Comprehensive Analysis of the INDETERMINATE DOMAIN (IDD) Gene Family and Their Response to Abiotic Stress in Zea mays

Abstract

Transcription factors (TFs) are important regulators of numerous gene expressions due to their ability to recognize and combine cis-elements in the promoters of target genes. The INDETERMINATE DOMAIN (IDD) gene family belongs to a subfamily of C2H2 zinc finger proteins and has been identified only in terrestrial plants. Nevertheless, little study has been reported concerning the genome-wide analysis of the IDD gene family in maize. In total, 22 ZmIDD genes were identified, which can be distributed on 8 chromosomes in maize. On the basis of evolutionary relationships and conserved motif analysis, ZmIDDs were categorized into three clades (1, 2, and 3), each owning 4, 6, and 12 genes, respectively. We analyzed the characteristics of gene structure and found that 3 of the 22 ZmIDD genes do not contain an intron. Cis-element analysis of the ZmIDD promoter showed that most ZmIDD genes possessed at least one ABRE or MBS cis-element, and some ZmIDD genes owned the AuxRR-core, TCA-element, TC-rich repeats, and LTR cis-element. The Ka:Ks ratio of eight segmentally duplicated gene pairs demonstrated that the ZmIDD gene families had undergone a purifying selection. Then, the transcription levels of ZmIDDs were analyzed, and they showed great differences in diverse tissues as well as abiotic stresses. Furthermore, regulatory networks were constructed through the prediction of ZmIDD-targeted genes and miRNAs, which can inhibit the transcription of ZmIDDs. In total, 6 ZmIDDs and 22 miRNAs were discovered, which can target 180 genes and depress the expression of 9 ZmIDDs, respectively. Taken together, the results give us valuable information for studying the function of ZmIDDs involved in plant development and climate resilience in maize.

1. Introduction

To survive better, terrestrial plants have evolved many mechanisms to accommodate a variety of complex environments, including drought, salt, and low temperature stress. Transcription factors (TFs) play dominant roles in plant growth and development and response to abiotic stress [1,2]. TFs can recognize and combine cis-elements, thus activating or repressing the transcription of downstream genes [3]. A number of important TF families have been identified in plants, including C2H2 (Cys2His2 zinc finger) [4], WRKY [5], bHLH (basic helix-loop-helix) [6], MYB [7], bZIP (basic region-leucine zipper) [8], and MADS-box [9]. Of these, the INDETERMINATE DOMAIN (IDD) family, a class of C2H2 zinc finger proteins, has been found only in land plants [10,11].

The IDD gene family is characterized by the INDETERMINATE (ID) domain, which contains four zinc fingers (ZFs) [10,12]. The four ZFs include two C2H2-type ZFs involved in DNA binding and two C2HC-type ZFs responsible for protein interaction [13]. The first IDD genes, ZmID1, were identified in Zea mays through transposon insertion, which regulates flowering time via a leaf-generated signal [14]. The IDD family gene has been discovered in numerous plants, including Arabidopsis [15], rice [16], Moso bamboo [17], and so on. So far, 16 IDD genes have been found in Arabidopsis, and 12 of them have been functionally verified [18]. In Arabidopsis thaliana, IDDs function in various metabolic and development processes, including root and leaf development, flowering time, seed maturation, hormone signaling, and abiotic stress [19]. For example, six AtIDD genes (IDD2, IDD3, IDD4, IDD5, IDD9, and IDD10) are involved in gibberellin (GA) homeostasis, thus modulating florescence through interactions with DELLA proteins [20]. Coelho et al. [18] reported that AtIDD14 could be alternatively spliced and produce two transcripts. AtIDD14α exists in normal conditions, but AtIDD14β accumulates when plants are subjected to cold stress. Moreover, AtIDD14 can also interact with ABFs/AREBs and cooperatively mediate ABA-dependent drought tolerance [21]. This suggests that IDDs are responsible for responding to different environmental stimuli in plants, but research concerning IDDs remains rare.

