, Volume 235, Issue 2, pp 253–266

Cloning and characterization of a maize bZIP transcription factor, ZmbZIP72, confers drought and salt tolerance in transgenic Arabidopsis


  • Sheng Ying
    • Institute of Crop ScienceChinese Academy of Agricultural Sciences
    • College of Biological SciencesChina Agricultural University
  • Deng-Feng Zhang
    • Institute of Crop ScienceChinese Academy of Agricultural Sciences
  • Jing Fu
    • Institute of Crop ScienceChinese Academy of Agricultural Sciences
    • College of Biological SciencesChina Agricultural University
  • Yun-Su Shi
    • Institute of Crop ScienceChinese Academy of Agricultural Sciences
  • Yan-Chun Song
    • Institute of Crop ScienceChinese Academy of Agricultural Sciences
    • Institute of Crop ScienceChinese Academy of Agricultural Sciences
    • Institute of Crop ScienceChinese Academy of Agricultural Sciences
Original Article

DOI: 10.1007/s00425-011-1496-7

Cite this article as:
Ying, S., Zhang, D., Fu, J. et al. Planta (2012) 235: 253. doi:10.1007/s00425-011-1496-7


In plants, the bZIP (basic leucine zipper) transcription factors regulate diverse functions, including processes such as plant development and stress response. However, few have been functionally characterized in maize (Zea mays). In this study, we cloned ZmbZIP72, a bZIP transcription factor gene from maize, which had only one copy in the maize genome and harbored three introns. Analysis of the amino acid sequence of ZmbZIP72 revealed a highly conserved bZIP DNA-binding domain in its C-terminal region, and four conserved sequences distributed in N- or C-terminal region. The ZmbZIP72 gene expressed differentially in various organs of maize plants and was induced by abscisic acid, high salinity, and drought treatment in seedlings. Subcellular localization analysis in onion epidermal cells indicated that ZmbZIP72 was a nuclear protein. Transactivation assay in yeast demonstrated that ZmbZIP72 functioned as a transcriptional activator and its N terminus (amino acids 23–63) was necessary for the transactivation activity. Heterologous overexpression of ZmbZIP72 improved drought and partial salt tolerance of transgenic Arabidopsis plants, as determined by physiological analyses of leaf water loss, electrolyte leakage, proline content, and survival rate under stress. In addition, the seeds of ZmbZIP72-overexpressing transgenic plants were hypersensitive to ABA and osmotic stress. Moreover, overexpression of ZmbZIP72 enhanced the expression of ABA-inducible genes such as RD29B, RAB18, and HIS1-3. These results suggest that the ZmbZIP72 protein functions as an ABA-dependent transcription factor in positive modulation of abiotic stress tolerance and may be a candidate gene with potential application in molecular breeding to enhance stress tolerance in crops.


Abiotic stressbZIP-transcription factorMaizeStress tolerance



Abscisic acid


ABA responsive element


Arabidopsis thaliana


Basic leucine zipper

CaMV 35S

Cauliflower mosaic virus 35S promoter


Green fluorescent protein


Polymerase chain reaction




Quantitative real-time PCR


Rapid amplification of cDNA ends


Reactive oxygen species


Sucrose non-fermenting-1 related protein kinase


Untranslated region




Plants are often exposed to various adverse environmental stresses (i.e. drought, high-salinity, and low temperature) and these abiotic stresses negatively affect crop productivity. Extensive molecular studies have demonstrated that abscisic acid (ABA) plays an important role in stress response and tolerance to abiotic stresses (Finkelstein et al. 2002; Xiong et al. 2002; Yamaguchi-Shinozaki and Shinozaki 2006; Wasilewska et al. 2008; Nakashima et al. 2009). Most of stress-responsive genes are regulated by ABA, but some of them are not, indicating the existence of distinctive molecular regulatory mechanisms in the expression of stress-responsible genes, an ABA-dependent and ABA-independent regulatory pathway.

In plants, by specifically recognizing the relevant cis-acting elements in promoter regions of ABA-responsive genes, a major cis-acting element (ABRE) was characterized to be important for the expression of these genes in response to ABA (Guiltinan et al. 1990; Mundy et al. 1990; Busk and Pages 1998; Hattori et al. 2002). Subsequently, a series of bZIP transcription factors were identified that bind to ABRE, and they were demonstrated to be critical for the activation of downstream ABA-inducible gene expression (Choi et al. 2000; Uno et al. 2000).

Basic leucine zipper (bZIP) is a big transcription factor family in plants, the members of which regulate especially stress response and hormone signal transduction (Kim 2006). The bZIP proteins contain a basic region binding to DNA and a leucine zipper dimerization motif. In Arabidopsis, 75 distinct bZIP genes have been identified and classified into 10 groups by their conserved regions (Jakoby et al. 2002), whereas in rice, 89 bZIPs were divided into 11 groups (Nijhawan et al. 2008). Guedes Correa et al. (2008) reported that green plant bZIP genes from Arabidopsis, rice, and black cottonwood could be divided, based on the bZIP domain and other conserved motifs similarities, into 13 groups of bZIP homologs in angiosperms. Most of ABRE-binding bZIPs belonged to group A, in which the expression of several members could be strongly induced by ABA and abiotic stresses (Jakoby et al. 2002; Lu et al. 2009).

