Characterization of the Histidine-Rich Loop of Arabidopsis Vacuolar Membrane Zinc Transporter AtMTP1 as a Sensor of Zinc Level in the Cytosol

Abstract

Introduction

Zinc (Zn) is an essential micronutrient that functions as a cofactor for many enzymes and proteins, including DNA polymerase, Cu/Zn superoxide dismutase and Zn-finger proteins (Greger 2004, Krämer 2005, Clemens 2010, Krämer 2010). It also leads to toxicity in excess concentrations (Rogers et al. 2000, Kobae et al. 2004, Fukao et al. 2011, Sinclair and Krämer 2012). Zn deficiency causes growth suppression of whole plants and chlorosis of leaves, whereas excess amounts of Zn²⁺ suppress root elongation. Zn in excess amounts enhances the generation of reactive oxygen species in the cell proliferation zone of roots and suppresses growth (Kawachi et al. 2009). Under excess Zn conditions, photosynthesis activity decreases caused by disorder of the Fe homeostasis (Sinclair and Krämer 2012). Zn homeostasis in cells, particularly cytosolic Zn²⁺ concentrations, is essential for the normal growth of plants. Excess Zn²⁺ is exported from the cytosol to the extracellular space and vacuoles via plasma membrane Zn transporters and vacuolar membrane Zn transporters, respectively. Arabidopsis thaliana has metal tolerance proteins (MTPs), zinc-regulated transporters/iron-regulated transporter-like proteins (ZIPs) and heavy-metal ATPases (HMAs) (P_1B subgroup of P-type ATPases), which differ in structure and transport mechanism (MacDiarmid et al. 2002, Krämer et al. 2007, Hanikenne et al. 2008, Morel et al. 2009, Kawachi et al. 2011, Shanmugam et al. 2013). ZIPs are transporters for translocation of Zn into the cytosol at the plasma membrane and membranes of intracellular organelles (Lin et al. 2010). MTP is a subfamily of the cation diffusion facilitator (CDF) family and uses a pH gradient across the membrane for active Zn transport. HMA actively transports Zn²⁺ using energy provided by ATP hydrolysis. Most organisms regulate the gene expression of these Zn transporters and channels by modulating Zn-binding transcription factors in response to Zn deficiency and toxicity (Eide 2009).

Among the 12 members of the MTP subfamily (Blaudez et al. 2003), AtMTP1 functions as an active transporter in the vacuolar membrane to sequester excess Zn²⁺ into the vacuole and maintain a low cytosolic Zn²⁺ concentration (Desbrosses-Fonrouge et al. 2004, Kawachi et al. 2009, Krämer et al. 2010). AtMTP1 is essential for basal tolerance of excess amounts of Zn²⁺ by A. thaliana, as demonstrated by the observation that the loss-of-function mutant mtp1 is phenotypically sensitive to excess Zn²⁺ (Kobae et al. 2004, Martinoia et al. 2007). The hyperaccumulator Arabidopsis halleri, which has up to five copies of MTP1 and a total MTP1 protein level higher than that of A. thaliana, shows extremely high tolerance to excess Zn (Becher et al. 2004, Dräger et al. 2004, Shahzad et al. 2010). The MTP subfamily members have six transmembrane helices and a relatively large and conserved C-terminal domain that binds Zn²⁺ in the cytosol. In addition to this common structure (Lu et al. 2009), AtMTP1 has a His-rich loop (His-loop), which faces the cytosol, between the fourth and fifth transmembrane helices (Kawachi et al. 2008). Other MTPs, such as A. halleri MTP1, Noccaea goesingense MTP1t1 (formerly TgMTP1t1), Populus MTP1, Saccharomyces cerevisiae Cot1 and Homo sapiens ZnT1, also have a His-loop, although the number of histidine residues and the length of the loop are different (Kawachi et al. 2008, Krämer 2010). Metal transporters with a His-loop were found in all plant species whose genomes have been sequenced. However, the His-loop is absent in Escherichia coli YiiP, which is a member of the CDF family and has been well characterized in terms of 3D structure and catalytic mechanisms (Lu and Fu 2007, Lu et al. 2009). Information on the higher order structure and physiological role of the His-loop is limited at present. The mutated AtMTP1 that lacked the first half of the His-loop (His-half) conferred an enhanced Zn²⁺ transport activity (Kawachi et al. 2008). The His-loop was recently found to comprise partly an intrinsically disordered structure binding four Zn²⁺ per molecule, with the Zn²⁺ binding inducing a structural change in the His-loop (Tanaka et al. 2013).

