ZmWRKY104 Transcription Factor Phosphorylated by ZmMPK6 Functioning in ABA-Induced Antioxidant Defense and Enhance Drought Tolerance in Maize

Simple Summary Current knowledge about the downstream substrate proteins of MAPKs is still limited. Our study screened a new WRKY IIa transcription factor as the substrate protein of ZmMPK6, and its phosphorylation at Thr-59 is critical to the role of ZmWRKY104 in ABA-induced antioxidant defense. Moreover, overexpression ZmWRKY104 in maize enhances the drought tolerance of transgenic plants. These findings define a mechanism for the function of ZmWRKY104 phosphorylated by ZmMPK6 in ABA-induced antioxidant defense and drought tolerance. Abstract Mitogen-activated protein kinase (MAPK) cascades are primary signaling pathways involved in various signaling pathways triggered by abiotic and biotic stresses in plants. The downstream substrate proteins of MAPKs in maize, however, are still limited. Here, we screened a WRKY IIa transcription factor (TF) in maize (Zea mays L.), ZmWRKY104, and found that it is a substrate of ZmMPK6. ZmWRKY104 physically interacts with ZmMPK6 in vitro and in vivo. Liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis results showed that threonine-59 (Thr-59, T59) was the major phosphorylation site of ZmWRKY104 by ZmMPK6. Subcellular localization analysis suggested that ZmWRKY104 acts in the nucleus and that ZmMPK6 acts in the nucleus and cytoplasmic membrane in the cytosol. Functional analysis revealed that the role of ZmWRKY104 in ABA-induced antioxidant defense depends on ZmMPK6. Moreover, overexpression of ZmWRKY104 in maize can enhance drought tolerance and relieve drought-induced oxidative damage in transgenic lines. The above results help define the mechanism of the function of ZmWRKY104 phosphorylated by ZmMPK6 in ABA-induced antioxidant defense and drought tolerance in maize.


Introduction
Abiotic stresses such as drought, salinity, oxidative stress, and temperature variations change the productivity of major crops and constitute the main cause of global crop losses [1,2]. Maize (Zea mays L.) is a major cereal crop species worldwide that is severely impacted by various abiotic stresses during its growth and productivity. Abiotic stresses also activate the emergence of reactive oxygen species (ROS) in plant cells. Abscisic acid (ABA) serves as a key endogenous messenger in the abiotic stress responses of plants [3], and can induce antioxidant defense-related enzymes such as ascorbate peroxidase (APX), superoxide dismutase (SOD), glutathione peroxidase (GPX) and catalase (CAT) [4]. Many signal molecules, such as calcium ion, nitric oxide (NO), and protein kinases, such as calcium-dependent protein kinase (CDPK), calcium/calmodulin-dependent protein kinase the Maize GDB (https://maizegdb.org/, accessed on 1 May 2020). Conserved domain analysis revealed that ZmWRKY104 has a highly conserved WRKYGQK motif and C2H2 (C-X5-C-X23-H-X1-H) zinc finger motif ( Figure S1A). Multiple alignment shows that WRKY was highly conserved among A. thaliana, O. sativa, and Z. mays ( Figure S1B). Phylogenetic analysis showed that ZmWRKY104 belongs to the WRKY IIa subfamily with a single WRKY domain, and is highly homogeneous to OsWRKY62 ( Figure S2) [37]. In a previous study, we screened the possible interacting proteins of ZmWRKY104 and identified a serine/threonine protein kinase-ZmMPK6. The full-length cDNA sequence of ZmMPK6 (NP_001105238) was obtained from GenBank (https://www.ncbi.nlm.nih.gov/genbank/, accessed on 1 May 2020), which encodes a polypeptide of 557 amino acid residues and a potential S_TKc. Multiple alignment revealed that MAPK was greatly conserved among O. sativa, A. thaliana, G. hirsutum, T. aestivum, and Z. mays, and a phylogenetic analysis showed that ZmMPK6 was highly homogeneous to OsMPK15 ( Figure S3B). ZmMPK6 belongs to Group D of plant MAPKs due to the presence of the T-D-Y activation motif in the T-loop via an extended C-terminal region ( Figure S3A) [19].

