Normalized X-ray fluorescence intensities are scaled from red (maximum) to blue (minimum) for individual elements. To generate the maps, a step size of 50 m was used. Dorsal side is at the top in all figures. Maps representative of three independent seeds are shown. RINO1: transgenic rice, in which antisense RINO1 gene was expressed under the control of the 18-kDa oleosin promoter. Bars = 300 m. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Elemental distributions of P, Zn, and Cu around the aleurone layer on the lateral side of a seed cross section at 30 days after flowering (DAF). (A and B) Elemental maps of P in a wild-type (WT) seed. Maximum detection limits of X-ray fluorescence intensity are 120 (A) and 22 (B). Normalized X-ray fluorescence intensities are scaled from red (maximum) to blue (minimum). Normalized intensities exceeding maximum limits of the scale bars are displayed as white. To generate the map, a step size of 1 m was used. Dorsal side is at the top in the figures. Bars = 20 m. Coat: seed coat, Aleu: aleurone layer, Endo: Endosperm. (C) Elemental line maps of P, Zn, and Cu from seed coat to endosperm in the lateral side of seed at 30 DAF. Two vertical lines in each figure represent both sides of the boundaries of the aleurone layer. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
rice seed development using an electron microprobe X-ray analyzer and reported that the accumulation of K in aleurone cells was delayed relative to that of P and Mg and was not observed until 19 DAF [31]. In our studies, however, K accumulation was not delayed and we detected K signals in aleurone cells at 10 DAF [23]. This inconsistency is probably due to differences in the sensitivity of the methods used. Since synchrotron-based -XRF analysis is a very sensitive method compared to other methods, we could detect a trace accumulation of K at the earlier stage. Using -XRF imaging analysis, we further revealed the dynamic changes in the mineral distributions of two independent lpa mutants and an lpa transformant in this study. The P content and the levels of four metal elements (K, Fe, Zn, and Cu) in the seeds of both lpa2623 and lpa3847 mutants, with 77.4 and 66.1% reduction in seed InsP6 , respectively, were the same as in the WT seeds (Figs. 2C and 3). Liu et al. also showed that there was no major difference in the total contents of P, K, Mg, Ca, Mn, Fe, and Zn between the seeds of the WT and an lpa1-1 mutant with a 45% reduction in seed phytate [32]. These results indicate that the storage form of phosphorus does not affect the translocation of minerals from vegetative organs into developing seeds, and that the changes in the distribution patterns of minerals observed in lpa seeds were not due to the changes in the amounts of minerals. The chromosome rough mapping of the lpa2623 and lpa3847 mutants revealed that the mutations occurred at two different loci, which were predicted to be in a different region from the RINO1 gene
176
H. Sakai et al. / Plant Science 238 (2015) 170–177
locus (Fig. S1). Therefore, the similar changes in the distribution patterns of minerals observed in the two mutants and the transformant might be caused by the low-phytate phenotype. However, we could not exclude the possibility that the changes in the mineral distribution were due in part to the reduction in the weights of lpa seeds (Fig. 1) and that the reduction in starch, which accumulates in the endosperm in rice, affected the mineral distribution. The accumulations of P and K in the aleurone layer of both lpa mutants were considerably less than those in the WT at all stages of seed development (Fig. 4). This tendency was also observed in the seeds of the lpa transformant. In a previous study, we concluded that translocated Pi was immediately synthesized into InsP6 and then accumulated in the aleurone layer cells of rice seeds [23]. In this study, because the amounts of P and K were identical to those in the WT (Figs. 2C and 3), and because P and K strong signals were not detected in the endosperm, P and K likely diffuse uniformly and thinly in the endosperm of lpa mutants. After translocation from vegetative organs, Pi that was not used for InsP6 biosynthesis in lpa mutants might immediately diffuse into the endosperm (Fig. S2). Also, K that was not captured by the reduced InsP6 might immediately diffuse into the endosperm. A decrease in the P and K concentrations in the bran of lpa mutants was also observed in rice and barley using milling products [21,33]. However, Ren et al. reported that concentrations of K, Ca, Zn, Mg, Fe, and Cu increased in milled rice (endosperm) of an lpa mutant, even though the distribution of P was not altered [22]. All of these studies used milled products, and therefore the precise distribution pattern of elements could not be analyzed. In this study, for the first time, the precise positions of the elements in lpa seeds could be determined using -XRF analysis. In contrast to K, the distribution pattern of Fe in seeds of both the lpa mutants and the lpa transformant were similar to those of the WT and NT, except that the region of Fe abundance in the aleurone layer of low-phytate seeds had a tendency to expand to the lateral and ventral directions (Figs. 4 and 6). The difference in the change in the distribution of Fe and K in lpa seeds (i.e., a minor change for Fe and a large change for K) may indicate a difference in the affinity of Fe and K for InsP6 . In the aleurone layer of the WT seeds, Fe was abundant in the dorsal side where vascular bundles existed; i.e., the translocation sites (Fig. S2)[34], while K was abundant in the lateral to ventral side of the aleurone layer, which is distant from the translocation sites. This suggests that Fe is captured by InsP6 immediately after its translocation into a seed. In contrast, in the aleurone layer K likely moves from the dorsal to the lateral and ventral sides before being captured by InsP6 (Fig. S2) [34]. Even in a low-phytate seed, the reduced InsP6 may preferentially bind to Fe, and K, having a lower affinity for InsP6 than Fe, then diffuses into the endosperm from the aleurone layer without forming a phytate salt (Fig. S2). The increased accumulations of Fe on the lateral and ventral sides of the aleurone layer of lpa mutants and the lpa transformant compared to the WT and NT occur because the reduced amount of InsP6 on the dorsal side of the aleurone layer could not capture all of the translocated Fe. In both the lpa mutants and the lpa transformant, the amount of Zn and Cu in the aleurone and subaleurone layers increased compared to the WT and NT (Figs. 4–6). Iwai et al. suggested that Zn is loosely bound and Cu is scarcely bound to InsP6 in rice WT seeds, and that large amounts of Zn and Cu may exist in a storage form other than phytate in the endosperm [23]. Iwai et al. also reported that Zn and Cu moved from the aleurone layer to the endosperm as seed development progressed [23]. During seed development, the aleurone cells are gradually filled with globoids and oil bodies, and therefore, metal elements that do not form phytate salt will be gradually extruded from the aleurone layer to the endosperm. A reduction in globoid size was observed in the aleurone layer of lpa seeds in rice, barley, and corn [32,35,36]. Because the estimated
width of the aleurone layers of the two lpa mutants were similar to that of WT, it is plausible that Zn and Cu in the lpa seeds were not moved out into the endosperm, but remained in the spaces between small globoids in the aleurone layer. As a result, the Zn and Cu signals in the aleurone layer were considerably higher in lpa seeds than in WT seeds (Figs. 4 and 5). A provocative question is why K that did not form phytate in low-phytate seeds did not remain in the aleurone layer as did Zn and Cu, but diffused uniformly into the endosperm. It is suggested that Zn in grains is mainly bound to S; i.e., it forms a metal thiolate complex with peptides and/or proteins [37,38], which is stored around the aleurone layer. Divalent and trivalent cations, such as Zn, can form a bridge between the negative charges of thiolate ligands. However, K cannot form a bridge between thiolate ligands because it is a monovalent cation; therefore, free K ions may diffuse promptly and uniformly into the endosperm of low-phytate seeds. Lpa2623 displayed a more severe phenotype than lpa3847, including greater changes in the distribution patterns of Fe and Cu. The lower concentration of InsP6 and/or starch in lpa2623 might result in greater changes in the distribution patterns of Fe and Cu in comparison with the lpa3847. Rose et al. examined the genetic variation of P partitioning in rice plants and showed that the partitioning of P between grain and straw at maturity (P harvest index) was highly correlated with the harvest index [39]. In comparison, in our study, the mineral contents, including P, of lpa seeds were not affected by the reduction in grain yields, which was due mainly to the reduction in starch content. This suggests that carbon loading and mineral loading into a seed are controlled by different regulatory mechanisms and that mineral nutrient contents might be improved without affecting grain yield. Reducing the phytate content of grains could enhance mineral absorption from mixed diets [40]. We found that reducing the InsP6 content of rice seeds did not simply result in all mineral elements diffusing into the endosperm without forming phytate. Fe might be bound to InsP6 even in low-phytate seeds, and hence low-phytate rice did not greatly increase Fe utilization. In addition, there was a dynamic change in the distribution patterns of many mineral elements in low-phytate seeds. Therefore, the mineral content may differ between the WT and mutants whether brown or white rice (milled rice) is used. Attention should be paid to both the storage form and the change in the amount of mineral elements in individual parts of the low-phytate seed. In conclusion, to our knowledge, this is the first study to identify the precise dynamics of mineral elements in low-phytate seeds during seed development by utilizing synchrotron-based -XRF fluorescence imaging. We showed differences in the affinity of each element for InsP6 , and the role of InsP6 in the accumulation pattern of metal elements. To understand the dynamics of mineral elements in seeds more precisely, the chemical forms of P and metal elements during seed development should be determined.
