Genome-wide identification of the AINTEGUMENTA-like gene family and functional characterization of ZmANT1 in regulating leaf length in maize

Abstract

Background

AINTEGUMENTA-LIKE (AIL) transcription factors belong to the APETALA2/ethylene-responsive factor (AP2/ERF) superfamily and play critical roles in plant growth and development. Although their functions have been characterized in several model and crop species, systematic knowledge of the AIL family in maize remains limited.

Results

Here, we performed a genome-wide identification of the maize AIL family and uncovered nine ZmAIL genes. These genes are asymmetrically distributed across six chromosomes and exhibit highly conserved gene and protein structures. Phylogenetic analysis grouped them into two clades: the ZmANT clade (ZmANT1-4) and the ZmAIL clade (ZmAIL1-5). Synteny analysis revealed that ZmAIL genes are more closely related to AIL genes from monocots than to those from dicots. Analysis of promoter cis-acting elements and expression profiles indicated that ZmAIL genes respond to environmental, hormonal, and developmental cues. Their expression patterns fall into two main categories: constitutively expressed and tissue-specific. Phenotypic analysis of a zmant1 mutant showed that ZmANT1 negatively regulates maize leaf length. Further experiments demonstrated that ZmANT1 directly binds to a specific cis-acting element in the promoters of target genes and regulates their expression, including four other ZmAIL members, thereby influencing multiple developmental pathways.

Conclusion

Overall, this study systematically identifies the maize AIL family and suggests a role for ZmANT1 in restricting leaf elongation. Our results provide a resource for future AIL functional studies and offer insights into the genetic mechanisms of leaf development in maize.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12870-026-09402-z.

Background

Transcription factors regulate plant growth, development, hormone signal transduction, and responses to environmental stresses [1, 2]. The AP2/ERF (APETALA2/ethylene-responsive factor) superfamily is one of the largest transcription factor families in plants [3]. It is divided into five subfamilies based on the number of AP2 domains and DNA-binding specificity: AP2, DREB, ERF, RAV, and Soloist [4]. The AP2 subfamily is characterized by two AP2 domains. It comprises two main lineages: euAP2 and ANT. These lineages are distinguished by the presence or absence of a miR172-binding site and specific amino acid insertions. The ANT lineage further diverges into basalANT and euANT lineages. The euANT lineage possesses a longer pre-domain region and four conserved motifs: euANT2–4 in the pre-domain and euANT1 in the AP2-R1 domain [5, 6]. In Arabidopsis thaliana, proteins of the euANT lineage are collectively defined as AINTEGUMENTA-like (AIL) transcription factors [7].

The first identified member of this family, AINTEGUMENTA (ANT), was initially characterized in Arabidopsis for its role in ovule and floral organ development [8]. Subsequent studies showed that ANT also functions in shoot and floral meristem maintenance, organ size and polarity, flower initiation, floral organ identity, and cell proliferation. In addition to ANT, seven other AIL transcription factors have been identified in Arabidopsis: AIL1, AIL2/BABY BOOM (BBM)/PLETHORA 4 (PLT4), AIL3/PLT1, AIL4/PLT2, AIL5/PLT5, AIL6/PLT3, and AIL7/PLT7 [7]. Functional analyses have revealed diverse roles for these genes. Overexpression of AIL2 in Arabidopsis induces somatic embryo development on leaf margins, petioles, and shoot apices [9]. Similarly, ectopic expression of AIL5 leads to embryoid formation on cotyledons, dedifferentiation of several tissues into callus, and impaired root development [10]. Mutants of ail5 exhibit reduced primary seed dormancy and decreased abscisic acid (ABA) sensitivity during seed germination [11–13]. AIL5, AIL6, and AIL7 also mediate responses to mechanical wounding, promoting wound healing and vascular regeneration [14]. Furthermore, the ail5/ail6/ail7 triple mutant shows alterations in phyllotaxis and root architecture [15, 16]. A key feature of this family is its dosage-dependent activity in plant development. For instance, single mutants of ail3, ail4, and ail6 show slight but significant reductions in root growth rate and meristem cell number. However, the ail3/ail4 double mutant exhibits severe root shortening, and the ail3/ail4/ail6 triple mutant displays a rootless phenotype [17, 18].

Beyond Arabidopsis, AIL transcription factors have been extensively studied and shown to play diverse roles in other plant species. In embryogenesis, rice (Oryza sativa) BBM1 not only regulates embryo development but has also been used to induce clonal seed production through synthetic apomixis [19]. During vegetative growth, AIL genes exhibit functional diversification while maintaining core roles in cell proliferation. In leaf development, maize ZmANT1 coordinates cell proliferation with vascular patterning, chloroplast biogenesis, and photosynthetic capacity [20]. Poplar LcAIL5 primarily modulates leaf morphology [21]. Medicago truncatula MtANTs determine leaf size by regulating cell proliferation during secondary morphogenesis [22]. Ectopic expression of Chinese cabbage BrANT-1 in Arabidopsis increases stomatal density and delays senescence [23], further supporting the conserved role of AIL genes in leaf development. In reproductive development, AIL genes display marked spatiotemporal specificity. OsPLT8 and OsPLT9 antagonistically modulate rice panicle architecture [24]. Similarly, AIL genes control organ size by regulating cell proliferation in fruits and flowers, as exemplified by MdANT1/2 in apple [25], VviANT1 in grape [26], and RrANT1 in rose [27]. Furthermore, AIL genes regulate dormancy cycles, as shown by VcANT-mediated regulation of VcTCP18 in blueberry bud dormancy [28]. Together, these findings reveal both conserved developmental modules and lineage-specific innovations in AIL-mediated growth regulation.