Previous studies have shown that transgenic plants overexpressing the group A of bZIP genes enhanced tolerance to abiotic stresses. In Arabidopsis, AREB1/ABF2 was reported to be an important component in glucose signaling and its overexpression affected multiple stress tolerance (Kim et al. 2004; Fujita et al. 2005). Overexpression of ABF3 or AREB2/ABF4 improved drought tolerance of transgenic plants (Kang et al. 2002). In rice, OsbZIP72 was reported to function as a positive regulator in ABA signal transduction and the seedlings overexpressing OsbZIP72 showed an increased drought tolerance (Lu et al. 2009).

Maize (Zea mays) is one of the most important crops for livestock and humans. Its growth and development are seriously affected by various abiotic stresses. Therefore, it is important to elucidate mechanisms of stress response in maize. A number of stress-related genes in maize have been identified and characterized well. However, only a few of maize bZIP transcription factor genes related to abiotic stress were reported (Nieva et al. 2005; Zhang et al. 2008, 2011; Jia et al. 2009). In present study, we isolated a putative bZIP gene, ZmbZIP72, analyzed its expression profile under different treatments, and characterized its role in stress tolerance. Transgenic Arabidopsis plants overexpressing ZmbZIP72 showed increased sensitivity to ABA and osmotic stress and exhibited enhanced tolerance to drought and salt stress compared with WT plants.

Materials and methods

Cloning of the ZmbZIP72 gene and sequence analysis

To identify maize bZIP genes, we searched for the mRNA sequences or expressed sequence tags (ESTs) that encode a basic leucine zipper domain in MaizeGDB database ( and PlantGDB ( using tblastn. Among the obtained sequences, one mRNA sequence (PUT-155a-Zea_mays-16601), which encoded a putative bZIP protein with the highest similarity to rice OsbZIP72 gene (AK065873), was chosen for the further study and designated as ZmbZIP72.

Based on this result, the full-length cDNA sequence of ZmbZIP72 was amplified from maize (Zea mays) inbred line “CN165” (seeds stored in our laboratory) PCR using SMART RACE cDNA Amplification Kit (Clontech). The amplified products were purified and cloned into pMD18-T vector (Takara) for sequencing. The promoter of ZmbZIP72 was amplified by PCR from maize CN165 genomic DNA, using Genome Walking Kit (Takara), and prediction of cis-acting element was performed by the software PlantCARE (, Lescot et al. 2002). The sequences of primers used in the present study are listed in Suppl. Table S1.

Sequence assembling was performed using DNAMAN (Version 5.0). Multiple protein alignment was performed with ClustalX (version 1.83; Thompson et al. 1997) using default parameters. The phylogenetic tree was constructed with the neighbor-joining (NJ) method in MEGA (version 3.1; Saitou and Nei 1987; Kumar et al. 2004) and presented using TreeView (Page 1996). Bootstrap analysis was performed using 1,000 replicates in MEGA to evaluate the reliability.

Plant growth and stress treatments

The seeds of CN165 were surface-sterilized with 70% ethanol for 5 min, followed by several rinses with tap water, and then sown in the pots containing a mixture of vermiculite and organic fertilizer (2:1, v/v). They were grown in a greenhouse under natural conditions and watered once every 3 days. The seedlings at three-leaf stage were removed from the pots and washed carefully with tap water.

The root, stalk, and leaf of seedlings at three-leaf stage, and the root (aerial root), stalk (below the highest node), leaf (at node where the ear grew), tassel, and immature ear of CN165 at pre-flowering stage were used for organ-specific expression analysis.

For NaCl and ABA treatment, the seedling roots were submerged into a water solution supplemented with 200 mM NaCl and 100 μM (±)-cis, trans-ABA (Sigma), respectively. For cold treatment, the seedlings at the same development stage were transferred to a growth chamber at 4°C. For drought treatment, the seedlings were taken immediately out from pots and placed on the Whatman 3MM paper to dry. The treated seedlings were harvested at the given time points, frozen immediately in liquid nitrogen, and stored at −76°C.

Quantitative real-time PCR analysis

Total RNAs were extracted using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. The DNase-treated RNA was reverse-transcribed using SuperScript®II Reverse Transcriptase (Invitrogen). Quantitative real-time PCR were performed on a Rotor-Gene 3000 instrument (Corbett), using SYBR Premix Ex Taq™ (Takara). The reaction procedures were as follows: denature at 95°C for 3 min, followed by 45 cycles of 95°C for 5 s, 59°C for 20 s, and 72°C for 15 s. The maize GAPDH gene (NM_001111943) and ArabidopsisActin (NM_112764) gene were used as internal controls for normalization. The data obtained were analyzed with Rotor-Gene 6000 Series Software 1.7 (Corbett). The relative expression of detected gene was calculated using the relative 2−ΔΔCt method (Livak and Schmittgen 2001) and the error bars indicate SD (n = 4). Two experiments on independently grown plant materials were performed to confirm the reproducibility of the results.

