OsATX1 Interacts with Heavy Metal P1B-Type ATPases and Affects Copper Transport and Distribution^,

OsATX1 controls root-to-shoot translocation of copper and its delivery to developing tissues in rice by interacting with the rice heavy metal P1B-ATPases HMA4, HMA5, HMA6, and HMA9.

Abstract

Copper (Cu) is an essential micronutrient for plant growth. However, the molecular mechanisms underlying Cu trafficking and distribution to different organs in rice (Oryza sativa) are poorly understood. Here, we report the function and role of Antioxidant Protein1 (OsATX1), a Cu chaperone in rice. Knocking out OsATX1 resulted in increased Cu concentrations in roots, whereas OsATX1 overexpression reduced root Cu concentrations but increased Cu accumulation in the shoots. At the reproductive stage, the concentrations of Cu in developing tissues, including panicles, upper nodes and internodes, younger leaf blades, and leaf sheaths of the main tiller, were increased significantly in OsATX1-overexpressing plants and decreased in osatx1 mutants compared with the wild type. The osatx1 mutants also showed a higher Cu concentration in older leaves. Yeast two-hybrid and bimolecular fluorescence complementation assays showed that OsATX1 interacts with the rice heavy metal P_1B-ATPases HMA4, HMA5, HMA6, and HMA9. These results suggest that OsATX1 may function to deliver Cu to heavy metal P_1B-ATPases for Cu trafficking and distribution in order to maintain Cu homeostasis in different rice tissues. In addition, heterologous expression of OsATX1 in the yeast (Saccharomyces cerevisiae) cadmium-sensitive mutant Δycf1 increased the tolerance to Cu and cadmium by decreasing their respective concentrations in the transformed yeast cells. Taken together, our results indicate that OsATX1 plays an important role in facilitating root-to-shoot Cu translocation and the redistribution of Cu from old leaves to developing tissues and seeds in rice.

Copper (Cu) is an essential micronutrient for the growth and development of all living organisms, serving as a cofactor in many enzymes. Cu deficiency endangers human health, causing immune defects (Collins and Klevay, 2011). In plants, Cu plays a key role in many biological processes, including photosynthetic and respiratory electron transport, cell wall remodeling, oxidative stress responses, and ethylene perception (Maksymiec, 1997; Pilon et al., 2006; Yruela, 2009). Visible symptoms of Cu deficiency include stunted growth, distortion or whitening of young leaves, damage to the apical meristem, and decreased seed setting and yield (Burkhead et al., 2009; Huang et al., 2016). Excess Cu, however, is extremely toxic to plants and generates reactive oxygen species, which cause cellular damage (Leng et al., 2015). Therefore, the uptake and distribution of Cu must be regulated strictly in plants.

Following Cu uptake by roots, Cu may be sequestered into the vacuoles or translocated to the shoots and then distributed/redistributed to different organs and tissues (Yamaji and Ma, 2014; Sasaki et al., 2016). Because Cu ions have high affinities for nitrogen- or sulfur-containing ligands, the amount of free Cu ions in the cytosol probably is limited to less than one ion per cell (Rae et al., 1999). In plant shoot tissues, a Cu concentration of 5 mg kg⁻¹ dry weight is considered to be sufficient for plant growth and development, whereas more than 20 mg kg⁻¹ can induce toxicity (Marschner, 1995; Burkhead et al., 2009), and plants have developed a complex Cu homeostatic network to facilitate its use and avoid toxicity in shoots. Cu homeostasis in plants depends on the control of root uptake (Sancenón et al., 2004), Cu trafficking via P-type ATPases and Cu chaperones (Himelblau et al., 1998; Hirayama et al., 1999; Abdel-Ghany et al., 2005a; Puig et al., 2007; Li et al., 2017), and regulation of the mRNAs of Cu proteins by microRNAs in response to Cu availability (Yamasaki et al., 2007).

The initial uptake of Cu into root cells is performed mainly by high-affinity transporters belonging to the Cu transporter (CTR) family, which are called COPT (Copper Transporter) proteins in plants (Puig and Thiele, 2002; Sancenón et al., 2003; Puig et al., 2007). There are six and seven COPT genes in the genomes of Arabidopsis (Arabidopsis thaliana) and rice (Oryza sativa), respectively (Sancenón et al., 2003; Yuan et al., 2010, 2011; Peñarrubia et al., 2015). In Arabidopsis, AtCOPT1, AtCOPT2, and AtCOPT6 are localized in the plasma membrane and involved in Cu uptake into the cytosol (Sancenón et al., 2004; Andrés-Colás et al., 2010; Jung et al., 2012; Garcia-Molina et al., 2013; Perea-García et al., 2013), whereas AtCOPT5 is localized intracellularly and functions in Cu mobilization from intracellular storage organelles (Garcia-Molina et al., 2011; Klaumann et al., 2011). In rice, OsCOPT1 and OsCOPT5 interact with themselves and each other, and the two proteins cooperate with the protein Xa13 (Bacterial Blight Resistance13) to affect Cu transport and distribution (Yuan et al., 2010). OsCOPT2, OsCOPT3, and OsCOPT4 interact individually with OsCOPT6 and cooperatively mediate highly efficient Cu transport in the ctr1Δ ctr3Δ yeast mutant (Yuan et al., 2011). In contrast, OsCOPT7 was found to function alone in Cu transport in the same yeast mutant (Yuan et al., 2011).

The P_1B-ATPases, also known as heavy metal ATPases (HMAs), play important roles in the transport of essential and potentially toxic metals (i.e. zinc [Zn], Cu, cadmium [Cd], and lead [Pb]) across cell membranes (Williams and Mills, 2005; Takahashi et al., 2012; Yan et al., 2016). There are eight and nine members of the P_1B-ATPase family in Arabidopsis and rice, respectively (Baxter et al., 2003). Of these, AtHMA5, AtHMA6, AtHMA7, and AtHMA8 are associated with Cu homeostasis (Burkhead et al., 2009). PAA1 (AtHMA6) and PAA2 (AtHMA8) are responsible for Cu delivery to the chloroplasts. PAA1 (AtHMA6) transports Cu across the plastid envelope, whereas PAA2 (AtHMA8) functions in Cu transport through the thylakoid membrane (Shikanai et al., 2003; Abdel-Ghany et al., 2005b). The Responsive to Antagonist (RAN1)/AtHMA7 transporter transports Cu to the ethylene receptors (Hirayama et al., 1999; Woeste and Kieber, 2000; Li et al., 2017). The AtHMA5 transporter, the closest homolog of RAN1 in the P_1B-type ATPase subfamily, is involved in Cu efflux and functions in loading Cu into the xylem for root-to-shoot translocation or Cu detoxification in Arabidopsis roots (Andrés-Colás et al., 2006; Kobayashi et al., 2008; Huang et al., 2016). Among the rice P_1B-ATPases, six members (OsHMA4–OsHMA9) belong to the Cu⁺/Ag⁺ transporter group (Deng et al., 2013). To date, only three out of six members have been characterized functionally. OsHMA4 is localized to the tonoplast and has been shown to function in sequestering Cu into the vacuoles in rice roots (Huang et al., 2016), whereas OsHMA5 is localized to the plasma membrane and involved in loading Cu to the xylem for root-to-shoot translocation (Deng et al., 2013). OsHMA9 is a metal efflux transporter responsible for Cu, Zn, Cd, and Pb efflux from the cells (Lee et al., 2007). Moreover, OsHMA9 and OsHMA6 are closely related to Arabidopsis RAN1 (Williams and Mills, 2005; Deng et al., 2013).