IDDs also play vital roles in the grass family, such as rice and barley. Ghd10, the ortholog of ZmID1, participates in mediating yield component characters via increasing plant height and tillering under short-day (SD) conditions [22]. OsIDD2 is a negative regulator in second cell wall formation through repressing target genes, which are related to sucrose metabolism and lignin synthesis [23]. OsIDD3/ROC1 enhances cold tolerance by directly targeting cis-elements of dehydration-responsive element-binding protein (DREB)/CBF1 [24]. OsIDD10 was reported to regulate ammonium absorption and nitrogen metabolism in roots [25]. In barley, BROAD LEAF1 (BLF1) was found to regulate leaf size by restraining longitudinal cell proliferation [26].

Maize is the most common cereal crop in the world. Nevertheless, genome-wide analysis of IDDs has not been conducted in maize. Thus far, three IDDs have been characterized in maize. For example, ZmID1 can regulate flowering time [11], and its paralogs ZmIDD9 and ZmIDDveg9 control endosperm development [27,28]. With the rapid development of sequencing technology, whole-genome analysis of ZmIDD gene families becomes feasible. In this study, we identified 22 candidate ZmIDD genes through bioinformatics analysis. Then, a systematic analysis of the phylogenetic tree, chromosome localization, gene structures, conserved motifs, and promoter cis-elements was conducted. In addition, tissue-specific and abiotic stress-mediated expression profiles of ZmIDDs were analyzed. At last, miRNA-target ZmIDDs and ZmIDDs-target genes were predicted. The results will give us valuable information for better comprehending the function of ZmIDDs in plant development and response to environmental stress in maize.

2. Results

2.1. Identification and Evolution Analysis of ZmIDD Family Genes in Zea mays

In this study, we identified 22 ZmIDDs. The gene and protein characteristics, such as gene names, ID, chromosomal locations, amino acid numbers, molecular weights (MW), and isoelectric points (pI), were shown in Table S1. For instance, the amino acid length of 22 ZmIDD proteins ranges from 354 to 815. The molecular weight ranges from 38 to 90 kDa. The coding sequence and protein sequence can be seen in Table S2. Evolutionary relationships of IDD genes were shown via a phylogenetic tree using 15 rice IDDs, 16 Arabidopsis IDDs, and 22 maize IDDs using the maximum likelihood method, the minimum evolution method, and the Neighbor-Joining method (Figure 1; Figures S1 and S2; Table S3). Three major clades were identified. Clades 1, 2, and 3 contained 4, 6, and 12 ZmIDDs, respectively (Figure 1; Figures S1 and S2).

2.2. Structure Analysis of ZmIDD

The DNA structure analysis showed that ZmIDDs had 1–4 exons distributed unevenly (Figure 2B; Table S2). Meanwhile, eight conserved motifs were found in ZmIDDs and named motifs 1–8 (Table S4). In detail, all ZmIDDs have motif 1, 2, and 3. Motifs 4, 5, and 7 were presented in all subfamilies. Motif 8 only exists in the members of clade 1. Motif 6 exists exclusively in the members of clade 2 and clade 3 (Figure 2C). Multi-sequence alignment of ZmIDDs demonstrated that all ZmIDD proteins owned two C2H2-type (ZF1 and ZF2) and two C2HC-type (ZF3 and ZF4) ZFs (Figure 3). In addition, 3D structures of ZmIDD1, ZmIDD2, and ZmIDD4 proteins, belonging to clade 3, clade 1, and clade 2, respectively, were predicted, indicating each ZmIDD had different 3D structures in spite of having four ZF fingers in all ZmIDDs (Figure 4).

2.3. Location and Duplication of ZmIDDs

The chromosome localization showed that 22 ZmIDDs were distributed across the maize genome except for chromosome 4 and 6 (Figure 5). In detail, ZmIDD clade 1 genes were distributed on chromosomes 1, 7, and 9; ZmIDD clade 2 genes were found on chromosomes 1, 2, 5, 7, and 10; and ZmIDD clade 3 genes were localized on chromosomes 1, 2, 3, 5, 7, and 8. Then, we detected gene duplication and found 8 segmental duplication events in ZmIDDs (Figure 6A; Table S5). It indicated that segmental duplication was the primary reason for the enlargement of ZmIDD genes in maize. Furthermore, a synteny analysis between the maize and rice genomes was also performed. It found that three IDD gene pairs (ZmIDD13/OsIDD9, ZmIDD1/OsIDD9, and ZmIDD17/OsIDD11) were found in Zea mays and Oryza sativa (Figure 6B; Table S6).

