ZmADF5, a Maize Actin-Depolymerizing Factor Conferring Enhanced Drought Tolerance in Maize

Drought stress is seriously affecting the growth and production of crops, especially when agricultural irrigation still remains quantitatively restricted in some arid and semi-arid areas. The identification of drought-tolerant genes is important for improving the adaptability of maize under stress. Here, we found that a new member of the actin-depolymerizing factor (ADF) family; the ZmADF5 gene was tightly linked with a consensus drought-tolerant quantitative trait locus, and the significantly associated signals were detected through genome wide association analysis. ZmADF5 expression could be induced by osmotic stress and the application of exogenous abscisic acid. Its overexpression in Arabidopsis and maize helped plants to keep a higher survival rate after water-deficit stress, which reduced the stomatal aperture and the water-loss rate, as well as improved clearance of reactive oxygen species. Moreover, seventeen differentially expressed genes were identified as regulated by both drought stress and ZmADF5, four of which were involved in the ABA-dependent drought stress response. ZmADF5-overexpressing plants were also identified as sensitive to ABA during the seed germination and seedling stages. These results suggested that ZmADF5 played an important role in the response to drought stress.


Introduction
Abiotic stress, especially drought, seriously affects crop growth, survival, and productivity.The steady growth of production is the primary goal for crop drought-tolerance breeding under the influence of climate change.As reduced variation in grain yield (GY) is a response to stress, the selection of superior genotypes for GY under drought-stressed conditions usually has limited and inconsistent progress [1].Linkage-based quantitative trait loci (QTL) mapping with bi-parent populations enables the identification of chromosomal regions associated with these quantitative traits of tolerance and their further application to quick genome selection.So far, numerous QTL that control stress tolerance have been identified [2][3][4][5].Meta-analysis helps to combine and compare these QTL identified from independent analyses to determine the number and location of existing, consensus QTL.Therefore, the confidence interval of an integrated QTL cluster can be reduced to improve the accuracy and effectiveness of QTL mapping with few confirmed QTL [2,6].Linkage Plants 2024, 13, 619 2 of 18 disequilibrium-based association analyses with natural populations is usually performed as a complement to linkage-based QTL mapping.So, joint linkage mapping and association analyses have been applied in the detection of QTL or genes for complex traits, including drought tolerance [7][8][9].Up to now, a lot of QTL/genes of important traits have been identified using combined linkage mapping and association analyses in maize, including genes related to plant height and ear height [10], male inflorescence size [11], gray leaf spot disease resistance [12], waterlogging tolerance [13], and drought tolerance [7].However, few of them have been cloned or functionally analyzed.
The actin-depolymerizing factor (ADF) gene family is an actin-binding protein family and a major regulator of actin dynamics in eukaryotes [14,15].ADFs can incorporate globular actin (G-actin) into filamentous actin (F-actin) to extend and maintain the stability of actin by nucleation; alternatively, they also can sever F-actin by dissociation [16].Both processes are ATP-dependent and affected by several factors such as pH [17] and phosphorylation of the ADFs [18].More ADFs have been identified in plants than in animals recently, accompanied by more diverse biochemical functions caused by mutations during evolution [15,16,19].Currently, the ADF gene family has been reported to play a crucial role in plant growth and development, including hypocotyl and root hair elongation, pollen germination and pollen tube growth, flowering time, cotton fibre development, etc. [20][21][22][23][24].For example, down regulation of AtADF1 gene expression affects flowering time [23].The AtADF5 loss-of-function mutants exhibited delayed pollen germination and defected pollen tubes [21].Moreover, several ADFs in plants have also been reported in relation to the regulation of biotic or abiotic stress responses.In Arabidopsis, AtADF2 is involved in resistance to root-knot nematode infection [25].AtADF4 links pathogen perception and defense activation [26], as well as drought stress tolerance [27].AtADF5 is another drought stress-responsive gene that responds to drought stress by regulating stomatal closure [28].The overexpression of OsADF3 in rice enhances drought tolerance in Arabidopsis [29].In wheat, TaADF3 [16], TaADF4 [30], and TaADF7 [31] play an important role in disease resistance.However, few ADF genes have been identified in maize, especially in association with biotic or abiotic stress resistance.
In this study, we identified and characterized a maize ADF gene, named ZmADF5, originally derived from a consensus drought-tolerance QTL in maize.To clarify whether and how ZmADF5 participates in drought stress, we analyzed the phylogenetic relationship of ADF in different species and the expression pattern of the ZmADF5 gene under drought stress.Meanwhile, we generated overexpressed and knockout ZmADF5 transgenic plants.We found that ZmADF5-overexpressing transgenic plants displayed enhanced tolerance to drought.Transcriptome data analysis and susceptibility to ABA in Arabidopsis thaliana with overexpression of ZmADF5 suggest that ZmADF5 also enhances drought tolerance through an ABA-dependent pathway.Our research shows that ZmADF5 could be a crucial candidate gene for improving drought tolerance in maize.

