Introduction

Trace amounts of nickel ion (Ni2+) are essential for plant nutrition. However, excessive concentrations of nickel in the environment become phytotoxic through disruption of various enzyme activities, photosynthesis and mineral nutrition (Pandey and Sharma 2002; Samarakoon and Rauser 1979). Therefore, plants possess a range of cellular mechanisms for Ni2+ homeostasis and tolerance (Hall 2002). A general mechanism for Ni2+ detoxification in plants is the chelation of Ni2+ by ligands such as glutathione (GSH), nicotianamine (NA), histidine (His), or organic acids, and subsequent compartmentalization of the ligand–metal complexes (Cobbett 2000; Hall 2002). For example, NA is synthesized from S-adenosylmethionine by the action of nicotianamine synthase (NAS) and heterologous expression of TcNAS1 and HvNAS1 in Arabidopsis resulted in increased NA production and improved tolerance to Ni2+ (Kim et al. 2005; Pianelli et al. 2005). Free His is another important Ni2+ chelator contributing to Ni2+ tolerance in plants. Some Ni2+-hyperaccumulating plant species contain high levels of free His which exists in coordination with the Ni2+ ions. Exogenous Ni2+ elicits a dose-dependent increase of free His concentration in xylem of these hyperaccumulators (Kerkeb and Kramer 2003; Krämer et al. 1996). Ni2+-nonaccumulating plant species were unable to uptake and accumulate exogenous Ni2+. However, the plants acquired the ability to uptake and accumulate external Ni2+ by supplementation of His (Kerkeb and Kramer 2003; Krämer et al. 1996). Overexpression of an enzyme in His biosynthetic pathway is sufficient to enhance Ni2+ tolerance through the increase of endogenous free His in Arabidopsis (Ingle et al. 2005; Wycisk et al. 2004). Meanwhile, several transporters involved in Ni2+ transport including Nicotiana calmodulin-binding protein 4 (NtCBP4), Arabidopsis iron-regulated protein 2 (AtIREG2) and Arabidopsis iron-regulated transporter 1 (AtIRT1) have been identified (Arazi et al. 1999; Nishida et al. 2011; Schaaf et al. 2006). However, the transcriptional regulations of these Ni2+ transporter genes remain unclear.

Cys2/His2-type zinc finger proteins (C2H2-ZFPs) represent one of the better-characterized classes of eukaryotic transcription factors (Ciftci-Yilmaz and Mittler 2008). Plant C2H2-ZFPs comprise a large family and play critical roles in developmental processes and environmental stress tolerance through transcriptional regulation, posttranslational regulation and protein–protein interactions (Laity et al. 2001). The C1-2i subclass of the Arabidopsis C2H2-ZFP family contains 20 members including ZAT6, ZAT7, ZAT10, ZAT11 and ZAT12. Members of this subclass are characterized by two zinc finger motifs, an invariant QALGGH motif in the zinc finger domain and an ERF-associated amphiphilic repression (EAR) motif in the C-terminal region. The EAR motif is believed to function as the transcriptional repression motif (Ciftci-Yilmaz and Mittler 2008; Takatsuji 1999). ZAT6 has been reported to be involved in root development and phosphate homeostasis. Overexpression of ZAT6 in Arabidopsis resulted in repression of primary root growth and a subsequent change in phosphate acquisition (Devaiah et al. 2007). Transgenic plants constitutively expressing ZAT7 showed enhanced tolerance to salt stress (Ciftci-Yilmaz et al. 2007). ZAT10 plays a dual role as a positive and negative regulator of numerous abiotic stresses (Mittler et al. 2006). ZAT12 plays a central role in reactive oxygen species (ROS) and abiotic stress signaling (Rizhsky et al. 2004). Recently, ZAT11 was shown to be involved in oxidative stress-induced programmed cell death (Qureshi et al. 2013). However, the physiological function of ZAT11 in plant development and stress tolerance still remains unclear. Here, we report that ZAT11 is a positive regulator in primary root growth at normal condition but a negative regulator in Ni2+ tolerance at high Ni2+ condition. We suggest that ZAT11 represses Ni2+ tolerance by the transcriptional repression of a vacuolar Ni2+ transporter gene.