Maize (Zea mays L.) is one of the most important global food and feed crops and is also widely used as a model species in genetics, developmental biology, and molecular biology research [29, 30]. Although the roles of AIL genes in plant growth and development have been demonstrated in diverse species [31–33], the AIL gene family in maize has not been systematically characterized. To fill this gap, we performed a genome-wide identification of ZmAIL genes. Nine members were identified, and their physicochemical properties, phylogenetic relationships, synteny, gene and protein structures, cis-acting elements, and expression patterns across tissues were systematically analyzed. We further examined the phenotype of the zmant1 mutant and explored the molecular mechanism by which ZmANT1 regulates leaf length. These findings provide valuable insights into the functions of ZmAIL genes and the genetic mechanisms underlying maize growth and development, and lay a foundation for future research and maize breeding.

Methods

Plant material and phenotyping

The maize inbred line B73 and the zmant1 mutant (B73 background) were grown under controlled conditions with a 16/8 h light/dark photoperiod at 25 ± 1 °C (light) and 20 ± 1 °C (dark). The zmant1 mutant, generated by ethyl methanesulfonate (EMS) mutagenesis of B73, was obtained from the Maize EMS Mutant Database (MEMD; https://maizeems.qlnu.edu.cn/) under accession EMS5-04bfea [34].

For phenotypic analysis, three biological replicates were performed, each consisting of ten plants per genotype. At the V3 stage (defined as the visibility of the collar on the third leaf) [35], the length of the second fully expanded leaf was measured from the auricle to the leaf tip.

Genome-wide identification of AIL genes in maize

A genome-wide identification of the AIL protein family in maize was performed using a systematic bioinformatics approach. The maize B73 reference genome (version v5) and the amino acid sequences of eight Arabidopsis AIL proteins were retrieved from the Maize Genetics and Genomics Database (MaizeGDB; https://maizegdb.org/) and The Arabidopsis Information Resource (TAIR; https://www.arabidopsis.org/), respectively [36, 37]. Using these Arabidopsis AIL sequences as queries, a BLASTP search against the maize proteome was conducted with TBtools-II software (v2.210) to identify candidate proteins [38].

To verify the presence of conserved domains, the candidate protein sequences were screened using a hidden Markov model (HMM) based on the AP2 domain (PF00847) [39, 40]. Domain architecture was then analyzed using the CD-search (Conserved Domain Database; https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) and SMART (Simple Modular Architecture Research Tool; https://smart.embl.de/) databases to confirm the existence of two complete AP2 domains in each candidate [41]. Only proteins that exhibited the typical characteristics of the AIL family, including two intact AP2 domains and conserved structural features, were retained for further analysis.

Additionally, the physicochemical properties (molecular weight and theoretical isoelectric point) and subcellular localization of the identified AIL proteins were predicted using the ExPASy ProtParam tool (https://web.expasy.org/protparam/) and the UniProt database (https://www.uniprot.org/), respectively [42, 43].

Phylogenetic analysis

AIL protein sequences from rice were retrieved from the Rice Genome Annotation Project database (https://rice.uga.edu/) [44]. A multiple sequence alignment (MSA) of AIL proteins from Arabidopsis, rice, and maize was performed using ClustalW in MEGA11 [45]. Based on this alignment, a phylogenetic tree was constructed using the neighbor-joining (NJ) method with the Jones-Taylor-Thornton (JTT) substitution model. To evaluate tree reliability, 1000 bootstrap replicates were performed. The resulting tree was visualized and annotated using iTOL (Interactive Tree of Life; https://itol.embl.de/) [46].

Gene structure, conserved motifs, and cis-acting element analysis

The exon–intron structures of ZmAIL genes were visualized using TBtools‑II. Conserved motifs in ZmAIL proteins were predicted with the MEME online tool (https://meme-suite.org/meme/), with the maximum number of motifs set to 10 and other parameters at default values [47]. Cis‑acting elements in the 2‑kb region upstream of the transcription start site (TSS) of each ZmAIL gene were identified using the PlantCARE database (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/) [48]. The distribution of these conserved motifs and cis‑acting elements was subsequently visualized using TBtools‑II.

