Function analysis of ZmZHD9, a positive regulator in drought stress response in transgenic maize


 Backgrounds: Drought stress is one of the major factors that affects maize yield. ZF-HD transcription factors have been proved to play pivotal roles in the regulation of plant growth, hormone conduction signaling and abiotic stress response. However, the molecular mechanism of ZmZHD9-mediated drought tolerance is not well understood. Results: In the present study, we analyzed the functions of ZmZHD9, a member of the maize ZF-HD family. ZmZHD9 is predominantly expressed in leaves, and was induced by drought, salinity, high temperature and abscisic acid (ABA). Subcellular localization indicated that ZmZHD9 protein was localized in the nucleus. ZmZHD9-overexpressing plants showed increased tolerance to drought stress compared with wild-type plants, evaluated by higher RWC and proline content, higher SOD and POD activity, lower REL and MDA content in transgenic plants under drought stress. In addition, the expression of six stress-responsive genes were significantly higher in ZmZHD9 transgenic plants than that in wild-type plants under drought stress.Conclusion: These results demonstrate that ZmZHD9 as a stress-responsive transcription factor which plays a positive regulatory role in response to drought.


Background
Transcription factors (TFs) are proteins that regulate the transcription of genes by binding speci c regions within the promoter [1,2], and play key roles in gene expression regulation in response to external or internal stimuli [3,4]. Zinc nger homeodomain (ZF-HD) proteins are homeodomain (HD) proteins which also contain conserved zinc nger domains at N-terminus regions and consist of two zinc nger motifs, CH2C and C3H2 [5]. Phylogenetic analyses showed that ZF-HD proteins were divided into two subfamilies, ZHD and MIF (Mini Zinc nger), and MIF genes only possess zinc nger structure but lacking HD domains at C-terminal, which similar with ZHD proteins without introns [6]. Since the rst ZF-HD TF was identi ed in the C4 plant Flaveria bidentis [7], a series of studies in plants have focused on ZF-HD TFs, including Arabidopsis thaliana [8], Oryza sativa L. [9], Solanum lycopersicum [10], Fagopyrum tataricum [11] and Glycine max [12].
ZF-HD genes play pivotal roles including the regulation of plant growth and development, hormone conduction and various environmental stresses [12,13]. Arabidopsis ZF-HD gene ZFHD1 was induced by drought, high salt, and ABA stress treatments, and its overexpression can enhance series of stressinduced gene expression and drought tolerance [14]. Arabidopsis ZF-HD protein AtHB33 was negatively regulated by ARF2 and involved in ABA response signal pathway. Over-expression AtHB33 transgenic plants were sensitive to ABA, while transgenic plants reducing AtHB33 by RNAi were more resistant to ABA [15]. Liu et al reported that overexpression of TsHD1 can improve the heat stress resistance of Thellungiella halophila and retarded its vegetative growth slightly. The co-overexpression of TsHD1 and TsNAC1 highly improved heat and drought stress resistance by increasing the accumulation of heat shock proteins and enhancing the expression levels of drought stress response genes [16].
Although ZF-HD TFs have been identi ed in many species, less information regarding the ZF-HD genes in maize is available. As one of the most widely cultivated crops, maize frequently suffers from drought stress, especially in Huanghuai region, so screening new genes resistant to drought stress is of great practical signi cance. We carried out the transcriptome analysis using a drought-resistant maize line Yu882 as experimental material under drought stress and rewatering treatment, and identi ed many TFs and functional protein families. In this study, In this study, we have analyzed the expression of the ZmZHD9 gene in different tissues of maize as well as under different abiotic stress. The subcellular localization and transactivation activity of ZmZHD9 protein were also determined. The ZmZHD9overexpression transgenic maize plants were generated to analyze the function in response to drought stress. This study would provide a candidate gene in drought resistance for molecular breeding.