The Cu chaperones are low-molecular-weight metal-receptor proteins that mediate intracellular Cu trafficking to specific Cu proteins and various cell organelles (O’Halloran and Culotta, 2000). In Arabidopsis, several Cu chaperones are involved in Cu delivery inside the cell, such as COX17 (Cu chaperone for cytochrome c oxidase), CCS (Cu chaperone for superoxide dismutase), and two homologs of the yeast ATX1 (Antioxidant Protein1) and CCH (ATX1-like Cu chaperone; Casareno et al., 1998; Chu et al., 2005; Puig et al., 2007). Among these Cu chaperones, CCS, CCH, and ATX1 all have an open-faced β-sandwich global fold with a conserved MXCXXC Cu-binding motif (Harrison et al., 1999; Puig et al., 2007). The AtCOX17 chaperone shows sequence similarity to the yeast COX17 that delivers Cu to cytochrome c oxidase, which is essential in the respiratory chain (Balandin and Castresana, 2002). The AtCCS chaperone shares a functional homolog with Ccs1p/Lys-7p from yeast that mediates the delivery of Cu to Cu/Zn-superoxide dismutase by protein-protein interaction (Abdel-Ghany et al., 2005a). Therefore, AtCCS, a cytosol and plastid dual-localized chaperone (Chu et al., 2005), was proposed to deliver Cu to both cytosolic and chloroplastic Cu/Zn-superoxide dismutase enzymes (Abdel-Ghany et al., 2005a). Both AtCCH and AtATX1 chaperones can complement the yeast atx1 mutant. The protein sequence of AtCCH also has a unique C-terminal extension (Mira et al., 2001), which might mediate plant-specific functions. Yeast two-hybrid experiments suggested that AtATX1 and AtCCHΔ (C-terminal extension-deleted AtCCH) interact directly with the N-terminal Cu-binding domain of AtHMA5 and RAN1 (AtHMA7), which includes MXCXXC residues thought to be involved in metal binding (Williams and Mills, 2005; Andrés-Colás et al., 2006; Puig et al., 2007). However, these interactions were abolished in the presence of the C terminus of AtCCH (Puig et al., 2007).

Cu⁺ ions do not access Cu⁺-ATPases in a free (hydrated) form but are bound to a chaperone protein (González-Guerrero and Argüello, 2008). Genetic evidence shows that AtATX1 delivers Cu⁺ to RAN1 for ethylene receptor biogenesis and signaling (Li et al., 2017). In Arabidopsis, AtATX1 but not AtCCH chaperone contributes to the tolerance of both excess and slightly deficient Cu (Woeste and Kieber, 2000; Shin et al., 2012). Overexpression of AtATX1 enhanced Cu tolerance in excess-Cu conditions and increased Cu sensitivity to severe Cu deficiency, causing growth inhibition (Shin et al., 2012; Shin and Yeh, 2012). However, the functions of Cu chaperones in rice remain unknown.

Cd is one of the most toxic heavy metals, disrupting several biochemical activities by displacing essential metals from their respective binding sites (Ueno et al., 2010; Heo et al., 2012; Mendoza-Cózatl et al., 2014; Clemens and Ma, 2016). The high affinity of Cd for thiols is considered the primary mechanism underlying Cd toxicity (Wei et al., 2014). In humans, Cd has a long biological half-life of 10 to 30 years and is retained in the kidneys. Chronic exposure to Cd can cause renal dysfunction, osteoporosis, and cancers (Nawrot et al., 2006). Because Cd can transfer readily from soil to food crops, dietary intake of Cd accounts for about 90% of the Cd exposure in the general nonsmoking population (Clemens et al., 2013). Therefore, it is important to reduce Cd accumulation in food crops, especially in rice.

In this study, we investigated the function and role of the OsATX1 chaperone in rice. We showed that OsATX1 plays an important role in promoting root-to-shoot Cu translocation and distribution to the developing tissues by interacting with the rice heavy metal P_1B-ATPases HMA4, HMA5, HMA6, and HMA9. In addition, heterologous expression of OsATX1 in the yeast Cd-sensitive mutant Δycf1 increased the tolerance to Cu and Cd.

RESULTS

Phylogenetic Analysis of OsATX1

The full-length coding region of OsATX1 (Os08g0205400) was amplified by PCR from cDNA of wild-type rice roots (var japonica, cv Zhonghua11) using primers designed according to the National Center for Biotechnology Information (NCBI; https://www.ncbi.nlm.nih.gov/). The sequence obtained was the same as that registered in the database. OsATX1 contains three exons and two introns, encoding a peptide of 81 amino acids.

Phylogenetic analysis shows that OsATX1 exhibits high amino acid sequence similarity (87% identity) to the Arabidopsis Cu chaperone AtATX1 (Supplemental Fig. S1). OsATX1 shares 75% and 41% amino acid sequence identity with rice OsCCH (the Cu chaperone) and OsATX, respectively (Supplemental Fig. S1). Protein sequence alignment of these ATX1-type metallochaperones suggests that OsATX1 possesses the predicted overall β-fold structure and an MXCXXC Cu⁺-binding motif (Supplemental Fig. S2).

Expression Pattern of OsATX1

Reverse transcription quantitative PCR (RT-qPCR) showed that OsATX1 was expressed in most rice organs throughout the growth period of cv Zhonghua11 grown in a paddy field (Fig. 1A). At the reproductive growth stage, the expression of OsATX1 was slightly higher in roots, leaves, and rachis compared with other organs, including basal stem, node, internode, and spikelet (Fig. 1A). The expression of OsATX1 was induced by the deficiency of iron (Fe), manganese (Mn), and Zn (Fig. 1B) and by a high concentration of Cu or Cd (Fig. 1, B–F). OsATX1 expression was significantly induced within 2 d of 20 µm Cu treatment (Fig. 1D) or 3 d of 20 µm Cd treatment (Fig. 1F) but was not significantly affected by high concentrations of Fe, Mn, or Zn (Supplemental Fig. S3). Spatial expression analysis showed that the expression of OsATX1 was highest in the mature root zone (1.5–2.5 cm from the apex) and lowest in the root tips (0–0.5 cm; Fig. 1G).