Identification of the Relationship between ZmADF5 and Two Linked Polymorphic Markers
Drought-tolerant consensus QTL 1 (DCQ1) was previously identified as located on maize chromosome bin 1.03 (Figure 1A, adapted from Hao et al., 2010 [2]).Association analysis with 538 known markers was used for the mapping of DCQ1 to identify the drought-tolerance genes near bin 1.03.In total, 517 protein-coding genes were identified with a mean of 1.04 markers per gene.Ten significant association signals were detected in relation to GY at the p = 10 −3 level, but only one single nucleotide polymorphism marker (PZE-101047611) was identified with the GY trait during exposure to the water-stressed (WS) condition in Xinjiang in 2009 (Figure 1B); this marker explained 9.55% of the GY variation under WS and was located on the 5 ′ -untranslated region of the GRMZM2G077942 gene.Notably, the tightly linked polymorphic marker umc1073 was identified on the same gene by linkage mapping within the DCQ1 confidential interval peak.Therefore, the gene GRMZM2G077942 might be a candidate gene for tolerance to drought stress in maize.GRMZM2G077942 encodes a conserved actin-binding protein (ABP) (referred to hereafter as ZmADF5), belonging to an ubiquitous family of low molecular weight (15~20 kDa).
Plants 2024, 13, 619 3 of 19 GRMZM2G077942 gene.Notably, the tightly linked polymorphic marker umc1073 was identified on the same gene by linkage mapping within the DCQ1 confidential interval peak.Therefore, the gene GRMZM2G077942 might be a candidate gene for tolerance to drought stress in maize.GRMZM2G077942 encodes a conserved actin-binding protein (ABP) (referred to hereafter as ZmADF5), belonging to an ubiquitous family of low molecular weight (15~20 kDa).

Phylogenetic Analysis of the ADF Gene Family in Maize and Other Plant Species
ZmADF5 belongs to the ADF family, which includes the main regulators of actin in plants.The functional divergence of ADFs has been discussed in Arabidopsis [19].To examine the phylogenetic relationships among ADFs in maize, Arabidopsis, and other plant species, an NJ tree was constructed after the alignment of 78 full-length ADF amino acid sequences in seven species (Supplementary Figure S1A).Except for ZmADFn-2 and GmADF1, all ADFs were clustered into four groups, which supported a previous classification [15].Groups I, II, and IV were composed of ADFs from eudicot and monocot plants clustered together.Moreover, Group III exclusively comprised monocots (Supplementary Figure S1A).Phylogenetic analysis showed that ZmADF1, ZmADF2-2, ZmADFn-3, ZmADFn-4, and ZmADFn-5 clustered in Groups I; ZmADF4 clustered in Groups II; ZmADF3-1, ZmADF3-2, and ZmADF3-3 clustered in Groups III; and ZmADF2-1, ZmADF5-1, ZmADF5-2, ZmADF6, and ZmADFn-1 clustered in Groups IV.A distinct difference in the length of the first exon of the ADFs was detected among the four groups.The first exon of the ADFs clustered in Groups I, II, and III contained only three bases (ATG) (Supplementary Figure S1C), which was the initiation codon; the length of the first exon varied in the Group IV ADFs [19].To further characterize the diversity of the ADF gene family, we analyzed the distributions of conserved motifs.Twenty motifs were identified using the MEME website (Supplementary Figure S1B).Motifs 1-4 were present

Phylogenetic Analysis of the ADF Gene Family in Maize and Other Plant Species
ZmADF5 belongs to the ADF family, which includes the main regulators of actin in plants.The functional divergence of ADFs has been discussed in Arabidopsis [19].To examine the phylogenetic relationships among ADFs in maize, Arabidopsis, and other plant species, an NJ tree was constructed after the alignment of 78 full-length ADF amino acid sequences in seven species (Supplementary Figure S1A).Except for ZmADFn-2 and GmADF1, all ADFs were clustered into four groups, which supported a previous classification [15].Groups I, II, and IV were composed of ADFs from eudicot and monocot plants clustered together.Moreover, Group III exclusively comprised monocots (Supplementary Figure S1A).Phylogenetic analysis showed that ZmADF1, ZmADF2-2, ZmADFn-3, ZmADFn-4, and ZmADFn-5 clustered in Groups I; ZmADF4 clustered in Groups II; ZmADF3-1, ZmADF3-2, and ZmADF3-3 clustered in Groups III; and ZmADF2-1, ZmADF5-1, ZmADF5-2, ZmADF6, and ZmADFn-1 clustered in Groups IV.A distinct difference in the length of the first exon of the ADFs was detected among the four groups.The first exon of the ADFs clustered in Groups I, II, and III contained only three bases (ATG) (Supplementary Figure S1C), which was the initiation codon; the length of the first exon varied in the Group IV ADFs [19].To further characterize the diversity of the ADF gene family, we analyzed the distributions of conserved motifs.Twenty motifs were identified using the MEME website (Supplementary Figure S1B).Motifs 1-4 were present in most of the ADF proteins, while the other conserved motifs were group specific.For example, motifs 5-7 were only found in Group IV.
Among the ADFs in maize, two were annotated with ZmADF5 at the NCBI: one was located on chromosome 1 (ZmADF5-1) and the other was located on chromosome 9 (ZmADF5-2).These two genes had an almost identical amino acid sequence with a similarity of 80.81%, except for the additional 29 amino acids in the N-terminal of ZmADF5-2 (Supplementary Figure S2).Both of the ZmADF5 genes were clustered in Group IV, where almost all ADF5 and ADF9 sequences were included (Supplementary Figure S1A).AtADF5 and AtADF9 have opposing biochemical properties compared to other ADFs in Arabidopsis because of N-terminal extensions and amino acid changes during evolution [19].Moreover, AtADF5 [32] and AtADF9 [33] control development in Arabidopsis; AtADF5 is involved in stress response to cold [32] and drought [28].These results indicate that ZmADF5 is involved in plant development and/or the stress response.