Materials and methods

Plant materials and growth conditions

All plants used in this study were in the Columbia-0 (Col-0) genetic background. T-DNA insertion mutants of ZAT11 (SALK_013996 and SALK_110012) were obtained from Arabidopsis Biological Resource Center. All seeds were surface sterilized, placed in the dark at 4 °C for 2 days, and then sown on 1/2 Murashige and Skoog (MS) agar plates containing 2 % sucrose. Seedlings were grown under a 16 h light/8 h dark cycle in a growth chamber at 22 °C, 75 % humidity and 100 μmol m−2 s−1 light intensity.

Construction of transgenic plants

ZAT11 cDNA fused in-frame to 3X FLAG was subcloned into the plant binary vector pCAMBIA 1300, resulting in 35S::ZAT11-FLAG plasmid. The plasmid was introduced into Agrobacterium tumefaciens GV3101, and the transformed Arabidopsis plants were obtained by the Agrobacterium-mediated floral dip method (Clough and Bent 1998). Seeds were collected from infiltrated plants and transformants were selected by plating on MS medium containing 40 μg ml−1 hygromycin. Homozygous lines of ZAT11-overexpressing transgenic plants were used for all experiments.

Histochemical GUS assay

The forward primer 5′-GAATTCGTAAAAAAGAAAGCTTATG-3′ and reverse primer 5′-GGATCCACTTCTTTGAGAATTAAGATC-3′ were used to amplify ZAT11 promoter from genomic DNA (−1,285 to +3 bp from the translational start site). The amplified fragment was digested with EcoR1/BamH1 and subcloned into pCAMBIA-1301 vector to yield a PromZAT11::GUS plasmid. The plasmid was transformed into Arabidopsis by the Agrobacterium-mediated floral dip method (Clough and Bent 1998). Histochemical localization of GUS activity was performed as previously described (Jefferson et al. 1987).

Subcellular localization of ZAT11

To ascertain the subcellular localization of ZAT11, ZAT11 cDNA was subcloned into the p35S::smGFP expression vector to construct the plasmid, 35S::ZAT11-smGFP, for the expression of ZAT11 tagged with soluble modified GFP (smGFP) at its C-terminus under the control of CaMV 35S promoter. The plasmid was introduced into Arabidopsis protoplasts by polyethylene glycol (PEG)-mediated transformation or onion epidermal cells using the particle bombardment method as previously described (Lee et al. 2007; Varagona et al. 1992). Expression of the fusion constructs was monitored at 24 h after transformation by fluorescence microscopy using a fluorescence microscope (Carl Zeiss, Germany) as previously described (Lee et al. 2007).

Semi-quantitative and quantitative RT-PCR

Total RNA was extracted using a LiCl method (Liu et al. 2010) and rendered free of genomic DNA by digestion with RNase-Free DNase I (Promega, USA). First-strand cDNA was synthesized from 5 μg of total RNA in a 20 μl reaction mixture using cDNA synthesis kit (Fermentas, Canada). Semi-quantitative RT-PCR was carried out following cDNA synthesis using gene-specific primers as previously described (Liu et al. 2010). Tubulin2 level was monitored as an internal control. Quantitative RT-PCR was performed as described (Han et al. 2012). Primer sequences are listed in Table S1.

Plant growth assays for stress tolerance

All Arabidopsis seeds used for experiments were grown and harvested at same condition. For stress tolerance analysis, seeds were grown on vertically placed 1/2 MS plates (with 2 % sucrose) for 4 days and then transferred to 1/2 MS plates containing H2O2, CdCl2, Ni(NO3)2 or KNO3 and grown for indicated time. The fresh weight and the root length of plants were measured.

The measurement of nickel ion content

Individual plants were harvested, rinsed in distilled water and dried at 80 °C for 48 h. The dried material was weighed and digested with concentrated HNO3 for 90 min at 200 °C. Samples were then centrifuged at 2,000×g for 10 min and the clear supernatant was diluted tenfold with deionized water. Ni2+ content was measured by inductively coupled plasma-optical emission spectroscopy (Perkin Elmer, CT, USA).