We investigated the physiological role of the His-loop in the ion selectivity and ion transport activity of AtMTP1 in growing plants. By heterologous expression of the mutated AtMTP1 in yeast cells, we found earlier that deletion of the whole sequence of the His-loop resulted in inactivation of AtMTP1. In contrast, the His-half AtMTP1, which lacks the first half with 32 residues (position 185–216 of AtMTP1) of the His-loop, retained the Zn²⁺ transport activity (Kawachi et al. 2008). The V_max value of His-half AtMTP1 was 11-fold that of the normal AtMTP1 and the K_m for Zn²⁺ was twice that of AtMTP1. The His-half AtMTP1 also transported Co²⁺ in yeast cells. Thus, the His-loop of AtMTP1 is not essential for Zn²⁺ transport and was proposed to act as a buffering pocket of Zn²⁺, a sensor of the Zn level and a Zn²⁺ selection domain (Kawachi et al. 2008, Kawachi et al. 2012). To evaluate the effect of the enhanced activity, lack of Zn sensor and altered ion selectivity of the His-half AtMTP1 on the growth of plants, we expressed the His-half AtMTP1 in mtp1-1 homozygous mutant plants and measured their growth, tolerance to Zn excess and deficiency, and amount of Zn incorporated in tissues. We obtained key information on the physiological role of the His-loop. Together with the results, we discuss the biochemical and physiological significance of the His-loop.

Results

Growth of 35S-His-half lines in excess Zn and Co

The sensitivity of the loss-of-function mutant mtp1-1 to excess Zn²⁺ indicates that AtMTP1 is important in Zn homeostasis, particularly the detoxification of excess cytosolic Zn into the vacuole (Kobae et al. 2004, Kawachi et al. 2009). A mutant AtMTP1 lacking the first half of the His-loop (32 residues; His-half; refer to membrane topology models in Supplementary Fig. S1) enhanced Zn²⁺ transport activity when expressed in S. cerevisiae (Kawachi et al. 2008). We accordingly investigated whether plants expressing the His-half AtMTP1 (35S-His-half lines) are more tolerant to excess Zn²⁺ than wild-type plants.

Suppression of root growth is a typical response of A. thaliana to excess Zn²⁺. There was no difference in root length among the plants under normal conditions (Fig. 1). The root length of wild-type A. thaliana accession Wassilewskija (Ws) on a medium containing 150 µM Zn was markedly decreased to <60% that under normal conditions. Roots of the mtp1-1 mutant were markedly shorter than those of Ws. Under the same conditions, the suppressed root length of plants expressing the normal AtMTP1 (35S-MTP1) and the His-half AtMTP1 (35S-His-half), which were expressed under the control of the 35S promoter, was the same as that of Ws (Fig. 1). Thus, the normal and the His-half AtMTP1 were functional in transgenic plants of mtp1-1. Complementation of the mtp1-1 phenotype by expressing AtMTP1 and His-half AtMTP1 was confirmed at 100, 200 and 250 µM Zn²⁺ (Supplementary Fig. S2). The mRNAs in the 35S-MTP1 and 35S-His-half lines grown at 0 and 5 µM Zn were quantified (Supplementary Fig. S3A). The mRNA levels were >4-fold higher than that of Ws, although their levels varied with lines. The mRNA level of AtMTP3, a vacuolar membrane Zn transporter, showed no significant change among the transgenic lines (Supplementary Fig. S3B).

Fig. 1

Open in new tabDownload slide

35S-His-half lines grew as well as the wild type (Ws) and 35S-MTP1 under Zn excess conditions. Seedlings of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines were grown in modified Hoagland medium in which the Zn concentration was modified to 150 µM or 5 µM (normal). Independent lines of 35S-MTP1 (nine lines) and 35S-His-half (11 lines) were examined. After 14 d, the root length was measured. For each line, 21 plants were independently analyzed. Values are expressed as mean ± SD. Significant differences from mtp1-1 (a loss-of-function line) are indicated by asterisks (*P < 0.01).

In an immunoblot (Fig. 2A), the expression and accumulation of AtMTP1 and His-half AtMTP1 protein were clearly demonstrated. The immunostained band of the His-half was slightly lower than that of normal AtMTP1. H⁺-pyrophosphatase (H⁺-PPase), a typical vacuolar membrane marker (Segami et al. 2014), was also detected in the same membrane fractions of Ws and mutant lines (Fig. 2B), suggesting correct localization of the His-half AtMTP1 in the vacuolar membrane. The fractions did not show significant change in the H⁺-PPase amount among the mutant lines.