ZmADF5 Expression Is Induced by Drought Stress
ZmADF5 was expressed in almost all tissues in maize (e.g., roots, stems, leaves, silk, and tassels) throughout the growth period; however, it had higher expression levels in leaf tips (Figure 2A).Under the water-deficit treatment, the gene was up-regulated in root and leaf tissues; the corresponding maxima were 3.60-fold and 5.36-fold greater than in the control (0 h), respectively (Figure 2B).Under the ABA treatment, the gene was up-regulated in root, stem, and leaf tissues at all time points; the corresponding peaks were present at 1, 3, and 1 h, respectively (Figure 2C).These results indicate that ZmADF5 might be induced in the drought-and ABA-stress responses.
in most of the ADF proteins, while the other conserved motifs were group specific.For example, motifs 5-7 were only found in Group IV.
Among the ADFs in maize, two were annotated with ZmADF5 at the NCBI: one was located on chromosome 1 (ZmADF5-1) and the other was located on chromosome 9 (ZmADF5-2).These two genes had an almost identical amino acid sequence with a similarity of 80.81%, except for the additional 29 amino acids in the N-terminal of ZmADF5-2 (Supplementary Figure S2).Both of the ZmADF5 genes were clustered in Group IV, where almost all ADF5 and ADF9 sequences were included (Supplementary Figure S1A).AtADF5 and AtADF9 have opposing biochemical properties compared to other ADFs in Arabidopsis because of N-terminal extensions and amino acid changes during evolution [19].Moreover, AtADF5 [32] and AtADF9 [33] control development in Arabidopsis; AtADF5 is involved in stress response to cold [32] and drought [28].These results indicate that ZmADF5 is involved in plant development and/or the stress response.

ZmADF5 Expression Is Induced by Drought Stress
ZmADF5 was expressed in almost all tissues in maize (e.g., roots, stems, leaves, silk, and tassels) throughout the growth period; however, it had higher expression levels in leaf tips (Figure 2A).Under the water-deficit treatment, the gene was up-regulated in root and leaf tissues; the corresponding maxima were 3.60-fold and 5.36-fold greater than in the control (0 h), respectively (Figure 2B).Under the ABA treatment, the gene was upregulated in root, stem, and leaf tissues at all time points; the corresponding peaks were present at 1, 3, and 1 h, respectively (Figure 2C).These results indicate that ZmADF5 might be induced in the drought-and ABA-stress responses.

Subcellular Localization of ZmADF5
To determine the subcellular localization of ZmADF5, a 35S::ZmADF5-GFP fusion construct was generated and transformed into maize protoplasts.Confocal scanning analysis showed that the fluorescence of 35S::GFP was distributed throughout the cell, while the fluorescence of 35S::ZmADF5-GFP was present in the nucleus and the cytosol in maize protoplasm (Figure 2D).

Drought Tolerance Is Enhanced in Arabidopsis Transgenic Plants through Overexpression of ZmADF5
Three independent ZmADF5-overexpressing transgenic lines of Arabidopsis were used for further analysis.After the cessation of watering for 3 weeks, most leaves in the wild type (WT) and transgenic plants began to dry and become dark because of water loss.Most WT plants gradually withered during the water-deficit stress, while the transgenic plants remained green and few withered (Figure 3A).Survival rate analysis in Arabidopsis exposed to water-deficit stress revealed that 82% of the transgenic plants were alive, whereas all WT plants had died (Figure 3B).As reported, Arabidopsis AtADF5 promotes stomatal closure in response to drought stress [28].Water-loss assays were conducted on detached leaves to test whether ZmADF5 is also involved in the regulation of stomatal movements under drought stress.As shown in Figure 3C, water loss was slower in ZmADF5-overexpressing leaves than in WT leaves under dehydrating conditions.No significant differences in the stomatal aperture were detected between the WT and transgenic plants under normal-watering conditions.However, the stomatal aperture after water-deficit treatment was larger in WT plants than in transgenic plants (Figure 3D,E).These results indicate that overexpression of ZmADF5 might improve drought tolerance in Arabidopsis.

Overexpression of ZmADF5 Enhances Drought Tolerance in Maize
To confirm that ZmADF5 enhanced drought tolerance, we constructed ZmADF5overexpressing (ZmADF5-OE) lines and ZmADF5-knockout (ZmADF5-KO) lines of maize, which were subjected to drought-tolerance assays (Supplementary Table S3 and Figure S3).The ZmADF5 transcript levels in the two controls (Zheng58 and C01), as well as the ZmADF5-OE and ZmADF5-KO lines, are presented in Figures 4B and 5B.After 10 days of water-deficit stress, most control plants (Zheng58) began to wither, while only a few leaves rolled and wilted in the ZmADF5-OE lines.The ZmADF5-OE lines appeared normal with leaves that remained green and grew better than the control plants did (Figure 4A); however, the ZmADF5-KO plants showed the opposite result (Figure 5A).The survival rates of the two controls and the transgenic lines were recorded 3 days after rehydration.The survival rates of the OE lines were 2.30-2.50-foldgreater than the survival rate of Zheng58 (Figure 4C), while the survival rates of the KO lines were 1.50-1.97-foldlower than that of C01 (Figure 5C).The water-loss rate of the OE lines was significantly lower than the waterloss rates of the Zheng58 plants at the beginning 1-4 h; the greatest difference happened at 4 h (Figure 4D), while the ZmADF5-KO plants showed opposite results (Figure 5D).Furthermore, we surveyed some physiological parameters (including malondialdehyde (MDA) and anthocyanin contents, as well as catalase (CAT) and superoxide dismutase (SOD) activities, chlorophyll content) of the two controls and the transgenic plants under normal-watering and water-deficit treatments at the same time.No significant differences in five physiological parameters were observed between the control and transgenic plants in the norma-watering treatment; under the water-deficit treatment, the activities of SOD and CAT, as well as the chlorophyll content, were significantly greater in ZmADF5-OE lines than in control Zheng58 plants (Figure 4E-G).These physiological parameters were significantly lower in ZmADF5-KO plants than in C01 plants (Figure 5E-G).Compared with the normal-watering treatment, anthocyanin and MDA contents significantly accumulated in the transgenic and control plants under continuous water-deficit treatment, but ZmADF5-OE plants exhibited less anthocyanin and MDA accumulation than Zheng58 (Figure 4H