Subcellular localization analysis

Full-length cDNA sequence except terminator codon (TGA) was amplified from CN165 and the sequencing-confirmed fragment was cloned into pZpIII vector, which contained the green fluorescent protein (GFP) reporter gene, to generate a ZmbZIP72-GFP fusion construct under the control of the CaMV 35S promoter. The construct was used for transient transformation of onion (Allium cepa) epidermis via a gene gun (Bio-Rad). After incubation on 1/2 MS medium for 24 h at 28°C, GFP fluorescence was observed under a confocal laser scanning system (Leica).

Transactivation assay

The full-length (W) and truncated fragments (N1-3, C1-2) of ZmbZIP72 (Fig. 5) were amplified by PCR and then fused into pGBKT7 vector (Clontech), respectively. pGBKT7 was used as negative control. These different constructs were individually transformed into yeast stain AH109 and the positive clones were selected on SD/-Trp/-His/-Ade. A colony-lift filter assay was performed according to the yeast handbook instructions (Clontech).

Generation of transgenic Arabidopsis plants

The sequencing-confirmed coding region of ZmbZIP72 cDNA was cloned into the pCAMBIA3301 vector (Cambia, Canberra, Australia), thus allowing the gene to be driven by the CaMV 35S promoter. The Arabidopsis thaliana L. ecotype Columbia plants (seeds stored in our laboratory) were grown under 16 h light/8 h dark condition at 22°C for transformation. The constructs were transferred into Agrobacterium tumefaciens GV3101 and were transformed into Arabidopsis wild type (WT) using the method by Zhang et al. (2006). The seeds of T0 generation were harvested and sown in soil. Two-week-old seedlings of T1 plants were screened by spraying with 0.5% (v/v) phosphinothricin (PPT) solution. The surviving transformants (T1) were confirmed by PCR to amplify 35S:ZmbZIP72 fragment and the bar gene. The confirmed transgenic plants were harvested individually. The T2 seeds were placed on MS (Murashige and Skoog 1962) agar medium (0.9%) containing 7 mg/L PPT and the transgenic lines with a 3:1 segregation ratio (resistant:sensitive) were selected to produce T3 seeds. The T3 lines displaying 100% PPT resistance were considered homozygous and used for further experiments. The Arabidopsis WT was used as control. All seeds of WT and transgenic plants were collected at the same stage.

Germination assays

Seeds were sterilized for 1 min with 70% (v/v) ethanol and 10 min with 5% NaClO (v/v), followed by several rinses with sterile distilled water. Fifty seeds of each transgenic lines and WT were placed on the MS agar medium (0.9%) supplemented with different concentrations of ABA (0.5, 1.0 or 1.5 μM), 150 mM NaCl, or 350 mM mannitol, respectively, maintained at 4°C in dark for 2 days, and then transferred to a controlled environment (16 h of light at 22°C and 8 h of darkness at 18°C). Germination (emergence of radicle) was calculated daily for 7 days after seeds were placed at controlled environment.

Treatments to transgenic plants

For the assays on the effects of dehydration on gene expression, the seeds of WT and transgenic plants were germinated and grown on the MS agar medium for 2 weeks. Then the seedlings were removed from the medium and dehydrated on dry Whatman 3MM paper and incubated at room temperature (22°C) for 6 h. Whole plants were used for RNA extraction.

For the drought tolerance evaluation, 30 seedlings were grown in pots filled with a soil mixture of vermiculite and organic fertilizer (1:1, v/v) for 2 weeks with constant watering before water was withheld. After 2 weeks without water, all pots were re-watered simultaneously, and then the seedlings were incubated under normal conditions. The survival rates were scored 7 days later. Plants were considered dead if there was no re-growth 7 days after re-watering.

For salt tolerance assay, 2-week-old Arabidopsis seedlings of each transgenic and WT plants were watered with NaCl solution (300 mM) applied at the bottom of the pots until the soil was completely saturated. Then the seedlings were incubated under normal conditions. After 2 weeks of salt treatment, all of the pots were re-watered simultaneously, and the phenotype was examined 7 days later. The 2-week-old seedlings were also subjected to high salt treatment by placing the pots in a tray containing 600 mM NaCl solution. After 10 days, the survival of plants was examined.

The experiments were repeated independently three times at least and the results were consistent. Results from one set of experiments are shown.

Measurement of leaf water loss, electrolyte leakage and proline content

For water loss analysis, ten fully expanded leaves at similar developmental stage (forth to seventh true rosette leaves) were detached from 4-week-old plants and put on Whatman 3MM paper at a constant temperature (22°C) and humidity (50%) for the indicated periods. The fresh weights of the leaves were measured using Sartorius Analysis Balance, and the percentage of fresh weight loss was calculated. Determination of electrolyte leakage was performed as described by Peng et al. (2010). Free proline content in plant was determined following the methods of Bate et al. (1973). The experiments were repeated independently three times and the results were consistent. Results from one set of experiments are shown.