Results
Cloning and Bioinformatics Analysis of ZmZHD9 Gene ZmZHD9 gene was cloned from Y882 and sequencing analysis showed that the open reading frame was 303 bp encoding a putative protein of 100 amino acids residues of 10.37 kDa with an isoelectric point of 8.93. Phylogenetic relationship analysis of ZmZHD9 with 17 Arabidopsis ZF-HD TFs and 14 Oryza sativa L. ZF-HD TFs showed that ZmZHD9 belongs subfamily of MIF (Mini Zinc nger). ZmZHD9 was clustered into group with OsMIF1, OsMIF3, AtMIF2 and AtMIF2 (Fig. 1A). Multiple alignments showed that ZmZHD9 proteins contained a Zinc nger domain at its N terminus (Fig. 1B), which suggested that ZmZHD9 was a member of subgroup MIF in maize.

Subcellular Localization and Transcriptional Activation Activity Analysis of ZmZHD9 protein
To detect the subcellular localization of the ZmZHD9 protein, the ORF of ZmZHD9 gene without the termination codon was cloned and the expression vector for pMDC83-ZmZHD9-GFP fusion protein was constructed. The recombinant vector pMDC83-ZmZHD9-GFP and pMDC83-GFP were introduced into Nicotiana benthamiana leaves and observed under a laser scanning confocal microscopy. As shown in Fig. 2, the pMDC83-ZmZHD9-GFP fusion protein was exclusively detected in the nucleus, whereas Green uorescent protein (GFP) control distributed evenly in the nucleus and the cytoplasm, indicating that ZmZHD9 gene encoded a nuclear localization protein.
To detect the transcriptional function of the ZmZHD9 as a transcription factor in maize, we performed the yeast two-hybrid procedure to evaluate its transactivation activity. The complete ORF region of ZmZHD9 was fused to the GAL4 DNA-binding domain in the pGBKT7 vector and transformed into Yeast strain AH109. As shown in Fig. 3, all transgenic yeasts grew well on SD/-Trp medium. The yeast cells harboring the pGBKT7-ZmZHD9 plasmid grew as well as the positive control and exhibited positive for α-galactosidase activity on SD/-Trp/-His/-Ade/X-α-gal medium, suggesting that the ZmZHD9 protein had transactivation activity to act as a transcriptional activator.

Stress-Related Regulatory Elements analysis in the Promoter of ZmZHD9
Cis-elements in promoter regions of genes always play crucial roles in stress responses. Cis-elements analysis showed that there were numerous stress-related regulatory elements such as TATA-box, ARE, low-temperature responsive (LTR), dehydration and salicylic acid (SA) and W box in the promoter region of 2000 bp upstream of the ATG start codon, indicating that ZmZHD9 was involved in plant hormonerelated signal transduction and response to abiotic stress (Table 1). Tissue-speci c expression was performed by qRT-PCR and the result indicated that ZmZHD9 showed high expression in leaf, root and ear, with weak expression detected in stem, and marginally expression observed in tassel (Fig. 4A). In addition, the expression patterns of ZmZHD9 of leaves under various treatments, including drought, salt, high temperature and ABA were also investigated. Under drought treatment, the ZmZHD9 transcript level was pronouncedly induced until reaching the highest level at 4 h, which greater than 2.6-fold of the initial level and followed by decreased at the last time point (Fig. 4B). ZmZHD9 was also induced by salt treatment and reached the highest level by 1.8-fold at 24 h after treatment (Fig. 4C). After treatment with high temperature, the expression of ZmZHD9 gradually decreased and reached the lowest level at 48 h, then sharply increased by 9-fold at 60 h than that 48 h (Fig. 4D). The expression of ZmZHD9 was also increased by exogenous ABA treatment and reached the highest level at 12 h (Fig. 4E). These results indicate that ZmZHD9 is highly expressed in leaves and that its expression is induced by abiotic stress, such as drought, high salinity and ABA.

