Search tips
Search criteria 


Logo of jexbotLink to Publisher's site
J Exp Bot. 2013 January; 64(1): 369–381.
Published online 2013 January 9. doi:  10.1093/jxb/ers315
PMCID: PMC3528025

Transport properties of members of the ZIP family in plants and their role in Zn and Mn homeostasis


A better understanding of the role of the Arabidopsis ZIP family of micronutrient transporters is necessary in order to advance our understanding of plant Zn, Fe, Mn, and Cu homeostasis. In the current study, the 11 Arabidopsis ZIP family members not yet well characterized were first screened for their ability to complement four yeast mutants defective in Zn, Fe, Mn, or Cu uptake. Six of the Arabidopsis ZIP genes complemented a yeast Zn uptake-deficient mutant, one was able partially to complement a yeast Fe uptake-deficient mutant, six ZIP family members complemented an Mn uptake-deficient mutant, and none complemented the Cu uptake-deficient mutant. AtZIP1 and AtZIP2 were then chosen for further study, as the preliminary yeast and in planta analysis suggested they both may be root Zn and Mn transporters. In yeast, AtZIP1 and AtZIP2 both complemented the Zn and Mn uptake mutants, suggesting that they both may transport Zn and/or Mn. Expression of both genes is localized to the root stele, and AtZIP1 expression was also found in the leaf vasculature. It was also found that AtZIP1 is a vacuolar transporter, while AtZIP2 is localized to the plasma membrane. Functional studies with Arabidopsis AtZIP1 and AtZIP2 T-DNA knockout lines suggest that both transporters play a role in Mn (and possibly Zn) translocation from the root to the shoot. AtZIP1 may play a role in remobilizing Mn from the vacuole to the cytoplasm in root stellar cells, and may contribute to radial movement to the xylem parenchyma. AtZIP2, on the other hand, may mediate Mn (and possibly Zn) uptake into root stellar cells, and thus also may contribute to Mn/Zn movement in the stele to the xylem parenchyma, for subsequent xylem loading and transport to the shoot.

Key words: Arabidopsis, AtZIP1, AtZIP2, Mn transport, Zip transporters, Zn transport.


The ZIP family of membrane transporters has been shown to play a role in the transport of four essential micronutrients: Zn, Fe, Mn, and Cu (Eide et al., 1996; Grotz et al., 1998; Pence et al., 2000; Wintz et al., 2003; Cohen et al., 2004; Pedas et al., 2008; Lin et al., 2009). ZIP, which stands for ZRT-IRT-like Proteins, was named for the two founding members, the high affinity yeast plasma membrane Zn uptake transporter, ZRT1, and the high affinity Arabidopsis thaliana plasma membrane Fe uptake transporter, IRT1 (Eide et al., 1996; Zhao and Eide, 1996). Subsequent studies of other plant ZIP family members demonstrated that not all the members of the ZIP family are involved in plasma membrane micronutrient uptake. An example of this in plants is AtIRT2, which was recently shown to play a role in Fe homeostasis by transporting Fe into endomembrane vesicles, and is not involved in cellular uptake as originally hypothesized (Vert et al., 2009).

ZIP family members have also been shown to transport heavy metals such as Cd, and hence the ZIP family may also play a significant role in how various heavy metals, both essential and toxic, are taken up and translocated throughout the plant (Guerinot, 2000; Pence et al., 2000; Rogers et al., 2000). A better understanding of the roles and functions of each of the 15 members of the Arabidopsis ZIP family should lead to new insights into micronutrient/heavy metal homeostasis. Certainly a primary goal of such an effort should be to identify the metals that each ZIP family member transports. Other important features of metal transporters that were focused on in this study are the membrane localization for ZIP transporters, whether they transport metals into or out of a specific organelle, and the regulation of ZIP gene expression by changes in plant micronutrient (Zn, Mn, Fe, or Cu) status. Gaining a better understanding of the Arabidopsis ZIP family as a whole should also help us better understand micronutrient nutrition in other biological systems as well, as the ZIP family of transport proteins is found in all branches of life, including plants, fungi, animals, and protists (Guerinot, 2000).

To date, only a limited number of members of the ZIP family have been characterized in plants (primarily in Arabidopsis) with regards to both their transport capabilities and their role in planta. Currently, there is a fairly detailed understanding of the roles and functions for three Arabidopsis ZIP transporters, AtIRT1, AtIRT2, and AtIRT3, with AtIRT1 being by far the most well studied based on its seminal role in root Fe uptake and transport (Eide et al., 1996; Rogers et al., 2000; Vert et al., 2001, 2002, 2009; Connolly et al., 2002; Lin et al., 2009). Very little or nothing is known about the function of the other 12 Arabidopsis ZIPs. Hence in this study, the aim was to characterize broadly the remaining ZIP family members to begin to gain insights into what metals they transport, and then two of these 11 ZIPs, AtZIP1 and AtZIP2, were functionally characterized in more detail for their possible roles in root Zn and Mn transport and homeostasis.

Materials and methods

Cloning of the Arabidopsis ZIP family genes

AtZIP2, ZIP7, and ZIP11 were requested from RIKEN from a scan of full-length clones ( ZIP3 was graciously given to us by Dr Guerinot, AtZIP1, 5, 6, 8, 9, 10, and 12 were cloned using full-length primers and cDNA from root and shoot tissue from Zn-deficient and Zn-replete Arabidopsis plants (Supplementary Table S1 available at JXB online).