Results and discussion

ZAT11 is a nuclear protein

ZAT11 has been expected to be localized in nucleus because ZAT11 was proposed to function as a transcriptional repressor (Ohta et al. 2001). However, the subcellular localization of ZAT11 has not yet been examined. To investigate the subcellular localization of ZAT11, we introduced 35S::ZAT11-smGFP plasmid into Arabidopsis protoplasts or onion epidermal cells for the expression of smGFP tagged ZAT11 fusion protein (Fig. 1). The fluorescence signal from ZAT11-GFP fusion protein was predominantly localized in the nucleus whereas the fluorescence signal of free GFP was detected throughout the whole cells in Arabidopsis protoplasts (Fig. 1b) as well as in onion epidermal cells (Fig. 1c). These results strongly indicate that ZAT11 is a nuclear localized transcriptional regulator. Most of C2H2-type ZFPs that contain two zinc finger motifs have been shown to bind directly to A(G/C)T repeats within an EP2 sequence (Sakamoto et al. 2004). However, we could not detect the binding of recombinant ZAT11 protein to the EP2 sequence by electrophoretic mobility shift assay (data not shown). This result implies that ZAT11 has different DNA binding property from other related C2H2-type ZFPs.

Fig. 1
figure 1

The subcellular localization of green fluorescence protein (GFP) tagged ZAT11. a Schematic representation of the constructs used for subcellular localization. smGFP and ZAT11::smGFP represent the constructs for constitutive expression of soluble modified GFP and ZAT11-smGFP fusion protein, respectively. b Subcellular localization of the indicated proteins in Arabidopsis protoplasts. c Subcellular localization of the indicated proteins in onion epidermal cells. DIC differential interference contrast, GFP green fluorescent protein, OVERLAP overlap of DIC and GFP images. Arrowheads indicate the positions of nuclei

Full size image

Tissue-specific pattern of ZAT11 gene expression

The tissue-specific expression pattern of ZAT11 was determined by evaluating GUS reporter activity in tissues of transgenic plants expressing PromZAT11::GUS. Strong GUS activity was detected in cotyledons, primary roots and hypocotyls of seedlings (Fig. 2). Strong expression of ZAT11 in root tip suggests that ZAT11 may be involved in development or stress tolerance in root.

Fig. 2
figure 2

The gene expression patterns of ZAT11 in tissues. Histochemical analysis of GUS activity in PromZAT11::GUS transgenic Arabidopsis plants. a Five-day-old whole seedling; b, c close-up views of the hypocotyl (b) and primary root tip (c) of a 5-day-old seedling; d 2-week-old whole seedling; e flowers; f silique

Full size image

Primary root growth was increased in ZAT11-overexpressing transgenic plants

To identify the physiological function of ZAT11, we obtained two independent T-DNA insertion mutant lines of ZAT11 (SALK_013996 and SALK_110012) from the Arabidopsis Biological Resource Center collection. However, these lines exhibited comparable transcript levels with wild type plant (Fig. S1) because of each T-DNA insertion in the 5′-untranslated region of ZAT11. Therefore, we performed a gain-of-function approach to investigate the physiological function of ZAT11. We constructed transgenic plants (ZAT11 OE) constitutively overexpressing FLAG-tagged ZAT11 under the control of CaMV 35S promoter (Fig. 3a). Three independent homozygous lines (OE #8, OE #11 and OE #20) showing high expression levels of ZAT11-FLAG protein were chosen for further study. RT-PCR analysis showed that ZAT11 transcripts were highly expressed in these transgenic lines (Fig. S2A). The primary roots were significantly longer in ZAT11 OE plants than in empty vector plants (EV) (Fig. 2b, c). The growth difference at primary roots was observed at 4 days after germination (Fig. 3b), and much obvious at 16 days after germination (Fig. 3c). Microscopic analyses suggested that the elongated primary root of ZAT11 OE was caused by the increased lengths of cells but not by the increased number of cells (Fig. S3). However, we could not detect significant difference in leaves in soil-grown plants (Fig. 3b, bottom panel). These results imply that ZAT11 may function as a positive transcriptional regulator for primary root growth at normal condition. The increased root growth phenotype of transgenic ZAT11 OE plants is opposite to those of ZAT6, ZAT7, ZAT10 and ZAT12 overexpressing transgenic plants (Ciftci-Yilmaz et al. 2007; Devaiah et al. 2007; Mittler et al. 2006; Rizhsky et al. 2004). This difference strongly indicates that the physiological function of ZAT11 could be different from those of other C2H2-type ZFPs in Arabidopsis.