To illustrate the evolutionary constraints acting, we calculated the Ks value, Ka value, Ka:Ks ratio, and divergence time of paralogous IDD genes. The majority of Ka:Ks ratios in segmental duplicated ZmIDD gene pairs were less than one, with the exception of ZmIDD15/ZmIDD3 with 1.1, and divergence time occurred between 6.788 Mya and 112.101 Mya ago (Table S5).

2.4. Cis-Elements Analysis in ZmIDDs Promoters

To figure out the function and regulatory pattern of ZmIDD genes, we scanned the promoter sequence of 22 ZmIDDs to analyze cis-elements, such as ABRE, AuxRR-core, MBS, TCA-element, TC-rich repeats, and LTR, related to ABA, auxin, drought-inducibility, salicylic acid, defense and stress, and low temperature responses (Figure 7; Table S7). Generally, 15 ZmIDDs (68%) had ABRE cis-elements, 7 ZmIDD genes (32%) owned AuxRR-core elements, 12 ZmIDD genes (46.7%) owned MBS cis-elements, and 3 (14%) ZmIDD genes carried LTR cis-elements. Six ZmIDD genes owned TCA elements, and four ZmIDD genes carried TC-rich repeats.

2.5. Tissue-Specific Expression Patterns of ZmIDDs

A number of reports have shown that IDD genes are expressed in many tissues of plants. For instance, some IDDs are mainly expressed in mature leaves [29], and some in immature leaves [10] or roots [30]. To determine the tissue-specific expression profiles of ZmIDDs, the transcription levels of 22 ZmIDDs in six tissues of B73, such as roots, leaves, stems, embryo, endosperm, and pericarp, were compared. Based on the difference in expression patterns, 22 ZmIDD genes were categorized into three groups (Figure 8A; Table S8). Group 1 consists of one gene (ZmIDD9), which is not expressed in six tissues. Group 2 contains 11 genes expressed only in certain tissues. For example, ZmIDD13 was expressed only in the root but not in other tissues. Group 3 had 10 genes expressed in all tissues. Moreover, group 3 could be divided into two subgroups. Subgroup 1 contains four ZmIDDs with a high transcription level (log2TPM + 1 > 1) in six tissues, including ZmIDD8, ZmIDD14, ZmIDD17, and ZmIDD22. The rest of the six genes belong to subgroup 2.

2.6. Stress-Induced Expression Patterns of ZmIDDs

We collected transcriptome data from ZmIDD genes in maizeGDB and compared gene expression patterns when plants were subjected to drought, salt, heat, cold, and ultraviolet light. Generally, the expression of ZmIDD genes displayed significant differences under diverse abiotic stresses (Figure 8B; Table S9). Interestingly, most ZmIDD genes showed down-regulation or no change in response to abiotic stress except for ZmIDD8 and ZmIDD21. For example, ZmIDD8 and ZmIDD21 exhibited up-regulation after salt and cold stress, respectively. Meanwhile, many ZmIDD genes were inhibited when plants were subjected to abiotic stress. For instance, the expression of ZmIDD6 and ZmIDD11 was decreased under cold and drought treatments, respectively. ZmIDD15 showed down-regulation under heat, salt, and cold stress. However, there are some ZmIDDs that were not in response to all the abiotic stresses analyzed in the current study, such as ZmIDD7, ZmIDD9, ZmIDD10, ZmIDD13, and ZmIDD19. In addition, ZmIDD8 displayed opposite expression patterns under heat and salt stress.

2.7. ZmIDDs-Regulated Genes and miRNA-Targeted ZmIDDs

There were 6 ZmIDDs (ZmIDD 1, 4, 5, 9, 10, and 13) that could bind the cis-elements and mediate the transcription of 180 downstream genes in maize. The detailed information is listed in Table S10. The GO enrichment analysis concerning downstream genes demonstrated that five ZmIDD-targeted genes were involved in protein dimerization activity (GO:0046983) and four ZmIDD-targeted genes were directed hydrolase activities, hydrolyzing O-glycosyl compounds (GO:0004553) (Figure 9; Tables S11 and S12).