Protein predictions

The translated AtZIP proteins were analysed using the online site, WoLF pSORT (, to estimate their predicted membrane localization. In addition, the AtZIP proteins were queried against the SubCellular Proteomic Database (SUBA;, for predicted subcellular localization. To estimate the secondary protein structure, including the number of predicted transmembrane domains, the translated proteins were analysed using the Mobyle, a bioinformatics analysis tool (von Heijne, 1992) (

Relative expression of ZIP family genes in Arabidopsis

Arabidopsis ZIP gene expression data were summarized from the AtGeneExpress database for global Arabidopsis gene expression through different developmental stages that was compiled by Schmid et al. (2005). The AtGeneExpress database was mined for the expression of 10 ZIP family members (At3g12750, At5g59520, At2g32270, At1g05300, At2g30080, At2g04032, At4g33020, At1g31260, At1g55910, and At5g62160) at day 7, 15, and 21 in both roots and shoots. The data are publically available at

Yeast studies

Four different yeast mutant strains were used in these studies: the Zn uptake-defective mutant zrt1/zrt2Δ (MATα ade6 can1 his3 leu2 trp1 ura3 zrt1::LEU2 zrt2::HIS3), the Fe uptake-defective mutant fet3/fet4Δ (MATa can1 his3 leu2 trp1 ura3 fet3::HIS3 fet4::LEU2), the Mn uptake-defective mutant smf1Δ (SLY8; MATa ura3 lys2 ade2 trp1 his3 leu2 smf1::HIS3), and the Cu uptake-defective mutant ctr1/ctr3Δ (MATa ctr1::ura3::Kanr ctr3::TRP1 his3 lys2-802 CUP1r). Each strain was maintained on YPD until introduction of either the gene of interest or the empty vector. Cultures of each yeast mutant strain containing one of the genes of interest were grown in liquid SC-URA to an optical density (OD) of 1 and then serially diluted 10-, 100-, and 1000-fold. Each dilution was plated out onto the specific restrictive media for that mutant. For the zrt1/zrt2Δ mutant, the restrictive medium contained SC-URA plus 1mM EDTA and 500 µM ZnCl2 for Zn-limiting growth conditions (Pence et al., 2000). For smf1Δ, the medium was SC-URA containing 20mM EGTA for Mn-limiting conditions (Supek et al., 1997). The ctr1/ctr3Δ mutant was assayed for growth on YPE medium (10g of yeast extract and 20g of peptone per litre, and 10% ethanol) for Cu-limiting conditions (Pena et al., 2000). For the fet3/fet4Δ mutant, cells were grown on SC-URA (pH 4.0) containing 80 µM bathophenanthroline disulphonate (BPS) which produced Fe-limiting growth conditions (Eide et al., 1996).

Quantitative RT–PCR

For quantitative PCR (qPCR), Arabidopsis Col-0 plants were grown on a modified Johnson’s solution containing 1.2mM KNO3, 0.8mM Ca(NO3)2, 0.1mM NH4H2PO4, 0.2mM MgSO4, 50 µM KCl, 12.5 µM H3BO3, 1 µM MnSO4, 0.1 µM NiSO4, 1 µM ZnSO4, 0.5 µM CuSO4, and 2mM MES (pH 5.5). For all metal treatments, plants were grown for 2 weeks on this medium and then switched to the same nutrient solution except that Zn, Cu, Mn, or Fe was omitted, and the plants were grown for an additional 7 d. Plants were then harvested, snap-frozen in liquid nitrogen, and ground to a fine powder using a mortar and pestle. Total RNA was isolated using the Plant RNeasy RNA mini kit (Qiagen, Valencia, CA, USA). cDNA was synthesized using the QuantiTect Reverse Transcription Kit (Qiagen). Transcript levels were measured using GoTaq® qPCR Master Mix (Promega, Madison, WI, USA) with the primer pairs GCATGCGGGTCATGTTCAC and CAACTCGGTCGAACCATGTG for AtZIP1 and GTTTGTAGCGGCTGGGAGTAA and GTCATCA TCCTCCCTCGACTCT for AtZIP2. A total of 25ng of purified cDNA was used in each reaction to compare transcript levels across treatments. 18S was used as an internal reference and was amplified using the primer pair CGCTATTGGAGCTGGAATTACC and AATCCCT TAACGAGGATCCATTG to normalize across treatments and species. A 1ng alquot of cDNA was used in these reactions since expression levels were high. Quantitative real-time reverse transcription–PCR (RT–PCR) was performed using an ABI 7500 real-time PCR system and SYBR Green kit (Applied Biosystems). PCR conditions used were 50 °C for 2min, 95 °C for 10min, followed by 40 cycles of 95 °C for 30 s, and 60 °C for 1min. A dissociation curve was performed after each of the two biological replicates to ensure only one product was being amplified.

Protein localization

AtZIP1 and AtZIP2 were cloned into the pSAT1 enhanced green fluorescent protein (eGFP) expression vector in-frame with the C-terminus of eGFP using the KpnI site, and transiently expressed in freshly isolated protoplasts from 3- to 4-week-old Arabidopsis seedlings according to Sheen et al. (2001). Confocal images were taken 15–18h post-transfection of the Arabidopsis protoplasts. Images were taken on a Leica TCS-SP5 confocal microscope (Leica Microsystems, Exton, PA, USA) with a ×63magnification, numerical aperture 1.2, water immersion objective. The eGFP was excited with a blue argon ion laser (488nm), and the emitted fluorescence was collected between 505nm and 545nm.

Promoter localization

To study the tissue-specific expression of both AtZIP1 and AtZIP2, 1kb of promoter sequence upstream of each start codon was cloned in front of a β-glucuronidase (GUS) reporter using the pGPTV:BAR vector and incorporated into A. thaliana ecotype Columbia via transformation with Agrobacterium tumefaciens, using the floral dip method for stable transformation. Primers used for amplification are listed in Supplementary Table S1 at JXB online. Transformed plants were selected after growth on soil using BASTA, and surviving seedlings were grown to seed. T2 seeds were then assayed for GUS activity by growing them for 10–14 d on a modified Johnson’s solution (composition listed above in the Quantitative RT–PCR section) under Zn-, Fe-, Mn-, or Cu-limiting conditions. AtZIP1p::GUS or AtZIP2p::GUS plants were then assayed for GUS activity in a solution containing 1mM X-Gluc (5-bromo-4-choloro-3-indolyl) β-d-glucuronic acid in 50mM Na2HPO4, pH 7.0, and 0.1% Triton X-100 for ~4h. The reaction was stopped by removing the GUS solution and replacing it with 95% ethanol. At least 10 independent transformation events for both ZIP1 and ZIP2 were tested for expression of the GUS reporter and, in each transgenic line, the image for the promoter::GUS reporter was similar.