ZmWRKY104 Is a Positive Regulator in ABA-Induced Antioxidant Defense
We first assessed the expression patterns of ZmWRKY104 in different tissues of maize. We isolated total RNA from different tissues, including roots, stems, leaves, pollen, and pistil. The results of qRT-PCR showed that ZmWRKY104 were expressed in all tested tissues, and the expression was the highest in leaves ( Figure S4). The expression of ZmWRKY104 increased at 15 min and reached a maximum at 90 min, with a sevenfold change, and maintained for up to 240 min after treatment in maize leaves, which suggested an association of this TF with the response to ABA ( Figure 1A). To investigate the part of ZmWRKY104 in ABA-induced antioxidant defense, we determined the effect of the transient expression or silencing of ZmWRKY104 in maize mesophyll protoplasts [38,39] on the activities of primary enzymes of APX and SOD in antioxidant defense. Firstly, we detected the efficiency of transiently overexpressing ZmWRKY104, silencing ZmWRKY104, and srdxZmWRKY104 in maize protoplasts. The results showed that compared with the control, the gene expression level of ZmWRKY104 was increased by about 3.5-fold after transient transformation ( Figure 1B). After transient transformation of ZmWRKY104-RNAi ( Figure 1C) and srdxZmWRKY104 ( Figure 1B), compared with the control, the gene expression level decreased by about 50%, indicating that the transient transformation of protoplasts can achieve the expected effect. As shown in Figure 1D,E, the activities of APX and SOD after transient expression of ZmWRKY104 were significantly higher than those in the control, and ABA treatment further enhanced these increased activities. On the contrary, transient silencing of ZmWRKY104 significantly reduced its activities compared with the control, and ABA treatment could not return these decreased activities to baseline levels ( Figure 1F,G). WRKY TFs have functional redundancy [40]. To validate the function of ZmWRKY104, chimeric repressor silencing technology was used [7,24,25]; we fused ZmWRKY104-SRDX-mCherry to the exogenous EAR motif repression domain SRDX and transformed the product (srdxZmWRKY104) into maize protoplasts. Similarly, transient expression of srdxZmWRKY104 also resulted in significant decreases in the activities of APX and SOD, and ABA treatment could not return their activities to the control levels ( Figure 1H,I). In summary, all the above-described results provide clear facts that ZmWRKY104 is a positive regulator in ABA-induced antioxidant defense. ZmWRKY104 and srdxZmWRKY104 in maize protoplasts; (C) detection of the efficiency of transiently silencing ZmWRKY104 in maize protoplasts. Gene expression analysis of ZmWRKY104 by RT−qPCR, with ZmActin as an internal control. Activities of APX (D) and SOD (E) of transiently expressing ZmWRKY104 in protoplasts. ZmWRKY104 (ubi: ZmWRKY104−mCherry) or Control (empty vector) were transfected into protoplasts. Activities of APX (F) and SOD (G) of transiently silencing ZmWRKY104 in protoplasts. ZmWRKY104−RNAi or Control (distilled water) were transfected into protoplasts. Activities of APX (H) and SOD (I) of transiently expressing srdxZmWRKY104 in protoplasts. srdxZmWRKY104 (ubi: ZmWRKY104−SRDX−mCherry) or Control (empty vector) were transfected into protoplasts. Culture medium (−ABA) or 10 μM ABA (+ABA) were treated with the protoplasts for 10 min. Values are the means ± SE of three different experiments. According to Duncan's multiple range test, means represented by the same letter did not significantly differ at p < 0.05.

ZmWRKY104 and ZmMPK6 Function in ABA−Induced Antioxidant Defense
Our previous research reported that MAPK was related to ABA−induced antioxidant defense [6]. Thus, we wondered whether ZmWRKY104 and ZmMPK6 function together in ABA−induced antioxidant defense. As shown in Figure 2, the activities of APX and SOD after transient expression of ZmWRKY104 or ZmMPK6 alone were significantly increased, and these increased activities were further enhanced by ABA treatment. Compared with ZmWRKY104 or ZmMPK6 alone, co−expression of ZmWRKY104 and ZmMPK6 further increased the activities. The activities of APX and SOD were also further enhanced by treatment with ABA. These results suggest that co−expression of ZmWRKY104 and ZmMPK6 improves ABA−induced antioxidant defense in maize.
Moreover, we investigated the subcellular localizations of ZmWRKY104 and ZmMPK6. ZmWRKY104 and ZmMPK6 fused to yellow fluorescent protein (YFP) were ZmWRKY104 and srdxZmWRKY104 in maize protoplasts; (C) detection of the efficiency of transiently silencing ZmWRKY104 in maize protoplasts. Gene expression analysis of ZmWRKY104 by RT-qPCR, with ZmActin as an internal control. Activities of APX (D) and SOD (E) of transiently expressing ZmWRKY104 in protoplasts. ZmWRKY104 (ubi: ZmWRKY104-mCherry) or Control (empty vector) were transfected into protoplasts. Activities of APX (F) and SOD (G) of transiently silencing ZmWRKY104 in protoplasts. ZmWRKY104-RNAi or Control (distilled water) were transfected into protoplasts. Activities of APX (H) and SOD (I) of transiently expressing srdxZmWRKY104 in protoplasts. srdxZmWRKY104 (ubi: ZmWRKY104-SRDX-mCherry) or Control (empty vector) were transfected into protoplasts. Culture medium (-ABA) or 10 µM ABA (+ABA) were treated with the protoplasts for 10 min. Values are the means ± SE of three different experiments. According to Duncan's multiple range test, means represented by the same letter did not significantly differ at p < 0.05.