Fig. 2

Open in new tabDownload slide

Immunoblotting of AtMTP1 and H⁺-PPase. Vacuolar membranes were prepared from 14-day-old plantlets of Ws and transgenic lines. Aliquots corresponding to 160 mg (A) and 40 mg (B) of fresh weight were subjected to SDS–PAGE and subsequent immunoblotting with anti-AtMTP1 (A) and anti-H⁺-PPase (B) antibodies.

As shown in Fig. 2A, the protein level of the His-half AtMTP1 was slightly different between two His-half lines, in which 35S-His-half #2 was a line with higher expression. The difference in the protein level was partially correlated with recovery of root length in the His-half-2 line under excess Zn conditions at 200 and 250 µM but not at 5 or 100 µM (Supplementary Fig. S2).

Yeast cells expressing the His-half were more tolerant to excess Co²⁺ compared with yeast expressing normal AtMTP1 (Kawachi et al. 2008). It was considered that the His-half AtMTP1 may remove excess Co²⁺ from the cytosol and translocate it into the vacuoles by Co²⁺/H⁺ exchange. We accordingly evaluated the tolerance of the mutant lines expressing the His-half and normal AtMTP1 to excess Co²⁺. Root lengths of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines were decreased to 65% in the presence of 30 µM Co²⁺ (Fig. 3A). However, there was no significant difference between these mutants and Ws under Co²⁺ excess conditions. To evaluate Co accumulation and Co distribution between tissues, we determined the Co content in the dry tissues by inductively coupled plasma atomic emission spectroscopy (ICP-AES). In every line, the Co content in roots was approximately 75% that in shoots (Fig. 3B). There was no significant difference in the Co content between Ws and mutant lines.

Fig. 3

Open in new tabDownload slide

35S-His-half lines grew and accumulated Co as well as Ws in the medium supplemented with Co. Seedlings of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines were grown in modified Hoagland medium supplemented with CoSO₄ at 0 or 30 µM (A) or in MS medium supplemented with CoSO₄ at 3 µM (B). (A) The root length of 14-day-old seedlings was measured. Values are expressed as mean ± SD; n = 21. (B) The content of Co in 21-day-old seedlings was measured by ICP-AES. Values are expressed as mean ± SD; n = 4.

Growth suppression in 35S-His-half lines under Zn-deficient conditions

AtMTP1 actively transports Zn²⁺ into vacuoles by acting as a Zn²⁺/H⁺ exchanger (Kobae et al. 2004, Kawachi et al. 2008) and is highly expressed in the apex of primary and lateral roots (Desbrosses-Fonrouge et al. 2005, Kawachi et al. 2009). Thus, Zn²⁺ may be accumulated at a high level in roots as compared with shoots. In contrast to AtMTP1, the His-half AtMTP1 does not suppress the transport activity under Zn-deficient conditions. Therefore, shoots of the 35S-His-half lines cannot receive sufficient Zn²⁺ from the roots when grown under Zn-deficient conditions. We examined this hypothesis by testing the tolerance of the seedlings of Ws and mutant lines under Zn-deficient conditions. To reduce the Zn concentration, we added a membrane-permeable, Zn²⁺ chelator N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) (Arslan et al. 1985, Shumaker et al. 1998) to modified Hoagland medium that was not supplemented with Zn²⁺. Seven-week-old seedlings showed symptoms of critical Zn deficiency, such as chlorosis and growth delay (Supplementary Fig. S4). Older leaves of Ws, mtp1-1 and 35S-MTP1 lines were light yellow and their younger leaves remained light green. The 35S-His-half lines suffered more severe damage and their younger and older leaves became light yellow. Furthermore, the growth of 35S-His-half lines was severely suppressed, and the fresh weight of 35S-His-half shoots tends to be lower than those of Ws, mtp1-1 and 35S-MTP1 lines (Supplementary Fig. S4).

To confirm these results, we established Zn deficiency without the use of excess chelator. We quantified the Zn content in agar products produced by different manufacturers by ICP-AES and used the one lowest in Zn content for seedling cultivation. Among type M agar (Sigma-Aldrich; 1.01 µg g^–1 dry matter), Ina agar (type BA-10, Funakoshi; 0.200 µg g^–1), purified agar (Nacalai Tesque; 0.530 µg g^–1) and gellan gum (Wako Pure Chemical; 1.35 µg g^–1), the Ina agar showed the lowest value of Zn content and was used in the experiments. Agar and related agents were dissolved in ultra-pure water to eliminate contamination of the agar by water.