Cloning and sequencing of the full-length cDNA of ZmbZIP72

By assembling the 5′ and 3′ RACE products, the coding region of ZmbZIP72 was deduced and the full-length cDNA was isolated from CN165 by RT-PCR. Sequence analysis revealed that ZmbZIP72 consisted of 894 bp open reading frame (ORF) encoding a polypeptide of 297 amino acids, 180 bp 5′-untranslated region (UTR), and 226 bp 3′-UTR. The deduced amino acids sequence of ZmbZIP72 had a calculated molecular weight of 32.4 kDa with a pI of 9.39. The deduced protein shared high amino acid sequence homology with rice OsbZIP72 (NP_001063362). Thus, the isolated gene was designated as ZmbZIP72 (HQ328839) for further study.

Structure and phylogenetic analysis of ZmbZIP72

The ZmbZIP72 protein contained a typical bZIP DNA-binding domain in its C-terminal region, and four conserved sequences disturbing in N- or C-region (Fig. 1a). Several protein kinase target sites, such as R/KXXS/T for CDPK, S/TXK/R for PKC, were involved in these conserved sequences (Furihata et al. 2006). To further analyze the phylogenetic relationship of ZmbZIP72 with the group A bZIP proteins from Arabidopsis and rice, a phylogenetic tree was constructed based on the amino acid sequences of these proteins (Fig. 1b).
Fig. 1

Sequence analysis of ZmbZIP72. a Comparison of amino acids of three bZIP proteins, i.e. Arabidopsis AREB1 and AREB2, rice OsbZIP72/OsAREB2 and OsbZIP66/TRAB1, and maize ZmbZIP72 (Accession number HQ328839). The black boxes indicate identical residues, and gray boxes indicate residues conserved among the three proteins. The conserved domain (i.e. basic-leucine zipper domain) is single-underlined, the nucleus localization signal (NLS) is double-underlined. The potential phosphorylated residues are represented by asterisk. b Phylogenetic relationship of ZmbZIP72 with Arabidopsis and rice group A bZIPs. Multiple sequence alignment was performed using ClustalX1.83, and the phylogenetic tree was constructed using the Neighbor-Joining method with amino acid sequences of each bZIP protein. The accession numbers of selected bZIPs are listed as follows: the Arabidopsis AtbZIP12/DPBF4, NP_565948; AtbZIP13, NP_199221; AtbZIP14, NP_195315; AtbZIP15, NP_199105; AtbZIP27, NP_179368; AtbZIP35/ABF1, NP_564551; AtbZIP36/ABF2/AREB1, NP_849777; AtbZIP37/ABF3, NP_849490; AtbZIP38/ABF4/AREB2, NP_566629; AtbZIP39/ABI5, NP_565840; AtbZIP40/GBF4, NP_171893; AtbZIP66/AREB3, NP_191244; AtbZIP67/DPBF2, NP_566870; the rice OsbZIP09, NP_001044598; OsbZIP10/OsABI5, NP_001044866; OsbZIP12, NP_001044913; OsbZIP23/OsAREB1, NP_001048225; OsbZIP29, ABF95681; OsbZIP40, NP_001055654; OsbZIP42, NP_001055893; OsbZIP46, NP_001057118; OsbZIP62, NP_001060681; OsbZIP66/TRAB1, NP_001062018; OsbZIP72/OsAREB2, NP_001063362

Genomic organization of ZmbZIP72

To analyze its genomic organization, genomic sequence of ZmbZIP72’s coding region was amplified from maize inbred line CN165. Sequencing results demonstrated that the genomic DNA of ZmbZIP72 was 2,164 bp (Fig. 2). Comparison of the genomic sequence with its corresponding cDNA sequence revealed that there were three introns, a fact which was similar to the distribution of introns in most of group A bZIP genes in Arabidopsis and rice. By searching for its genomic DNA from the database of MaizeGDB, ZmbZIP72 was mapped in bin 2.07 (BAC ID: AC190490.1) of the genetic linkage map IBM 2 near the SSR marker bnlg1329.
Fig. 2

Gene structure of ZmbZIP72. Filled boxes indicate exons and lines between boxes indicate introns

In addition, we cloned the promoter sequence of ZmbZIP72 (1,670 bp upstream from the base next to the start codon), and searched for the putative cis-acting element in the sequence using the program PlantCARE (Lescot et al. 2002). As a result, various stress-responsive cis-acting elements were identified in this 1.7 kb promoter region (see Suppl. Table S2) and a number of light-responsive cis-elements were also found (data not shown). These results suggested that ZmbZIP72 might be involved in response to various stresses and controlled by a complicated regulatory mechanism.