ZmWRKY104 and ZmMPK6 Function in ABA-Induced Antioxidant Defense
Our previous research reported that MAPK was related to ABA-induced antioxidant defense [6]. Thus, we wondered whether ZmWRKY104 and ZmMPK6 function together in ABA-induced antioxidant defense. As shown in Figure 2, the activities of APX and SOD after transient expression of ZmWRKY104 or ZmMPK6 alone were significantly increased, and these increased activities were further enhanced by ABA treatment. Compared with ZmWRKY104 or ZmMPK6 alone, co-expression of ZmWRKY104 and ZmMPK6 further increased the activities. The activities of APX and SOD were also further enhanced by treatment with ABA. These results suggest that co-expression of ZmWRKY104 and ZmMPK6 improves ABA-induced antioxidant defense in maize.
Moreover, we investigated the subcellular localizations of ZmWRKY104 and ZmMPK6. ZmWRKY104 and ZmMPK6 fused to yellow fluorescent protein (YFP) were transiently expressed by Agrobacterium infiltration in tobacco leaves. Subcellular localization results showed that ZmWRKY104 was specifically located in the nucleus of tobacco mesophyll cells, and ZmMPK6 was specifically located in the cytoplasmic membrane and nucleus of tobacco mesophyll cells ( Figure 3). transiently expressed by Agrobacterium infiltration in tobacco leaves. Subcellular localization results showed that ZmWRKY104 was specifically located in the nucleus of tobacco mesophyll cells, and ZmMPK6 was specifically located in the cytoplasmic membrane and nucleus of tobacco mesophyll cells ( Figure 3).

ZmMPK6 Interacts with and Phosphorylates ZmWRKY104
To identify ZmWRKY104 as a target of ZmMPK6, a GST pull−down assay was first used. Full−length ZmWRKY104 fused to GST protein and full−length ZmMPK6 tagged with poly−His. His−tagged ZmMPK6 was maintained on beads with immobilized GST−ZmWRKY104 but not with immobilized GST ( Figure 4A), which suggests that ZmMPK6 interacted with ZmWRKY104 in vitro. Subsequently, we performed Y2H assays

ZmMPK6 Interacts with and Phosphorylates ZmWRKY104
To identify ZmWRKY104 as a target of ZmMPK6, a GST pull-down assay was first used. Full-length ZmWRKY104 fused to GST protein and full-length ZmMPK6 tagged with poly-His. His-tagged ZmMPK6 was maintained on beads with immobilized GST-ZmWRKY104 but not with immobilized GST ( Figure 4A), which suggests that ZmMPK6 interacted with ZmWRKY104 in vitro. Subsequently, we performed Y2H assays to further identify the interaction in vitro. Full-length ZmWRKY104 was inserted into the pGADT7 (AD) vector, and ZmMPK6 was inserted into the PGBKT7 (BD) vector. Both vectors were then transformed and selected on SD/Trp/Leu/His/Ade with X-α-galactosidase, and the interaction between ZmWRKY104 and ZmMPK6 was observed ( Figure 4B). We then performed Co-IP to validate the interaction between ZmWRKY104 and ZmMPK6 in Nicotiana benthamiana leaves. Proteins were immunoprecipitated with protein-A/G agarose beads and analyzed by immunoassay using anti-Myc and anti-Flag antibodies. Immunoblot (IB) analyses using an anti-Myc antibody showed an interaction between Flag-ZmWRKY104 and Myc-ZmMPK6 ( Figure 4C). An LCI assay in tobacco leaves was then performed to further confirm the interaction between ZmWRKY104 and ZmMPK6, and the results showed that the ZmWRKY104 protein could interact with the ZmMPK6 protein ( Figure 4D). Thus, the above results indicated that ZmMPK6 interacts with ZmWRKY104.   GST or GST-ZmWRKY104 fusion protein was incubated with His-ZmMPK6 in GST beads. His-ZmMPK6 was detected using an antibody of anti-His by Western blot. GST-ZmWRKY104 and GST were detected using an anti-GST antibody by Western blot. Molecular masses are marked on the right; (B) Y2H assay. AD-vector and BD-vector were Y2H vectors with no insert. To test the interaction, the medium of SE-Trp-Leu-His-Ade/X-α-gal was used. pGBKT7-53/pGADT7-T was used as a positive control, and pGBKT7lam/pGADT7-T was used as a negative control; (C) Co-IP assay. 35S:Myc-ZmMPK6 and 35S:Flag-ZmWRKY104 were co-expressed in four-week-old Nicotiana benthamiana leaves. Proteins were immunoprecipitated (IP) with Flag antibody and analyzed by immunoblot (IB) using anti-Flag and anti-Myc antibodies. Molecular masses are marked on the right; (D) LCI assay. ZmMPK6 was fused to nLUC, and ZmWRKY104 was fused to cLUC. ZmMPK6-nLUC and cLUC-ZmWRKY104 were then co-expressed in Nicotiana benthamiana leaves. nLUC/cLUC-ZmWRKY104, ZmMPK6-nLUC/cLUC, and nLUC/cLUC were used as negative controls. Luciferase signals were captured using the Tanon 5200 image system. Scale bar = 1 cm. All experiments were repeated at least three times with similar results.
To examine which region(s) of ZmWRKY104 mediates its binding to ZmMPK6, we generated two truncations, ZmWRKY104 1-85AA and ZmWRKY104 86-266AA ( Figure S5A), and tested these truncations both in vitro and in vivo through GST pull-down and LCI assays. As shown in Figure S5B,C, the interaction region was located at the 1-85 AA region at the N-terminus of ZmWRKY104. To verify that ZmMPK6 phosphorylates ZmWRKY104, we performed an in vitro gel kinase assay. Results showed that ZmMPK6 has autophosphorylation and substrate phosphorylation activity and can phosphorylate myelin basic protein (MBP) and ZmWRKY104 ( Figure 5A). Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis results showed that Threonine-59 (Thr-59, T59) was the major phosphorylation site of ZmWRKY104 by ZmMPK6 ( Figure 5B).