Chromosomal location and syntenic analyses of ZmAIL genes

The chromosomal locations of ZmAIL genes were visualized using TBtools-II based on their positions in the maize reference genome. Syntenic relationships among ZmAIL genes were analyzed using MCScanX implemented in TBtools-II [49]. To examine synteny between maize and other species, reference genome data for rice, peanut (Arachis hypogaea), soybean (Glycine max), potato (Solanum tuberosum), barley (Hordeum vulgare), and common wheat (Triticum aestivum) were retrieved from the Ensembl Plants database (http://plants.ensembl.org) [50]. Syntenic relationships between maize AIL genes and those in these six species were then systematically analyzed.

Transcriptome analysis

RNA-seq data from ten maize tissues were obtained from the ArrayExpress database under accession number E-MTAB-8628 [51]. According to the original study, each tissue was represented by two biological replicates, each composed of mRNA pooled from three individual plants. Expression levels of ZmAIL genes were quantified using Salmon (v1.9.0) [52], and transcript abundance was normalized to transcripts per million (TPM). For each gene and tissue, the mean TPM of the two replicates was calculated and visualized as log₂ (TPM + 1) in a bubble plot using TBtools‑II. The TPM values for all ZmAIL genes across tissues are provided in Additional file 14: Table S10.

RNA extraction and qPCR of ZmAIL genes

Maize samples were flash-frozen in liquid nitrogen and ground to a fine powder. Total RNA was extracted using RNA isolater Total RNA Extraction Reagent (Vazyme Biotech Co., Ltd., Nanjing, China). RNA concentration and purity were assessed using a NanoDrop spectrophotometer (Thermo Fisher Scientific, USA). For each sample, 1 µg of total RNA was used for genomic DNA removal and first-strand cDNA synthesis with the HiScript II Q RT SuperMix for qPCR (+ gDNA wiper) kit (Vazyme Biotech Co., Ltd., Nanjing, China). Gene-specific primers were designed using Primer Premier 5 software. Quantitative real-time PCR (qPCR) was performed on a QuantStudio 6 Real-Time PCR System (Thermo Fisher Scientific, USA) using ChamQ SYBR qPCR Master Mix (Low ROX Premixed) (Vazyme Biotech Co., Ltd., Nanjing, China) in 20-µL reaction volumes. Three technical replicates were included for each biological replicate. The maize Cullin gene was used as an internal reference, and relative expression levels were calculated using the 2⁻^ΔΔCt method.

Subcellular localization

To determine the subcellular localization of ZmANT1, its full-length coding sequence (CDS) without the stop codon was amplified by PCR and cloned into the pBin-eGFP vector using the MonClone™ Single Assembly Cloning Mix kit (Monad Biotech Co., Ltd., Nanjing, China). The resulting ZmANT1-eGFP fusion construct was introduced into Agrobacterium tumefaciens strain GV3101 by the freeze–thaw method. For transient expression, Agrobacterium-mediated transformation was performed by infiltrating fully expanded leaves of four-week-old Nicotiana benthamiana plants. After incubation at 25 °C for 48 h, eGFP fluorescence was observed under a confocal laser scanning microscope (LSM880; Zeiss). An empty pBin-eGFP vector was used as a control in all experiments.

Yeast one-hybrid assays

The full-length CDS of ZmANT1 was amplified by PCR and cloned into the pB42AD vector to generate the AD-ZmANT1 construct. Three tandem repeats of the core motif sequence (TGGCACAGTTCCCGAGGTGAA) were inserted into the pLacZi-2µ reporter vector to generate the 3×motif-LacZ reporter construct. A mutated version of the motif (TGCATATGAACAAGTGCAGAA, mutated nucleotides in italics) was similarly used to construct the Mu-3×motif-pLacZ reporter as a negative control. The following four combinations were co‑transformed into EGY48 yeast competecleotidnt cells using the PEG/LiAc method: AD‑ZmANT1 + 3×motif‑LacZ; AD‑ZmANT1 + Mu‑3×motif‑pLacZ; empty pB42AD + 3×motif‑LacZ; and empty pB42AD + Mu‑3×motif‑pLacZ. Transformed cells were plated on SD/‑Trp‑Ura selective medium and incubated at 28 °C for 3–5 days. Positive colonies were picked and streaked onto SD/‑Trp‑Ura medium containing galactose/raffinose (Gal/Raf) and X‑gal. After incubation at 28 °C for 1–2 days, LacZ activity was assessed based on blue color development.

Binding motif and downstream gene analysis

The DNA-binding motif of ZmANT1 was derived using the Plant TF Binding Motif Shift plugin in TBtools-II and visualized with LogoJS (https://logojs.wenglab.org/app/). Promoter sequences of all maize genes were extracted using TBtools-II. Putative target genes of ZmANT1 were predicted by scanning these promoters for the identified motif using FIMO (Find Individual Motif Occurrences; https://meme-suite.org/meme/tools/fimo) [53]. Functional enrichment analysis of the predicted target genes was performed using TBtools-II, including KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway and GO (Gene Ontology) enrichment. The enrichment results were visualized on the OmicShare Tools platform [54].