Phenotypic analysis of AtZIP1 and AtZIP2 T-DNA knockout lines

Seeds corresponding to the AtZIP1 (At3g12750) and AtZIP2 (At5g59520) genomic regions were requested from the ABRC (Alonso et al., 2003). Genotyping of individual plants was done using one of the two primers listed in Supplementary Table S1 at JXB online, with one primer anchored in the open reading frame of each ZIP member to allow for the possibility of the T-DNA insertion being incorporated in either direction. The primer GCTGTTGCCCGTCTCACTGGTG was used for the T-DNA insert. PCR conditions were 95 °C for 30 s; 55 °C for 30 s; 72 °C for 4min; 35 cycles were run for all lines tested. The PCR product was then purified using a Qiagen PCR clean up kit (Qiagen) and sequenced with both the primers listed above. To verify the lack of AtZIP1 transcript in the AtZIP1 T-DNA line, the 5’ portion of the AtZIP1 transcript was assayed for using the primer pair AAAAAGGATTCCGGCGTTACAA and CCCAATCTCAAGCACCTGCGACACTAT; the 3’ end of the gene was studied using the primer pair TCGCAGGTGCTTGAGATTGGGATA GTT and AGATAAGCAGCCAAGTGAAGCCAGAGA. To verify the lack of AtZIP2 transcript in the AtZIP2 T-DNA line, the 5’ portion of the gene was assayed with the primer pair GCGACGAAGAAGAG GAGACCAACCAG and TATCTCACGACGTCCTTCCTCTTTCAC, and the 3’ end of the gene was studied using the primer pair, AAGAG GAAGGACGTCGTGAGATAAAGA and TAACCGCAACGTACAC AAAAACTCCAC.

Growth conditions and determination of mineral content for T-DNA knockout lines

Nutrient-replete T-DNA lines were grown on 1× Murashige and Skoog (MS) medium plus 1% sucrose. To provide Zn-deficient or high Zn plants, either Zn was omitted from the medium or ZnSO4 was added to give a final Zn concentration of 300 µM. To alter the plant Mn status, plants were grown on 1× MS medium where either the Mn was omitted or MnSO4 was added to give a final concentration of 250 µM. For inductively coupled plasma atomic emission spectroscopy (ICP-AES) determination of plant Zn content, plants were grown on 1× MS+1% sucrose with a total Zn concentration of 300 µM. For ICP determination of plant Mn content, plants were grown on 1× MS+1% sucrose with a total Mn concentration of 250 µM. Following growth for 7 d, plants were harvested and metals adsorbed in the root cell wall were desorbed in 5mM CaCl2 solution for 15min using well established protocols in our lab (Hart et al., 1998a , b ), and then rinsed in 18 MΩ water. Roots and shoots were separated, dried, weighed, and analysed for Zn and Mn content using ICP-AES.


Isolating the 11 remaining Arabidopsis ZIP family members

In an attempt to characterize the 11 remaining uncharacterized Arabidopsis ZIP family members functionally, public repositories were searched for full-length cDNA clones of the Arabidopsis ZIP genes. Only ZIP2, ZIP7, and ZIP11 were found as full-length cDNA clones in the RIKEN database. AtZIP3 was obtained from the Guerinot Lab, which they originally used in the research reported in Grotz et al. (1998). To clone the remaining eight members of the Arabidopsis ZIP family, primers were designed based on the predicted open reading frame listed in TAIR, and the ZIP genes were cloned from RNA isolated from roots and shoots of plants grown under Zn-replete and Zn-deficient conditions. Both Zn-replete and Zn-deficient conditions were used to increase the chance that a given ZIP transporters was expressed. The DNA sequences for all isolated genes were as listed in TAIR except ZIP6 and ZIP8. The version of AtZIP6 used here harbours a point mutation which changes amino acid 65 from an isoleucine to a methionine. This mutation is in the beginning of the second predicted transmembrane domain. The version of AtZIP8 which was isolated was significantly different from the reference gene with regards to the predicted genomic structure presented on TAIR, in that the last predicted intron failed to be removed, which led to a truncation in the predicted protein, causing the loss of the last two transmembrane domains. The experiments were not successful in isolating any other versions of ZIP8 despite repeated efforts to clone this gene from either the roots or shoots of Arabidopsis plants grown under control and Zn-, Fe-, Mn-, or Cu-limiting conditions. Otherwise all other ZIP family members were as predicted by the gene models in TAIR.

Yeast complementation

To determine the metal specificities for transport of the proteins encoded by these 11 ZIP genes, each gene was tested for its ability to complement one of four different yeast metal uptake mutants. The four different yeast lines consisted of mutants defective in Zn uptake (zrt1/zrt2Δ), Cu uptake (ctr1/ctr3Δ), Fe uptake (fet3/fet4Δ), or Mn uptake (smf1Δ). Of the 11 different genes tested, six ZIP genes, ZIP1, ZIP2, ZIP3, ZIP7, ZIP11, and ZIP12, were able to complement the zrt1/zrt2Δ yeast mutant fully or partially under Zn-limiting conditions (Fig. 1). Six AtZIP genes were also able to complement the Mn uptake mutant smf1Δ fully or partially under Mn-limiting conditions; these genes are ZIP1, ZIP2, ZIP5, ZIP6, ZIP7, and ZIP9. None of the 11 genes tested was able to complement the Cu uptake mutant, ctr1/ctr3Δ, and only ZIP7 was able to complement the Fe uptake mutant, fet3/fet4Δ partially. Of the 11 genes tested, only ZIP8 failed to complement any of the four uptake mutants.