Acknowledgments The ICP–OES analysis was supported by Japan Advanced Plant Science Network. The -XRF analysis was performed at the BL37XU in the SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2010B1695 and No. 2013B1209). The authors are grateful to Dr. M. Takahashi, Ms. H. Aihara, Mr. N. Morishita, Mr. Y. Tagashira, Mr. G. Hayakawa, and the members of Laboratory of Plant Nutrition and Physiology, Utsunomiya University for the technical assistance for the -XRF analysis. We are grateful to Ms. T. Shimizu for her help in rough mapping of lpa2623 mutant.
H. Sakai et al. / Plant Science 238 (2015) 170–177
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.plantsci.2015.06. 006 References [1] V. Raboy, myo-Inositol-1,2,3,4,5,6-hexakisphosphate, Phytochemistry 64 (2003) 1033–1043. [2] J.N.A. Lott, J.S. Greenwood, G.D. Batten, Mechanisms and regulation of mineral nutrient storage during seed development, in: J. Kigel, G. Galili (Eds.), Seed Development and Germination, Marcel Dekker, New York, 1995, pp. 215–235. [3] L. Bohn, L. Josefsen, A.S. Meyer, S.K. Rasmussen, Quantitative analysis of phytate globoids isolated from wheat bran and characterization of their sequential dephosphorylation by wheat phytase, J. Agric. Food Chem. 55 (2007) 7547–7552. [4] P.B. Holm, K.N. Kristiansen, H.B. Pedersen, Transgenic approaches in commonly consumed cereals to improve iron and zinc content and bioavailability, J. Nutr. 132 (2002) 514S–516S. [5] V. Raboy, Approaches and challenges to engineering seed phytate and total phosphorus, Plant Sci. 177 (2009) 281–296. [6] L. Bohn, A.S. Meyer, S.K. Rasmussen, Phytate: impact on environment and human nutrition. A challenge for molecular breeding, J. Zhejiang Uni. Sci. B 9 (2008) 165–191. [7] J.N.A. Lott, Accumulation of seed reserves of phosphorus and other minerals, in: D.R. Murray (Ed.), Seed Physiology, Development, vol. 1, Academic Press, Sydney, 1984, pp. 139–166. [8] S.R. Larson, K.A. Young, A. Cook, T.K. Blake, V. Raboy, Linkage mapping of two mutations that reduce phytic acid content of barley grain, Theor. Appl. Genet. 97 (1998) 141–146. [9] S.K. Rasmussen, F. Hatzack, Identification of two low-phytate barley (Hordeum vulgare L.) grain mutants by TLC and genetic analysis, Hereditas 129 (1998) 107–112. [10] S.R. Larson, J.N. Rutger, K.A. Young, V. Raboy, Isolation and genetic mapping of a non-lethal rice (Oryza sativa L.) low phytic acid 1 mutation, Crop Sci. 40 (2000) 1397–1405. [11] Q.L. Liu, X.H. Xu, X.L. Ren, H.W. Fu, D.X. Wu, Q.Y. Shu, Generation and characterization of low phytic acid germplasm in rice (Oryza saiva L.), Theor. Appl. Genet. 114 (2007) 803–814. [12] M. Guttieri, S.J. Barker, L. Perez-Grau, Identification and characterization of a low phytic acid wheat, Crop Sci. 44 (2004) 418–424. [13] V. Raboy, P.F. Gerbasi, K.A. Young, S.D. Stoneberg, S.G. Pickett, A.T. Bauman, P.P.N. Murthy, W.F. Sheridan, D.S. Ertl, Origin and seed phenotype of maize low phytic acid 1-1 and low phytic acid 2-1, Plant Physiol. 