Results

Identification and characterization of ZmAIL genes in maize

Using eight Arabidopsis AIL proteins as queries, we performed BLASTP and HMM searches against the maize genome to identify candidate AIL genes. CD-search and SMART analyses confirmed the presence of two complete AP2 domains in each candidate, leading to the definitive identification of nine ZmAIL genes (Table 1 and Additional file 1: Table S1). Based on their chromosomal positions and a previous nomenclature [20], these genes were designated ZmANT1–ZmANT4 and ZmAIL1–ZmAIL5.

Table 1.

Characteristics of AIL transcription factors in maize

Gene	Gene ID	Chromosome	AA	MW (kDa)	pI	Subcellular localization
ZmANT1	Zm00001eb058850	chr1:288413629–288,417,775	642	67.40	6.68	Nuclear
ZmANT2	Zm00001eb299300	chr7:3283453–3,288,207	616	65.05	6.80	Nuclear
ZmANT3	Zm00001eb117830	chr2:241705784–241,708,890	603	63.70	6.89	Nuclear
ZmANT4	Zm00001eb399440	chr9:151104405–151,108,720	652	70.46	6.74	Nuclear
ZmAIL1	Zm00001eb005740	chr1:16251121–16,255,748	543	57.67	7.17	Nuclear
ZmAIL2	Zm00001eb067980	chr2:4892377–4,897,638	492	52.57	6.06	Nuclear
ZmAIL3	Zm00001eb144510	chr3:171481058–171,485,592	706	73.22	6.41	Nuclear
ZmAIL4	Zm00001eb247080	chr5:188009935–188,014,600	679	71.79	5.89	Nuclear
ZmAIL5	Zm00001eb389400	chr9:112080950–112,084,516	488	52.30	5.83	Nuclear

The encoded proteins range from 488 to 706 amino acids in length, with predicted molecular weights of 52.30–73.22 kDa and isoelectric points (pI) of 5.83–7.17. Notably, only ZmAIL1 has a pI above 7 (7.17), whereas the other eight ZmAIL proteins are acidic (pI 5.83–6.89). Subcellular localization predictions consistently placed all nine proteins in the nucleus.

Phylogenetic and classification analysis of ZmAIL genes

To investigate the evolutionary relationships of AIL proteins across species, we constructed a phylogenetic tree using sequences from Arabidopsis (n = 8), rice (n = 10), and maize (n = 9) with MEGA11 (Fig. 1 and Additional file 2: Table S2). The tree resolved these proteins into five distinct clades (Groups 1–5). Group 3 was the largest, containing three AtAILs, four OsAILs, and two ZmAILs. Group 1 contained AtANT, four ZmANTs, and three OsPLTs. Groups 2 and 5 comprised five and three members, respectively, whereas Group 4 contained only two AtAILs, representing the smallest clade.

Fig. 1.

Phylogenetic analysis of AIL proteins from Arabidopsis, rice, and maize. An NJ tree was constructed in MEGA11 based on a multiple sequence alignment of AIL protein sequences. Bootstrap values were calculated from 1000 replicates

Furthermore, maize AIL proteins clustered more closely with rice AIL proteins than with those from Arabidopsis, indicating closer evolutionary relationships and potentially conserved functions between maize and rice.

Gene structure and conserved motifs analysis

To characterize the structural features of the ZmAIL family, we analyzed the exon–intron organization and conserved protein motifs. ZmANT3 and ZmAIL3 each contain eight exons and seven introns, whereas the other seven ZmAIL genes each contain nine exons and eight introns (Fig. 2B). Using the MEME online tool, ten conserved motifs were identified across the nine ZmAIL proteins (Fig. 2C, D and Additional file 3: Table S3). Motifs 1–9 are present in all ZmAIL proteins, and motifs 1–7 correspond to the AP2-R1–linker–AP2-R2 domain region characteristic of AP2-like proteins. Notably, motifs 2, 9, 8, and 4 contain the euANT1, euANT2, euANT3, and euANT4 motifs, respectively (Additional file 4: Figure S1). Motif 10 is present exclusively in ZmANT1, ZmANT2, and ZmANT3, which cluster within the same evolutionary clade (Fig. 2A). Together, these results indicate that the gene and protein structures of the ZmAIL family are highly conserved.

Fig. 2.

Phylogenetic relationships, gene structures, and conserved motifs of ZmAIL proteins A An NJ tree constructed from full-length ZmAIL protein sequences. Bootstrap values are based on 1000 replicates B Exon–intron structures of ZmAIL genes. UTRs, CDS (exons), and introns are shown as green rectangles, yellow rectangles, and black lines, respectively C Conserved motifs in ZmAIL proteins identified by MEME. The ten motifs are represented by differently colored boxes D Sequence logos of the ten motifs

Chromosomal location and synteny analysis of ZmAIL genes

The nine ZmAIL genes were unevenly distributed across six maize chromosomes. Chromosomes 3, 5, and 7 each carried a single gene, whereas chromosomes 1, 2, and 9 each contained two genes (Fig. 3A). The physical distances between gene pairs on the same chromosome varied substantially. The closest pair, ZmANT4 and ZmAIL5, was located on chromosome 9 and separated by approximately 39 Mb, while the most distant pair, ZmAIL1 and ZmANT1, was located on chromosome 1 and separated by approximately 272 Mb. A similarly large distance of approximately 236 Mb separated ZmAIL2 and ZmANT3 on chromosome 2. This distribution pattern suggests that the ZmAIL gene family may have undergone chromosomal rearrangements and local duplication events during evolution.