Fig. 1.
Complementation of yeast metal uptake mutants with AtZIP genes. Complementation of yeast mutants defective in the uptake of Zn (zrt1/zrt2Δ), Mn (smf1Δ), Fe (fet3/fet4Δ), or Cu (ctr1/ctr3Δ) expressing any one of the 11 ...

From these yeast complementation studies, it was determined that, potentially, ZIP7 could transport Zn, Mn, and Fe, ZIP1 and ZIP2 could transport Zn and Mn, ZIP3, ZIP11, and ZIP12 could transport Zn, and ZIP5, ZIP6, and ZIP9 could transport Mn.

Relative expression of ZIP family members

To begin to better understand the role of the ZIP family of transporters in plants, the expression of 10 of the 11 plant ZIP genes studied here was compiled from a global gene expression map for Arabidopsis development (Schmid et al., 2005), and is summarized in Table 1 for 7-, 15-, and 21-day-old plants. For the study of Schmid et al. (2005), all of the ZIPs studied here were analysed for expression, except AtZIP8, which is not on the Affymetrix array used for that study. ZIP gene expression in both roots and shoots is presented in Table 1 to determine if there is an organ-/tissue-specific bias in gene expression for specific ZIPs. From this analysis, it was found that ZIP1, ZIP2, ZIP3, ZIP5, and ZIP6 are all expressed at higher levels in the roots than in the shoots. Also ZIP1, ZIP2, ZIP3, and ZIP5 increase in relative expression in the roots as the plant gets older. The ZIP genes that showed significantly higher expression in the shoots are ZIP7 and, to a lesser extent, ZIP11, which showed high shoot expression at day 7 in the shoots but then its expression decreased as the plant aged. ZIP9, ZIP10, and ZIP12 showed relatively equal expression in both roots and shoots, and did not show any difference in expression as the plant aged.

Table 1.
Relative expression of 10 members of the ZIP family of transporters in Arabidopsis in 7-, 15-, and 21-day-old plants compiled from the AtGenExpress database ( Schmid et al., 2005) Listed are the relative expression of 10 Arabidopsis ZIP family members ...

It was beyond the scope of this study to characterize all 11 ZIP genes in detail. Thus for the subsequent more detailed research, it was decided to focus on two ZIP transporters, AtZIP1 and AtZIP2. These two ZIP transporters were two of the three ZIPs (along with ZIP7) that appear to mediate both Zn and Mn uptake, which was of interest, especially since very little is known about plant Mn transporters in particular. Also, it was decided to focus on candidate root metal uptake transporters for this more detailed study, and ZIP1 and ZIP2 were two of four ZIPs (along with ZIP3 and ZIP6) that are expressed more strongly in roots.

Investigation of AtZIP1 and AtZIP2 expression in response to different metal treatments

The relative transcript levels of AtZIP1 and AtZIP2 were quantified in Arabidopsis plants grown under either nutrient-replete conditions, Zn deficiency, Fe deficiency, Cu deficiency, or Mn deficiency. In nutrient-replete plants, AtZIP1 was found to be expressed ~50% more in the roots than in the shoots when expression was normalized to either 18S or actin2 transcript levels (actin data not shown). AtZIP1 transcript abundance in roots was affected most by Zn deficiency, which significantly increased transcript abundance >4-fold relative to roots of replete-grown plants (Fig. 2A). Expression of AtZIP1 was also increased in the roots by Fe deficiency, which resulted in a 2.5-fold increase in AtZIP1 expression; however, Cu and Mn deficiency had no significant effects on AtZIP1 transcript levels. Increased shoot expression of AtZIP1 was also seen in response to Mn deficiency, which increased transcript levels ~80% relative to levels of AtZIP1 expression in shoots of replete-grown plants (Fig. 2B). Cu and Fe deficiency had no effect on AtZIP1 expression, and Zn deficiency significantly reduced AtZIP1 expression by ~50% relative to expression in shoots of replete-grown plants.

Fig. 2.
Relative transcript levels of AtZIP1. Quantitative real-time PCR analysis of AtZIP1 expression in: (A) roots or (B) shoots of A. thaliana plants grown on nutrient-replete, –Zn, –Mn, –Fe, or –Cu nutrient solution. The average ...

AtZIP2 was found to be expressed ~10-fold more in the roots than the shoots of nutrient-replete plants (data not shown). In roots, AtZIP2 expression was decreased in response to Fe deficiency, while Zn, Mn, and Cu deficiency had no effect on expression (Fig. 3A). Shoot expression of AtZIP2 was found to be very low, and this low level of expression was decreased by ~60–70% in response to Mn and Fe deficiency (Fig. 3B). AtZIP2 expression was not altered by Zn or Cu deficiency.

Fig. 3.
Relative transcript levels of AtZIP2. Quantitative real-time PCR analysis of AtZIP2 expression in: (A) roots or (B) shoots of A. thaliana plants grown on nutrient-replete, –Zn, –Mn, –Fe, or –Cu nutrient solutions. The average ...

Localization of the AtZIP1 and AtZIP2 protein

Both AtZIP1 and AtZIP2 were cloned in-frame with an N-terminal eGFP and transiently expressed in Arabidopsis protoplasts. The eGFP:ZIP1 fusion protein localized to the vacuole (Fig. 4AD). The arrow in Fig. 4 indicates the characteristic invaginations associated with a vacuolar localization. The image for expression of the eGFP:ZIP2 fusion was consistent with a plasma membrane localization (Fig. 4EH).

Fig. 4.
Subcellular localization of eGFP:AtZIP1 and eGFP:AtZIP2. Both AtZIP1:eGFP and AtZIP2:eGFP fusion proteins were transiently expressed in Arabidopsis protoplasts for 16h. Each set of four panels consists of the GFP–AtZIP image (upper left), chlorophyll ...