Overexpression of ZmZHD9 Enhances Drought Tolerance in Transgenic Maize Plants
To investigate the function of ZmZHD9, we performed Agrobacterium-mediated transformation and obtained single-copy homozygous T 3 transgenic lines. Finally, three independent ZmZHD9-OE transgenic lines (OE8, OE13, and OE17) were selected for subsequent experiments. qRT-PCR analysis showed that the expression levels of ZmZHD9 were higher in three transgenic plants than that in control plants (WT), respectively, indicating that the transgenic plants were successfully generated as designed (Fig. 5A).
Under normal conditions, there was no visible morphological differences in phenotype between 3 transgenic lines (OE8, OE13, OE17) and control (WT), while after natural drought stress for 10 days, the leaves of WT plants appeared severe dehydration and wilting symptoms, whereas the leaves of ZmZHD9-OE transgenic plants had signi cantly delayed leaf rolling (Fig. 5B). the leaf relatively water content (RWC) and proline of three transgenic lines exhibited higher levels than that of WT plants ( Fig. 5C and   5D), respectively. Moreover, the relative electrolyte leakage levels (REL) and malondialdehyde (MDA) contents of ZmZHD9-OE plants were lower than WT plants ( Fig. 5E and 5F). Compared to the WT plants, antioxidant reductases activity of SOD and POD were signi cantly higher in ZmZHD9-OE plants ( Fig. 5G and 5H). Taken together, the results indicated that the overexpression of ZmZHD9 can regulate physiological processes that increase tolerance of transgenic maize plants to drought stress.

Expression Patterns Analysis of ZmZHD9 in Transgenic Plants under Different Abiotic Stress
The expression patterns of ZmZHD9 gene under abiotic stress in transgenic plants were performed by qRT-PCR. Under drought and salt stress, the expression of ZmZHD9 was upregulated in transgenic lines that was increased 2.2-fold and 3.3-fold as compared to those in the wild-type plants, respectively ( Fig. 6A, 6B). In contrast, ZmZHD9 was inhibited with much lower expression level under high temperature stress (Fig. 6C). Under ABA treatment, ZmZHD9 was induced, with 2.7-fold increased expression level than that in wild-type (Fig. 6D). These results are consistent with the expression of ZmZHD9 gene in nontransgenic plants, indicated that ZmZHD9 play important role in response to abiotic stress.

Altered Expression of Drought Stress Responsive Genes in Transgenic Plants
To further investigate the pathway regulated by ZmZHD9 under drought stress, transcript levels of six stress-responsive genes were veri ed in WT and transgenic maize plants under normal and drought stress. The transcript levels of ZmPC5S1, ZmABI2, ZmED22, ZmSNC1, ZmDRE2A, and ZmLEA3 had no signi cant difference between WT and OE plants under normal conditions. Under drought stress, the expression level of ZmDREB2A were upregulated in ZmZHD9 transgenic lines that was increased over 3fold as compared in the wild-type plants (Fig. 7E). ZmP5CS1, ZmSAC1, and ZmLEA3 were induced by ZmZHD9 in transgenic plants with about 2.5-fold increased expression level in comparison to that of the WT plants (Fig. 7A, 7D, and 7F). Among all the detected genes, ZmABI2 and ZmRD22 were upregulated in ZmZHD9 transgenic lines by less than 2-fold as compared to those in wild-type plants (Fig. 7B, 7C). All these results suggest that ZmZHD9 may enhance maize resistance to drought through activate the expression of stress-responsive genes.