To determine whether a miRNA-mRNA regulatory network exists in maize, 321 known miRNAs were scanned. Finally, 22 miRNAs were predicted to inhibit the transcription level of ZmIDDs (Figure 10; Table S13). In short, miRNA-targeted ZmIDDs were divided into 5 networks: group 1/4/5/6 involving ZmIDD 4/7/18/5, respectively; group 2 involving two ZmIDDs (ZmIDD8 and ZmIDD17); and group 3 involving three ZmIDDs (ZmIDD 2/3/21) (Figure 10; Table S13).

3. Discussion

Transcription factors have regulatory and functional roles in plant growth and development [31]. The indeterminate domain (IDD) genes exist universally in all plants. In Zea mays, only three IDDs have been functionally verified: INDETERMINATE1 (ID1), the dominant regulator in flowering [14,32], and Naked Endosperm 1 and 2 (NKD1 and NKD2) associated with seed development [26,28]. Hence, whole-genome analysis and function prediction of the ZmIDD genes were necessary in maize. In this study, we identified 22 ZmIDD genes, which were named ZmIDD1 to ZmIDD22 according to their chromosome localization in maize (Figure 5; Tables S1 and S2). Three IDDs (ID1, NKD1, and NKD2) correspond to ZmIDD5, ZmIDD7, and ZmIDD22 identified in this work, respectively. Gene duplication and diversification play an essential role in plant evolution [33]. Prochetto et al. [19] reported that IDD derived from Chalorophyta and experienced a gene duplication event about 470 million years ago (MYA). Several times, duplication led to the emergence and diversification of terrestrial plants. Hence, monocots and dicots owned the same IDD gene numbers (16–23) [10,34,35]. In the current study, eight segmental duplication events were found in ZmIDDs (Figure 6; Table S5). Meanwhile, divergence time occurred between 6.788 Mya and 112.101 Mya ago, and most Ka/Ks values were less than one (Table S5), implying that the ZmIDD gene family may have suffered robust purifying selective pressure in the course of evolution.

Plant hormones, as important signaling molecules, can control plant growth, development, and stress responses [36]. Previous research has reported that many IDDs are related to hormone homeostasis [12,37,38]. For example, IDD14, IDD15, and IDD16 can cooperatively function in lateral organ morphogenesis and gravitropism through accelerating auxin synthesis and transport in Arabidopsis [39]. As shown in Figure 1, Figures S1 and S2, AtIDD14 was a paralog of AtIDD15 and AtIDD16, and ZmIDD2, ZmIDD3, ZmIDD6, ZmIDD15, and ZmIDD21 were orthologous genes of IDD14, IDD15, and IDD16 in Arabidopsis. Meanwhile, the promoter of 7 ZmIDDs has an AuxRR-element involved in auxin biosynthesis, including ZmIDD1, ZmIDD2, ZmIDD6, ZmIDD7, ZmIDD9, ZmIDD13, and ZmIDD15 (Figure 7; Table S7). It was suggested that although ZmIDD3 and ZmIDD21 were orthologs of AtIDD14, the promoter of ZmIDD3 and ZmIDD21 does not have the AuxRR-element, thus developing different functions in plant growth. Wang et al. [40] reported that miRNA167 could directly regulate the auxin response factors GmARF8a and GmARF8b and participate in lateral root development in soybean. In this study, miR167 could target and lead to the degradation of ZmIDD7 (Figure 10; Table S13). In addition, xyloglucan endotransglucosylase/hydrolase (XTH) is regulated by auxin and functions in plant developmental plasticity [41]. Gullner et al. [42] reported that glutathione-S-transferases (GSTs) were involved in intracellular auxin transport. Here, ZmIDD1 and ZmIDD13 could target xyloglucan endotransglucosylase/hydrolase protein 32 and glutathione S-transferase GST 21 (Table S10). The results demonstrated that ZmIDD may also be responsible for auxin biosynthesis, thus regulating plant growth.