Overexpression of ZmADF5 Enhances Drought Tolerance in Maize
To confirm that ZmADF5 enhanced drought tolerance, we constructed ZmADF5overexpressing (ZmADF5-OE) lines and ZmADF5-knockout (ZmADF5-KO) lines of maize, which were subjected to drought-tolerance assays (Supplementary Table S3 and Figure S3).The ZmADF5 transcript levels in the two controls (Zheng58 and C01), as well as the The survival rates of the OE lines were 2.30-2.50-foldgreater than the survival rate of Zheng58 (Figure 4C), while the survival rates of the KO lines were 1.50-1.97-foldlower than that of C01 (Figure 5C).The water-loss rate of the OE lines was significantly lower than the water-loss rates of the Zheng58 plants at the beginning 1-4 h; the greatest difference happened at 4 h (Figure 4D), while the ZmADF5-KO plants showed opposite results (Figure 5D).

ABA Dependence of Enhanced Drought Tolerance in Transgenic Plants
RNA-Seq was performed to further explore the mechanism of the improved drought tolerance mediated by ZmADF5 overexpression.We compared gene expression patterns in the ZmADF5-overexpressing (OE10) transgenic line and WT plants of Arabidopsis under normal-watering and water-deficit treatments (WT-N and OE-N under normal-watering conditions; WT-D and OE-D under water-deficit conditions).In total, 319 differentially Plants 2024, 13, 619 9 of 18 expressed genes (DEGs) were identified in the normal-watering condition, with 252 upregulated and 67 down-regulated in the OE10 line relative to the WT (Figure 6A).Under water-deficit stress conditions, 55 genes were differentially expressed (34 up-regulated and 21 down-regulated) (Figure 6B).Subsequently, Gene Ontology (GO) analysis was performed to identify the potential functions of these genes.Biological processes, such as responses to stress and external, abiotic, and chemical stimuli were greatly enriched among the identified DEGs, as were catalytic and oxidoreductase activities (Figure 6D).These transcriptomic changes could contribute to the high survival rate observed in transgenic Arabidopsis plants that had been subjected to drought stress.Among the 55 DEGs, 17 genes that overlapped between OE-N/OE-D and WT-D/OE-D were regulated by drought stress and ZmADF5 (Figure 6A, Table 1).Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis showed that these genes were enriched in various pathways, including carotenoid biosynthesis (ko00906) and circadian rhythms (ko04712).Two genes (CYP707A3 and PHYB) were annotated in both pathways.CYP707A3 is involved in carotenoid biosynthesis (ko00906), which is an important ABA signal transduction pathway.PHYB is involved in ko04712; it is also an ABA-dependent drought-resistant gene [34].Six significantly altered genes were chosen for qRT-PCR analysis to confirm the DEGs identified by RNA-Seq.As shown in Supplementary Figure S4, the expression levels of all the tested genes in the different materials revealed a transcription expression pattern that was similar to the RNA-Seq data.Previous experiments have shown that ABA can induce the expression of ZmADF5 (Figure 2C).ABA plays an important role in plant responses to drought stress as a plant hormone, and it is directly associated with both CYP707A3 and PHYB, so we examined the ABA sensitivities of ZmADF5-overexpressing plants.In the absence of ABA, the seed germination rates of the ZmADF5-overexpressing lines and WT were not significantly different.With the increase in ABA concentration, the seed germination rates of WT and ZmADF5-overexpressing lines decreased, and ZmADF5-overexpressing lines displayed lower seed germination rates than WT (Supplementary Figure S5A,B).For root length measurements, the WT and ZmADF5-overexpressing lines had similar root length growth rates in the absence of ABA.However, the root length of ZmADF5-overexpressing lines was significantly longer than that of the WT under 20 µM ABA treatment (Supplementary Figure S5C,D).These experiments demonstrate that ZmADF5-overexpressing lines were more sensitive to ABA.Overall, these results suggest that drought tolerance in the transgenic plants might have been enhanced through an ABA-dependent pathway that led to increased expression levels of many stress-responsive genes.

The Drought-Tolerant Candidate Gene ZmADF5 belongs to the Maize ADF Gene Family
Drought stress resistance in plants is a complex trait controlled by numerous QTL.Although many drought-resistance-related QTL have been identified, few genes have been cloned through QTL mapping in maize.Association analysis is widely used to identify traits and genes; several drought-related genes have been identified by association analysis, such as ZmNAC111 [35], ZmVPP1 [36], and ZmTIP1 [37].In this study, we confirmed the QTL identified by a meta-analysis and association mapping; we identified the droughttolerance gene ZmADF5 in maize bin 1.03, where a hotspot region for maize drought tolerance has been reported in previous studies [2,5].ZmADF5 belongs to the ADF gene family, which reportedly has an important role in the plant stress response.Phylogenetic analysis showed that ZmADF5 was clustered in Group IV, where all ADF5s from different plant species were in the same branch (Supplementary Figure S1A).Among these genes, AtADF5 [28] and OsADF5 [38] conferred drought tolerance; thus, we initially presumed that ZmADF5 might have similar drought tolerance in maize.In our study, ZmADF5 was induced by water-deficit and ABA treatments (Figure 2B,C).The overexpression of ZmADF5 increased drought tolerance in maize (Figure 4A,C).These results suggest that ZmADF5 is a candidate gene for the QTL in bin 1.03 and is involved in the drought stress response.