Fig. 3.

Chromosomal location and synteny of ZmAIL genes A Chromosomal distribution of ZmAIL genes. Chromosome numbers and gene names are shown on the left and right, respectively. The vertical scale indicates chromosome length (0–350 Mb) B Syntenic relationships among ZmAIL genes. Red lines connect syntenic gene pairs

To further explore the evolutionary history of ZmAIL genes, we performed synteny analysis within the maize genome and between maize and six representative species. Within maize, syntenic relationships were detected for only three ZmAIL genes, namely ZmANT1 (Chr1), ZmANT2 (Chr7), and ZmANT3 (Chr2) (Fig. 3B). Cross-species analysis revealed substantially more syntenic gene pairs between maize and monocots (barley, rice, wheat) than between maize and dicots (peanut, soybean, potato). Specifically, only 2, 4, and 3 syntenic pairs were identified with peanut, soybean, and potato, respectively, whereas 11, 11, and 33 pairs were identified with barley, rice, and wheat (Fig. 4 and Additional file 5: Table S4). Among monocots, all ZmAIL genes except ZmAIL1 had syntenic pairs in all three species; ZmAIL1 lacked a syntenic pair in barley. In contrast, among dicots, ZmANT3, ZmANT4, ZmAIL2, and ZmAIL3 had no syntenic pairs in any of the three species, while the remaining five ZmAIL genes possessed only one or two pairs in certain dicot species. Conservation analysis further showed that ZmAIL5 had syntenic pairs in five of the six species (all except soybean); ZmANT1, ZmANT2, and ZmAIL4 had pairs in four species; and the other five genes had pairs in only three species (Additional file 6: Figure S2).

Fig. 4.

Synteny analysis of AIL genes between maize and six plant species. Gray lines represent collinear blocks across the genomes, and orange lines highlight syntenic AIL gene pairs

These findings indicate that the ZmAIL gene family underwent lineage-specific diversification after the divergence of monocots and dicots. Notably, ZmANT3, ZmANT4, ZmAIL2, and ZmAIL3 had syntenic gene pairs exclusively in monocots and were absent from all tested dicot species, suggesting that these genes may have originated after the monocot lineage diverged.

Cis-acting element analysis of ZmAIL genes

To investigate the potential biological functions of ZmAIL genes, we analyzed the cis-acting elements in their 2-kb promoter regions upstream of the TSS using PlantCARE and visualized the results with TBtools-II (Fig. 5A). A total of 226 cis-acting elements, representing 19 distinct types, were identified across the ZmAIL promoters (Additional file 7: Table S5). The number of element types varied among genes, ranging from 5 in ZmANT4 to 12 in ZmANT3. The total number of elements also differed, from 16 in ZmANT2 to 37 in ZmAIL1. Light-responsive elements were the most abundant type in every gene and were present in all ZmAIL genes.

Fig. 5.

Cis-acting elements in the 2-kb promoter regions of ZmAIL genes A Distribution of cis-acting elements. Different colors represent distinct element types B Heatmap showing the abundance of cis-acting elements. Color intensity reflects occurrence frequency

Based on their functions, the 19 element types were grouped into five major classes: hormone response, abiotic stress response, environmental signal response, developmental regulation, and metabolic pathway regulation (Fig. 5B). Environmental signal-responsive elements were the most abundant (86 elements, 38.05%), followed by hormone-responsive elements (69 elements, 30.53%). Abiotic stress-responsive elements accounted for 44 elements (19.47%), developmental regulation-related elements for 23 (10.18%), and metabolic pathway-related elements for only four (1.77%).

These results suggest that ZmAIL genes are primarily involved in light and hormone signaling, with additional roles in abiotic stress responses, developmental regulation, and metabolic processes.

Expression patterns of ZmAIL genes

To characterize the expression patterns of the nine ZmAIL genes, we analyzed RNA-seq data from ten maize (B73) tissues: seedling root, seedling shoot, V11 leaf base, V11 leaf middle, V11 leaf tip, V18 tassel, V18 ear, R1 anther, 16 days after pollination (DAP) endosperm, and 16 DAP embryo (Fig. 6, Additional file 15: Figure S5 and Additional file 16: Figure S6). Two distinct expression patterns were observed. ZmANT1–ZmANT4 clustered together and showed low-to-moderate expression across multiple tissues, indicating a broad expression profile. In contrast, ZmAIL1–ZmAIL5 formed a separate cluster and displayed high expression in specific tissues. Among all ZmAIL genes, ZmANT4 was the most broadly expressed, detected in all ten tissues, whereas ZmAIL3 exhibited the most restricted expression, with transcripts detected only in seedling root, seedling shoot, and 16 DAP embryo. All nine genes were expressed in seedling shoot and 16 DAP embryo. In V11 leaf tip, only ZmANT2, ZmANT3, and ZmANT4 were detected, all at low levels.