Discussion
ZF-HD genes play important role in plant growth, development and abiotic stress response. Signi cant progress has been made in identifying and characterizing ZF-HD genes in a number of plant species, but only a few maize ZF-HD TF genes were reported. To further study the function of the ZF-HD TFs in maize, we isolated and characterized ZmZHD9 gene from the former transcriptome result and systemic analyses was conducted in this study. Structural analysis showed that ZmZHD9 proteins only contained a Zinc nger domain at its N terminus, belonging to subgroup MIF. Tissue expression pattern analysis showed that ZmZHD9 was highly expressed in leaves, and responsive to drought, high salinity, heat and ABA treatment (Fig. 2B-2E). Cis-elements are key molecular switches involved in the transcriptional control of dynamic networks of stress-induced gene and play crucial role in response to abiotic stress, such as ABRE, ARE, DRE, LTR, W box and TATA-box [17][18][19][20]. In our study, ABRE, ARE, DRE, LTR and W box were detected in the ZmZHD9 promoter region (Table 1). DREs (dehydration-responsive element) are wellknown speci c cis-elements that are regulated by ABA-independent drought-induced OsDREB2 transcription factors [21,22]. Thus, the drought-induced expression of ZmZHD9 could be regulated in an ABA-independent manner.  [4,23]. In this study, MDA content was lower from ZmZHD9-OE transgenic lines than wildtype plants under drought stress, indicating that ZmZHD9-OE can improve tolerance to oxidative stress caused by drought (Fig. 6F). And the REL exhibited the same result, keep a lower level in ZmZHD9-OE transgenic plants than in WT (Fig. 6E), illustrating that drought stress have less membrane damage to transgenic plants than wild-type plants. Previous studies have proved that proline content increased could enhance the concentration of cell protoplasm to maintain normal membrane function under abiotic stresses, which could be improved tolerance to environment stress [24,25]. In the present study, the proline content in ZmZHD9-OE transgenic lines increased, and the increasing rate were signi cantly higher than wild-type plants under drought stress, the same with Tang's report [26].
Drought and salt stresses often lead to oxidative stress, reactive oxygen species (ROS) accumulation, leading to protein structure damage, membrane peroxidation [4,[27][28][29]. Scavenging ROS can reduce oxidative damage and improve the plants tolerance to abiotic stresses [30]. Previous study showed that under salt and drought stresses, the ROS scavenging enzyme genes such as SOD, glutathione peroxidase (GPX), catalase (CAT), ascorbate peroxidase (APX), and DHAR were systematically up-regulated in the overexpressing transgenic plants [31][32][33]. In the current study, the activities of SOD and POD in the ZmZHD9-OE lines were higher than in the wild-type plants under drought stress (Fig. 6G, 6H), which contributed to the increased drought tolerance of transgenic maize plants. The results suggesting that ZmZHD9 gene might protect the cell me membrane integrity by regulating the cellular levels of ROS to under drought stress.
Many reports have revealed that gene-overexpression in transgenic plants lead to induced expression of stress-responsive genes, which in turn leads to enhance tolerance to various stress [34,35]. For example, P5CS1 encodes a rate-limiting enzyme, which is upregulated at transcriptional level and necessary for proline accumulation under drought stress treatment [36]. The overexpression of OsP5CS in rice can promote the accumulation of proline and increase the resistance to abiotic stress [37]. Furthermore, DREB proteins belong to AP2/ERF TF family, which were indeed implicated responses to drought stress in plants [38]. In Arabidopsis, DREB1A and DREB2A speci cally interact with dehydration-responsive element (C-repeat) involved in drought stress-responsive gene expression [21]. Subsequent studies showed that ZmLEA3, ZmSNAC1 and RD22 are stress-inducible genes that improve resistance to abiotic stress in transgenic plants [39][40][41]. Similarly, in our study, the expression of stress-responsive genes including ZmP5CS1, ZmABI2, ZmED22, ZmSNC1, ZmDRE2A and ZmLEA3 were higher in transgenic lines than in wild-type plants under drought stress (Fig. 8). However, the expression of these genes were no signi cant differences under normal condition, despite the fact that constitutive promoter was to drive the gene expression. One possible to explanation is that other stress-responsive regulators are required to activate ZmZHD9-dependent stress responsive gene under drought stress. A similar observation has been reported in rice OsMYB6 transgenic lines [26]. Taken together, ZmZHD9 overexpression transgenic plants can enhance drought tolerance possible due to the reinforced expression of these stress-responsive genes.