Under normal conditions in the modified Hoagland medium containing 5.0 µM ZnSO₄, all lines, Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines, grew well and showed no difference in the shoot fresh weight (Fig. 4). Under Zn-deficient conditions in the Ina agar using ultra-pure water, the growth of seedlings was markedly suppressed. Shoot fresh weights of Ws were <60% those under normal conditions. 35S-MTP1 and mtp1-1 lines grew as vigorously as Ws under Zn-deficient conditions. However, the growth of 35S-His-half lines was severely suppressed to 42% of that under normal conditions. The difference in the values between 35S-His-half and the three other lines, Ws, mtp1-1 and 35S-MTP1, suggested that the expression of the His-half AtMTP1 had a negative effect on growth under Zn deficiency.

Fig. 4

Open in new tabDownload slide

35S-His-half lines grew poorly under Zn-deficient conditions. Plants of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines were grown in modified Hoagland medium lacking Zn (0 µM Zn) or with 5 µM Zn (normal conditions). Independent lines of 35S-MTP1 (nine lines) and 35S-His-half (11 lines) were examined. After 21 d, the shoots were weighed. Values are expressed as mean ± SD; n = 36. Significant differences from WS are indicated by asterisks (*P < 0.05; ***P < 0.005). (B) Three shoots each from Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines grown under normal (5 µM Zn) and Zn-deficient conditions (0 µM Zn) for 27 d were photographed.

In this experiment, damage of the leaves by Zn deficiency was mild compared with that in the medium containing TPEN (Supplementary Fig. S4), because leaf color remained green even after 27 d (Fig. 4). We concluded that this experimental condition was suitable for the evaluation of Zn deficiency in living plants and used it in subsequent experiments.

Shoot–root ratio of Zn content was reduced in 35S-His-half lines

Zn contents in roots and shoots from 21-day-old plants were determined by ICP-AES. Under normal growth conditions, the Zn content in roots was approximately 7-fold higher than that in shoots in each line (Fig. 5A). There was no difference in the value among Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines. In contrast, the Zn content in roots and shoots of Ws, mtp1-1 and 35S-MTP1 lines was markedly decreased to <10% (roots) and 30% (shoots) under Zn-deficient conditions compared with that under normal conditions (Fig. 5B). In 35S-His-half lines, the Zn content was significantly high in roots and low in shoots under the same conditions. These results indicate that Zn²⁺ accumulated in the roots of 35S-His-half plants with high activity. For this reason, Zn²⁺ could not be provided to shoots in 35S-His-half lines.

Fig. 5

Open in new tabDownload slide

35S-His-half lines accumulated higher amounts of Zn in roots than the wild type even under Zn-deficient conditions. Seedlings of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines were grown in modified Hoagland medium lacking Zn (0 μM Zn) or with 5 μM Zn. After 21 d, the contents of Zn in shoots and roots were measured by ICP-AES. Values are expressed as mean ± SD; n ≥ 4. Significant differences from Ws are indicated by asterisks (*P < 0.05).

Next, we examined by ICP-AES the effect of the expression of normal and mutant AtMTP1 on the accumulation of other mineral ions under Zn-deficient conditions, given that cross-regulation among multiple metals such as Zn²⁺ and Fe²⁺ has been reported in plants (Fukao et al. 2011). For example, Ca was preferentially accumulated in shoots and the content in roots was approximately 10% that of shoots in all the lines (Fig. 6A). The content of Fe was higher in roots than in shoots (Fig. 6B). No difference was observed in the contents of Ca and Fe between Ws and the three mutant lines, suggesting little contribution of AtMTP1 to Ca and Fe uptake and accumulation in plants. No significant difference in the content of any other metals was detected in these experiments.

Fig. 6

Open in new tabDownload slide

The 35S-His-half mutant accumulated the same amounts of Ca and Fe as the wild type under Zn-deficient conditions. Seedlings of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines were grown in modified Hoagland medium lacking Zn. The contents of Ca (A) and Fe (B) in shoots and roots of 21-day-old plants were measured by ICP-AES. Values are expressed as the mean ± SD; n ≥ 4.

Chloroplast number and Chl content in 35S-His-half lines were reduced under Zn-deficient conditions

The low Zn content in 35S-His-half lines led us to examine the effect on the leaf cells. In the cell layer just under the leaf epidermis, the number of chloroplasts visualized by autofluorescence was markedly decreased in Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines under Zn-deficient conditions (Fig. 7A, B). The size of chloroplasts was reduced in all lines examined under Zn-deficient conditions. Furthermore, we quantified the Chl a content in the acetone extracts from leaves. On a shoot fresh weight basis, Chl a contents in Ws, mtp1-1 and 35S-MTP1 lines were reduced under Zn-deficient conditions by >25% (Fig. 7C). The reductions in chloroplast number and Chl a content were significant in 35S-His-half plants under Zn-deficient conditions. For example, the Chl a content in 35S-His-half lines under Zn-deficient conditions was half that under Zn-sufficient conditions.