Figure 1.

Expression pattern of OsATX1 in rice. A, Expression levels of OsATX1 in different rice organs at different growth stages. Samples were taken from cv Zhonghua11 grown in a paddy field. B, Organ-dependent expression of OsATX1. Seedlings were exposed to a nutrient solution with or without Fe, Mn, Zn, or Cu for 7 d. C and E, OsATX1 expression in response to Cu and Cd. Seedlings were exposed to different concentrations of Cu (C) or Cd (E) for 2 weeks. D and F, Time-course expression of OsATX1 under Cu (D) or Cd (F) treatment. Seedlings were treated with 20 μm Cu or Cd for the indicated days. G, Root spatial expression of OsATX1. Different root segments (0–0.5, 0.5–1.5, and 1.5–2.5 cm from the root tip) were excised. The expression level was determined by RT-qPCR. Data are means ± sd (n = 3). Statistical comparison was performed by Tukey’s test: ** P < 0.01.

Tissue Specificity of OsATX1 Expression

To examine the tissue and cell specificity of OsATX1 expression, transgenic plants were generated with GUS and GFP reporter gene expression driven by the OsATX1 promoter. GUS activity was detected preferentially in the root elongation and mature zones and in lateral roots but not in the root caps (Fig. 2, A–C). Transverse sections of roots showed that the exodermis and the stele had the highest GUS activity (Fig. 2, F–H), whereas the GUS activity also was found in the leaf vascular bundles (Fig. 2, D, E, I, and J). GFP fluorescence indicated strong expression of OsATX1 in the leaf vascular bundles and in the exodermis and the stele of roots (Fig. 2, K–S), consistent with the pattern revealed by GUS staining.

Figure 2.

Tissue specificity of OsATX1 expression. A to J, Expression pattern of OsATX1 detected by histochemical straining of GUS activity in plants transformed with the construct OsATX1_Pro:GUS. A, Part of a seedling. B, Lateral roots. C Main root. D, Leaf. E, Magnified image of part of D. F, Transverse section of a root. G, Enlarged view of a vascular bundle from F. H, Transverse section of a lateral root. I, Transverse section of a leaf. J, Enlarged view of a vascular bundle from I. K to S, Expression pattern of OsATX1 detected by confocal microscopy analysis of GFP fluorescence in plants transformed with the construct OsATX1_Pro:GFP. K to M, Transverse section of a leaf. N to P, Transverse section of a root. Q to S, Enlarged view of a vascular bundle from N to P. GFP fluorescence (K, N, and Q), bright-field images (L, O, and R), and merged images (M, P, and S) are shown. Bars = 1 mm for A to E, 50 µm for F, G, and I to P, and 20 µm for H and Q to S.

Subcellular Localization of OsATX1

To determine the subcellular localization of OsATX1, the coding sequence of full-length OsATX1 protein was fused to GFP. OsATX1-GFP was cloned into a transient expression vector under the control of the constitutive cauliflower mosaic virus 35S promoter. We observed that the OsATX1-GFP signal was diffuse and largely colocalized with AtATX1-mCherry in rice mesophyll protoplasts (Supplemental Fig. S4A). It has been shown that AtATX1-mCherry is localized to the cytoplasm and nucleus and adheres to endoplasmic reticulum membranes (Puig et al., 2007; Li et al., 2017).

To confirm this observation, we also used a transient expression system to express OsATX1-GFP and OsNRAMP5-RFP in Nicotiana benthamiana leaf epidermal cells. OsNRAMP5 is an Mn and Cd transporter in rice that localizes to the plasma membrane (Ishikawa et al., 2012; Sasaki et al., 2012; Yang et al., 2014). We observed that the OsATX1-GFP signal also was partially colocalized with OsNRAMP5-RFP in the plasma membrane (Supplemental Fig. S4B). Together, our subcellular observations indicate that OsATX1 is localized to the cytoplasm, nucleus, and plasma membrane and also may adhere to endomembranes.

Phenotypic Analysis of OsATX1 Transgenic Lines

To investigate the physiological role of OsATX1 in rice, we generated OsATX1 knockout mutants (CRISPR lines) and OsATX1 overexpression plants in the cv Zhonghua11 background (Supplemental Fig. S5). When grown in normal hydroponic culture, the OsATX1 transgenic lines showed no apparent phenotypes compared with the wild type (cv Zhonghua11; Fig. 3, A–F). Analysis using inductively coupled plasma mass spectrometry (ICP-MS) showed that the Cu concentrations were significantly higher in the shoots but lower in the roots of three independent OsATX1-overexpressing lines compared with wild-type plants (Fig. 3, G and H). The Cu concentrations in the roots of the osatx1 mutants were significantly higher than that of the wild-type plants (Fig. 3H), but no significant changes were found in the shoots (Fig. 3G). There were no consistent differences in the concentrations of other analyzed essential metals (Fe, Mn, and Zn) in either shoots or roots between OsATX1 transgenic lines and the wild type (Supplemental Fig. S6, A–F). The concentrations of Zn in the shoots of two out of the three osatx1 mutants were slightly lower than that in the wild-type plants (Supplemental Fig. S6E). In addition, the concentrations of Cd in the shoots of the OsATX1-overexpressing plants and the roots of osatx1 mutants also were higher than that in the wild type (Supplemental Fig. S6, G and H). These results indicate that OsATX1 affects the transport of Cu and Cd.

Figure 3.

Characterization of OsATX1 transgenic lines grown in normal nutrient solution. A, Phenotype comparison of wild-type plants (cv Zhonghua11 [ZH11]), three OsATX1-overexpressing (OX) lines, and the osatx1 mutants (three different CRISPR lines) at the vegetative stage. B to D, Crown root number (B), plant height (C), and primary root length (D). E and F, Dry weight of shoots (E) and roots (F). G and H, Cu concentrations in shoots (G) and roots (H) of OsATX1 transgenic lines and the wild type. All plants were cultivated in normal nutrient solution containing 0.12 µm Cu for 5 weeks. All data were compared with cv Zhonghua11. DW, Dry weight. Data are means ± sd of five independent biological replicates. Statistical comparison was performed by Tukey’s test: ** P < 0.01.