Conclusions
In this study, we isolated and characterized the maize ZmZHD9 gene, phylogenetic tree and sequence analyses con rmed that ZmZHD9 belongs to MIF subgroup which contain Zinc nger domain at its N terminus. The promoter region of ZmZHD9 gene contains multiple core elements responsive to abiotic stresses and hormones. qRT-PCR results showed that ZmZHD9 genes was up-regulated in response to PEG, NaCL and ABA treatment. Subcellular localization assay showed that ZmZHD9 protein is localized in the nucleus. Over-expression of ZmZHD9 in transgenic maize remarkably improved drought resistance. Moreover, over-expression of ZmZHD9 enhanced the expression of stress-responsive genes indicating that ZmZHD9 may as a stress-responsive transcription factor which plays a positive regulatory role in response to drought. These results suggested that ZmZHD9 could act as a potential candidate gene for genetic engineering to improve drought and other abiotic stress.

Plant materials, Growth Condition and Abiotic Stress
Maize (Zea mays L. Yu882), tobacco (Nicotiana benthamiana) were used in this study. All seeds were provided by laboratory of Professor Li xia Ku of Henan Agricultural University. Seeds of Yu882 were selected and sown in soil and vermiculite mixture (3:1) in a growth chamber at 25 ± 2℃ under a long-day conditions, with 16 h/8h (light/dark) photoperiod cycles, 70% relative humidity and a light density of approximately 300 µmol/ (m − 2 ·s − 1 ). The nutrient solution was replaced every 2 days. When the seedlings have 3-fully expanded leaves, they were transferred into Hoagland nutrient solution with drought stress (20% PEG6000), salinity (200 mmol/L NaCL), ABA (5 µmol/L) and heat stress (37 °C) treatments. The second fully expanded leaves were sampled at 0, 4, 12, 24, 48 and 60 h were collected, frozen in liquid nitrogen immediately and stored at -80℃ for further expression analysis. Three plants from different treatment were used as biological replicates. Gene Cloning and Sequence Analysis Total RNA was extracted from leaves using Trizol reagent (TaKaRa, Dalian, China) following the manufacturer's instructions. RNA integrity and purity were analyzed by spectrophotometry and 1% agarose gel electrophoresis. 1 µg RNA was took for rst-strand cDNA synthesis using the Prime Script™ RT reagent Kit (TaKaRa, Dalian, China) according to the manufacturer's protocol. Based on the sequence of ZmZHD9, the RT-PCR primers (Additional le 1: Table S1) were designed to amplify the full length of ZmZHD9 gene. The PCR product was cloned into the pMD19-T vector, and sequenced to con rm the accuracy.
The nucleotide and amino acid sequence of ZmZHD9 were used to search its homologous genes and proteins by using the database of National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/). DNAMAN software (Version 5.2.2.0; Lynnon Biosoft, USA) was used to alignment the amino acid sequences of ZmZHD9 and its homologs. A phylogenic analysis of ZF-HD proteins from maize and other species were performed by MEGA 6.0 by the neighbor-joining method [42]. The online database Expasy (http://web.expasy.org/protparam/) was used to predicted molecular weight (MW) and isoelectric point (PI) of ZmZHD9 protein [43]. The online website: PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) was used to predict Cis-acting elements that respond to abiotic stresses in the promoter region (the 5 upstream regions 2000 bp) [44] . Expression Pro le Analysis by qRT-PCR qRT-PCR analyses were performed according TB Green PremixEx Taq™ II (TaKaRa, Dalian, China) manufacturer's instructions with following program: 95 °C for 30 s, 40 cycles of 95 °C for 5 s, 60 °C for 30 s and 72 °C for 30 s. Quantitative RT-PCR was carried out with a CFX96 system (Bio-Rad, CA, USA).
The 18S (GenBank No. AF168884.1) gene was used as an internal control. The method of 2 −∆∆CT was used to calculate the relative expression level [45]. All qRT-PCR experiments contained three biological replicates. And all primers were list in Additional le 1: Table S1.