Fig. 6.

Expression patterns of ZmAIL genes across ten maize tissues. Bubble plot showing expression levels [log2(FPKM + 1)] based on RNA-seq data from the ArrayExpress database. Bubble size and color (red, high; blue, low) represent expression magnitude

These results indicate that ZmAIL genes display a spectrum of expression patterns, from broad (e.g., ZmANT4) to tissue-specific (e.g., ZmAIL3), suggesting diverse roles in maize growth and development, ranging from housekeeping functions to specialized regulatory processes.

Functional analysis of ZmANT1 in maize development

To investigate the role of ZmAIL genes in maize growth and development, we selected ZmANT1 as a representative member for functional characterization. The zmant1 mutant was obtained from the MEMD and confirmed by PCR and Sanger sequencing. Sequencing revealed a single nucleotide substitution in the third exon of ZmANT1 that introduced a premature stop codon (Additional file 8: Figure S3A). At the V3 stage, the second fully expanded leaf of the zmant1 mutant was significantly longer than that of wild-type B73 (Fig. 7A, B and Additional file 8: Figure S3B), suggesting that ZmANT1 acts as a negative regulator of leaf length.

Fig. 7.

Functional characterization of ZmANT1 A Whole-plant phenotypes of wild-type B73 and zmant1 seedlings. Scale bar = 2 cm B Quantification of the second fully expanded leaf length in B73 and zmant1 C Yeast one-hybrid assay confirming direct binding of ZmANT1 to the identified cis-acting element D Subcellular localization of the ZmANT1-eGFP fusion protein in Nicotiana benthamiana leaves. Scale bar = 10 μm E GO enrichment analysis of putative target genes of ZmANT1 F KEGG enrichment analysis of putative target genes of ZmANT1. G Relative expression of ZmANT2, ZmANT3, ZmANT4, and ZmAIL1 in B73 and zmant1. Data are means ± SD (three biological replicates). P‑values were determined by two‑tailed Student’s t‑test and are indicated above the bars

We next determined the subcellular localization of ZmANT1 by transiently expressing the ZmANT1-eGFP fusion protein in Nicotiana benthamiana leaves. Confocal microscopy showed that the eGFP signal co-localized with a nuclear marker, confirming that ZmANT1 localizes to the nucleus (Fig. 7D).

To further explore the molecular mechanism, we derived the DNA-binding motif of ZmANT1 using TBtools-II and confirmed its specific binding to this motif by yeast one-hybrid assays (Fig. 7C and Additional file 9: Figure S4). Scanning the promoters of all maize genes with this motif using FIMO identified 5,956 putative target genes (Additional file 10: Table S6). GO enrichment analysis revealed significant over-representation of terms related to cell proliferation and energy metabolism (Fig. 7E and Additional file 11: Table S7). KEGG enrichment analysis further identified eight significantly enriched pathways, including sulfur metabolism, pyruvate metabolism, and brassinosteroid biosynthesis, representing key metabolic and hormonal pathways (Fig. 7F and Additional file 12: Table S8). These findings suggest that ZmANT1 may influence leaf development by regulating genes associated with cell proliferation, energy metabolism, and hormone signaling.

Notably, among the putative target genes identified by genome‑wide motif scanning, four ZmAIL members (ZmANT2, ZmANT3, ZmANT4, ZmAIL1) were found to harbor the ZmANT1 binding motif in their promoters (Additional file 17: Figure S7). qPCR analysis confirmed that these genes were significantly downregulated in the zmant1 mutant compared with B73. ZmANT2 was reduced to 0.36‑fold (p = 0.037), ZmANT3 to 0.31‑fold (p = 0.014), ZmANT4 to 0.24‑fold (p = 0.016), and ZmAIL1 to 0.15‑fold (p = 0.025) of the wild‑type level (Fig. 7G). This suggests that ZmANT1 may directly regulate the expression of other ZmAIL genes, thereby participating in an endogenous regulatory network that controls leaf growth.

Discussion

AIL transcription factors belong to the AP2 subfamily of the AP2/ERF superfamily and are key regulators of plant growth and development. Although these factors have been identified and characterized in multiple species, systematic knowledge of the AIL family in maize has been lacking. In this study, we performed a comprehensive characterization of the ZmAIL family, revealing its evolutionary relationships, structural features, and expression patterns. Our results further suggest that ZmANT1 likely functions as a negative regulator of maize leaf length.