Fig. 3
figure 3

Increased growth of primary root in ZAT11 OE plants. a Western blot analysis of ZAT11 protein levels in empty vector (EV) and several independent Prom35S::ZAT11-FLAG transgenic plant lines (OE). Total proteins (50 μg) from 2-week-old seedlings were analyzed by immunoblot with an anti-FLAG antibody (Anti-FLAG). The Coomassie brilliant blue-stained membrane (CBB) is shown as a loading control. b The indicated lines were grown on 1/2 MS (top and middle rows) or in soil (bottom row). The photographs were taken at 4 days, 8 days and 4 weeks after germination as indicated. c Measurements of primary root length in plants grown on 1/2 MS-agar for 4, 8 and 16 days. Data are the mean ± SE of three independent experiments. Asterisks indicate statistically significant differences compared with the corresponding EV as determined by Student’s t test (*P < 0.05; **P < 0.01)

Full size image

ZAT11 overexpressing transgenic plants are sensitive to Ni2+

It has been reported that ZAT11 transcript is highly induced by H2O2 (Gechev et al. 2005), and ZAT11 may be involved in oxidative stress-mediated programmed cell death (Qureshi et al. 2013). To determine whether ZAT11 can confer the resistance against H2O2, we treated H2O2 to ZAT11 OE. The elongated primary root phenotype of ZAT11 OE was disappeared by the treatment of H2O2 (Fig. S2B), suggesting that ZAT11 cannot protect the toxic effect of H2O2. Several metal ions are well known to induce free radicals including H2O2 and result in oxidative stress in plants (Maksymiec 2007). To determine whether ZAT11 can confer the resistance against toxic metal ions, we examined the growth of ZAT11 OE in the media containing different concentrations of CdCl2 or Ni(NO3)2. The root growth of both plants was inhibited by the increase of CdCl2 concentration. However, the primary roots of ZAT11 OE were still longer than those of EV in media containing CdCl2 (Fig. S4), indicating that ZAT11 may not function in the resistance against CdCl2 (Fig. S4). In contrast, three ZAT11 OE lines had shorter primary roots than EV plants in media containing 60, 80 or 100 μM Ni(NO3)2, whereas they had longer roots than EV plants in control media (Figs. 4a, d, S5). In addition, ZAT11 OE plants showed increased numbers of lateral roots compared to EV plants in medium containing 60 μM Ni(NO3)2 (Fig. 4b). In addition, the first leaves of ZAT11 OE plants showed chlorosis in media containing 80 μM Ni(NO3)2 whereas EV plants did not (Fig. 4c). Fresh weights of ZAT11 OE plants were significantly reduced by the addition of Ni(NO3)2 to media (Figs. 4e, S5C).To rule out the possibility of toxicity due to high nitrate levels, we repeated the experiment under the same media containing twice the concentration of KNO3 instead of Ni(NO3)2 (Fig. S6). The primary roots were longer in ZAT11 OE than in EV plants regardless of the concentration of KNO3. These data strongly suggest that ZAT11 is a negative regulator of Ni2+ tolerance in Arabidopsis. Next, we compared Ni2+ content in roots of ZAT11 OE and EV plants grown in media containing 60 or 80 μM Ni2+. The Ni2+ contents were significantly lower in ZAT11 OE compared to EV (Fig. 5), suggesting ZAT11 may be a negative regulator of Ni2+ accumulation in the internal compartments.