ZmADF5 Contributes to the Regulation of Drought Tolerance through an ABA-Dependent Pathway Involving ROS Scavenging in Maize
Abiotic stress usually induces the accumulation of ROS in plant cells, causing damage to plants.The removal of excess ROS improves abiotic stress tolerance in plants.Antioxidants and ROS-scavenging enzymes (e.g., SOD and CAT) are essential for detoxifying ROS during normal metabolism, particularly during stress [39,40].In our study, we found that SOD and CAT activities, as well as chlorophyll content, were significantly greater in the ZmADF5-OE lines than in the control plants, while the ZmADF5-KO lines showed contrasting results (Figures 4E-G and 5E-G).Moreover, MDA is the final decomposition product of membrane lipid peroxidation caused by ROS, and its content can reflect the degree of membrane damage.Compared with the wild type, ZmADF5-OE plants accumulated less MDA after water-deficit treatment (Figure 4H), which indicated that the membrane damage of overexpressed ZmADF5 plants was lower.These results demonstrate that drought stress caused less damage to the ZmADF5-OE lines than to Zheng58.
In our study, transgenic Arabidopsis overexpressing ZmADF5 exhibited greater sensitivity to exogenous ABA than the WT plants did (Supplementary Figure S5).Moreover, there is evidence that AtADF5 [28] and OsADF5 [38] enhance plant drought tolerance through an ABA-dependent pathway.Therefore, we presumed that ZmADF5 may also be involved in the ABA-dependent pathway in response to drought stress.By RNA-Seq in Arabidopsis, among the DEGs after the water-deficit treatment, 17 (0.4%) were regulated by ZmADF5 and the drought treatment.Among these 17 genes, 4 were involved in the ABA-dependent drought stress response.RAP2-6 has been identified as an ABA-dependent abiotic response gene; the overexpression of RAP2-6 enhances drought and salt stress resistance in transgenic plants [41,42].MSL3 is an osmotic stress-response gene; msl2 msl3 mutants express proline and ABA metabolism genes under drought or osmotic stress [43].CYP707A3 is a key gene in the ABA catabolic pathway; it has an important role in the abiotic stress response, including the response to drought [44].Moreover, the overexpression of ZmADF5 makes transgenic Arabidopsis hypersensitive to ABA (Supplementary Figure S5).These results show that ZmADF5 the may be involved in ABA-dependent signaling pathway in response to drought stress.
In conclusion, our results show that ZmADF5 is a drought stress-responsive gene in maize.ZmADF5 can improve drought tolerance in transgenic Arabidopsis and maize.Under water-deficit stress, overexpression of ZmADF5 increased SOD and CAT activities, and decreased MDA and anthocyanin accumulation.The stomatal aperture of transgenic ZmADF5 Arabidopsis plants decreased after water-deficit treatment, the germination rate decreased, and the root length increased after different concentrations of ABA treatment.These findings suggest that ZmADF5 serves as a positive regulator of drought stress and has good potential to improve the genetics of drought stress tolerance in crops.

Association Analysis of Grain Yield in Maize under Drought Stress
In our previous study, a consensus drought-tolerance QTL was identified in maize by meta-analysis; it was named DCQ1.This QTL integrated 21 drought-tolerant QTLs with a series of backgrounds deriving from different genetic populations [2].This consensus QTL region is located on chromosome bin 1.03; the linked marker corresponding to drought tolerance at the peak value of the confidence interval (i.e., the most likely position) was umc1073, which belongs to the gene GRMZM2G077942 encoding an actin-depolymerizing factor 5 (ZmADF5) protein (Figure 1A).Therein, association analysis of maize grain yield (GY) under WW and WS conditions was performed using a maize-association panel that consisted of 210 inbred lines collected from China [45,46].The plant GY was phenotyped in 2009 and 2011 in Xinjiang, China (43 • 54 ′ N, 87 • 28 ′ E), and in 2010 in Hainan, China (18 • 14 ′ N, 109 • 31 ′ E).The plants were given sufficient water during the growth period; watering was stopped before pollen dispersal for drought-stress treatment [45].The experiments were repeated twice with both water conditions each year.The best linear, unbiased prediction of GY was evaluated for each individual in the six environments (year × location × treatment) using META-R software (http://hdl.handle.net/11529/10201)(accessed on 1 February 2018), with water management as a grouping factor.
The panel was genotyped using the Illumina MaizeSNP50 BeadChip; 41,087 single nucleotide polymorphisms with a minor allele frequency > 0.05 were included in the genome-wide association study.Mixed linear modeling with population structure (Q) and kinship (K) was conducted in TASSEL 5.0 software [47].Principal component analysis and kinship of the panel were also calculated in TASSEL 5.0; the first four principal components were used to estimate the population structure, which explained 22.06% of the variance.Single nucleotide polymorphisms with a p-value < 10 −3 were significantly associated with GY.