Fig. 7

Open in new tabDownload slide

Chloroplasts and Chl content in the leaves of 35S-His-half lines were lower than those in other lines. Plants including Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines were grown for 27 d in modified Hoagland medium lacking Zn (0 µM Zn) or with 5 µM Zn. The second layers of leaf epidermis were observed using a confocal laser scanning microscope. (A) Merged images of autofluorescence (green) of chloroplasts and differential interference contrast images. (B) Images of autofluorescence of chloroplasts of four lines. (C) Chl a content in leaves of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines grown under normal (5 µM Zn) and Zn-deficient (0 µM) conditions.

Discussion

AtMTP1 functions as a scavenger of excess Zn²⁺ in the cytosol by translocating Zn²⁺ into vacuoles. This function is evident from the phenotype property of mtp1, which is extremely sensitive to excess Zn²⁺ in culture medium (Kobae et al. 2004, Desbrosses-Fonrouge et al. 2005, Krämer 2010). Furthermore, mtp1-1 lines accumulated less Zn²⁺ in their roots and shoots than the wild type (Kawachi et al. 2009). We focused attention on the His-loop, which is distinctive for a subgroup of so-called Zn-CDF proteins from plants (Montanini et al. 2007). The His-half AtMTP1 conferred an enhanced activity and transported Co²⁺ when expressed in yeast cells (Kawachi et al. 2008). The His-loop of AtMTP1 is not essential for Zn²⁺ transport and was proposed to play the roles of a buffering pocket for Zn²⁺ and a sensor of the Zn²⁺ level, and to be involved in ion selectivity (Kawachi et al. 2008, Kawachi et al. 2012). This hypothesis was tested by characterization of transgenic lines expressing the His-half AtMTP1.

First, it should be noted that the His-half AtMTP1 is functional in transgenic plants, given that the His-half AtMTP1 complemented the lack of normal AtMTP1. The 35S-His-half lines recovered their tolerance to excess Zn, as shown by root elongation (Fig. 1). Furthermore, the His-half AtMTP1 protein was clearly detected in an immunoblot (Fig. 2). The His-half AtMTP1 efficiently transports excess cytosolic Zn²⁺ into vacuoles. We accordingly inferred that 35S-His-half lines show higher tolerance to excess Zn than Ws. However, the root length of 35S-His-half lines was the same as that of Ws under Zn excess conditions. We may suggest two possible explanations for this phenomenon: (i) the cytosol Zn²⁺ concentrations were too high to be removed by the His-half AtMTP1, although the mutated AtMTP1 actively transported Zn²⁺ into vacuoles; or (ii) the vacuolar Zn²⁺ concentration and/or a gradient of Zn²⁺ concentration across the vacuolar membrane was too high for active Zn²⁺ transport by the His-half AtMTP1. These explanations are supported by the observation that 35S-MTP1 also showed no extra tolerance to excess Zn²⁺ (Fig. 1).

The expression of the His-half AtMTP1, which was shown to transport Co²⁺ (Kawachi et al. 2008), did not change the tolerance to Co or content of Co as in the 35S-MTP1 lines (Fig. 3). A. thaliana probably has other Co-selective transporter(s) in the vacuolar membrane, and the contribution of the His-half AtMTP1 is negligible in plants.

Secondly, we discuss the physiological role of the His-loop of AtMTP1 in plants in view of the results of growth and Zn content in tissues under Zn-deficient conditions. Growth suppression of plants and a decrease in Chl a content are typical symptoms of Zn deficiency. These symptoms were observed with the medium containing a low concentration of TPEN, and were most severe in the 35S-His-half lines (Supplementary Fig. S4). However, cultivation in the medium containing TPEN seemed to damage plant tissues, probably due to removal of not only free Zn²⁺ but also other divalent metal ions from tissues, given that TPEN has relatively high affinity for Zn²⁺, Fe²⁺ and Mn²⁺ (Arslan et al. 1985, Thamatrakoln and Hildebrand 2008). In the other experiments, the agar with the lowest content of Zn and highly purified water were used to evaluate the effects of Zn deficiency. This cultivation medium may be suitable judging from the low Zn content in plants under the conditions mentioned below. Plants showed reductions of shoot fresh weight due to Zn deficiency (Fig. 4). In 35S-His-half lines, additional significant suppression was observed in shoot fresh weight. Suppression of growth in 35S-His-half shoots may be caused by the shortage of Zn in shoots.