OsATX1-Overexpressing Plants Are More Sensitive to Excess Cu

In Arabidopsis, overexpression of AtATX1 was found to enhance Cu tolerance in plants grown under excess Cu conditions and to increase Cu sensitivity to severe Cu deficiency (Shin et al., 2012; Shin and Yeh, 2012). We compared the Cu tolerance of OsATX1 transgenic lines with that of the wild type under different Cu concentrations. Under Cu deficiency, the growth of all OsATX1 transgenic lines was similar to that of the wild type (Fig. 4A). However, in the presence of excess Cu (2–20 µm), the growth of OsATX1-overexpressing plants was stunted, with significantly reduced plant height, primary root length, dry weight of shoots and roots, crown root number, and crown root length compared with that of wild-type plants (Fig. 4). The growth of the osatx1 mutants also was affected slightly when Cu concentration was higher than 10 µm (Fig. 4). At 20 µm Cu, the crown root length of the osatx1 plants was significantly lower than that of the wild type, but there was no significant difference in the crown root number (Fig. 4, F and G). At 50 µm Cu, the growth of the crown root in all plants was inhibited almost completely (Fig. 4, A–F). When supplied with excess or deficient levels of Fe, Mn, and Zn, the OsATX1 transgenic lines showed no significant difference compared with the wild type (Supplemental Fig. S7). These results indicate that OsATX1-overexpressing plants, different from AtATX1 overexpression in Arabidopsis, are more sensitive to Cu excess.

Figure 4.

Phenotypes of OsATX1 transgenic lines under Cu stress and Cu deficiency. Plant lines are as described in Figure 3. A, Cu tolerance test. Ten-day-old seedlings were exposed to a nutrient solution containing 20 µm Cu chelator bathocuproine disulfonate (BCS) or different Cu concentrations. After 4 weeks, the plants were photographed. B, Plant height. C, Primary root length. D and E, Dry weight of shoots (D) and roots (E). F, Crown root number. G, Crown root length of treated plants with 20 µm CuSO₄. Data are means ± sd of five independent biological replicates. Statistical comparison was performed by Tukey’s test: *, P < 0.05 and **, P < 0.01.

OsATX1 Affects the Root-to-Shoot Translocation of Cu

To further dissect the role of OsATX1 in Cu transport, a Cu concentration gradient experiment was performed. Plants were grown initially in normal nutrient solution containing 0.12 µm Cu for 4 weeks and then transferred to nutrient solutions containing 2 µm Cu chelator bathocuproine disulfonate or 0.4 or 2 µm Cu for another week. The Cu concentrations were significantly higher in shoots and lower in roots in three independent OsATX1-overexpressing plants compared with the wild type grown with each of the Cu treatments (Fig. 5, A and B). A higher root-to-shoot Cu translocation was found in OsATX1-overexpressing plants (Fig. 5C). The Cu concentrations in roots of the osatx1 mutants were significantly higher than those in the wild-type plants at each Cu treatment. However, no significant difference in the Cu concentration of shoots was found between the wild-type plants and the osatx1 mutants (Fig. 5, A and B). The root-to-shoot translocation of Cu in osatx1 mutants was significantly lower than that in the wild type (Fig. 5C). When treated with 0.02, 0.2, or 2 µm Cd for 1 week, the OsATX1-overexpressing plants contained higher levels of Cd in the shoots than the wild-type plants (Fig. 5D), whereas osatx1 mutants accumulated higher Cd concentrations in the roots (Fig. 5G). However, the Cd concentration did not change significantly in the roots of OsATX1-overexpressing plants or in the shoots of the osatx1 mutants compared with the wild type (Fig. 5, E and F).

Figure 5.

Uptake and translocation of Cu and Cd in OsATX1 transgenic lines. Plant lines are as described in Figure 3. A and B, Cu concentrations in the shoots (A) and roots (B). C, Translocation of Cu from roots to shoots. D to G, Cd concentrations in shoots (D and F) and roots (E and G). H and I, Dose responses of Cu (H) and Cd (I) in the xylem sap. Xylem sap was collected from cv Zhonghua11 and the OsATX1 transgenic lines exposed to different Cu or Cd concentrations for 3 d. All data were compared with cv Zhonghua11. BCS, Bathocuproine disulfonate; DW, dry weight. Data are means ± sd of three independent biological replicates. Statistical comparison was performed by Tukey’s test: *, P < 0.05 and **, P < 0.01.

To confirm these results, we further determined Cu and Cd concentrations in the xylem sap. The Cu and Cd concentrations in the xylem sap increased with increasing external Cu or Cd concentration in all plants (Fig. 5, H and I). In the OsATX1-overexpressing plants, the Cu and Cd concentrations in xylem sap were significantly higher than those in the wild type at each treatment (Fig. 5, H and I). However, there was no significant difference in the concentration of Cu and Cd in the xylem sap between the osatx1 mutants and the wild type (Fig. 5, H and I). These results indicate that OsATX1 is involved in the translocation of Cu from roots to shoots and that the overexpression of OsATX1 also increases Cd translocation from roots to shoots.

OsATX1 Is Required for Cu Distribution to Developing Tissues at the Reproductive Stage

At the reproductive stage, we also analyzed the Cu distribution in different organs. The OsATX1-overexpressing plants accumulated significantly higher concentrations of Cu in the panicles, upper nodes and internodes, as well as younger leaf blades and leaf sheaths of the main tiller but lower Cu concentrations in the basal stem and the roots compared with the wild-type plants (Fig. 6A). In contrast, the osatx1 mutants showed significantly lower Cu concentrations in panicles, upper nodes, internodes, and younger leaf sheaths but higher concentrations in the older leaf blades and leaf sheaths (Fig. 6B). In addition, overexpression of OsATX1 markedly increased the concentration of Cu in the brown rice, by 188% to 282% compared with the wild type, when exposed to 0.4 µm Cu in hydroponic culture until ripening (Fig. 6C). By contrast, the osatx1 mutants accumulated less Cu (64%–65%) in the brown rice compared with the wild type (Fig. 6D). Under 0.2 µm Cd treatment, overexpression of OsATX1 increased the Cd concentrations in all aboveground tissues at the flowering stage (Supplemental Fig. S8A). OsATX1-overexpressing plants also contained 32% to 37% higher Cd in the brown rice (Supplemental Fig. S8B), whereas no significant difference was shown in the brown rice of osatx1 mutants and the wild type (Supplemental Fig. S8C). In addition, there were no significant differences in the concentrations of Fe, Zn, and Mn in the brown rice from OsATX1 transgenic lines and the wild type (Supplemental Fig. S9, A and B). These results suggest that OsATX1 plays a vital role in the Cu distribution from older leaf blades and leaf sheaths to developing tissues such as younger leaves and the panicle.

Figure 6.