Phylogenetic Analysis of ADF Genes
In total, 78 ADF genes in maize and six other plant species (sorghum [Sorghum bicolor], millet [Setaria italica], rice [Oryza sativa], Arabidopsis [Arabidopsis thaliana], soybean [Glycine max], and potato [Solanum tuberosum]) were used for the alignment and phylogenetic analyses (Supplementary Table S1).The amino acid sequences of all ADF genes were downloaded from the National Center for Biotechnology Information (NCBI, https:// www.ncbi.nlm.nih.gov/accessed on 10 February 2019).The evolutionary analysis was conducted in MEGA7 [48] using the neighbor-joining (NJ) method [49] and 78 amino acid sequences.The optimal tree with the sum of branch length = 7.14 is shown.The percentages of replicate trees in which the associated taxa were clustered together in the bootstrap test (1000 replicates) are shown next to the branches [50].The evolutionary distances were computed using the p-distance method [51]; the units comprised the number of amino acid differences per site.Conserved motifs in 78 ADF amino acid sequences were analyzed using Multiple Em for Motif Elicitation software (https://meme-suite.org/tools/meme accessed on 20 April 2020), and the maximum number of motifs was 20.All genome sequences were downloaded from NCBI.The conserved motif and gene structure diagrams were drawn for all ADF genes using TBtools software (https://github.com/CJ-Chen/TBtoolsaccessed on 25 April 2020).

Expression Profiles of Maize ZmADF5
The gene expression data of the ADF genes in maize were downloaded from MaizeGDB (https://www.maizegdb.org/accessed on 10 January 2019).The expression levels of seeds, roots, stems, leaves, and the flowering parts at different growth stages were analyzed using the heatmap2 package in R software (R Foundation for Statistical Computing, Vienna, Austria).To further characterize the ZmADF5 expression patterns among stress treatments, we conducted quantitative real-time polymerase chain reaction (qRT-PCR) analyses.Maize seeds were grown to the V2 stage in quartz sand, and then transferred to a hydroponic system in Hoagland nutrient solution.The seedlings were treated with 20% PEG6000 and 100 µM ABA for 12 h at the V3 stage.The roots, stems, and leaves of maize were sampled at 0, 1, 3, 6, and 12 h after treatment; they were compared with the corresponding controls planted in normal conditions.Tissues equally mixed from five plants were regarded as a single replicate; two replicates were collected for the treated and control materials.Total RNA was extracted from the samples using TranZol UP (TransGen Biotech, Beijing, China), in accordance with the manufacturer's instructions.First-strand cDNA was synthesized using the FastQuant RT Kit (Tiangen).qRT-PCR was conducted using the SuperReal PreMix Plus kit (SYBR Green) (Tiangen) in IQ5 (Bio-Rad Laboratories Inc., Hercules, CA, USA).Relative expression levels were calculated using the 2 −∆∆Ct method [52].The specific primers used in this study are listed in Supplementary Table S2.

Subcellular Localization of ZmADF5
The ZmADF5 coding region without the terminal codon (TGA) was amplified using a pair of primers that contained BamHI sites and 18 bp overlapping homologous ends of the pAN580 vector.Using the Seamless Assembly Cloning Kit (Clone Smarter), the PCR product was fused upstream of the GFP gene to generate a ZmADF5-GFP fusion construct driven by the CaMV35S promoter.Maize protoplasts were extracted and transformed in accordance with the method established by Yoo et al. (2007) [53].The (35S:ZmADF5-GFP) vector and the control vector were transformed into maize mesophyll protoplasts using a PEG-mediated transformation method.The subcellular location of ZmADF5 was detected by fluorescence microscopy (LSM980; ZEISS, Jena, Germany) after 16 h of incubation in the dark.The specific primers used in this study are listed in Supplementary Table S2.

Transgenic-Positive Plant Construction and Screening
In this study, A. thaliana ecotype Col-3 was used as the wild type (WT).To construct the pCHF3-ZmADF5 vector, the full-length coding sequence of the ZmADF5 gene was amplified using KOD FX polymerase (Toyobo, Tokyo, Japan); genomic DNA from the B73 inbred line served as the template.The vector was digested with SmaI (New England Biolabs, Ipswich, MA, USA); the linearized vector and the ZmADF5 DNA fragment were assembled using the Seamless Assembly Cloning Kit (Clone Smarter), in accordance with the manufacturer's instructions.The reconstructed vector was introduced into Agrobacterium tumefaciens after sequencing, and then transferred to Arabidopsis by the floral dip method [54].T 0 generation seeds of transgenic plants were screened on 1/2 Murashige and Skoog (MS) medium with kanamycin.The positive plants were then transplanted into pots filled with a 2:1 mixture of vermiculite and nutritional soil.Three parallel lines of T 3 transgenic generation plants were used for further analysis.
The ZmADF5 coding sequence was amplified from the B73 cDNA library and inserted into the SmaI site of the CUB vector (this vector is controlled by the maize UBI promoter).The reconstructed UBI:ZmADF5-GFP plasmid was transformed into A. tumefaciens, and then introduced into the maize HiII hybrid line.Transgenic T 0 plants were cultured in a 16 h light/8 h dark greenhouse; transgene-positive plants were detected by PCR.T 0 transgene-positive plants were backcrossed to the Zheng58 inbred line to a purified genetic background; two independent ZmADF5 transgenic lines were obtained for further analysis of drought resistance.CRISPR/Cas9 was used to simultaneously edit the ZmADF5, and then obtain knockout lines.The same guide RNA with high specificity was designed based on the ZmADF5 coding region.The sequence was cloned into the CPB-Cas9 vector.T 0 transgene-positive plants were identified by specific primer PCR.T 2 seeds were obtained from T 0 transgene-positive plants through self-crossing for further analyses.The constructed transgenic maize is listed in Supplementary Table S3.