The Zn contents in roots and shoots of 21-day-old plants cultivated under normal conditions were approximately 70 and 10 µg g DW^–1, respectively (Fig. 5). This imbalance of Zn between roots and shoots, representing Zn immobility in plants, is consistent with previous reports (Talke et al. 2006, Kawachi et al. 2009, Fukao et al. 2011). Zn may be predominantly sequestered in root cells, particularly in vacuoles, and transferred to shoots as needed. Under Zn-sufficient conditions, root cells absorb sufficient Zn²⁺ and the excess Zn²⁺ was transported into the vacuoles in roots via AtMTP1 and AtMTP3. Part of the Zn²⁺ in the root cells was translocated to shoots (Fig. 5A). Under Zn-sufficient conditions, there may be no difference in the Zn contents between Ws and 35S-His-half lines because normal AtMTP1 may express its full activity at high Zn²⁺ concentrations.

However, under Zn-deficient conditions, 35S-His-half plants sequestered more Zn²⁺ in roots and the level was 3-fold higher than that in Ws (Fig. 5B). The increase of Zn²⁺ content in the 35S-His-half roots is not due to side effects of other metals such as Fe, because there was no difference in contents of other metals including Fe between Ws and mutant lines (Fig. 6). Excess accumulation of Zn²⁺ in 35S-His-half roots led to the reduction of the Zn concentration in shoots, which was <50% that of Ws. This overaccumulation of Zn in roots may be due to the lack of the His-loop, which suppresses Zn transport activity of AtMTP1 under low-Zn conditions (Fig. 8D). However, the Zn transport activity of AtMTP1 in Ws and 35S-MTP1 plants may be adequately suppressed by the function of the His-loop under Zn-deficient conditions, resulting in a well-balanced distribution of Zn²⁺ in roots and shoots (Figs. 5, 8C). The reason why mtp1-1 plants accumulated Zn²⁺ in roots and shoots at the same level as Ws in the presence of sufficient Zn²⁺ (Fig. 5) may be the relatively high activity of AtMTP3 in the vacuolar membrane of 21-day-old plants (Arrivault et al. 2006, Ricachenevsky et al. 2013) (Fig. 8A), although AtMTP1 plays an essential role in the accumulation of Zn²⁺ in the vacuoles in young seedlings (Kobae et al. 2004, Krämer 2005). There was no significant difference in the mRNA levels of AtMTP3 between Ws and mutant lines (Supplementary Fig. S3). The present observations support our hypothesis that the His-loop senses the Zn level in the cytoplasm and suppress its transport activity under low-Zn conditions (Kawachi et al. 2008). This physiological function is essential for the growth of plants under Zn-deficient conditions (Figs. 4, 5). In roots, Zn²⁺ is absorbed by root hairs and epidermal cells (Tanaka et al. 2014) and then stored in the vacuoles via AtMTP1, which is expressed in most root tissues including the root tip and elongating and mature regions (Desbrosses-Fonrouge 2005, Kawachi et al. 2009). Thus, AtMTP1 may play a key role in Zn²⁺ distribution among tissues co-operatively with AtMTP3.

Fig. 8

Open in new tabDownload slide

Schematic model of the physiological effect of the AtMTP1 His-loop in plants under Zn-deficient conditions. Flows of Zn ions in plants expressing normal AtMTP1 (A and C) and the His-half AtMTP1 with high transport activity (B and D) under normal (A and B) and Zn-deficient conditions (C and D) are shown. AtMTP1 and the His-half AtMTP1 function as Zn²⁺/H⁺ exchangers with the help of proton pumps (blue circles). In the vacuolar membrane, another Zn²⁺/H⁺ exchanger, AtMTP3 (yellow ovals), functions in sequestration of Zn into vacuoles. Part of Zn taken from the culture medium or soils by roots is compartmentalized into the vacuoles in root cells by normal AtMTP1 (green ovals) or the His-half AtMTP1 (red ovals), and the remainder of Zn is transported to the vascular tissues and then transferred to shoots. The thickness of arrows corresponds to the flow amount of Zn in tissues. Zn contents in roots and tissues are shown according to the results in Fig. 5. The letter size of Zn²⁺ indicates the difference of Zn²⁺ levels.