Tissue-specific expression

To study the tissue-specific expression of both AtZIP1 and AtZIP2, 1kb of promoter sequence upstream of each start codon was cloned in front of a GUS reporter gene and stably transformed into Arabidopsis plants. Plants harbouring the AtZIP1p::GUS reporter were grown under nutrient-replete conditions as well as in response to Zn, Fe, Mn, and Cu deficiency. Analysis of T3 homozygous lines showed that AtZIP1 is predominantly expressed in the root stele and in the shoot vasculature (Fig. 5AF). It was only possible to see AtZIP1::GUS expression in seedlings grown under Zn deficiency conditions. In either nutrient-replete plants or plants subjected to Fe, Mn, or Cu deficiency, GUS staining was seen, indicating that Zn deficiency induced the greatest increase in AtZIP1 expression, which is consistent with what was seen from the qPCR data in Fig. 2A, at least for expression in roots.

Fig. 5.
Tissue localization of AtZIP1 and AtZIP2 expression in transgenic Arabidopsis. Arabidopsis seedlings were transformed with either 1kb of the AtZIP1 promoter (A–F) or 1kb of the AtZIP2 promoter (G–K), driving a GUS reporter. The seedlings ...

AtZIP2p::GUS lines were also studied under the same conditions as for AtZIP1, and AtZIP2 was found to be expressed at lower levels than AtZIP1, with the strongest expression occurring under nutrient-replete conditions. Like AtZIP1, AtZIP2 is expressed most strongly in the root stele. As one moves from the younger to more mature root regions, strong stelar AtZIP2 expression was seen in the primary root elongation zone, with lower expression in the stele of lateral roots (Fig. 5GK). Then, particularly in the primary root, AtZIP2 stelar expression increases in more mature regions of the root, especially close to the root–shoot junction (Fig. 5I, ,K).K). No expression of the AtZIP2p::GUS reporter could be seen in the shoots.

Phenotypic analysis of AtZIP1 and AtZIP2 T-DNA knockout lines

To better understand the role of AtZIP1 and AtZIP2 in Zn and Mn nutrition, seeds of plants harbouring projected T-DNA insertions in these two genes were obtained from the ABRC. Two independent homozygous lines were isolated with a disruption in an exon of AtZIP1. Conformation of the insertions identified two distinct T-DNA lines for AtZIP1 and one for AtZIP2 with T-DNA insertions in an exon. As depicted in Fig. 6A, the T-DNA insertions for the two AtZIP1 T-DNA lines are located towards the 3’ end of the gene. One of these lines, SALK_023634, results in a predicted truncated AtZIP1 protein that has lost the last two transmembrane domains; the SALK_109897 line is predicted to generate a truncated protein that has lost half of the eighth transmembrane domain. As mentioned above, only one T-DNA line could be found that had an insertion located in an AtZIP2 exon. The T-DNA line (SALK_094937) with a T-DNA insertion located in the AtZIP2 coding region is predicted to encode a protein that has lost the last three transmembrane domains (Fig. 6B). To confirm that the lines were true knockouts and not knockdown lines, primers were designed for both the 5’ and 3’ ends of the AtZIP1 and AtZIP2 transcripts. As seen in Fig. 6, only partial transcripts could be obtained for both AtZIP1 and AtZIP2; no full-length transcripts were found for any of the three T-DNA lines tested.

Fig. 6.
T-DNA insertion lines. Diagrammatic model depicting the genomic structure of (A) AtZIP1 and (B) AtZIP2 along with the location of the T-DNA insertions. Grey boxes represent either the 5’- or 3’-untranslated region, black boxes represent ...

When the T-DNA Arabidopsis lines were grown on nutrient-replete media, no obvious phenotype was observed for the three lines. Subsequently, plants were grown first on either Zn-deficient (–Zn) or high Zn (300 µM Zn) media, and root tolerance (relative root growth) was quantified for each T-DNA knockout line. There was no difference observed in relative root growth for either AtZIP1 T-DNA knockout line compared with Col-0 in response to Zn deficiency or high Zn growth conditions (Fig. 7A). However, the AtZIP2 T-DNA knockout line exhibited significant responses to growth under both Zn growth conditions compared with Col-0 plants, showing increased tolerance to Zn deficiency and increased sensitivity to growth on high Zn (Fig. 7A).

Fig. 7.
Tolerance to Zn deficiency and high Zn, as well as Zn accumulation in AtZIP1 and AtZIP2 T-DNA knockout lines. (A) Tolerance to Zn deficiency (–Zn nutrient solution) and high Zn (nutrient solution +300 µM Zn) based on relative root growth ...

The T-DNA lines were also assayed for root and shoot Zn accumulation in response to growth for 2 weeks on high Zn (300 µM Zn). ICP analysis of root and shoot Zn content for the T-DNA lines showed that roots of AtZIP2 knockout plants accumulated significantly more Zn than roots of Col-0 plants (Fig. 7B), which is consistent with the increased sensitivity to high Zn seen in Fig. 7A. It should be noted that the relatively high root Zn concentrations (1000–2000 ppm) are due in part to Zn2+ binding to negatively charged cation exchange sites in the cell walls that is not totally removed by any desorption protocol including the standard protocol (15min desorption in 5mM CaCl2; Hart et al., 1998a , b ). When micronutrient metal uptake in roots was first studied, detailed desorption studies were conducted and for the divalent cations Zn2+, Cu2+, Mn2+, and Cd2+ it was found that no desorption protocol removed all of the cell wall-bound metal. The only way this could be accomplished was with longer desorption times that also allowed significant efflux of the absorbed metal from the cytoplasm. Hence the 15min desorption period using 5mM CaCl2 (the 5mM Ca2+ relatively effectively desorbs other metal cations bound to the cell wall) is a compromise in that it is the longest desorption period which can be run without allowing for significant loss of Zn or Mn ions that have been previously absorbed into the root symplasm. The T-DNA lines were also grown on Mn-deficient (–Mn) and high Mn (250 µM) conditions; both the AtZIP1 and AtZIP2 knockout lines were more sensitive to Mn deficiency compared with Col-0 plants, with the AtZIP1 T-DNA lines exhibiting considerably greater sensitivity to Mn deficiency (Fig. 8A). When high Mn stress was imposed, AtZIP2 lines were somewhat more tolerant to high Mn compared with Col-0 plants (based on root growth), whereas AtZIP1 lines showed no difference in tolerance to high Mn relative to Col-0 plants.