Fig. 4
figure 4

Decreased Ni2+ tolerance in ZAT11 OE plants. Four-day-old EV and ZAT11 OE plants grown on 1/2 MS were transferred to plates containing the same medium supplemented with 0, 60, 80 or 100 μM Ni(NO3)2. a Photographs illustrating phenotypes of ZAT11 OE plants at 4 days after transfer. b Increased lateral root numbers in ZAT11 OE plants at 4 days after transfer to growth medium containing 60 μM Ni(NO3)2. Arrowheads indicate the position of lateral roots. c Primary leaves in ZAT11 OE at 4 days after transfer to growth medium containing 80 μM Ni(NO3)2. Arrowheads indicate chlorosis in first leaves. d Measurements of primary root length of seedlings 4 days after transfer to medium containing indicated concentrations of Ni(NO3)2. Data are the mean ± SE of three independent experiments. Asterisks indicate statistically significant differences compared with the corresponding EV as determined by Student’s t test (*P < 0.05; **P < 0.01). e Measurements of fresh weight of seedlings 4 days after transfer to medium containing indicated concentrations of Ni(NO3)2. Data are the mean ± SE of three independent experiments. Asterisks indicate statistically significant differences compared with the corresponding EV as determined by Student’s t test (*P < 0.05; **P < 0.01)

Full size image
Fig. 5
figure 5

Less accumulation of Ni2+ in ZAT11 OE plants. Four-day-old EV and ZAT11 OE plants grown on 1/2 MS were transferred to 1/2 MS containing 60 or 80 μM Ni(NO3)2. Ni2+ content was measured in roots at 2 weeks after transfer. Data are the mean ± SE of three independent experiments. Asterisks indicate statistically significant differences compared with the corresponding EV as determined by Student’s t test (**P < 0.01)

Full size image

The transcript levels of a vacuolar Ni2+ transporter is decreased in ZAT11 OE plants

An important mechanism contributing to Ni2+ resistance is the chelation of Ni2+ by GSH, NA or His. Genes encoding key enzymes involved in biosynthesis of GSH, NA or His molecules are GSH1, NAS1 and ATP-PRT1/2, respectively. To test whether ZAT11 is involved in the gene expression of these genes, we performed RT-PCR. However, transcript levels of GSH1, NAS1 and ATP-PRT1/2 were not significantly different in ZAT11 OE and EV plants and not changed by the supplement of Ni(NO3)2 (Fig. 6a).

Fig. 6
figure 6

Decreased expression of a vacuolar Ni2+ transporter in ZAT11 OE plants. Four-day-old EV and ZAT11 OE plants grown on 1/2 MS were transferred to fresh 1/2 MS without (Control) or with 100 µM Ni(NO3)2 supplement. Total RNA was extracted from the plants at 2 weeks after transfer. Tubulin2 (TUB2) transcript level was measured as an internal standard for all PCR analyses. Shown in (a) are results of RT-PCR analysis of NAS1, GSH1, ATP-PRT1 and ATP-PRT2 levels. Shown in (b) is the quantitative PCR analysis of IREG2 transcripts. Data are the mean ± SE of three independent experiments. Asterisks indicate statistically significant differences compared with the corresponding EV as determined by Student’s t test (**P < 0.01)

Full size image

Sequestration of Ni2+ in the vacuole is one of key mechanisms to confer Ni2+ resistance in plant (Schaaf et al. 2006). Vacuolar Ni2+ transporters include members of IREG family. Overexpression of an IREG family, IREG2, resulted in enhanced nickel tolerance and increased Ni2+ accumulation in roots (Schaaf et al. 2006). To test whether the reduced Ni2+ accumulation in ZAT11 OE plants was correlated with the reduced gene expression of IREG2, we measured IREG2 transcript level by quantitative PCR (Fig. 6). The level of IREG2 transcript was significantly lower in ZAT11 OE than in EV plant regardless of the presence of 100 µM Ni(NO3)2 in media (Fig. 6b). These results suggest that ZAT11 is negatively involved in the transcription of IREG2 gene. Further study would be required to examine whether ZAT11 directly or indirectly regulates the transcription of IREG2.

In this study our results showed that ZAT11 positively regulates the elongation of primary root and negatively regulates the resistance against Ni2+ through the transcriptional control of a vacuolar Ni2+ transporter gene, IREG2. Our finding suggests the new biological function of a C2H2-type zinc finger protein that is involved in the metal ion tolerance of plants.