Cu concentrations in different rice tissues. Plant lines are as described in Figure 3. Plants were initially grown in normal nutrient solution. At the booting stage, OsATX1-overexpressing and wild-type plants were transferred to a nutrient solution containing 0.4 µm Cu (A and C), and at the late tillering stage, osatx1 mutant and wild-type plants were exposed to 0.4 µm Cu (B and D). Different organs at the flowering stage (A and B) and mature grains (C and D) were sampled, and Cu concentrations were determined. DW, Dry weight. Data are means ± sd of three independent biological replicates. Statistical comparison was performed by Tukey’s test: *, P < 0.05 and **, P < 0.01.

OsATX1 Interacts with Heavy Metal P_1B-Type ATPases

Previous studies indicated that Arabidopsis ATX1-like Cu chaperones interact with the N-terminal metal-binding domains of AtHMA5 and RAN1 (AtHMA7; Andrés-Colás et al., 2006; Puig et al., 2007; Shin et al., 2012). Because our data strongly support a role for OsATX1 in Cu translocation and distribution, it is possible that OsATX1, like AtATX1, also recruits Cu to heavy metal P_1B-ATPases. To test this hypothesis, we first performed yeast two-hybrid assays to investigate whether OsATX1 interacts in vivo with the heavy metal P_1B-ATPases. As shown in Figure 7A, cells coexpressing OsATX1 with OsHMA4, OsHMA5, OsHMA6, or OsHMA9 all were able to grow in the medium without Leu, Trp, and His, indicating that there are interactions between OsATX1 and heavy metal P_1B-ATPases (OsHMA4, OsHMA5, OsHMA6, and OsHMA9). Following this, to confirm the interaction in planta, we used bimolecular fluorescence complementation (BiFC) assays in N. benthamiana. When split yellow fluorescent protein (YFP) constructs containing OsATX1 and heavy metal P_1B-ATPase (OsHMA4, OsHMA5, OsHMA6, or OsHMA9 individually) were cotransformed into N. benthamiana leaves, strong YFP signals were observed (Fig. 7B). These results demonstrate that OsATX1 is able to interact with the rice heavy metal P_1B-ATPases HMA4, HMA5, HMA6, and HMA9.

Figure 7.

Functional interaction between OsATX1 and heavy metal P-type ATPases. A, Yeast two-hybrid assays of OsATX1 interaction with OsHMA4, OsHMA5, OsHMA6, and OsHMA9. AH109 cells cotransformed with pDEST22 (AD) and pDEST32 (BD) plasmids were grown in liquid cultures and spotted on synthetic defined (SD)-Trp-Leu and SD-Trp-Leu-His plates. X-gal, 5-Bromo-4-chloro-3-indolyl-β-d-galactopyranoside acid. B, BiFC assays of OsATX1 interaction with OsHMA4, OsHMA5, OsHMA6, and OsHMA9. The indicated split YFP constructs were transiently coexpressed in leaves of N. benthamiana for 3 d, and fluorescent images were obtained using a confocal laser scanning microscope (Olympus FV1000). Bars = 20 μm.

OsATX1 Confers Cd and Cu Tolerance in the Yeast Mutant Strain Δycf1

To gain a better understanding of the roles of OsATX1, we expressed OsATX1 in the yeast (Saccharomyces cerevisiae) wild-type strain BY4741 and the Cd-sensitive mutant strain Δycf1. The resulting Cd and Cu tolerance conferred by OsATX1 was determined by growth assays. Expression of OsATX1 in the Cd-sensitive yeast mutant strain Δycf1 increased its Cd tolerance at 60 µm CdCl₂, whereas the expression of OsNRAMP5, an Mn/Cd transporter (Ishikawa et al., 2012; Sasaki et al., 2012; Yang et al., 2014) included as a positive control, resulted in an increased sensitivity to Cd (Fig. 8A). When grown in liquid minimal medium containing 10 µm CdCl₂, the yeast cells expressing OsATX1 were less sensitive to Cd and showed increased growth compared with cells transformed with the vector control (Fig. 8C). In contrast to yeast expressing OsATX1, the yeast cells expressing OsNRAMP5 showed significantly greater growth inhibition (Fig. 8, A and C). Expression of OsATX1 in both wild-type (BY4741) and mutant (Δycf1) yeast strains increased their tolerance to 4 mm CuSO₄ (Fig. 8B). These results suggest that the heterologous expression of OsATX1 in yeast increased Cd and Cu tolerance.

Figure 8.

Functional analysis of OsATX1 by heterologous expression in yeast. A and B, Effects of the expression of OsATX1 on Cd (A) and Cu (B) tolerance in yeast. The wild-type yeast strain BY4741 and its mutant strain Δycf1 were transformed with the empty vector pYES2 or OsATX1. Serial dilutions (1:10) of overnight cell suspensions were spotted on SD-Ura medium containing different Cd concentrations (A) or 4 mm CuSO₄ (B) in the presence of Gal. OsNRAMP5 was used as a positive control. The strains treated with 0, 30, or 45 µm CdCl₂ or 0 mm CuSO₄ were grown at 30°C for 3 d and then treated with 60 µm CdCl₂ for 5 d or with 4 mm CuSO₄ for 10 d. C, Growth rate of yeast cells. The yeast mutant strain Δycf1 transformed with empty vector pYES2 or OsATX1 was grown in SD-Ura medium supplemented with 0 or 10 µm CdCl₂. D and E, Cd (D) and Cu (E) uptake in yeast. The yeast mutant strain Δycf1 transformed with empty vector pYES2 or OsATX1 was treated with 60 µm CdCl₂ (D) or 1, 3, 5, or 10 mm CuSO₄ (E) for 45 min. DW, Dry weight. Data are means ± sd of three independent biological replicates. Statistical comparison was performed by Tukey’s test: ** P < 0.01.

To further confirm the above result, the Cd and Cu accumulation in the yeast mutant strain Δycf1 was determined. When treated with excess Cd or Cu, the Cd and Cu concentrations in cells expressing OsATX1 were significantly lower than those in cells expressing the empty vector control (Fig. 8, D and E). These results suggest that OsATX1 increases the tolerance to Cd and Cu by decreasing their concentrations in the yeast cells.

DISCUSSION

Cu is one of the least abundant micronutrients, and its homeostasis must be strictly controlled in all living organisms. Cu chaperones are known to assist the intracellular homeostasis of Cu by their Cu-chelating ability. In this study, we characterized the function of OsATX1, which has a high similarity to AtATX1 and yeast ATX1 (Supplemental Fig. S1). Functional characterization of OsATX1 in rice showed that it plays a role in regulating the transport of Cu and Cd, especially the root-to-shoot translocation of Cu and the distribution of Cu to the developing tissues.