Stress Treatment and Phenotyping of Transgenic Arabidopsis
The drought treatment was conducted in medium and soil.Seeds from the T 3 generation of transgenic plants and the WT were sterilized in 10% sodium hypochlorite for 10 min, and then washed 10 times in distilled water.The seeds were planted on 1/2 MS medium plate with 1% sucrose and 0.8% agar.After 3 days of vernalization at 4 • C, the plants were grown under a long-day condition (16 h light/8 h dark) at 22 • C.
Plants for the drought-tolerance assay were transferred from 1/2 MS medium into pots filled with a 2:1 mixture of vermiculite and nutritive soil at the two-leaf stage.After 3 weeks of growth in the normal-watering condition, the plants were exposed to drought stress by the cessation of watering for 2 weeks.Then, the plants were allowed to recover via rehydration; their growth state was recorded 3 days later.The leaves of the WT and transgenic plants under drought stress and the normal-watering control were sampled.Leaves sampled from three plants were mixed as a single replicate; WT and transgenic Arabidopsis were subjected to five replicate water treatments.The leaves were used for stomatal aperture analysis, where leaves were floated in solutions containing 30 mM KCl and 10 mM MES-Tris, and the irradiance was set to 150 µmol•m −2 •s −1 for 3 h.The width and length of the stomatal pores, as determined by using ImageJ software (http://rsbweb.nih.gov/ij accessed on 15 May 2021), were used to calculate the stomatal apertures (ratio of width to length).

Germination Assay and Root Growth Measurement
The sterilized Arabidopsis seeds were plated on 1/2 MS medium with 0 µM ABA, 0.5 µM ABA, 1.0 µM ABA, and 1.5 µM ABA.After vernalization for 3 days at 4 • C, the plates were transferred to 22 • C for culture and the germination rate of the seeds was recorded.For root length measurement, the vernalized seeds were grown on 1/2 MS medium for 4 days.Then, roots with the same length were transferred to 1/2 MS medium containing 20 µM ABA.During the growth, the root length was observed and photographed.

Phenotypic Analyses in Transgenic Maize
The seedlings of control and transgenic maize were germinated in Petri dishes for 3 days under dark conditions.After germination, control and transgenic maize seeds were transplanted in seedling trays filled with fertile soil and placed in a greenhouse (16 h light/8 h dark photoperiod, 25 • C).When the seedlings grew to the three-leaf stage, they were treated with drought stress by withholding watering.The plants received water after approximately 14 days, and the survival rate of the plants was determined 3 days later.Leaves of the 14-day-old maize were collected.The leaves were weighed after 0, 0.5, 1, 1.5, 2, and 4 h at room temperature, respectively.The leaves were then dried in an oven at 65 • C for 24 h; their dry weights were recorded and the water-loss rates were calculated as water loss (%) = (initial fresh weight − final fresh weight)/initial fresh weight × 100.Chlorophyll content was determined using the SPAD-502 Chlorophyll Meter.The MDA content, CAT activity, and SOD activity were determined using biochemical assay kits (Solarbio, Beijing, China).Anthocyanin content was measured using the method established by Drumm-Herrel and Mohr (1982) [55].All samples had three biological replicates.Normally watered plants were used as the control in all experiments.

RNA-Seq Analysis of Transgenic Plants
Two-week-old transgenic Arabidopsis and WT seedlings were treated without watering for 10 days.Four groups of pooled tissues from three plants in the transgenic and WT lines were collected before and after the drought treatment.Three replicates and 12 total samples were collected for the RNA-Seq analysis.Total RNA was isolated using the RNAprep pure Plant Kit (Tiangen); the concentration and integrity of the RNA were evaluated

Figure 1 .
Figure 1.Identification of the ZmADF5 gene through linkage mapping and association mapping meta-analysis.(A) Meta-analysis of constitutive QTL for drought tolerance on maize chromosome 1 (adapted from Hao et al., 2010 [2]).(B) Manhattan plot of the GWAS result of maize grain yield in the well-watered (WW) and water-stressed (WS) conditions.

Figure 1 .
Figure 1.Identification of the ZmADF5 gene through linkage mapping and association mapping meta-analysis.(A) Meta-analysis of constitutive QTL for drought tolerance on maize chromosome 1 (adapted from Hao et al., 2010 [2]).(B) Manhattan plot of the GWAS result of maize grain yield in the well-watered (WW) and water-stressed (WS) conditions.

Figure 2 .
Figure 2. Expression patterns of ZmADF5 in maize and subcellular localization of ZmADF5.(A) Expression levels of ZmADF5 in different tissues and growth stages in the normal condition.Data

Figure 2 .
Figure 2. Expression patterns of ZmADF5 in maize and subcellular localization of ZmADF5.(A) Expression levels of ZmADF5 in different tissues and growth stages in the normal condition.Data were downloaded from MaizeGDB (https://www.maizegdb.org/)(accessed on 10 January 2019).(B) Expression pattern of ZmADF5 at the seedling stage in the water-deficit treatment.(C) Expression pattern of ZmADF5 at the seedling stage in the ABA treatment.(D) Localization of GFP and the ZmADF5-GFP fusion protein in maize protoplasts.Scale bar, 10 µm.