Fig. 8.
Tolerance to Mn deficiency and high Mn, as well as Mn accumulation in AtZIP1 and AtZIP2 T-DNA knockout lines. (A) Tolerance to Mn deficiency (–Mn nutrient solution) and high Mn (nutrient solution +250 µM Mn) based on relative root growth ...

With regards to root and shoot Mn accumulation in response to growth on high Mn, both T-DNA knockout lines exhibited much greater root Mn accumulation compared with Col-0 plants (6- to 7-fold greater), but no differences in shoot Mn accumulation were seen (Fig. 8B).

Because root Fe uptake can sometimes interact and interfere with root Zn uptake, Fe accumulation was also looked at in the two knockout lines under the same replete, high and –Zn, and high and –Mn growth conditions. It was found that there were no differences in root and shoot Fe accumulation between the AtZIP1 and AtZIP2 knockout lines and the Col-0 plants grown under all three nutrient regimes, as seen for nutrient-replete plants in Supplementary Fig. S1 at JXB online (data for high Zn/–Zn and high Mn/–Mn grown plants are not shown for the sake of clarity).


The ZIP family has been implicated in the transport of the essential micronutrients/heavy metals Fe, Zn, Mn, and Cu (Eide et al., 1996; Grotz et al., 1998; Pence et al., 2000; Wintz et al., 2003; Cohen et al., 2004; Pedas et al., 2008; Lin et al., 2009). When the remaining 11 members of the ZIP family which to date have not been characterized were tested for their ability to complement the various yeast uptake mutants, six of the ZIPs were able to transport Zn, six were able to transport Mn, and one ZIP was able to transport Fe. Ten of the 11 were able to complement at least one mutant, three were able to complement two different mutants and one transporter gene, ZIP7, was able to complement the Zn-, Fe-, and Mn-deficient uptake mutants. Only ZIP8 failed to complement any of the yeast transport mutants. This may be due to: (i) the alternative transcript for ZIP8 repeatedly isolated in this study possibly encoding a non-functional protein; (ii) possible mislocalization of the ZIP8 protein in yeast to an internal membrane; or (iii) the possibility that ZIP8 may be a pseudogene. It was not possible to isolate the predicted ZIP8 sequence listed on TAIR, but it should be noted that one of the two introns was edited out in the ZIP8 sequence which was isolated here, and most probably represents a real transcript. Also, the lack of the predicted ZIP8 mature transcript may suggest that if ZIP8 is a true gene, it must be expressed at very low levels or may only be expressed under specific conditions or in a tissue not studied herein. This inability to clone ZIP8 was not entirely surprising as there are no listed ZIP8 expressed sequence tags in the TAIR database; the only evidence that ZIP8 is expressed is two peptide matches.

AtZIP1 transcript levels increase in the roots as the Arabidopsis plant ages, and its expression decreases in the shoot during the same developmental time sequence. AtZIP1 transcript levels were higher in the roots under both Zn deficiency (4-fold increase) and Fe deficiency (2.5-fold increase) relative to expression in nutrient-replete roots, while no significant changes in AtZIP1 expression were seen in response to the other micronutrient deficiencies. This increase in relative transcript levels under Zn deficiency in the roots is similar to what was seen when AtZIP1 was first cloned by Grotz et al. (1998). In the shoots, Zn deficiency resulted in reduced AtZIP1 transcript abundance, while Mn deficiency resulted in increased expression levels of AtZIP1. A region that is very similar in sequence to the zinc-responsive element identified in the AtZIP4 promoter (Assunção et al., 2010) was found in the AtZIP1 promoter, although it differs by 2bp from the AtZIP4 ZRE; this element is located ~700bp upstream of the start codon in the AtZIP1 promoter. Assunção et al. (2010) showed that the lack of expression of the transcription factors driving Zn responsiveness in the ZIPs (bZIP19 and bZIP23) did lead to altered AtZIP1 expression and thus AtZIP1 may be an additional target of the two bZIP proteins. This may also suggest that a somewhat broader promoter sequence is recognized by the bZIP transcription factors in controlling Zn homeostasis in plants.

Root and shoot AtZIP2 transcript abundance was decreased in response to Zn, Fe, and Mn deficiency. The AtZIP2 promoter does not appear to harbour any ZRE elements similar to those identified by Assunção et al. (2010), which would explain the lack of activation of AtZIP2 expression under Zn-limiting conditions. However, the observed lower transcript levels under Zn and Mn deficiency suggest that AtZIP2 is not a primary transporter involved in Zn and Mn uptake from the soil under Zn- and Mn-limiting conditions. Also, the AtZIP2 expression pattern fits well with what was presented for the same gene in the Atgenexpress database for global gene expression using the Affymetrix ATX array, which also showed ~10-fold higher transcript levels in the roots relative to the shoots, and that older plants exhibit much higher expression than is seen in younger plants (Table 1). AtZIP2 also appears to be one of the most highly expressed ZIP genes in Arabidopsis, but only in mature roots.

To better understand the role that both AtZIP1 and AtZIP2 have in Zn homeostasis in plants, plant growth and Zn accumulation were quantified in AtZIP1 and AtZIP2 T-DNA insertion lines in plants grown under both high and low Zn conditions. From these findings presented in Fig. 7, which were a lack of an effect of knocking out AtZIP1 on tolerance and tissue accumulation in response to low or high Zn, it appears that AtZIP1 probably does not play a major role in Zn uptake. Possibly because of its vacuolar localization, it could play a role in remobilizing Zn from the vacuole. Several members of the ZIP family in animals and yeast have been shown to be involved in transporting Zn from the vacuole to the cytoplasm (MacDiarmid et al., 2000; Liuzzi and Cousins, 2004). The localization to the vacuole along with the lack of a phenotype when knocked out under high or low Zn conditions suggests that either AtZIP1 plays a very specific role in Zn homeostasis, or there may be redundancy with other ZIP family members or other Zn transporters in the genome. However, the increased sensitivity to low Mn and the increased accumulation of Mn in roots in the AtZIP1 knockout suggests that AtZIP1 probably plays a role in remobilizing Mn from the vacuole into the cytosol (Fig. 8A, ,B).B). In the current study, the tissue-specific expression of AtZIP1 in the stele along with the in planta data suggests that AtZIP1 plays a role in helping to remobilize Mn for translocation to the shoot when the plant is responding to low Mn status.