ZmMPK6 Phosphorylation of ZmWRKY104 Plays a Key Role in ABA−Induced Antioxidant Defense
To identify whether the Thr−59 site is the major site of ZmWRKY104, we mutated the T59 site of ZmWRKY104 to non−phosphorylated alanine (T59A) and simulated phosphorylated state aspartic acid (T59D) for an in vitro gel kinase assay. Results showed that ZmWRKY104 T59A could be weakly phosphorylated by ZmMPK6, indicating that the Thr−59 site of the ZmWRKY104 protein blocked most of the phosphorylation of ZmWRKY104 by ZmMPK6 ( Figure S6). To explore the role of the Thr−59 site of ZmWRKY104 in ABA−induced antioxidant defense, the maize transient expression system of protoplasts was used to test its effect on the activities of APX and SOD. As shown in Figure 6, compared with the control, transient expression of ZmWRKY104 T59A alone did

ZmMPK6 Phosphorylation of ZmWRKY104 Plays a Key Role in ABA-Induced Antioxidant Defense
To identify whether the Thr-59 site is the major site of ZmWRKY104, we mutated the T59 site of ZmWRKY104 to non-phosphorylated alanine (T59A) and simulated phosphorylated state aspartic acid (T59D) for an in vitro gel kinase assay. Results showed that ZmWRKY104 T59A could be weakly phosphorylated by ZmMPK6, indicating that the Thr-59 site of the ZmWRKY104 protein blocked most of the phosphorylation of ZmWRKY104 by ZmMPK6 ( Figure S6). To explore the role of the Thr-59 site of ZmWRKY104 in ABA-induced antioxidant defense, the maize transient expression system of protoplasts was used to test its effect on the activities of APX and SOD. As shown in Figure 6, compared with the control, transient expression of ZmWRKY104 T59A alone did not significantly influence the activities, whereas transient expression of ZmWRKY104 T59D alone significantly increased the activities. The activities in ZmWRKY104 T59A /ZmMPK6-co-expressed protoplasts were equal to those found in protoplasts expressing ZmMPK6 alone. Compared with the control, co-expression of ZmWRKY104 T59D /ZmMPK6 significantly increased the activities, and ABA treatment further increased these activities ( Figure 6). These data clearly suggest that the phosphorylation of ZmWRKY104 at Thr-59 by ZmMPK6 plays a key role in ABA-induced antioxidant defense. According to Duncan's multiple range test, means represented by the same letter did not significantly differ at p < 0.05.