Except for the zinc finger domain, the IDD proteins owned two conserved amino acid residues (Ser73 and Ser182): Ser73 could be phosphorylated by MPK6 [38], and Ser182 has been reported to be modified by AKIN10 (the catalytic subunit of SnRK1) [37]. Here, all 22 ZmIDDs have Ser73 and Ser182 (Figure 3). Meanwhile, the amino acid length of 22 ZmIDD proteins ranged from 354 to 815 (Table S1). It indicated that although the ZmIDDs contained a highly similar ID domain in their N-terminal, the flank sequences varied greatly. Jeong et al. [37] reported that phosphorylation of AtIDD8 at Ser-182 obviously decreased its transcriptional activity, and atidd8 mutants or over-expression of AKIN10 led to a delay in flowering in Arabidopsis. In the current study, protein-protein interactions (PPI) between ZmIDDs and ZmSnRKs were predicted using the STRING web server. Only one ZmSnRK protein (ZmSnRK1.1) that could interact with the ZmIDD protein (ZmIDD5) was found (Figure S3). ZmIDD5 exhibited expression in most tissues except for the root and showed down-regulation under cold and heat stress (Figure 6; Tables S8 and S9). Meanwhile, miR156 might bind and cause the cleavage of ZmIDD5 (Figure 10; Table S13). It was reported that miR156 could enhance cold and heat stress tolerance by repressing the expression level of transcription factors in plants [43,44]. It suggests that ZmIDD5 may also respond to abiotic stresses and regulate flowering time. The starch degradation in guard cells can cause stomatal opening [45], which is likely to increase guard cell turgor pressure by supplying soluble sugars. Furthermore, Seo et al. [30] discovered that AtIDD14 directly activated qua-quine starch, thereby mediating starch metabolism in Arabidopsis. Hence, IDD14-regulated starch degradation is likely to be responsible for stomatal opening. In this study, ZmIIDD6 and ZmIDD11 were down-regulated under drought stress, but other ZmIDDs do not respond to drought responses (Figure 8; Table S9). The promoters of ZmIDD6 and ZmIDD11 have an ABRE element and a MBS element, respectively (Figure 7; Table S7). It indicated that ZmIDD could also respond to drought stress in maize. The Indeterminate Domain Protein ROC1 (the ortholog of ZmIDD21) enhanced chilling tolerance by activating DREB1B/CBF1 in rice [24]. Phytochrome A signal transduction 1 (PAT1) could interact with IDD3 to activate lipoxygenase 3 (LOX3) and increase JA-Ile accumulation in grape calli under low temperature stress [46]. In the current study, ZmIDD21 was up-regulated under cold stress (Figure 8; Table S9), implying ZmIDD21 was also a positive regulator of cold stress.

Generally, we conducted a genome-wide analysis of IDD genes in maize. It has a vital implication for further comprehending the biological functions of ZmIDDs. However, a lot of functional verification work needs to be conducted in future studies.

4. Materials and Methods

4.1. Identification of ZmIDDs in Zea mays

The genome and protein sequences of maize B73 were obtained from MaizeGDB (https://maizegdb.org/ (accessed on 12 October 2022)). With default parameters, the Hidden Markov Model (HMMER3.0) profile of the C2H2 protein domain (PF00096) was used to search the protein database in the maize genome [47]. Then, we used the maize genome database to conduct BLASP. Gene and protein sequences in Arabidopsis were downloaded from the Ensemble Plants database [48]. The longest transcripts were retained, and incomplete sequences were deleted. In addition, we used the SMART database to reconfirm sequences [49]. Putative ZmIDD genes were named ZmIDD1 to ZmIDD22 based on their chromosome localization. Moreover, protein characteristics were determined by the ExPASy tool (http://www.expasy.org/tools/ (accessed on 20 October 2022)) [50].

4.2. Phylogenetic Tree, Gene Structure, Conserved Motifs, and Cis-Elements Analysis

Multiple protein sequence alignments were conducted using ClustalW [51]. A phylogenetic tree was constructed by MEGA version 7.0 through the maximum likelihood method, Neighbor-Joining method, and Minimum Evolution method [52]. Firstly, the evolutionary history was inferred by using the Maximum Likelihood method based on the Jones-Taylor-Thornton (JTT) matrix-based model. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using a JTT model and then selecting the topology with the highest log likelihood value. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. Secondly, evolutionary history was inferred using the Minimum Evolution method. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches. The evolutionary distances were computed using the p-distance method and are in units of the number of amino acid differences per site. The ME tree was searched using the Close-Neighbor-Interchange (CNI) algorithm at a search level of 1. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. Finally, evolutionary history was inferred using the Neighbor-Joining method. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches. The evolutionary distances were computed using the p-distance method and are in units of the number of amino acid differences per site. In addition, we used iTOL to beautify the evolutionary tree (https://itol.embl.de/ (accessed on 21 October 2022)) [53].