OsATX1 Is Involved in the Transport and Distribution of Cu in Rice

Cu taken up by the roots has to be loaded into the xylem for translocation to the shoots. OsATX1 was found to be expressed in the exodermis and the pericycle cell layer of the root mature zone (Fig. 2, F–H). The strong expression in the pericycle cells suggests a possible role in controlling the loading of Cu into the xylem vessels. Overexpression of OsATX1 increased Cu accumulation in the shoots and xylem sap and the root-to-shoot translocation of Cu, whereas knockout of OsATX1 decreased the root-to-shoot translocation of Cu and increased root Cu concentrations (Fig. 5, A–C). However, the Cu concentrations in the shoots and xylem sap in osatx1 mutants did not differ significantly from those in the wild type (Fig. 5, A and I). Because of a narrow range of Cu from deficiency to toxicity in shoots, plants have developed a complex Cu homoeostatic network to maintain the Cu concentration in shoots. In general, osatx1 mutants and the wild type treated with 2 µm Cu contain 279% to 379% higher Cu in roots but only 11% to 25% higher Cu in shoots compared with that in plants treated with 0.4 µm Cu (Fig. 5, A and B). The lack of a strong phenotype in the shoots of osatx1 mutants also may be due to the functional redundancy of other Cu chaperones in rice. In addition, the Cu deficiency signals in the shoots of osatx1 mutants may stimulate the absorption of Cu by roots and, subsequently, increase Cu accumulation in the shoots. Similarly, Arabidopsis atx1 mutants did not show any significant difference in shoot Cu concentration from that of wild-type plants, although the atx1 mutants were hypersensitive to excess Cu (one-half-strength Murashige and Skoog medium; Shin et al., 2012).

OsATX1 was found to be highly expressed in the vascular bundles of leaves (Fig. 2, I–M), implying a role of the chaperone in Cu distribution. Cu is distributed preferentially to the expanded leaves and then redistributed to the developing tissues through phloem transport, a process termed phloem-kickback mode (Yamaji and Ma, 2014). In general, the mutants had increased Cu concentrations in the older leaf blades and leaf sheaths (Fig. 6B) but decreased Cu concentrations in the developing tissues, including panicles, upper nodes and internodes, younger leaf sheaths, and the brown rice (Fig. 6, A and C). These results suggest that knockout of OsATX1 resulted in decreased remobilization of Cu from older leaves to the developing tissues and the seeds in rice.

Additionally, AtATX1 has been shown to play an essential role in Cu homeostasis in Arabidopsis. Overexpression of AtATX1 enhances the accumulation of Cu in the shoots and elevates the tolerance to excess Cu but the hypersensitivity to severe Cu deficiency (Shin et al., 2012; Shin and Yeh, 2012). In contrast, we found that the OsATX1-overexpressing plants became hypersensitive to excess Cu (Fig. 4). This is probably because OsHMA4 was shown to sequester Cu into root vacuoles and enhance Cu tolerance. oshma4 mutants accumulated more Cu in the shoots and less Cu in the roots and showed a higher sensitivity to excess Cu compared with that in wild-type plants (Huang et al., 2016), which is consistent with the phenotypes of the OsATX1-overexpressing plants observed here. Overexpression of OsATX1 increased Cu concentration in the shoots but decreased Cu concentration in the roots (Fig. 5, A and B). Overaccumulation of Cu in shoots may induce toxicity in rice. In contrast, none of the HMA proteins in Arabidopsis have been shown to be involved in the efflux of Cu into the vacuoles; thus, more Cu is retained in the cytoplasm of root cells, which may induce toxicity by inhibiting root growth. Overexpression of AtATX1 stimulated Cu allocation to shoots (Shin et al., 2012), thus enhancing Cu tolerance. It appears that the overexpression of OsATX1 and AtATX1 in rice and Arabidopsis, respectively, causes contrasting effects regarding Cu tolerance.

OsATX1 Affects Cu Transport and Distribution Probably by Interacting with Heavy Metal P_1B-Type ATPases in Rice

As a Cu chaperone, ATX1 cannot directly transport Cu across membranes, but it may deliver Cu to the target proteins for Cu trafficking. Yeast ATX1 and AtATX1 in Arabidopsis have been reported to chelate Cu in the binding site containing the MXCXXC conserved sequence (Pufahl et al., 1997; Shoshan and Tshuva, 2011; Shin et al., 2012), which is important for Cu-dependent protein-protein interaction (Pufahl et al., 1997; Andrés-Colás et al., 2006; Puig et al., 2007; Shoshan and Tshuva, 2011; Shin et al., 2012). The alignment of protein sequences revealed that OsATX1 contains an MXCXXC Cu⁺-binding motif (Supplemental Fig. S2), and yeast two-hybrid and BiFC assays showed that OsATX1 is able to interact with the rice heavy metal P_1B-ATPases HMA4, HMA5, HMA6, and HMA9 (Fig. 7), which contain two heavy metal-associated domains in the N terminus (Williams and Mills, 2005; Andrés-Colás et al., 2006; Lee et al., 2007).

In Arabidopsis, AtATX1 was proposed to deliver Cu to AtHMA5 for Cu detoxification in roots (Shin et al., 2012; Huang et al., 2016). Using triplin as a Cu ion chelator, AtATX1 was shown to act upstream of RAN1 for Cu⁺ delivery (Li et al., 2017). In rice, the expression of OsHMA4, OsHMA5, OsHMA6, and OsHMA9 was induced under a high concentration of Cu (Lee et al., 2007; Deng et al., 2013; Huang et al., 2016). OsHMA5 is localized mainly at the root pericycle cells and involved in Cu xylem loading, whereas OsHMA9 is a metal efflux protein and was suggested to function in metal mobilization from mature leaves to young leaves (Lee et al., 2007). At the reproductive stage, OsHMA5 also is localized at the xylem region of DVBs (Diffuse vascular bundles)in node I, which suggests that OsHMA5 also is involved in releasing Cu from nodes to the xylem connected to the grains (Deng et al., 2013; Mitani-Ueno et al., 2018). Therefore, our results suggest that OsATX1 may deliver Cu to OsHMA5 for Cu translocation from roots to shoots and Cu distribution from nodes to the grains as well as deliver Cu to OsHMA9 for redistribution in different shoot tissues. Such roles of OsATX1 are consistent with the altered Cu distribution in the osatx1 mutants and OsATX1-overexpressing lines. In addition, OsATX1 was widely localized in cells, including the cytoplasm, nucleus, and plasma membrane, and may adhere to endomembranes of rice cells (Supplemental Fig. S4), where it can conveniently mediate Cu trafficking to the target proteins. Further studies are required to investigate the role of these P_1B-ATPases and to validate the model that OsATX1 delivers Cu ions to OsHMA4, OsHMA5, OsHMA6, and OsHMA9 for Cu homeostasis in different rice tissues.