Phylogenetic analysis of AIL proteins from Arabidopsis, maize, and rice revealed a clear divergence between monocot and dicot lineages. Maize AIL proteins clustered more closely with rice AIL proteins in multiple branches, whereas Arabidopsis AIL proteins formed largely independent clades. This pattern provides initial evidence for lineage-specific evolution of the AIL gene family, consistent with the broader phenomenon of lineage-specific expansion in plant gene families as an adaptive response to diverse environmental pressures [55]. Synteny analysis further supports this divergence: maize retained substantially more syntenic gene pairs with monocots (barley, rice, and wheat) than with dicots (peanut, soybean, and potato) [56]. Notably, ZmANT3, ZmANT4, ZmAIL2, and ZmAIL3 had no syntenic gene pairs in any of the three dicot species examined and were also positioned distantly from their Arabidopsis homologs in the phylogenetic tree. Integrating phylogenetic and syntenic evidence, we propose that the AIL gene family underwent a pronounced divergence between monocots and dicots. Furthermore, ZmANT3, ZmANT4, ZmAIL2, and ZmAIL3 likely represent members that evolved or experienced lineage-specific diversification within the Poaceae after the monocot–dicot split. Expanding the species sampling in future studies will help validate and refine this evolutionary scenario.

Phylogenetic analysis of maize AIL proteins resolved two major clades, corresponding to the ANT and AIL subclades. In addition to the known euANT1–4 motifs [5], we identified a novel conserved sequence (SMPLKSDGSLCI) within the pre-motif region upstream of the euANT3 motif in ZmANT1–3 (Additional file 3: Table S3). This motif was also present in OsPLT7 and OsPLT8, which clustered within the same evolutionary branch as ZmANT1–3, but was absent from Arabidopsis AIL proteins. Given that similar lineage-specific motifs contribute to neofunctionalization in other transcription factor families [57], this newly identified motif may play an important biological role. However, the biological function of this motif has not been experimentally validated, and any specific role remains hypothetical.

Expression patterns of ZmAIL genes closely mirrored their evolutionary relationships. For example, the closely related ZmAIL1 and ZmAIL2 shared promoters enriched in light-, MeJA-, and ABA-responsive elements, and their expression profiles were highly similar: both were highly expressed in 16 DAP embryos and broadly detected across multiple tissues. This dual expression pattern suggests that they function both in general growth regulation and in specific developmental processes such as seed development. By contrast, ZmANT1–4 were expressed at low to moderate levels across diverse tissues without marked tissue preference, pointing to broader, housekeeping-like functions. ZmAIL3–5, however, exhibited restricted expression in only a few tissues, indicative of strong tissue specificity. This narrowed expression scope may reflect neofunctionalization during evolution, enabling these genes to acquire more specialized roles [58].

Collectively, these findings demonstrate that the ZmAIL gene family has diversified at both the sequence level and the expression level, with expression patterns tightly linked to evolutionary history. The ANT subclade likely retains ancestral, broadly acting regulatory functions, whereas certain members of the AIL subclade, such as ZmAIL3–5, may have evolved specialized functions through restricted expression domains. This functional divergence between the ANT and AIL subclades has also been reported in cotton [59], suggesting a conserved pattern of functional evolution within the AIL family.

AIL transcription factors have been shown to promote leaf growth in species such as Medicago truncatula and Setaria viridis [20, 22]. However, our results suggest that ZmANT1 acts as a negative regulator of leaf length in maize. This functional divergence among orthologs is not unusual. Notably, in rice, OsPLT9 and OsPLT8 both cluster in the same phylogenetic group (Group 1) as ZmANT1. OsPLT9 negatively regulates axillary meristem establishment and secondary branch formation in rice panicles, and its loss-of-function mutant displays increased secondary branches, spikelets, and elongated primary branches. OsPLT8 exerts a dual role, acting as a positive regulator of primary branch number yet a negative regulator of primary branch length and secondary branch number [24]. These observations suggest that AIL genes do not merely act as simple promoters or inhibitors of growth; rather, they exert precise, context-dependent restrictive or permissive effects on distinct meristem types at different developmental stages. We therefore hypothesize that ZmANT1 plays a role analogous to OsPLT9, restricting leaf growth by negatively regulating cell division or elongation. Loss of ZmANT1 function would relieve this restriction, leading to the excessive leaf elongation observed in the zmant1 mutant.

Transcription factors regulate target gene expression by specifically recognizing and binding cis-acting elements [60]. Identifying downstream target genes is therefore essential for understanding their functions. In Arabidopsis, ANGUSTIFOLIA3 (AN3), a member of the GRF-interacting factor (GIF) family, controls cell proliferation; loss-of-function an3 mutants produce smaller leaves, whereas AN3 overexpression results in enlarged leaves [61–64]. Here, we identified ZmGIF3 (Zm00001eb409680), a maize homolog of AN3, as a putative downstream target of ZmANT1, indicating that ZmANT1 may influence leaf development by regulating cell proliferation-related pathways.