ZmWRKY104 Overexpression Enhances Drought Tolerance in Transgenic Maize Plants
We used PEG6000 to imitate drought stress to investigate the expression of ZmWRKY104 in maize plants. Results revealed that ZmWRKY104 was powerfully induced at 15 min by PEG ( Figure 7A). The above results indicate that ZmWRKY104 was induced by drought stress.
To further understand the part of ZmWRKY104 in drought stress, we generated transgenic maize plants under the control of the ubiquitin promoter via Agrobacterium tumefaciens−mediated transformation. We selected two independent lines (#15 and #17; Figure  S7) that exhibited similar ZmWRKY104 gene and protein levels for further analysis. Compared with wild−type (WT) lines, ZmWRKY104−OE transgenic lines displayed no morphological changes under normal growth conditions ( Figure 7B). After natural drought, the wild−type lines were restricted, and the leaves were curled and yellow, while the ZmWRKY104−OE transgenic lines grew better, and the leaves were stretched; after rewatering, the leaves of the ZmWRKY104−OE transgenic lines spread out evenly, while the wild−type lines only partially survived and could not fully restore the flat state ( Figure  7B). Under drought stress, compared with the WT lines, the ZmWRKY104−OE transgenic lines had higher survival rates ( Figure 7C), leaf water retention capacity ( Figure 7D), and activity of antioxidant defense enzymes ( Figure 7E). In addition, the electrolyte leakage rate ( Figure 7F) and malondialdehyde (MDA) content ( Figure 7H) of the ZmWRKY104−OE transgenic lines were lower than those of the wild−type lines; the relative water content ( Figure 7G) and accumulated proline content ( Figure 7I) were higher than those of the wild−type lines. The above results indicate that ZmWRKY104 enhances drought stress tol- ZmMPK6-mCherry (ZmWRKY104/ZmMPK6) simultaneously, ubi: ZmWRKY104 T59A -mCherry and ubi: ZmMPK6-mCherry (ZmWRKY104 T59A/ ZmMPK6) simultaneously, and ubi: ZmWRKY104 T59D -mCherry and ubi: ZmMPK6-mCherry (ZmWRKY104 T59D /ZmMPK6) simultaneously were transiently expressing in protoplasts. Culture medium (-ABA) or 10 µM ABA (+ABA) were treated with the protoplasts for 10 min. Values are the means ± SE of three different experiments. According to Duncan's multiple range test, means represented by the same letter did not significantly differ at p < 0.05.

ZmWRKY104 Overexpression Enhances Drought Tolerance in Transgenic Maize Plants
We used PEG6000 to imitate drought stress to investigate the expression of ZmWRKY104 in maize plants. Results revealed that ZmWRKY104 was powerfully induced at 15 min by PEG ( Figure 7A). The above results indicate that ZmWRKY104 was induced by drought stress.
To further understand the part of ZmWRKY104 in drought stress, we generated transgenic maize plants under the control of the ubiquitin promoter via Agrobacterium tumefaciens-mediated transformation. We selected two independent lines (#15 and #17; Figure S7) that exhibited similar ZmWRKY104 gene and protein levels for further analysis. Compared with wild-type (WT) lines, ZmWRKY104-OE transgenic lines displayed no morphological changes under normal growth conditions ( Figure 7B). After natural drought, the wild-type lines were restricted, and the leaves were curled and yellow, while the ZmWRKY104-OE transgenic lines grew better, and the leaves were stretched; after rewatering, the leaves of the ZmWRKY104-OE transgenic lines spread out evenly, while the wild-type lines only partially survived and could not fully restore the flat state ( Figure 7B). Under drought stress, compared with the WT lines, the ZmWRKY104-OE transgenic lines had higher survival rates ( Figure 7C), leaf water retention capacity ( Figure 7D), and activity of antioxidant defense enzymes ( Figure 7E). In addition, the electrolyte leakage rate ( Figure 7F) and malondialdehyde (MDA) content ( Figure 7H) of the ZmWRKY104-OE transgenic lines were lower than those of the wild-type lines; the relative water content ( Figure 7G) and accumulated proline content ( Figure 7I) were higher than those of the wild-type lines. The above results indicate that ZmWRKY104 enhances drought stress tolerance.

Discussion
TFs play key roles in plants, allowing them to adjust to new environmental factors, such as drought. As one of the largest families of TFs in higher plants, WRKY families play crucial roles in response to various stresses [12,14,26]. More and more evidence show that WRKY protein physically interacts with a variety of proteins, and these proteins play a role in controlling or regulating various processes. MAP kinase is one of the main proteins that interact with WRKY TFs [41,42]. The MAP kinase pathway is related to regulate the activities of WRKY33 to manage plant defense responses [43]. The cascade may act as