Subcellular Localization of ZmZHD9 Protein
The coding region of ZmZHD9 was ampli ed by PCR using primers containing Asc I and Spe I site (Additional le 1: Table S1) and the PCR products were digested with Asc I and Spe I and then inserted into the same enzymes digested pMDC83-GFP to product a fusion protein (pMDC83-ZmZHD9-GFP). The fusion vector was introduced into EHA105 and transiently expressed in leaves of N. benthamiana (1month-old) according in ltration method [46]. The infected tobacco was cultured for 24 h in the dark and then transferred to a light incubator. The GFP-associated uorescence of fusion constructs was detected after 48 h using a uorescence microscopy (LSM 700, Carl Zeiss, and Jena, Germany).

Transactivation Activity of ZmZHD9 in Yeast
The coding region of ZmZHD9 were ampli ed from cDNA templates from leaves sample using a pair of gene-speci c primers (Additional le 1: Table S1) and then ligated into EcoR I and BanH I-digested pGBKT7 vector. The vector pGBKT7-ZmZHD9, pGADT7 empty vector (negative control) and pGBKT7-53 vector (positive control) were transformed into Yeast strain AH109 according the lithium acetatemediated method, respectively. The transformed Yeast cells were examined on the solid medium plates of SD/-Trp and SD/-Trp/-Ade/-His/X-α-gal (5-Bromo-4-chloro-3-indolyl-α -D-galactoside) at 30℃ for 3-5d.

Generation of Transgenic Maize of ZmZHD9 Gene
The CDS of ZmZHD9 was ampli ed using the primers containing the Asc I and BamH I sites (Additional le 1: Table S1). The PCR product and the expression vector pFGC5941 were double digested with Asc I and BamH I, and the vector pFGC5941-ZmZHD9 was obtained with T4 DNA ligase. The expression vector was transferred into the Agrobacterium tumefaciens strain EHA105 for maize transformation by using Agrobacterium-mediated transformation method [47]. Drought stress Tolerance Analysis of ZmZHD9-overexpressing Transgenic Maize Three transgenic line (OE8, OE13 and OE17) and wild-type (WT) plants were grown in pots containing soil and vermiculite mixture (3:1) in greenhouse. When seedlings were at the three-leaf stage, transgenic lines and WT plants stopped watering to practice natural drought stress for 10 days. The control groups were watered normally. Leaves from transgenic and WT plants were collected for qRT-PCR and physiological traits analysis. Drought stress treatment contained three biological replicates.
For other abiotic stress, the seedings of 3-fully expanding leaves were transferred into Hoagland nutrient solution with drought stress (20% PEG6000), salinity (200 mmol/L NaCL), ABA (5 µmol/L) and heat stress (37 °C) treatments. The second fully expanded leaves were sampled at 24 h were collected, frozen in liquid nitrogen immediately and stored at -80℃ for further expression analysis. Three plants from different treatment were used as biological replicates.

Physiological and Biochemical Analysis of Transgenic Maize
Leaves fresh weight (FW), dry weight (DW) and saturated weight (SW) were measured to calculate the relative water content (RWC) based on the formula: RWC = [(FW − DW)/ (SW − DW)] × 100%. The relative electrolyte leakage (REL) was measured referring to the method described by with some modi cations [48]. Proline content was detected as described method by Zhao [49]. The malondialdehyde (MDA) content was detected according to the method reported by Xia [50]. The superoxide dismutase (SOD) and peroxidase (POD) activities were estimated according to described method by Zhang [51]. All of the measurements have three replicates.

Statistical Analysis
All values reported in this study were the means of three independent replicate measurements. Statistical signi cance of the differences was analyzed by SAS software using Duncan's multiple-range test with a signi cance level of 0.05 (P < 0.05).

Consent for publication
Not applicable.

Availability of data and materials
The dataset supporting the conclusions of this article is included within the article and its additional les.

Competing interests
The authors declare that they have no competing interests.     The experiment included three biological replicates. Data represent the mean value ± standard deviation (SD) (n=3). Signi cant differences from the WT control are indicated by asterisks (Student′s t-teat, *p≤0.05;**p≤0.01) Figure 7 The expression of stress-responsive genes in the WT and transgenic plants under normal (CK) and drought stress (Drought) condition. The experiment included three biological replicates. Data represent