At present, the molecular mechanism of the suppression of Zn transport activity specifically under low-Zn conditions in contrast to high-Zn conditions remains to be resolved. Several amino acid residues in the His-loop are involved in Zn transport and ion selectivity (Podar et al. 2012). The His-loop partly comprises a polyproline type II structure and binds four Zn²⁺ per molecule (Tanaka et al. 2013). The conformational change of the His-loop by Zn²⁺ binding to the loop has also been shown by circular dichroism spectroscopy (Tanaka et al. 2013). Kawachi et al. (2012) emphasized the functional importance of Asn258 for regulation of Zn transport activity. This Asn258 is located at the cytosolic side of the fifth transmembrane domain and linked to the His-loop, as inferred from the 3D structure of E. coli YiiP (Lu et al. 2009). It is accordingly considered that the His-loop is not occupied by Zn under low-Zn conditions and that the entrance to the Zn transport pore is obstructed by Asn258. Under high-Zn conditions, the His-loop changes its conformation and opens the translocation pore by changing the conformational position of Asn258 at the cytosolic side. This hypothesis should be tested by determination of the complete structures of AtMTP1 with or without Zn²⁺ bound to the His-loop.

In conclusion, this study revealed that the His-loop is involved in suppression of the Zn transport activity of AtMTP1 under Zn-deficient conditions. This suppression may be essential for plant growth under Zn-deficient conditions. MTP1 and related transporters in the CDF family in plants and various organisms have the His-loop, and this loop may, like AtMTP1, play a key role in Zn²⁺ homeostasis in the cytosol.

Materials and Methods

Plant materials and growth conditions

Arabidopsis thaliana strain Ws was used as the wild type. A homozygous loss-of-function mutant line mtp1-1 was prepared as described previously (Kobae et al. 2004). Seeds were surface-sterilized and germinated on sterile gel plates containing 2.5 mM MES-KOH, pH 5.7, 1% (w/v) sucrose, 1.2% Ina agar (Funakoshi) and modified Hoagland medium (Haydon et al. 2012) for the preparation of Zn-deficient agar plates. The modified Hoagland medium contained 1.5 mM Ca(NO₃)₂, 0.28 mM KH₂PO₄, 0.75 mM MgSO₄, 1.25 mM KNO₃, 0.5 µM CuSO₄, 5 µM ZnSO₄, 5 µM MnSO₄, 25 µM H₃BO₃, 0.1 µM Na₂MoO₄, 50 µM KCl, 5 µM Fe-HBED and 3 mM MES-KOH, pH 5.7. Standard agar plates contained Murashige and Skoog (MS) salt (Wako Pure Chemical), 2.5 mM MES-KOH, pH 5.7, 1% (w/v) sucrose and 1.2% Ina agar. ZnSO₄ and CoSO₄ were added to the medium at the indicated concentrations to evaluate the tolerance to excess Zn or Co. Seedlings were grown in the medium containing TPEN, a Zn chelator, to evaluate growth under Zn-deficient conditions (Thamatrakoln and Hildebrand 2008). Seedlings were grown at 22°C under long-day conditions (16 h light/8 h dark at 80–110 µmol m^–2 s^–1). In all experiments, the seeds were incubated at 4°C for 3 d in the dark before culture on medium.

Production of transgenic plants

cDNAs for an AtMTP1 (Kobae et al. 2004) and a Δ185–216 mutant AtMTP1 (Kawachi et al. 2008) in pKT10 vectors were prepared as described previously. The cDNAs for the pENTR entry vector were cloned into pGWB2 binary vector with the 35S promoter of Cauliflower mosaic virus, introduced into the Agrobacterium tumefaciens GV3101::PM90 strain and then used to transform mtp1-1 homozygous mutant plants by the floral dip method (Clough and Bent 1998). The resulting homozygous plants were used as mtp1-1:35S::AtMTP1 (35S-MTP1) and mtp1-1:35S::Δ185–216 AtMTP1 (35S-His-half) lines after selection with kanamycin and hygromycin. Plants of 9–11 independent T₄ lines were used for physiological experiments.

Determination of mineral contents in plant tissues and agars

Arabidopsis thaliana seedlings were grown on agar plates, which were set vertically. After 21 d, shoots and roots were collected from seedlings, washed once in 10 mM EDTA and three times in ultra-pure water, and then dried at 70°C for 48 h. Dried tissues (40–100 mg) were digested with pure concentrated HNO₃ for 25 min at 130°C in Teflon vessels (Ethos-1600). The mineral contents of the seedlings were then determined by ICP-AES (IRIS ICAP, Nippon Jarrell Ash).

The Zn content in agar products produced by different manufacturers was measured by ICP-AES. Agar products were from Sigma-Aldrich (type M), Funakoshi (Ina agar type BA-10) and Nacalai Tesque (agar purified). The Zn content of gellan gum (Wako Pure Chemical) was also determined.