Discussion
TFs play key roles in plants, allowing them to adjust to new environmental factors, such as drought. As one of the largest families of TFs in higher plants, WRKY families play crucial roles in response to various stresses [12,14,26]. More and more evidence show that WRKY protein physically interacts with a variety of proteins, and these proteins play a role in controlling or regulating various processes. MAP kinase is one of the main proteins that interact with WRKY TFs [41,42]. The MAP kinase pathway is related to regulate the activities of WRKY33 to manage plant defense responses [43]. The cascade may act as a molecular hub, integrating different signaling networks of WRKY TFs with various upstream proteins. The fact that most WRKYs are distinguishingly regulated by bacterial pathogens or SA treatment further confirms this, which also activates the MPK3/6 pathway [44]. In this study, we demonstrate that ZmWRKY104, which belongs to group IIa of the WRKY superfamily, is a positive component of the ABA-induced antioxidant defense (Figure 1).
The function of TFs or their related proteins could be regulated by MAPK of its phosphorylation [45]. In defense signaling, some studies have shown that WRKY protein is associated with defense-induced MAPK signaling cascade. The NtMEK2-SIPK/WIPK signaling pathway is involved in regulating the expression of defense genes in tobacco [41]. Here, our data show that ZmMPK6-phosphorylated ZmWRKY104 increases ABA-induced antioxidant defense in maize, and phosphorylation at the Thr59 position plays an important role in this process (Figures 5 and 6). Therefore, the modulation of WRKY protein phosphorylated by MAPK seems to be a common phenomenon for regulating plant defense responses.
Subcellular localization experiments revealed that ZmWRKY104 focuses on the nucleus (Figure 3), which accords with the truth that it acts as a TF, potentially modulating the transcription of defense-related genes. However, how Thr-59 phosphorylation affects the DNA binding of ZmWRKY104 is unclear. In addition, the cascade upstream of ZmMPK6 in ABA signaling in maize also needs to be further explored.
Overexpression of ZmWRKY104 modified the drought tolerance of maize plants (Figure 7), indicating that this gene is a positive regulator of drought resistance. Detailed analysis revealed that ZmWRKY104 overexpression alleviates the degree of oxidative damage in response to drought in maize. First, ZmWRKY104 overexpression reduced the electrolyte leakage, which is an immediate stress marker of damage in cell membrane under drought conditions [46]. Second, ZmWRKY104 overexpression lines accumulated higher proline than WT plants. The accumulation of high levels of proline was beneficial to plant resistance to environmental stress [47]. Third, ZmWRKY104 overexpression lines possessed lower MDA contents, and MDA is typically used as an important indicator of oxidative damage under environmental stresses [48]. Fourth, ZmWRKY104 overexpression lines had higher relative water content than WT plants, which is necessary for plants to survive under drought conditions. Furthermore, the alleviation of oxidative damage might be caused by enhanced antioxidant defense.
In summary, our data show that ZmWRKY104 interacts with and is phosphorylated by ZmMPK6, Thr-59 phosphorylation is critical to the role of ZmWRKY104 in ABA-induced antioxidant defense, and overexpression of ZmWRKY104 enhances drought tolerance and alleviates drought-induced oxidative damage.

Plant Materials and Treatments
In this study, we use maize (Zea mays L., cv. B73) and tobacco (Nicotiana benthamiana) seeds. The maize seed was sown in trays placed in an environmental chamber with a photosynthetically active radiation (200 µmol m −2 s −1 ), a temperature (22-28 • C), and a 14 h/10 h (day/night) photoperiod and watered every 2 days [49].
Maize plants were grown under dark conditions at 25 • C, and the second leaf was fully extended to extract protoplasts. We cut the plant from the base of the stem and put it in distilled water for 2 h to relieve the wound pressure. After this treatment, we put the cut end of the stem into a beaker wrapped in aluminum foil, which contained a solution of 100 µM ABA and 10% (w/v) polyethylene glycol (PEG6000) solution. Plants immersed in distilled water for the entire period under the same conditions served as controls. After the treatment, subsequent leaves of the separated maize plants were sampled and immediately frozen in liquid nitrogen.
Tobacco seeds were sown in a soil pot with a photosynthetically active radiation (120 µmol m −2 s −1 ), a temperature (23 • C), and a photoperiod of 16 h/8 h (day/night) and watered every two days. After 4 weeks of growth, we use the plants for subsequent experiments.

Yeast Two-Hybrid (Y2H) Assay
According to the yeast protocol handbook (Clontech, Shiga, Japan), Y2H assays were performed. ZmWRKY104 was cloned into pGADT7 using EcoRI/BamHI sites, ZmMPK6 was cloned into pGBKT7 using BamH/PstI sites, and Table S1 shows the primers. The prey vector was converted into Y187 of yeast strain, and the bait vector was transformed into Y2HGold of yeast strain using the lithium acetate method. The prey and bait strains were mated and spread on stringent selective medium plates containing X-α-gal (40 µg mL −1 ), then incubated for 3-5 days at 30 • C and verified every day.