It was somewhat surprising to find that AtZIP1 is localized to the vacuole in Arabidopsis but was still able to complement a yeast mutant defective in plasma membrane Zn uptake. This is not the first time a plant micronutrient transporter complemented yeast mutants defective in plasma membrane metal uptake but were found to reside in an endomembrane in plants. The ZIP family member, IRT2, was found in an initial study to complement yeast plasma membrane Fe and Zn uptake-defective mutants (Vert et al., 2001). In a subsequent publication, AtIRT2 was found to localize to endomembrane vesicles in Arabidopsis (Vert et al., 2009). Also, AtNramp3 and AtNramp4 were both shown to complement the fet3/fet4 Fe uptake mutant (Thomine et al., 2000), but were later found to localize to the vacuole in plants and are involved in the efflux of Fe out of the vacuole (Lanquar et al., 2005). There are two possible explanations for the ability of the ZIPs that localize to plant endomembranes to complement yeast plama membrane transporter mutants. It is possible in yeast, particularly for AtZIP1, that it mediates vacuolar Zn transport from the vacuole to the cytoplasm when the yeast are grown on low Zn, and this provides enough cytoplasmic Zn to promote yeast growth on the low Zn media. Alternatively, it may be that in yeast, the transporters and particularly AtZIP2, which appears to be a Mn/Zn uptake transporter, are mislocalized to the yeast plasma membrane and mediate Mn and Zn uptake into the cell.

AtZIP2 most probably plays a role in Mn (and possibly Zn) transport into the root vasculature for translocation to the shoot and not root Mn uptake from the soil, based on its high root expression in the stele, as well as the phenotypes observed for the T-DNA AtZIP2 knockout line. The T-DNA insertion line lacking AtZIP2 expression was more tolerant to high Mn and less tolerant to low Mn, which is consistent with AtZIP2 playing a role in ultimately providing Mn to the shoot. The suggestion is that this transporter mediates Mn uptake into the cells of the root stele, which assists in providing Mn to the xylem parenchyma, where other transporters such as AtHMA2/4 mediate xylem loading of the metal for transport to the shoot in the transpiration stream. This would also be consistent with the increased root Mn (and Zn) accumulation in this knockout line resulting from a lack of or reduced transport into the root stele, and ultimately into the xylem and up to the shoot. This could maintain root Zn and Mn at high levels. It is also unlikely that AtZIP2 plays a role in root Zn uptake from the soil, given the tissue in which ZIP2 is expressed. When previous findings had shown that AtZIP2 was a high affinity Zn uptake transporter with an appropriately low K m for Zn uptake from the soil, it did seem like a logical candidate for this role (Grotz et al., 1998). However, the data presented in the current study support a role for AtZIP2 in metal transport into the stele which presumably promotes the accumulation of the metal in xylem parenchyma. This is supported by the observation that root expression of AtZIP2 under Zn deficiency is reduced compared with nutrient-replete conditions. It is likely that expression of a major root Zn uptake transporter should be up-regulated under Zn deficiency conditions, similar to IRT1 being up-regulated by Fe deficiency. This lower level of expression in the roots under deficiency conditions again suggests a role for ZIP2 in the transport of Zn and Mn to or into the xylem parenchyma, which would be associated with higher gene expression under conditions of sufficient Zn or Mn to help facilitate the movement of the metals to the xylem for transport to the shoots. AtZIP2 expression was higher in more mature regions of the root and lower in the growing root tip within the regions of cell division and differentiation. The literature is lacking in definitive studies quantifying which regions of the root contribute most to Zn translocation to the shoot. However, it is generally accepted, as shown in Shi et al. (2009), that Zn (or another divalent cation) uptake for translocation to the shoot is greater in the mature regions of the root. Future studies mapping which portions of the roots are the most active in Zn uptake from the soil as well as how Zn moves throughout the root, when compared with the cell- and tissue-specific expression of these transporters, will be useful in determining which root micronutrient transporters are involved in root Zn uptake from the soil.

Supplementary data

Supplementary data are available at JXB online.

Figure S1. Root and shoot Fe accumulation determined via ICP-AES for 2-week-old AtZIP1 or AtZIP2 T-DNA insertion lines and Col-0 plants grown on MS nutrient solution+1% sucrose.

Table S1. A list of primers used to amplify the predicted open reading frames of 11 AtZIP family members and to isolate the promoters of AtZIP1 and AtZIP2.