124 (2000) 355–368. [14] J. Shi, H. Wang, Y. Wu, J. Hazebroek, R.B. Meeley, D.S. Ertl, The maize low-phytic acid mutant lpa2 is caused by mutation in an inositol phosphate kinase gene, Plant Physiol. 131 (2003) 507–515. [15] J. Shi, H. Wang, J. Hazebroek, D.S. Ertl, T. Harp, The maize low-phytic acid 3 encodes a myo-inositol kinase that plays a role in phytic acid biosynthesis in developing seeds, Plant J. 42 (2005) 708–719. [16] W.D. Hitz, T.J. Carlson, P.S. Kerr, S.A. Sebastian, Biochemical and molecular characterization of a mutation that confers a decreased raffinosaccharide and phytic acid phenotype on soybean seeds, Plant Physiol. 128 (2002) 650–660. [17] J. Shi, H. Wang, K. Schellin, B. Li, M. Faller, J.M. Stoop, R.B. Meeley, D.S. Ertl, J.P. Ranch, K. Glassman, Embryo-specific silencing of a transporter reduces phytic acid content of maize and soybean seeds, Nat. Biotechnol. 25 (2007) 930–937. [18] M. Kuwano, T. Mimura, F. Takaiwa, K.T. Yoshida, Generation of stable ‘low phytic acid’ transgenic rice through antisense repression of the 1D-myo-inositol 3phosphate synthase gene (RINO1) using the 18-kDa oleosin promoter, Plant Biotechnol. J. 7 (2009) 96–105. [19] C. Mendoza, F.E. Viteri, B. Lonnerdal, K.A. Young, V. Raboy, K.H. Brown, Effect of genetically modified, low-phytic acid maize on absorption of iron from tortillas, Am. J. Clin. Nutr. 68 (1998) 1123–1127.
177
[20] T.L. Veum, D.R. Ledoux, D.W. Bollinger, V. Raboy, A. Cook, Low-phytic acid barley improves calcium and phosphorus utilization and growth performance in growing pigs, J. Anim. Sci. 80 (2002) 2663–2670. [21] R.J. Bryant, J.A. Dorsch, K.L. Peterson, J.N. Rutger, V. Raboy, Phosphorus and mineral concentrations in whole grain and milled low phytic acid (lpa) 1-1 rice, Cereal Chem. 82 (2005) 517–522. [22] X.L. Ren, Q.L. Liu, H.W. Fu, D.X. Wu, Q.Y. Shu, Density alteration of nutrient elements in rice grains of a low phytate mutant, Food Chem. 102 (2007) 1400–1406. [23] T. Iwai, M. Takahashi, K. Oda, Y. Terada, K.T. Yoshida, Dynamic changes in the distribution of minerals in relation to phytic acid accumulation during rice seed development, Plant Physiol. 160 (2012) 2007–2014. [24] S.I. Kim, C.B. Andaya, S.S. Goyal, T.H. Tai, The rice OsLpa1 gene encodes a novel protein involved in phytic acid metabolism, Theor. Appl. Genet. 117 (2008) 769–779. [25] X.H. Xu, H.J. Zhao, Q.L. Liu, T. Frank, K.H. Engel, Mutations of multi-drug resistance-associated protein ABC transporter gene 5 result in reduction of phytic acid in rice seeds, Theor. Appl. Genet. 119 (2009) 75–83. [26] M. Kuwano, F. Takaiwa, K.T. Yoshida, Differential effects of a transgene to confer low phytic acid in caryopses located at different positions in rice panicles, Plant Cell Physiol. 50 (2009) 1387–1392. [27] K.T. Yoshida, T. Wada, H. Koyama, R. Mizobuchi-Fukuoka, S. Naito, Temporal and spatial patterns of accumulation of the transcript of myo-inositol-1phosphate synthase and phytin-containing particles during seed development in rice, Plant Physiol. 