Co-Immunoprecipitation (Co-IP) Assay
ZmWRKY104 or ZmMPK6 were fused to Flag or Myc tags cloned into 1300-221 vectors using KpnI/BamHI sites or BstBI/BamHI sites. Table S1 shows the primers. 35S: Myc-ZmMPK6, 35S: Flag-ZmWRKY104, 35S: Myc-ZmMPK6, and 35S: Flag-ZmWRKY104 were expressed in Nicotiana benthamiana leaves. After incubation for 3 days, proteins were extracted from the leaves with buffer as described previously and centrifuged at 4 • C at 10,000 g for 30 min [6]. The soluble proteins were incubated with an anti-Flag (Abmart) or anti-Myc antibody (Abmart) bound to IP buffer at 4 • C for 3 h. We washed the beads with immunoprecipitation buffer 3 times and boiled for 8 min in 1 × SDS buffer to elute proteins. After centrifugation, the supernatant was analyzed with anti-MyC antibody (Abmart) by immunoblotting.

Luciferase Complementation Imaging (LCI) Assay
We fused full-length and truncated sequences of ZmWRKY104 to the pC1300-cLUC vector utilizing KpnI/BamHI sites, and fused full-length sequences of ZmMPK6 to the pC1300-nLUC vector utilizing SacI/BamHI sites. Table S1 shows the primers. Then we transformed them into GV3101 of Agrobacterium tumefaciens strain and incubated under dark conditions at 30 • C for 2-3 days, and selected the positive clone and incubated in yeast extract broth (YEB) liquid medium for 16 h at 28 • C [49]. The bacteria were collected in 2 mL tubes and centrifuged at 5000 rpm for 5 min at 25 • C. Then we removed the supernatant and resuspended the bacteria in buffer (10 mM MgCl 2 , 10 mM MES, pH 5.7, and 100 µM acetosyringone) to a final OD600 of 0.5. After 3 to 5 h, we injected the bacteria into Nicotiana benthamiana leaves, and 72 h after injection, we sprayed with 1 mM D-Luciferin (Thermo Fisher Scientific, Waltham, MA, USA) at the abaxial sides of the leaves. The leaves were then stored for 30 min in the dark, and we used a camera (Tanon 5200 Multi, Tanon Biomart, Beijing, China) to capture the signal of LUC.

Expression and Purification of Recombinant Proteins
ZmWRKY104 and its truncated mutants (ZmWRKY104 1-85AA and ZmWRKY104 86-266AA ), were amplified, cloned into the vector of pGEX-4T-1 with a GST tag, and expressed in BL21 (DE3) of E. coli strain. The expression of plasmids or the empty vector was induced for 6 h at 28 • C with isopropyl β-D-1-thiogalactopyranoside (IPTG) of 0.2 mM in LB broth on a shaker at 160 rpm. Full-length ZmWRKY104, ZmWRKY104 T59A or ZmWRKY104 T59D was transformed into BL21 (DE3) of E. coli with the His-tagged expression vector pET-30a, and induced by incubation for 6 h at 22 • C with 0.2 mM IPTG. Full-length ZmMPK6 was cloned into the pET-30a vector in frame with the His tag, transformed into BL21 (DE3) of E. coli strain, and induced by incubation for 6 h at 26 • C with 0.5 mM IPTG.

In Vitro Kinase Assay
Recombinant His-ZmWRKY104, His-ZmWRKY104 T59A or His-ZmWRKY104 T59D was incubated with 0.01 mg mL −1 ZmMPK6 protein in buffer of phosphorylation (20 mM MgCl 2 , 25 mM Tris-HCl, pH 7.5, 2 mM MnCl 2 , and 1 mM DTT) for 30 min at 30 • C in a final volume of 50 µL including 1 µg of MBP, 10 µM ATP, and 10 µCi [γ-32 P] ATP. We added SDS sample buffer to stop the reaction and boiled the product for 10 min in a boiling water bath and resolved by 12% SDS-PAGE. We removed the unincorporated [γ-32 P] ATP by washing with 5% trichloroacetic acid (w/v) and 1% sodium pyrophosphate (w/v) more than three times. The phosphorylated substrates were visualized by autoradiography.

Isolation of Total RNA and qRT-PCR Analysis
The analysis was performed as performed as described previously [7]. The primers are described in detail in Table S1.

Vector Construction and In Vitro Synthesis of Double-Stranded RNA
The construction and synthesis were performed as described previously [7]. The primers are given in Table S1.

Protoplast Preparation and Transfection with DNA Constructs or dsRNAs
The experiments were performed as indicated previously [7]. The transformed protoplasts were then incubated overnight at 25 • C in the dark in culture solution, treated for 10 min with or without 10 µM ABA, and utilized for the subsequent analysis.