Preparation of vacuolar membrane fraction and immunoblot analysis

Vacuolar membrane fractions were prepared from 14-day-old seedlings (n = 500) of Ws, mtp1-1, 35S-MTP1 and 35S-His-half plants grown in MS medium as described previously with some modifications (Maeshima and Yoshida 1989). Plants were homogenized in medium containing 0.25 M sorbitol, 50 mM Tris-acetate, pH 7.5, 1 mM EGTA, 20 µM p-(amidinophenyl)methanesulfonyl fluoride hydrochloride, 1% (w/v) polyvinylpyrrolidone and 2 mM dithiothreitol. The homogenates were centrifuged at 7,000×g at 4°C for 10 min, and the supernatants were centrifuged at 100,000×g at 4°C for 35 min. The pellets, which were crude membrane fractions, were suspended in the buffer containing 0.7 M sucrose, 20 mM Tris-acetate, pH 7.5, 1 mM EGTA and 2 mM MgCl₂ using a Teflon homogenizer. The suspension was overlaid with 0.25 M sorbitol, 20 mM Tris-acetate (pH 7.5), 1 mM EGTA, 2 mM MgCl₂ and 2 mM dithiothreitol. After centrifugation at 100,000×g for 30 min, the turbid interface portion was collected and diluted in 3 vols. of the homogenization medium. The suspension was then centrifuged at 100,000×g for 35 min, and the resulting precipitate (vacuolar membrane) was suspended in the buffer.

The proteins in the fractions were separated by SDS–PAGE and transferred to an Immobilon-P membrane (Millipore). After blocking with 5% defatted milk, the membrane filter was incubated with primary antibody and then with horseradish peroxidase-conjugated protein A (1 : 2,000) (GE Healthcare). Chemiluminescent reagents of ECL (GE Healthcare) were used for the detection of antigens. Antibodies specific to the N-terminal region of AtMTP1 (At2g46800, position 1–12, MESSSPHHSHIV) (Tanaka et al. 2013) and the substrate-binding region of H⁺-PPase (At1g15690, position 257–273, DLVGKIERNIPEDDPRN) (Segami et al. 2010) were used as anti-AtMTP1 antibodies (1 : 1,000 dilution) and anti-H⁺-PPase antibodies (1 : 4,000 dilution), respectively.

Microscopic observations

Images of chloroplast autofluorescence in cells of the second layer of leaf epidermis of a 27-day-old plant were obtained using an upright FV1000-D confocal laser scanning microscope (Olympus). The excitation wavelength and transmission range for emission were 473 nm and 630–730 nm, respectively.

Chl quantification

RNA extraction and real-time RT–PCR

Total RNA fractions were prepared from 22-day-old whole plants of Ws, mtp1-1, 35S-MTP1 and 35S-His-half lines using a QIA shredder and RNeasy Mini kit (Qiagen). RNA (0.5 µg) was converted into cDNA using a ReverTra Ace qPCR RT Master Mix with gDNA Remover (Toyobo). Real-time reverse transcription–PCR (RT–PCR) analysis was performed with a Thermal Cycler Dice™ Real Time System TP800 (TAKARA) using a THUNDERBIRD™ SYBR qPCR Mix (Toyobo). The primer sets used for the real-time RT–PCR were ATGGAGTCTTCAAGTCCCCAC (Fw) and CCGGAAGCATTCTTAGAATCTG (Rv) for AtMTP1, CTGATGTTGCAGCCTTTGCA (Fw) and TTTGCCTTCCATCCCGAA (Rv) for AtMTP3 and GTCGAGGTTTACGCATCCATG (Fw) and CTCGCCAAAAACCATTTCAG (Rv) for AtTIP4;1. Relative mRNA contents were normalized to the AtTIP4;1 transcript.

Funding

This work was supported by the Japan Society for the Promotion of Science (JSPS) [Grants-in-Aid Nos: 25650093 and 26252011]; the Steel Foundation for Environmental Protection Technology (2010-29); and the Japan Science and Technology Agency (JST) A-Step [AS242Z02907N] to M.M. N.T. received Research Fellowship for Young Scientists (doctoral course students) and Program for Leading Graduate School from the JSPS.

Acknowledgments

We are grateful to Dr. Yoichi Nakanishi and Dr Shoji Segami (Nagoya University) for their fruitful advice.

Disclosures

The authors have no conflicts of interest to declare.

Abbreviations

Abbreviations
 
• CDF

  cation diffusion facilitator

 
• His-loop

  histidine-rich loop

 
• His-half

  a deletion mutant lacking the first half of the histidine-rich loop

 
• H⁺-PPase; H⁺-pyrophosphatase; ICP-AES

  inductively coupled plasma atomic emission spectroscopy

 
• MS

  Murashige and Skoog

 
• MTP

  metal tolerance protein

 
• TPEN

  N, N, N′, N′-tetrakis(2-pyridylmethyl)ethylenediamine

References