Expression patterns of ZmbZIP72

To speculate on the functional relevance of the ZmbZIP72 gene, we performed expression analysis under different abiotic stresses. As shown in Fig. 3a, the expression of ZmbZIP72 began to increase at 6 h after ABA treatment, followed by a reduction to the background level at 24 h. Under salt stress, the transcripts of ZmbZIP72 were strongly induced after 3 h of NaCl treatment, and continuously accumulated up to 24 h of treatment. When the seedlings were subjected to drought stress, a rapid accumulation of ZmbZIP72 was observed at 1 h after treatment and reached the highest level at 3 h, then decreased gradually. No significant changes of ZmbZIP72 were detected at all tested time points as compared with non-treated controls under cold stress.
Fig. 3

Expression patterns of ZmbZIP72 gene in maize. a Expression of ZmbZIP72 in seedlings at three-leaf stage under ABA, high-salinity, drought and cold treatments. The transcript level of ZmbZIP72 at 0 h (untreated control) was used as calibrator. b Expression of ZmbZIP72 and ZmDREB2A response to cold, salt and drought stress. The transcript level of ZmbZIP72 at 0 h (untreated control) was used as the calibrator. c Organ-specific expression of ZmbZIP72 in mature plants and in three leaf seedlings. The transcript level of ZmbZIP72 in the mature stalk was used as calibrator. The maize GAPDH gene (NM_001111943) was used as the internal control for normalization. Error bars indicate SD (n = 4)

To validate this experiment, we used the ZmDREB2A, a gene encoding a typical DREB2-type protein in maize, as a positive control. ZmDREB2A was reported to be induced by cold, salt, and drought stresses (Qin et al. 2007). Here, our data were consistent with previous studies (Fig. 3b).

Furthermore, we examined the tissue-specific expression of ZmbZIP72 in maize seedlings and mature plants. The ZmbZIP72 transcripts were detected in all organs tested, but the highest expression level was found in roots, both in young seedlings and mature plants, and the lowest in stalks (Fig. 3c).

Subcellular localization of ZmbZIP72 protein

A nuclear localization signal was found in the bZIP conservative domain of ZmbZIP72 (Fig. 1a). To examine the localization of the ZmbZIP72 protein, a ZmbZIP72-GFP fusion construct, driven by the CaMV 35S promoter, was introduced into onion (Allium cepa) epidermal cells by particle bombardment. The results showed that ZmbZIP72-GFP protein was detected only in the nucleus, while the control (transformation of GFP construct) was observed in the entire region of the cell (Fig. 4).
Fig. 4

Subcellular localization of ZmbZIP72 in onion epidermal cells. GFP and ZmbZIP72-GFP fusion protein were driven under the control of the CaMV 35S promoter. The onion epidermal cells were bombarded as described in the “Materials and methods”. Bar indicates 91 μm

Transactivation activity of ZmbZIP72

To check if ZmbZIP72 has transcriptional activity, the entire coding region was fused to the GAL4 DNA-binding domain in the vector pGBKT7 and the construct was transformed into yeast. The transformants containing pGBK-ZmbZIP72 grew well on the selection medium. Colony-lift filter assay supported these results (data not shown). These results indicated that ZmbZIP72 had transcriptional activity in yeast. Furthermore, several N-terminal regions of ZmbZIP72 with partial deletions and their complemented regions were constructed and were introduced into yeast. It was found that the amino acid sequence at the N terminus (amino acids 23–63) was necessary for the transactivation activity of ZmbZIP72 in yeast (Fig. 5).
Fig. 5

Transactivation analysis of ZmbZIP72. Fusion proteins of the GAL4 DNA-binding domain and different portion of ZmbZIP72 were checked for their transactivation activity in the yeast stain. Plus and minus indicate positive and negative, respectively, for the transactivation activity

Sensitivity of ZmbZIP72-overexpressed transgenic plants to ABA and osmotic stress

To investigate the in vivo function of ZmbZIP72 in plants, we generated transgenic Arabidopsis expressing the ZmbZIP72 cDNA under the control of the CaMV 35S promoter, and 15 homozygous transgenic lines were obtained. The expression level was checked by qRT-PCR analysis (see Suppl. Fig. S1), and two independent lines, TL6-1 and TL12-3, were selected for further analysis. To examine the possible phenotype of transgenic lines, T3 progeny of the ZmbZIP72-overexpressed lines and the wild type were grown in the greenhouse under well-watered conditions (16 h light/8 h dark, 22°C). Compared with WT plants, transgenic plants exhibited no retarded growth under normal conditions (see Suppl. Fig. S2).

To check whether overexpression of ZmbZIP72 affected the ABA sensitivity of the transgenic plants, the seeds from the transgenic plants and the WT plants were germinated on the medium containing different concentrations of ABA and the germination rates were compared. There was no difference between the WT and the transgenic plants when germinating under normal conditions (Fig. 6a). In the presence of exogenous ABA, the germination of both WT and ZmbZIP72-overexpressing seeds was inhibited significantly, but the inhibition of transgenic seeds was much more serious than that of WT (Fig. 6b, c). For example, at the third day after sown on MS agar medium containing 1.5 μM ABA, approximately 75% of the WT seeds germinated comparing with only 40% seeds of the transgenic lines (Fig. 6d).
Fig. 6

Germination assays of ZmbZIP72-overexpressed plants. a Germination assays in absence of ABA. b Germination assays in presence of 1.5 µM ABA. c Germination assays on MS medium with different concentrations of ABA. d Photograph of germination on MS medium with 1.5 µM ABA at day 3 after incubation at 22°C. e Germination assays in presence of 150 mM NaCl or 350 mM mannitol, respectively. Each measurement consisted of 50 seeds. Error bars indicate SD (n = 4). Significantly different values from WT at *P < 0.05 and **P < 0.01 level, respectively, by student’s t test

We also observed that the germination of ZmbZIP72-overexpressing seeds was more sensitive to osmotic stress than the WT (Fig. 6e). These results indicated that the ZmbZIP72-overexpressing transgenic plants were hypersensitive to ABA and osmotic stress at seed germination stage.