Supplementary Data:


  • Alonso JM, Stepanova AN, Leisse TJ, et al. 2003. Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301, 653–657. [PubMed]
  • Assunção AG, Herrero E, Lin YF, Huettel B, Talukdar S, Smaczniak C, Immink RG, van Eldik M, Fiers M, Schat H. 2010. Arabidopsis thaliana transcription factors bZIP19 and bZIP23 regulate the adaptation to zinc deficiency. Proceedings of the National Academy of Sciences, USA 107, 10296–10301. [PubMed]
  • Cohen CK, Garvin DF, Kochian LV. 2004. Kinetic properties of a micronutrient transporter from Pisum sativum indicate a primary function in Fe uptake from the soil. Planta 218, 784–792. [PubMed]
  • Connolly EL, Fett JP, Guerinot ML. 2002. Expression of the IRT1 metal transporter is controlled by metals at the levels of transcript and protein accumulation. The Plant Cell 14, 1347–1357. [PubMed]
  • Eide D, Broderius M, Fett J, Guerinot ML. 1996. A novel iron-regulated metal transporter from plants identified by functional expression in yeast. Proceedings of the National Academy of Sciences, USA 93, 5624–5628. [PubMed]
  • Grotz N, Fox T, Connolly E, Park W, Guerinot ML, Eide D. 1998. Identification of a family of zinc transporter genes from Arabidopsis that respond to zinc deficiency. Proceedings of the National Academy of Sciences, USA 95, 7220–7224. [PubMed]
  • Guerinot ML. 2000. The ZIP family of metal transporters. Biochimica et Biophysica Acta 1465, 190–198. [PubMed]
  • Hart JJ, Norvell WA, Welch RM, Sullivan LA, Kochian LV. 1998a. Characterization of zinc uptake, binding, and translocation in intact seedlings of bread and durum wheat cultivars. Plant Physiology 118, 219–226. [PubMed]
  • Hart JJ, Welch RM, Norvell WA, Sullivan LA, Kochian LV. 1998b. Characterization of cadmium binding, uptake and translocation in intact seedlings of bread and durum wheat cultivars. Plant Physiology 116, 1413–1420. [PubMed]
  • Lanquar V, Lelièvre F, Bolte S, et al. 2005. Mobilization of vacuolar iron by AtNRAMP3 and AtNRAMP4 is essential for seed germination on low iron. EMBO Journal 24, 4041–4051. [PubMed]
  • Lin YF, Liang HM, Yang SY, Boch A, Clemens S, Chen CC, Wu JF, Huang JL, Yeh KC. 2009. Arabidopsis IRT3 is a zinc-regulated and plasma membrane localized zinc/iron transporter. New Phytologist 182, 392–404. [PubMed]
  • Liuzzi JP, Cousins RJ. 2004. Mammalian zinc transporters. Annual Review of Nutrition 24, 151–172. [PubMed]
  • MacDiarmid CW, Gaither LA, Eide DJ. 2000. Zinc transporters that regulate vacuolar zinc storage in Saccharomyces cerevisiae . EMBO Journal 19, 2845–2855. [PubMed]
  • Pedas P, Ytting CK, Fuglsang AT, Jahn TP, Schjoerring JK, Husted S. 2008. Manganese efficiency in barley: identification and characterization of the metal ion transporter HvIRT1. Plant Physiology 148, 455–466. [PubMed]
  • Pena MM, Puig S, Thiele DJ. 2000. Characterization of the Saccharomyces cerevisiae high affinity copper transporter Ctr3. Journal of Biological Chemistry 275, 33244–33251. [PubMed]
  • Pence NS, Larsen PB, Ebbs SD, Lasat MM, Letham DLD, Garvin DF, Eide D, Kochian LV. 2000. The molecular basis for heavy metal hyperaccumulation in Thlaspi caerulescens . Proceedings of the National Academy of Sciences, USA 97, 4956–4960. [PubMed]
  • Rogers EE, Eide DJ, Guerinot ML. 2000. Altered selectivity in an Arabidopsis metal transporter. Proceedings of the National Academy of Sciences, USA 97, 12356–12360. [PubMed]
  • Schmid M, Davison TS, Henz SR, Pape UJ, Demar M, Vingron M, Schölkopf B, Weigel D, Lohmann JU. 2005. A gene expression map of Arabidopsis thaliana development. Nature Genetics 37, 501–506. [PubMed]
  • Sheen J. 2001. Signal transduction in maize and Arabidopsis mesophyll protoplasts. Plant Physiology 127, 1466–1475. [PubMed]
  • Shi J, Gras MA, Silk WK. 2009. Laser ablation ICP-MS reveals pattern of copper differing from zinc in growth zones of cucumber roots. Planta 229, 945–954. [PubMed]
  • Supek F, Supekova L, Nelson H, Nelson N. 1997. Function of metal-ion homeostasis in the cell division cycle, mitochondrial protein processing, sensitivity to mycobacterial infection and brain function. Journal of Experimental Biology 200, 321–330. [PubMed]
  • Thomine S, Wang R, Ward JM, Crawford NM, Schroeder JI. 2000. Cadmium and iron transport by members of a plant transporter gene family in Arabidopsis with homology to NRAMP genes. Proceedings of the National Academy of Sciences, USA 97, 4991–4996. [PubMed]
  • Vert G, Barberon M, Zelazny E, Séguéla M, Briat JF, Curie C. 2009. Arabidopsis IRT2 cooperates with the high-affinity iron uptake system to maintain iron homeostasis in root epidermal cells. Planta 229, 1171–1179. [PubMed]
  • Vert G, Briat JF, Curie C. 2001. Arabidopsis IRT2 gene encodes a root–periphery iron transporter. The Plant Journal 26, 181–189. [PubMed]
  • Vert G, Grotz N, Dédaldéchamp F, Gaymard F, Guerinot ML, Briat JF, Curie C. 2002. IRT1, an Arabidopsis transporter essential for iron uptake from the soil and for plant growth. The Plant Cell 14, 1223–1233. [PubMed]
  • von Heijne G. 1992. Membrane protein structure prediction: hydrophobicity analysis and the ‘positive inside’ rule. Journal of Molecular Biology 225, 487–494. [PubMed]
  • Wintz H, Fox T, Wu YY, Feng V, Chen W, Chang HS, Zhu T, Vulpe C. 2003. Expression profiles of Arabidopsis thaliana in mineral deficiencies reveal novel transporters involved in metal homeostasis. Journal of Biological Chemistry 278, 47644–47653. [PubMed]
  • Zhao H, Eide D. 1996. The yeast ZRT1 gene encodes the zinc transporter of a high affinity uptake system induced by zinc limitation. Proceedings of the National Academy of Sciences, USA 93, 2454–2458. [PubMed]

Articles from Journal of Experimental Botany are provided here courtesy of Oxford University Press