Overexpression of OsATX1 Increases Cd Accumulation in the Shoots

Cd is a highly toxic environmental contaminant. Organisms are equipped with Cd detoxification mechanisms, such as extrusion, chelation, and compartmentalization (Wysocki and Tamás, 2010; Wei et al., 2014; Luo et al., 2018). Yeast ATX1 was reported to chelate Cd²⁺ as well as Cu⁺ (Pufahl et al., 1997; Heo et al., 2012). In the Cd-sensitive yeast mutant Δycf1, lacking the ability to compartmentalize Cd into vacuoles, heterologous expression of OsATX1 increased Cu and Cd tolerance and reduced their concentrations in the yeast cells (Fig. 8). There may be specific plasma membrane-localized transporters in yeast for Cu and Cd efflux, and overexpression of OsATX1 may enhance Cu and Cd efflux to the external medium through these transporters, thus enhancing the tolerance to Cu and Cd.

At the vegetative growth stage, the Cd concentrations of the shoots and the xylem sap were increased in the OsATX1-overexpressing plants, whereas root Cd accumulation was higher in the osatx1 mutants (Fig. 5, D, G, and I). These phenotypes are similar to those of Cu, suggesting that OsATX1 also is involved in regulating Cd translocation to the shoots in rice. This also is supported by the Cd accumulation pattern in all aboveground tissues at the flowering stage (Supplemental Fig. S8). Our results suggest that OsATX1 may deliver Cd to other P-type ATPases for Cd trafficking or form a complexation with thiols; however, the exact mechanism behind this remains to be investigated in the future.

In conclusion, our study has demonstrated that OsATX1 is a Cu chaperone with a major role in mediating the intracellular Cu trafficking to the P_1B-ATPases HMA4, HMA5, HMA6, and HMA9 for Cu homeostasis in different rice tissues.

MATERIALS AND METHODS

Plant Materials and Growth Conditions

To generate the overexpression construct, the full-length coding sequence of OsATX1 was amplified using rice (Oryza sativa ‘Zhonghua11’) total cDNA as a template, and its primers were designed based on the sequence of Os08g0205400 from the NCBI (https://www.ncbi.nlm.nih.gov/). The entry clone was obtained through recombination of the PCR product with pDONR207 (Invitrogen). Error-free clones were then introduced into the destination vector pJC034 to produce expression vectors by LR recombination (Gong et al., 2013). The CRISPR-Cas system was used as described previously (Miao et al., 2013) to obtain osatx1 mutants. Briefly, a spacer sequence was cloned into the entry vector pOs-sgRNA and then into the expression vector of Cas9 by Gateway systems. The constructs were introduced into Agrobacterium tumefaciens strain EHA105 and then transferred into cv Zhonghua11 as described previously (Hiei et al., 1994). For each construct, at least three independent OsATX1 transgenic lines were selected for the targeted elemental analysis.

Seeds of the wild type (cv Zhonghua11) and OsATX1 transgenic lines were germinated for 3 d at 37°C on filter paper soaked with distilled water. After germination, the seeds were transferred to a net floating on distilled water in a greenhouse for 7 d. The seedlings used in following experiments were transferred to a plastic container for hydroponic culture in a greenhouse. Hydroponic experiments were performed using a standard rice culture solution as described previously (Yoshida et al., 1976), and the nutrient solution was renewed every 3 d. All experiments were performed with three biological replicates.

Phylogenetic Analysis and Sequence Alignment

The amino acid sequences of rice ATX1 and its related Cu chaperones in this study were extracted from the NCBI (https://www.ncbi.nlm.nih.gov/), RGAP (http://rice.plantbiology.msu.edu/), TAIR (http://www.arabidopsis.org/), and SGD (https://www.yeastgenome.org/) databases. The amino acid sequences were aligned using the ClustalW program. The neighbor-joining trees were constructed using MEGA7 (http://www.megasoftware.net/) with default parameters. The reliability of the reconstructed tree was evaluated by a bootstrap test with 1,000 replicates. The sequence alignment was built with MultAlin (Corpet, 1988) and displayed using the ESPript 3.0 sever (Robert and Gouet, 2014).

RNA Extraction and Expression Analyses

RT-qPCR was performed using total RNA extracted with the TransZol RNA Extraction Kit (TransGen). Three micrograms of RNA was used to synthesize the first-strand cDNAs in 20 μL of reaction mixture using EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen) according to the manufacturer’s instructions. The quantitation of transcript abundance was performed using the SYBR Premix Ex Taq kit (TaKaRa) on the ABI 7500 Real-Time PCR system (Applied Biosystems). The rice UBIQUITIN5 gene was used as the internal reference. The primer sequences are listed in Supplemental Table S1. Gene expression was quantified using the relative quantification method. Tissues were taken from at least three plants and pooled for each biological replicate at various stages as described in Figure 1.

Genotyping of osatx1 Mutant Plants

Genotyping of osatx1 was performed by PCR using a PCR Mix (2xTSINGKE Master Mix; TSE004). PCR was performed in an ABI 9700 thermocycler (Applied Biosystems) with the following cycling profile: 94°C for 5 min, followed by 35 cycles of 94°C for 30 s, 58°C for 30 s, and 72°C for 40 s, and a final 10-min extension at 72°C. The primer sequences are listed in Supplemental Table S1.

OsATX1 Promoter Fusion with GUS and GFP

A sequence of 1,593 bp preceding the translation initiation site of the OsATX1 gene was amplified using genomic DNA from cv Zhonghua11 as a template. The PCR product was inserted first into the entry vector pDONR207 and then into the destination vector pHGWFS7.0, which allows the fusion of the OsATX1 promoter sequence with two reporter genes, namely GFP and GUS (Jin et al., 2015). The OsATX1_Pro:GUS and transgenic plant tissues were incubated in a X-Gluc staining buffer at 37°C for 4 h (Jefferson et al., 1987) and then washed with 75% (v/v) ethanol overnight. The cleared samples were observed by light microscopy. The primer sequences are listed in Supplemental Table S1.

Subcellular Localization

To investigate the tissue and cellular specificity of OsATX1 expression, full-length cDNA of OsATX1 was amplified using cv Zhonghua11 total cDNA as a template. The PCR product was cleaved using EcoRI and KpnI and cloned into pM999-GFP vector to generate an OsATX1-GFP fusion construct driven by the cauliflower mosaic virus 35S promoter (35S:OsATX1-GFP), which was cotransformed with 35S:AtATX1-mCherry into rice protoplasts via polyethylene glycol-mediated transformation as described previously (Xie and Yang, 2013). The florescence signal was observed using a confocal laser scanning microscope (Olympus FV1000) after incubation at 22°C for 12 to 24 h. Moreover, 35S:OsATX1-GFP and 35S:OsNRAMP5-RFP were used to transform A. tumefaciens EHA105, then the A. tumefaciens were infiltrated into leaves of Nicotiana benthamiana. After a 2-d incubation, the transformed plant leaves were observed and imaged using confocal microscopy (Leica TCS SP2).