Performance of the ZmbZIP72-overexpressed transgenic plants under drought stress

The 14-day-old soil-grown plants were not watered for 2 weeks and then re-watered and grown under normal conditions for 7 days (Fig. 7a). When water was withheld for 7 days, phenotypic changes of these plants were observed. Leaf rolling and shrinking were substantially delayed in the ZmbZIP72-overexpressing plants compared with WT. When subjected to drought stress for 2 weeks, the WT plants showed severe wilting and chlorosis on the rosette leaves, whereas the transgenic plants were turgid and their leaves remained green. After 7 days of watering resumed, the transgenic plants showed a stronger growth recovery and higher survival rate than the WT plants (Fig. 7b). For example, more than 80% of ZmbZIP72-overexpressing plants recovered and grew normally; however, only about 12% of the WT plants survived. Furthermore, we verified that the ZmbZIP72-overexpressing plants exhibited lower electrolyte leakage and slower water loss in detached rosette leaves in comparison with WT under water-deficit stress (Fig. 7c, d). These results indicated that the overexpression of ZmbZIP72 in transgenic Arabidopsis could significantly increase drought tolerance.
Fig. 7

Performance of ZmbZIP72-overexpressed transgenic plants under drought stress. a Phenotypic comparison of the plants under drought stress. Transgenic plants and WT were grown for 2 weeks under normal conditions, withheld from water for 2 weeks, and then re-watered for 7 days. b Quantitative analysis of the survival rate of the plants as shown in a. Error bars indicate SD (n = 3). c Water loss in WT and transgenic plants. d Relative electrolyte leakage of transgenic plants and WT in response to drought stress. Error bars indicate SD (n = 3); significantly different from values of WT at *P < 0.05 and **P < 0.01 level, respectively, by student’s t test

Performance of the ZmbZIP72-overexpressed transgenic plants under salt stress

With the results that ZmbZIP72 expression was up-regulated by high-salinity treatment in maize, we investigated the effect of salt stress on the performance of ZmbZIP72-overexpressing transgenic plants. The 2-week-old soil-grown plants were subjected to the treatment with 300 mM NaCl solution. Following re-watering, the WT plants exhibited growth inhibition, whereas the transgenic plants thrived well (Fig. 8) and the transgenic line 12-3 (TL12-3) could even blossom (see Suppl. Fig. S4). The soil-grown seedlings were also treated with high salt (600 mM NaCl) for 10 days. However, the survival of all the transgenic plants showed no significant difference in comparison with WT plants under the stress (data not shown). These results indicated that ZmbZIP72 might play partial roles in salt tolerance.
Fig. 8

Performance of ZmbZIP72-overexpressed transgenic plants under salt stress. Phenotypic comparison of the plants under salt stress. Two-week-old seedlings grown in soil were treated with 300 mM NaCl solution for 2 weeks, and then recovered for 7 days

Effects on the expression of stress-responsive genes

To elucidate the regulatory function of ZmbZIP72, quantitative RT-PCR experiments were conducted to compare the expressions of six stress-responsive genes identified in regulated pathways between WT and the transgenic lines using gene-specific primers (see Suppl. Table S1). All six genes showed higher expression levels in both the WT and transgenic plants under dehydration treatment, but the transcript levels of three ABA-inducible genes (i.e. RD29B, RAB18 and HIS1-3) in the transgenic seedlings were significantly higher than that in WT (Table 1). For example, under dehydration treatment, the relative expression level of RD29B drastically increased, reaching 544.05 fold in the transgenic seedlings, but only 171.54 fold in WT, relative to the untreated control. Nevertheless, the expression of DREB2A and P5CS1 were slightly increased in the transgenic plants compared with those of WT.
Table 1

Relative expression levels of stress-responsive genes in response to dehydration treatment


Plant line

Dehydration treatmenta

0 h

6 h



1.00 ± 0.05

70.03 ± 6.22



1.15 ± 0.10

126.88 ± 7.17



1.00 ± 0.08

171.54 ± 11.05



0.84 ± 0.04

544.05 ± 18.24



1.00 ± 0.11

37.46 ± 2.13



1.35 ± 0.22

90.81 ± 7.62



1.00 ± 0.07

96.88 ± 8.43



1.58 ± 0.10

351.10 ± 12.27



1.00 ± 0.16

15.63 ± 0.93



1.05 ± 0.12

22.19 ± 1.79



1.00 ± 0.11

37.74 ± 4.18



1.19 ± 0.24

47.29 ± 3.09

a0 h and 6 h represent samples at 0 h and 6 h after dehydration treatment, respectively. Results are mean values ± SD (n = 4)