19 Figure 3 .
Figure 3. Phenotype of the 35S:ZmADF5 transgenic Arabidopsis.(A) Drought tolerance of transgenic Arabidopsis plants overexpressing ZmADF5.(B) Statistical analysis of the survival rate after waterdeficit stress treatment.(C) Water loss from detached rosettes of WT and 35S:ZmADF5 transgenic plants.Water loss was expressed as the percentage of initial fresh weight.(D) Stomatal closure in WT and 35S:ZmADF5 transgenic plants.(E) Values are mean ratios of width to length.Error bars represent standard errors of three independent experiments.Bars, 10 mm.Significant differences were determined by t-tests.** p < 0.01.

Figure 3 .
Figure 3. Phenotype of the 35S:ZmADF5 transgenic Arabidopsis.(A) Drought tolerance of transgenic Arabidopsis plants overexpressing ZmADF5.(B) Statistical analysis of the survival rate after waterdeficit stress treatment.(C) Water loss from detached rosettes of WT and 35S:ZmADF5 transgenic plants.Water loss was expressed as the percentage of initial fresh weight.(D) Stomatal closure in WT and 35S:ZmADF5 transgenic plants.(E) Values are mean ratios of width to length.Error bars represent standard errors of three independent experiments.Bars, 10 mm.Significant differences were determined by t-tests.** p < 0.01.

Figure 4 .
Figure 4. ZmADF5 enhances drought tolerance in transgenic maize.(A) Drought tolerance in ZmADF5-OE lines.(B) Relative expression levels of ZmADF5 in the transgenic maize (ZmADF5-OE1 and ZmADF5-OE2) lines under normal conditions.(C) Statistical analysis of survival rates after water-deficit stress treatment.(D) Water loss from detached rosettes of wild-type Zheng58 plants and the ZmADF5-OE lines.Water loss was expressed as the percentage of initial fresh weight.(E-I) Comparison of chlorophyll content (E), SOD activity (F), CAT activity (G), MDA content (H), and anthocyanin content (I) between leaves from wild-type Zheng58 plants and ZmADF5-OE lines under water-deficit stress conditions.Significant differences were determined by t-tests.* p < 0.05, ** p < 0.01.

Figure 5 .Figure 5 .
Figure 5. Knocking out ZmADF5 reduces drought tolerance in transgenic maize.(A) Drought tolerance in ZmADF5-KO lines.(B) Relative expression levels of ZmADF5 in ZmADF5 knockout (ZmADF5-KO1 and ZmADF5-KO2) lines under normal conditions.(C) Statistical analysis of the survival rate after water-deficit stress treatment.(D) Water loss from detached rosettes of wild-type CO1 plants and the ZmADF5-KO lines.Water loss was expressed as the percentage of initial fresh weight.(E-I) Comparison of chlorophyll content (E), SOD activity (F), CAT activity (G), MDA content (H), and anthocyanin content (I) between leaves from wild-type CO1 plants and the Figure 5. Knocking out ZmADF5 reduces drought tolerance in transgenic maize.(A) Drought tolerance in ZmADF5-KO lines.(B) Relative expression levels of ZmADF5 in ZmADF5 knockout (ZmADF5-KO1 and ZmADF5-KO2) lines under normal conditions.(C) Statistical analysis of the survival rate after water-deficit stress treatment.(D) Water loss from detached rosettes of wild-type CO1 plants and the ZmADF5-KO lines.Water loss was expressed as the percentage of initial fresh weight.(E-I) Comparison of chlorophyll content (E), SOD activity (F), CAT activity (G), MDA content (H), and anthocyanin content (I) between leaves from wild-type CO1 plants and the ZmADF5-KO lines under water-deficit stress conditions.Significant differences were determined by t-tests.* p < 0.05, ** p < 0.01.

Plants 2024, 13 , 619 10 of 19 Figure 6 .
Figure 6.Transcriptomic analysis of 35S:ZmADF5 transgenic Arabidopsis under normal and waterdeficit conditions.(A) Venn diagrams of differentially expressed genes in OE plants relative to WT plants under normal-watering and water-deficit conditions, using a significant cutoff of p-value < 0.05.(B) Volcano plot of differentially expressed gene distributions in ZmADF5-overexpressing and wild-type Arabidopsis plants under water-deficit stress.The red and green dots indicate significantly differentially up-regulated and down-regulated genes, respectively.The blue dots represent non differentially expressed genes.(C) Heatmap and cluster analysis of differentially expressed genes.(D) GO annotation of differentially expressed genes in ZmADF5-overexpressing and wild-type Arabidopsis plants under water-deficit stress.

Figure 6 .
Figure 6.Transcriptomic analysis of 35S:ZmADF5 transgenic Arabidopsis under normal and waterdeficit conditions.(A) Venn diagrams of differentially expressed genes in OE plants relative to WT plants under normal-watering and water-deficit conditions, using a significant cutoff of p-value < 0.05.(B) Volcano plot of differentially expressed gene distributions in ZmADF5-overexpressing and wildtype Arabidopsis plants under water-deficit stress.The red and green dots indicate significantly differentially up-regulated and down-regulated genes, respectively.The blue dots represent non differentially expressed genes.(C) Heatmap and cluster analysis of differentially expressed genes.(D) GO annotation of differentially expressed genes in ZmADF5-overexpressing and wild-type Arabidopsis plants under water-deficit stress.

Table 1 .
Differentially expressed genes are regulated by ZmADF5 overexpression and drought stress.