In Arabidopsis, AUXIN-REGULATED GENE INVOLVED IN ORGAN SIZE (ARGOS) is strongly induced by auxin and promotes organ growth by extending the expression duration of ANT [65–67]. Similarly, overexpression of ZAR1, a maize ARGOS homolog, increases leaf area by promoting cell number [68], supporting functional conservation of the ARGOS-ANT regulatory module in organ growth. Further analysis identified additional ZmANT1 targets involved in auxin biosynthesis and transport, including YUCCA8 (Zm00001eb362000), PIN1 (Zm00001eb372180) and PIN4 (Zm00001eb196240) [69, 70]. In maize, PIN1 is directly regulated by LG1, and its overexpression leads to increased auricle area and leaf angle, while PIN4 accumulates in the preligule band and mediates polar auxin transport in the leaf epidermis [71, 72]. Therefore, by potentially regulating these auxin‑related genes, ZmANT1 may modulate auxin biosynthesis and distribution, thereby influencing leaf length. Taken together, we propose a working model in which auxin‑induced ARGOS may influence ZmANT1 function, and ZmANT1 may in turn modulate auxin signaling and cell proliferation through its predicted targets, ultimately affecting leaf length. (Additional file 18: Figure S8)

This study has several limitations that should be considered. First, our phylogenetic analysis included only three species. Broader sampling of monocot and dicot AIL sequences would help clarify the evolutionary divergence of the family. Second, the novel conserved motif identified in ZmANT1–3 remains functionally uncharacterized. Its biological relevance awaits experimental testing. Third, the possible origin of ZmANT3, ZmANT4, ZmAIL2 and ZmAIL3 after the monocot–dicot split was inferred from synteny analysis with three monocot and three dicot species. Extending this analysis to additional taxa is needed to test this hypothesis. Fourth, the RNA‑seq dataset used for expression profiling contained only two biological replicates. This limited the ability to perform robust statistical comparisons. Future studies using higher‑replicate transcriptomics combined with RT‑qPCR are necessary to confirm the expression patterns. Fifth, the functional characterization of ZmANT1 relied on a single EMS allele. Generating additional knockout and overexpression lines, and analyzing their phenotypes, would strengthen the conclusions. Sixth, our analysis of the zmant1 mutant was restricted to the second leaf at the V3 stage. The cellular basis of the increased leaf length—whether through increased cell number, cell size, or both—remains undetermined. Moreover, potential roles of ZmANT1 in other organs or developmental stages were not explored. Seventh, the direct binding of ZmANT1 to the promoters of candidate target genes was inferred from motif scanning and yeast one‑hybrid assays for the core motif. However, this binding has not been validated by in vivo assays such as ChIP‑qPCR or in vitro assays such as EMSA for individual promoters. Finally, the proposed regulatory model is based partly on cross‑species literature (auxin–ARGOS, ARGOS–ZmANT1) and on computationally predicted target genes (ZmANT1–PIN1/PIN4/YUCCA8 and other ZmAILs). All of these interactions require experimental verification in maize.

Conclusions

In this study, we systematically identified and characterized the ZmAIL gene family in maize and revealed a key role for ZmANT1 in regulating leaf development. Comprehensive analysis of the nine ZmAIL genes, which are unevenly distributed across six chromosomes, showed a close correspondence among their phylogenetic relationships, structural features, and expression patterns. Synteny analysis further indicated that ZmAIL genes share higher conservation with monocot orthologs than with dicot orthologs. This pattern highlights both evolutionary conservation and potential functional diversification within the family. Functional characterization showed that ZmANT1 is a nuclear-localized transcription factor and suggested that it acts as a negative regulator of leaf length. ZmANT1 directly binds to a specific cis-acting element and modulates the expression of downstream genes, including four other ZmAIL members. Collectively, our findings provide valuable insights into the maize AIL gene family and suggest a potential mechanism by which ZmANT1 regulates leaf development. They also offer a valuable resource for future functional studies of AIL transcription factors in plants.

Abbreviations

AIL

Aintegumenta-like

ANT

Aintegumenta

AP2/ERF

Apetala2/ethylene-responsive factor

BBM

Baby boom

PLT

Plethora

ABA

Abscisic acid

EMS

Ethyl methanesulfonate

MEMD

Maize EMS Mutant Database

MaizeGDB

Maize Genetics and Genomics Database

TAIR

The Arabidopsis Information Resource

HMM

Hidden Markov model

SMART

Simple Modular Architecture Research Tool

MSA

Multiple sequence alignment

NJ

Neighbor-joining

JTT

Jones-Taylor-Thornton

iTOL

Interactive Tree of Life

qPCR

Quantitative real-time polymerase chain reaction

CDS

Coding sequence

Gal

Galactose

Raf

Raffinose

FIMO

Find Individual Motif Occurrences

KEGG

Kyoto Encyclopedia of Genes and Genomes

GO

Gene Ontology

DAP

Days after pollination

AN3

Angustifolia3

GRF

Growth-Regulating Factor

GIF

GRF-interacting factor

ARGOS

Auxin-regulated gene involved in organ size

Authors’ contributions

LD and CC conceived and designed the experiments. LD performed the experiments, analyzed the data, and drafted the manuscript. XX and WY performed phenotyping. ZY and YX cultivated the plant materials. GZ and SJ collected plant samples. All authors participated in manuscript revision and approved the final version.

Funding

This research was supported by the National Key R&D Program of China (grant no. 2022YFD1201700 to C.C.) and the National Key R&D Program of China (grant no. 2023YFD1200501 to C.C.).

Data availability

The datasets supporting the conclusions of this article are included within the article and its additional files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.