The DNA structure, such as exon and intron arrangement, was detected by the Gene Structure Display Server (GSDS) tool (http://gsds.cbi.pku.edu.cn/ (accessed on 21 October 2022)) [54]. The conserved motifs were analyzed by the MEME program with default parameters [55] and annotated using the InterProScan database (http://www.ebi.ac.uk/Tools/pfa/iprscan/ (accessed on 1 November 2022)) [56]. The gene structures and motifs were then imaged using the TBtools software (v1.09832) [57].

Promoter sequences (−1.5 kb) of each ZmIDD gene were used to scan any potential cis-elements by PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/ (accessed on 12 November 2022)) [58]. Then, we used TBtools to visualize the architecture [57].

4.3. Chromosome Distribution and Synteny Block of ZmIDDs

We used the Circos tool to locate ZmIDD genes on maize chromosome [59]. The collinearity of the orthologous IDD genes between maize and rice was analyzed using MCScanX software [60]. The nonsynonymous substitution rate (Ka) and synonymous substitution rate (Ks) of ZmIDDs were calculated by the ParaAT tool [61]. The Ka:Ks ratio was then calculated using Calculator 2.0 software [62]. In addition, Ks/2λ was used to estimate gene duplication time [63], where λ = 1.5 × 10 − 8.

4.4. Modeling of 3D Structures of ZmIDDs and Zinc Finger Alignment

We used Swiss-Model (https://swissmodel.expasy.org/interactive/ (accessed on 12 January 2023)) to predict the 3D structure of ZmIDD proteins and the SAVES server to assess the 3D structure (http://saves.mbi.ucla.edu/ (accessed on 12 January 2023)) [64]. Meanwhile, the amino acid alignment of the zinc finger was performed by DNAMAN.

4.5. Prediction of ZmIDDs-Regulated Genes

To better comprehend the regulatory mechanism of ZmIDDs, the target genes regulated by ZmIDDs were predicted through the online PlantRegMap tool [65], where the species and parameters were set to Zea mays, Organ:All, Method:FunTFBS, and Mode:TF (retrieve targets). Then, Gene Ontology (GO) enrichment of target genes was analyzed by an online bioinformatics server (http://www.bioinformatics.com.cn/?p=1 (accessed on 1 February 2023)) with a p-value < 0.05.

4.6. Prediction of miRNA-ZmIDD Regulatory Networks

The miRNAs, which can regulate the ZmIDD gene expression, were predicted using the psRNATarget server with the following parameters: the penalty for a G:U pair was one, and the number of mismatches allowed in the seed region was zero [66]. Then, we used Cytoscape V3.8.2 software (https://cytoscape.org/download.html (accessed on 1 February 2023)) to image the interaction networks.

4.7. Expression Profile Analysis of ZmIDDs

The tissue-specific and abiotic stress-induced transcriptional data of ZmIDDs in maize B73 were downloaded from qTeller in MaizeGDB [67,68]. We used the DSEeq2 R package to conduct differential expression analysis, and the heatmaps were imaged by TBtools software. The genes with a |log2 ratio|≥1 were considered differentially expressed genes (DEGs).

5. Conclusions

In the current study, we identified and characterized 22 ZmIDD proteins that contained a complete IDD domain. On the basis of their amino acid sequences, the 22 ZmIDDs were categorized into three clades. In the course of the evolution of ZmIDD genes, segmental duplication events played a dominant role. Moreover, the cis-acting elements, gene expression, and regulatory network of ZmIDD families were also analyzed. These findings add to our understanding of the ZmIDD gene family’s characteristics and provide useful information for further functional characterization of ZmIDDs in maize climate resilience.