Although whole-genome sequencing of maize has made it possible to characterize the entire family of maize transcription factors (Schnable et al. 2009), only a small number of bZIP transcription factors have been found to possess apparent functions (Foley et al. 1993; Nieva et al. 2005). Especially, bZIP genes which are involved in abiotic stress signaling remain obscure. In the present study, we identified a novel bZIP gene from maize. Amino acid alignment demonstrated that ZmbZIP72, which contained highly conserved motif in the basic and hinge region, was similar to other bZIP proteins (i.e. TRAB1 and OsbZIP72) and was also classified as group A bZIP protein (Fig. 1b), suggesting their functional similarity. Earlier reports indicated that the ABA-dependent phosphorylation by protein kinases (such as SnRK2 s) were necessary for bZIPs activation (Lopez-Molina et al. 2001; Kagaya et al. 2002; Furihata et al. 2006; Yoshida et al. 2010). Here, we found several phosphorylatable serine residues in the conserved motifs (Fig. 1a). These residues were demonstrated to function as in vivo targets for kinases, allowing the fine-tuning of bZIP factor activity under stress conditions (Kirchler et al. 2010).

In previous studies, information of the published QTLs relevant to drought-tolerance was integrated and 79 Meta-QTLs were screened out (Li et al. 2010). According to the results, the region in bin 2.07 was predicted to contain several drought-tolerance genes. In the present study, ZmbZIP72 was mapped in bin 2.07 of the genetic linkage map IBM 2 based on the alignment of genomic sequence in maize genome databases. Our present results partially confirmed the former speculation.

We provided evidence that the nucleus-located ZmbZIP72 protein functioned as a transcriptional activator in yeast and its N-terminal region was required for the activation (Figs. 4, 5). These observations were consistent with previous reports (Fujita et al. 2005; Xiang et al. 2008; Zou et al. 2008; Lu et al. 2009). We further verified that ZmbZIP72 could activate the expression of several ABA-inducible genes in transgenic Arabidopsis (Table 1). Thus, these results indicated that ZmbZIP72 functioned as ABA-mediating transcription factor. Previously, it was reported that the bZIP-type transcription factors could form hetero- or homodimers to function cooperatively and required ABA-dependent modification for full activation in Arabidopsis and rice (Hobo et al. 1999; Uno et al. 2000; Kobayashi et al. 2004, 2005; Ehlert et al. 2006; Furihata et al. 2006; Liao et al. 2008a, b; Yoshida et al. 2010; Zhang et al. 2011). We speculated that ZmbZIP72 might undertake its transcriptional activity through the homodimerization in vivo, and further studies are needed to investigate the mechanism of these interactions.

Most of the group A bZIP genes (i.e. AtAREB2, OsABF1, and OsbZIP23) played important roles in ABA signal transduction (Uno et al. 2000; Kang et al. 2002; Xiang et al. 2008; Hossain et al. 2010). Here, we demonstrated that the expression of ZmbZIP72 was significantly induced by ABA, salt, and drought stress in maize (Fig. 3a), and the ZmbZIP72-overexpressing transgenic Arabidopsis exhibited increased sensitivity to ABA and osmotic stress at seed germination stage (Fig. 6), suggesting that the ZmbZIP72 protein might be involved in response to drought and salt stress in ABA-dependent manner.

Extensive previous studies showed that overexpression of bZIP transcription factor genes improved abiotic stress tolerance (Kim et al. 2004; Oh et al. 2005; Vanjildorj et al. 2005; Liao et al. 2008a, b; Zou et al. 2008; Hossain et al. 2010; Hsieh et al. 2010). In Arabidopsis, overexpression of ABF4/AREB2 resulted in enhanced drought tolerance (Kang et al. 2002). Overexpression of OsbZIP23 in rice increased plant tolerance to drought and salt stress (Xiang et al. 2008). Recently, it was reported that overexpression of a bZIP gene, isolated from Poncirus trifoliate, conferred dehydration and drought tolerance in transgenic tobacco through scavenging ROS and modulating expression of stress-responsive genes (Huang et al. 2010). In addition, constitutive expression of maize ABP9 in transgenic Arabidopsis remarkably enhanced the tolerance to abiotic stresses (Zhang et al. 2011). In the present study, we found that overexpression of ZmbZIP72 in Arabidopsis significantly improved drought and partial salt tolerance, which were supported not only by phenotypic performance but also by physiological changes, such as leaf water loss and electrolyte leakage (Figs. 7, 8). Our findings suggested that ZmbZIP72 might participate in regulation of stress tolerance.

In order to dissect the enhanced drought tolerance at the molecular level, expression of six stress-responsive genes were monitored before and after dehydration treatment. It was well known that RD29B, RAB18 and HIS1-3 could be induced by different stresses through an ABA-dependent pathway, and the higher expression of these genes improved the plant tolerance to multiple stresses (Ascenzi and Gantt 1997; Shinozaki and Yamaguchi-Shinozaki 1997; Zhu 2002; Fujita et al. 2005). Up-regulation of the above genes suggested that they might be activated by ZmbZIP72. It could perhaps explain the enhanced tolerance of the transgenic plants to drought stress. Moreover, we noticed that no obvious differences existed under normal conditions, suggesting that ZmbZIP72 might require post-translational modification for full activation (i.e. phosphorylation), like other transcription factors (Kagaya et al. 2002; Furihata et al. 2006; Nakashima et al. 2009; Yoshida et al. 2010). Extensive analysis will be necessary to investigate the particular regulatory mechanism. Our present results suggested that the ZmbZIP72 protein could function as a positive transcription activator in an ABA-dependent stress-response pathway.