119 (1999) 65–72. [28] P.S. Chen, T.Y. Toribara, H. Warner, Microdetermination of phosphorus, Anal. Chem. 28 (1956) 1756–1758. [29] M. Kuwano, A. Ohyama, Y. Tanaka, T. Mimura, F. Takaiwa, K.T. Yoshida, Molecular breeding for transgenic rice with low-phytic-acid phenotype through manipulating myo-inositol 3-phosphate synthase gene, Mol. Breed. 18 (2006) 263–272. [30] M. Okamura, T. Hirose, Y. Hashida, T. Yamagishi, R. Ohsugi, N. Aoki, Starch reduction in rice stems due to a lack of OsAGPL1 or OsAPL3 decreases grain yield under low irradiance during ripening and modifies plant architecture, Funct. Plant Biol. 40 (2013) 1137–1146. [31] M. Ogawa, K. Tanaka, Z. Kasai, Accumulation of phosphorus, magnesium and potassium in developing rice grains: followed by electron microprobe X-ray analysis focusing on the aleurone layer, Plant Cell Physiol. 20 (1979) 19–27. [32] J.C. Liu, I. Ockenden, M. Truax, J.N.A. Lott, Phytic acid-phosphorus and other nutritionally important mineral nutrient elements in grains of wild-type and low phytic acid (lpa1-1) rice, Seed Sci. Res. 14 (2004) 109–116. [33] K. Liu, K.L. Peterson, V. Raboy, Comparison of the phosphorus and mineral concentrations in bran and abraded kernel fractions of a normal barley (Hordeum vulgare) cultivar versus four low phytic acid isolines, J. Agric. Food Chem. 55 (2007) 4453–4460. [34] K. Hoshikawa, Anthesis, fertilization and development of caryopsis, in: T. Matsuo, S. Shimizu, K. Hoshikawa, E. Maeda, K. Yamazaki (Eds.), Science of the Rice Plant, vol. 1, Morphology, Nobunkyo, Tokyo, 1990, pp. 273–306. [35] I. Ockenden, J.A. Dorsch, M.M. Reid, L. Lin, L.K. Grant, V. Raboy, J.N.A. Lott, Characterization of the storage of phosphorus, inositol phosphate and cations in grain tissues of four barley (Hordeum vulgare L.) low phytic acid genotypes, Plant Sci. 167 (2004) 1131–1142. [36] L. Lin, I. Ockenden, J.N.A. Lott, The concentrations and distribution of phytic acid-phosphorus and other mineral nutrients in wild-type and low phytic acid 1-1 (lpa 1-1) corn (Zea mays L.) grains and grain parts, Can. J. Bot. 83 (2005) 131–141. [37] E. Lombi, K.G. Scheckel, J. Pallon, A.M. Carey, Y.G. Zhu, A.A. Meharg, Speciation and distribution of arsenic and localization of nutrients in rice grains, New Phytol. 184 (2009) 193–201. [38] D.P. Persson, T.H. Hansen, K.H. Laursen, J.K. Schjoerring, S. Husted, Simultaneous iron, zinc, sulfur and phosphorus speciation analysis of barley grain tissues using SEC–ICP–MS and IP–ICP–MS, Metallomics 1 (2009) 418–426. [39] T.J. Rose, J. Pariasca-Tanaka, M.T. Rose, Y. Fukuta, M. Wissuwa, Genotypic variation in grain phosphorus concentration, and opportunities to improve P-use efficiency in rice, Field Crop. Res. 119 (2010) 154–160. [40] V. Raboy, Seed phosphorus and the development of low-phytate crops, in: B.L. Turner, A.E. Richardson, E.J. Mullaney (Eds.), Inositol Phosphates: Linking Agriculture and the Environment, CAB International, Oxfordshire, 2006, pp. 111–132.