Site-Directed Mutagenesis
For the mutation of ZmWRKY104, we utilized the M5 HiPer Site-Directed Mutagenesis Kit (Mei5bio, Beijing, China) based on the manufacturer's instructions. Table S1 shows the DNA oligonucleotides sequences.

Subcellular Localization
Tobacco leaves were transfected with GV3101 of Agrobacterium strain carrying the fusion construct of 35S: ZmWRKY104-YFP, the 35S: ZmMPK6-YFP fusion construct, or 35S: YFP. Yellow fluorescence was observed as indicated previously under a Zeiss LSM710 device [7]. The nucleus was stained with a nuclear marker (RFP-H2A), and the plasma membrane was stained with a membrane marker (PM-RK) [50].

Mass Spectrometry Analysis
The His-ZmWRKY104 fusion protein was reacted with the His-ZmMPK6 fusion protein in kinase assay buffer in vitro with ATP. Then we enriched the phosphorylated ZmWRKY104 and digested by trypsin followed by mass spectrometry/liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis as indicated previously [7]. The results of mass spectrometry were analyzed using the software pFind Studio (http://pfind. ict.ac.cn/software/pLink/index.html, accessed on 23 Jul 2020).

Generation of Transgenic Maize Plants
Insert ZmWRKY104 into the HindIII/KpnI sites of the pCUN-N-HF vector with ubiquitin promoter; Table S1 shows the primers. We used the maize inbred line B73 for genetic transformation. The recombinant vector was introduced into maize by LBA4404 of Agrobacterium tumefaction strain via maize shoot tip transformation [54,55]. Agrobacteria containing the recombinant strain were cultured separately and then mixed in infiltration buffer to a final OD600 between 0.65 and 0.75. The mixture of Agrobacterium cultures were infiltrated with young maize stems by a vacuum pump (0.045 to 0.060 MPa) for 3 to 5 min and then sown in pots containing a 1:1 mixture of organic nutrient soil vermiculite. The pots were placed in an environmental chamber with 22-28 • C, photosynthetically active radiation (200 µmol m −2 s −1 ), and a 14 h/10 h (day/night) photoperiod and watered every 3 days. The positive plants were identified by PCR analysis and then transferred to the greenhouse for planting and harvesting, and Table S1 shows the primers. qRT-PCR analyzed the expression of ZmWRKY104 in transgenic lines, and two independent homozygous T2 lines, ZmWRKY104 #15 and ZmWRKY104 #17, were selected for subsequent experiments.

Drought Tolerance and Oxidative Damage Analysis
After 4 weeks of growth, seedlings were subjected to native drought conditions (water was withheld) for 10 days and then rewatered using normal protocols for 3 days. The seedling phenotype was photographed, and the survival rate was calculated. The experiments were repeated 3 times, each individual line with more than 35 plants employed in every repeated experiment, and one typical image is shown. The relative water content (RWC), malondialdehyde (MDA) levels, electrolyte leakage, the content of proline (Pro), and APX and SOD activities were measured as previously described [7,[56][57][58].

Phylogenetic Analysis
To construct a phylogenetic tree, the sequences of plant WRKY proteins and MAPK proteins were retrieved from the NCBI database and aligned using the ClustalW program, a neighbor-joining phylogenetic tree of these sequences was then constructed, and bootstrap analysis was performed with 500 iterations using the MEGA7 program [52,59].

Conclusions
In this study, we identified a WRKY IIa TF, ZmWRKY104, as the substrate protein of ZmMPK6 in maize. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis results showed that threonine-59 (Thr-59, T59) was the major phosphorylation site of ZmWRKY104 by ZmMPK6. Functional analysis revealed that the role of ZmWRKY104 in ABA-induced antioxidant defense depends on ZmMPK6. Genetic analysis showed that overexpression of ZmWRKY104 in maize can improve drought tolerance and alleviate drought-induced oxidative damage in transgenic plants.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/biology10090893/s1, Figure S1. Multiple alignment of ZmWRKY104 with other orthologs in rice, Arabidopsis. Figure S2. Phylogenetic relationships of ZmWRKY104 with other orthologs in rice, Arabidopsis. Figure S3. Multiple alignment and phylogenetic relationships of ZmMPK6 with other orthologs in rice, Arabidopsis, wheat, cotton. Figure S4. RT-qPCR analysis of ZmWRKY104 expression in different tissues of maize, Figure S5. ZmMPK6 interacts with the truncations of ZmWRKY104, Figure S6. Phosphorylation of ZmWRKY104 T59A and ZmWRKY104 T59D by ZmMPK6 in vitro, Figure S7. Analysis the gene expression and protein level of the ZmWRKY104 transgenic lines, Table S1: PCR Primers Used.