Proline accumulation was beneficial for stress tolerance (Verbruggen and Hermans 2008). P5CS1 was the rate-limiting enzyme in the biosynthesis of proline and induced via ABA-dependent and -independent pathways under drought stress (Hu et al. 1992; Yoshiba et al. 1999). Here, the transcript level of P5CS1 exhibited minor change in the seedlings; however, the ZmbZIP72-overexpressing transgenic plants accumulated more free proline than that in WT (see Suppl. Fig. S3). These results indicated that the proline accumulation in transgenic plants was not directly dependent on the increased expression of P5CS1, but possibly indirectly through activation of other relevant genes.

Notably, the expression of DREB2A exhibited no significant differences between the transgenic plants and WT, suggesting that ZmbZIP72 had no impact on the expression of DREB2A. Because of the increased expression of RD29A in the transgenic plants, we speculated that ZmbZIP72 might co-operate with DREB2A in ABA-dependent gene expression, as suggested by Narusaka et al. (2003).

Recently, Zhu and his colleagues successfully identified all essential core components of an ABA response pathway from hormone perception to gene function in Arabidopsis and attested the signal transduction through the RCAR/PYR-PP2C-SnRK2 complex in vitro (Fujii et al. 2009). They pointed out that the SnRK2 s and bZIPs functioned as critical mediators in regulating the ABA-responsive gene expression. In a previous study, we identified a SnRK2 protein from maize and verified its phosphorylatory ability in vitro (Ying et al. 2011). However, no positive interaction was observed between ZmSAPK8 and ZmbZIP72 (data not shown). In maize, there were at least 11 SnRK2 genes identified (Huai et al. 2008), which could potentially interact with ZmbZIP72. The exact regulation pattern in maize needs to be further investigated.

Since transcription factors act as master regulators that coordinate the expression of stress-responsive genes, the TF-based technologies are considered to be a prominent part of the next generation of successful biotechnology crops (Century et al. 2008). However, several reports pointed out that constitutive overexpression of transcription factor genes frequently results in unwanted phenotypes, such as growth retardation (Kasuga et al. 1999; Kim et al. 2004; Dai et al. 2007). Here, ZmbZIP72-overexpressing transgenic plants exhibited no obvious adverse effects on growth and development, such as plant height and plant phenotype (see Suppl. Fig. S2), compared with WT, indicating the potential application of this gene in crops for stress tolerance improvements.

In conclusion, we cloned and characterized the ZmbZIP72 gene from maize. Overexpression of ZmbZIP72 resulted in enhanced tolerance to drought and salt stress. Although the detailed mechanism of ZmbZIP72 function in response to abiotic stress is not clear, our current data provides valuable information for molecular breeding that could lead to improved stress tolerance in crops. At present, the work of transforming ZmbZIP72 into maize is in progress in our laboratory.


We are grateful to Dr. Feng Qin (Institute of Botany, CAS) for critical reading on the manuscript. This work was partly supported by grants provided by the Ministry of Science and Technology of China (2009CB118401, 2011CB100100) and China Natural Science Foundation (30730063).

Supplementary material

425_2011_1496_MOESM1_ESM.doc (24 kb)
Expression analysis of ZmbZIP72 in transgenic lines. qRT-PCR analysis of ZmbZIP72 expression in WT and transgenic lines TL5-2, TL6-1, TL11-3 and TL12-3. The transcript level of ZmbZIP72 in TL11-3 was used as the calibrator. The Arabidopsis actin gene (NM_112764) was used as the internal control for normalization. Error bars indicate SD (n = 4) (DOC 23 kb)
425_2011_1496_MOESM2_ESM.doc (138 kb)
Phenotype of ZmbZIP72-overexpressed transgenic Arabidopsis plants under normal growth conditions. a Morphological comparisons of 2-week-old plants of WT and ZmbZIP72-overexpressed transgenic plants under normal growth conditions. b The plant height of WT and ZmbZIP72-overexpressed transgenic plants as described in a. Error bars indicate SD (n = 8) (DOC 137 kb)
425_2011_1496_MOESM3_ESM.doc (49 kb)
Free proline content in plants. Relative content of proline in 2-week-old transgenic seedlings. Error bars indicate SD (n = 6). * indicates significant difference from WT at P < 0.05 by student’s t-test (DOC 49 kb)
425_2011_1496_MOESM4_ESM.doc (130 kb)
Salt-stress tolerance of WT and ZmbZIP72-overexpressed Arabidopsis seedlings. Two-week-old seedlings grown in soil were treated with 300 mM NaCl solution for 2 weeks, and then recovered for 7 days (DOC 130 kb)
425_2011_1496_MOESM5_ESM.doc (58 kb)
Supplementary material 5 (DOC 57 kb)
425_2011_1496_MOESM6_ESM.doc (30 kb)
Supplementary material 6 (DOC 29 kb)

Copyright information

© Springer-Verlag 2011