Determination of Metals in Plant Tissues

The tissues were washed with distilled water three times before sampling. All samples were dried at 80°C for 3 d and then digested in 65% nitric acid in a MARS6 microwave (CEM) with a gradient of temperatures from 120°C to 180°C for 1 h. After dilution in deionized water, the metal contents of the samples were determined by ICP-MS (Agilent 7700 series). The root-to-shoot translocation of Cu was calculated as the amount of Cu accumulated in the shoots as a percentage of the total Cu uptake in plants, as described previously (Miyadate et al., 2011).

Collection and Analysis of Xylem Sap

Wild-type and OsATX1 transgenic lines were cultured hydroponically until the late vegetative stage. For the analyses of Cu and Cd concentrations, plants were exposed to 5 µm CuSO₄ or 0.5 µm CdSO₄ for 3 d. The shoots (2 cm above the roots) were excised with a razor, and the xylem sap was collected with a micropipette for 2 h after decapitation of the shoots. The Cu and Cd concentrations of the xylem sap were diluted with 2% HNO₃ and determined by ICP-MS.

Yeast Two-Hybrid and BiFC Assays

For the yeast two-hybrid assay, the coding region of OsATX1 was cloned into the pDEST32 vector using Gateway technology (Invitrogen). OsHMA4, OsHMA5, OsHMA6, and OsHMA9 coding sequences were cloned separately into pDEST22 vector (Invitrogen). The combined plasmids were transformed into yeast (Saccharomyces cerevisiae) strain AH109 for yeast two-hybrid assays (Invitrogen manufacturer’s instructions). The positive clones were selected from synthetic dropout medium lacking Leu and Trp. The interactions were tested on a selective medium lacking Leu, Trp, and His. The β-galactosidase activity was assayed by X-gal after 4 d.

The BiFC assays were performed as described previously (Tian et al., 2011). N. benthamiana plants were grown in a growth chamber at 23°C with a 16-h/8-h light/dark cycle. OsATX1 was cloned into the pEXP-NA vector and fused to the N terminus of YFP, and OsHMA4, OsHMA5, OsHMA6, and OsHMA9 were cloned into the pEXP-NB vector and fused separately to the C terminus of YFP. The fusion constructs were introduced into A. tumefaciens strain EHA105 by electroporation. The tested construct pairs were expressed in leaves of N. benthamiana for 2 d before microscopic observation. The YFP fluorescence signal in the transformed leaves was detected by confocal microscopy (Olympus FV1000).

Yeast Complementation Assay

The S. cerevisiae strain BY4741 (MATa; his3DΔ1; leu2DΔ0; met15DΔ0; ura3DΔ0) and its mutant strain Δycf1 (MATa; his3DΔ1; leu2DΔ0; met15DΔ0; ura3DΔ0; YDR135c::kanMX4), which lacks the function for the compartmentalization of Cd into vacuoles (Li et al., 1997), were used for heterologous expression of OsATX1. To evaluate the role of OsATX1 in Cu and Cd detoxification, the coding region of OsATX1 was cloned into the EcoRI and XbaI sites of a yeast expression vector, pYES2 (Invitrogen), and then introduced into yeast strains using the standard procedures (Invitrogen). Wild-type yeast strain BY4741 and its mutant strain Δycf1 transformed with the empty vector pYES2 or OsATX1 were grown in SD-Ura liquid medium, which contained 2% Gal, 0.67% yeast nitrogen base without MnSO₄ (Krackeler Scientific), 0.2% appropriate amino acid, and 50 mm 2-morpholinoethanesulfonic acid, and overnight cell suspension with serial dilutions (1:10) were spotted onto SD-Ura agar supplemented with 0 or 4 mm CuSO₄ or 0, 30, 45, or 60 µm CdCl₂ in the presence of Gal. OsNRAMP5 was used as a positive control. In a further experiment, the yeast mutant strain Δycf1 transformed with different vectors was precultured in liquid SD-Ura medium to a cell density corresponding to an OD₆₀₀ of 0.8 to 1.2 at 30°C. A total of 200 µL of culture was used to inoculate 50 mL of the liquid medium supplemented with 0 or 10 µm CdCl₂, and their cell densities were measured using a spectrometer at 600 nm.

To determine Cu and Cd uptake, yeast cells precultured on liquid SD-Ura medium to a cell density corresponding to an OD₆₀₀ of 0.8 to 1 were treated with 1, 3, 5, or 10 mm CuSO₄ or 60 µm CdCl₂ and cultured at 30°C for 45 min. After harvesting and washing five times with deionized water (MilliQ; Millipore), all cells were dried at 80°C for 3 d and then used to determine metal contents by ICP-MS.

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers XM_015792604.1 (OsATX1), XM_015770686.1 (OsHMA4), XM_015780452.1 (OsHMA5), XM_015771427.1 (OsHMA6), XM_015788093.1 (OsHMA9), and XM_015778855.1 (OsATX).

Supplemental Data

The following supplemental materials are available.

• Supplemental Figure S1. Phylogenetic relationship of ATX proteins in rice, Arabidopsis, and yeast.

• Supplemental Figure S2. Protein sequence alignment of plant ATX1 protein sequences with the sequence of the known ATX in yeast.

• Supplemental Figure S3. Time-course expression of OsATX1 under high concentrations of Fe, Mn, or Zn.

• Supplemental Figure S4. Subcellular localization of OsATX1 protein.

• Supplemental Figure S5. Generation of osatx1 knockout mutants and OsATX1-overexpressing plants in rice.

• Supplemental Figure S6. Mineral concentrations of the shoots and roots at the vegetative stage.

• Supplemental Figure S7. Phenotypes of OsATX1 transgenic lines and the wild type under metal stress and deficiency.

• Supplemental Figure S8. Cd concentrations in different rice tissues at the reproductive stage.

• Supplemental Figure S9. Metal concentrations of the brown rice.

• Supplemental Table S1. Primers used in this study.

Dive Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

• ATX1 Gramene: AT1G66240

• ATX1 Araport: AT1G66240

• CCH Gramene: AT3G56240

• CCH Araport: AT3G56240

• COX17 Gramene: AT3G15352

• COX17 Araport: AT3G15352

• HMA4 Gramene: AT2G19110

• HMA4 Araport: AT2G19110

• HMA5 Gramene: AT1G63440

• HMA5 Araport: AT1G63440

• RAN1 Gramene: AT5G44790

• RAN1 Araport: AT5G44790

• copper CHEBI: CHEBI:30052

• mineral CHEBI: CHEBI:46662

• xylem AmiGo: PO:0005352

• zinc CHEBI: CHEBI:27363