Regulatory role of AGC genes in heat stress adaptation in maize (Zea mays)

Identification and sequence analysis of AGC gene family in Zea mays

The Arabidopsis Information Resource (TAIR) database (https://www.arabidopsis.org/) (Garnacho-Montero and Amaya-Villar 2010) and the National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/) (Geer et al. 2010) were used to classify members of the AGC gene family. Protein Blast is a widely accepted tool for identifying homologous sequences. The TAIR is specifically designed for Arabidopsis thaliana L., making it a primary resource for researchers studying this model plant and NCBI databases, being comprehensive repositories of genetic information, offer a diverse set of sequences for comparative analysis. Protein blast from NCBI (https://blast.ncbi.nlm.nih.gov) was used (default parameters) to retrieve homologues for the protein sequences of the AGC gene family of Z. mays L.

Phylogenetic analysis of AGC gene family in Z. mays

The phylogenetic tree was built using protein sequences of the AGC gene family from A. thaliana and homologues from Z. mays. The choice of A. thaliana as a reference for Z. mays AGC genes holds significance due to several reasons. A. thaliana is a well-studied model plant with a fully sequenced genome, and its AGC gene repertoire is relatively well-characterised. Using A. thaliana as a reference allows for comparative analysis, leveraging the existing knowledge base to infer potential functions and regulatory mechanisms of Z. mays AGC genes. Furthermore, the evolutionary conservation of AGC genes between A. thaliana and Z. mays provides insights into the functional and structural aspects of these genes. The maximum likelihood tree was generated through default parameters in IQ-TREE (http://iqtree.cibiv.univie.ac.at/) (Nguyen et al. 2015). IQ-TREE, a prominent tool for phylogenetic tree reconstruction, excels in maximum likelihood inference, robustly estimating evolutionary trees for accurate reconstructions. Its advanced model selection and provision of reliable branch support values, alongside a user-friendly interface, make it accessible and efficient for researchers in phylogenetic analyses. In addition, the tree was modified with iTOL (https://itol.embl.de/) (Letunic and Bork 2019).

Gene structures and motif analysis of AGC gene family in Z. mays

The gene strctures of AGC gene family in Z. mays was identified using the online server called Gene Structure Display Server (GSDS, http://gsds.gao-lab.org), which is a bioinformatics tool utilised for visualising and analysing the gene structure of DNA sequences, aiding in the identification of exon–intron boundaries and other structural features. Motifs and their locations were identified using the protein sequences of ZmAGC members. MEME (Multiple Em for Motif Elicitation) is a bioinformatics tool designed for discovering conserved motifs in DNA, RNA or protein sequences. It employs statistical algorithms to identify significant motifs, providing insights into potential functional elements across sequences. Therefore, MEME was employed to perform motif analysis using default settings (https://meme-suite.org/meme/tools/meme) (Bailey et al. 2015).

Physiochemical properties of AGC gene family in Z. mays

ProtParam (https://web.expasy.org/protparam/) was used to predict the physiochemical properties of AGC proteins, such as molecular weight, theoretical pI, number of amino acids and aliphatic index (Garg et al. 2016). Softberry ProtComp tool was used to predict the proteins’ sub-cellular localisation (http://www.softberry.com/berry.phtml?topic=protcomppl&group=programs&subgroup=proloc).

Domain analysis and protein-protein interaction analysis

The identification of conserved domains was carried out using the NCBI database CDD (Conserved Domain Database) available at https://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml, and the Pfam database (Marchler-Bauer et al. 2015). The results obtained from the analysis were visualised using a tool called Tbtools. The protein–protein interaction network (PPI) was built with the help of a STRING database (https://string-db.org) (Mering et al. 2003). The entered sequence data was transferred to this database and a protein interaction with other proteins in the family was built. In addition, the proteins were also divided into three groups using k-means clustering algorithm.

Promoter region and regulatory elements analysis of ZmAGC genes

The promoter and untranslated region (UTR) sections of every single gene in the AGC gene family were gathered and saved in a file from the NCBI nucleotide database. Cis-regulatory elements were subsequently predicted using the PlantCare tool, an online server-based tool that employs the utilising the gene’s promoter and UTR sequences for this analysis (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/) (Lescot et al. 2002). Because PlantCare can only handle only one sequence at a given moment, the process took around an hour or even longer to accomplish for all the genes. Subsequently, the outcomes were aggregated and visualised by utilising TBtools, which is an open-source software available on GitHub (https://github.com/CJ-Chen/TBtools) (Chen et al. 2020).

Gene expression analysis of ZmAGC members under heat stress

Data on heat stress transcriptomes were retrieved from the NCBI SRA database, accession number SRX19031355 (https://www.ncbi.nlm.nih.gov/sra) (Benson et al. 2010). The transcriptome data was then used to map the reads against the Z. mays reference genome Zm-B73-REFERENCE-NAM-5.0 using the Galaxy server. The data retrieved and organised by the RNA-seq tool Galaxy was used, and gene expression levels were assessed using the fragment mapping tool Cufflinks (which specialises in transcription, estimation, and testing of DEGs in RNA samples). This data was then used for expression profiling using another tool called TBtools.

RNA extraction, cDNA synthesis and real-time qPCR of AGC gene family in Z. mays

RNA isolation from Z. mays leaves was conducted using the Trizol method, and quantification was performed with the Thermo Nanodrop 2000. Subsequently, for the 1 mg of the First-Strand Synthesis (FSS) kit was used to synthesise cDNA from RNA. The resulting cDNA was stored at −20°C for future experiments. For quantitative real-time polymerase chain reaction (qPCR), the iTaq Universal SYBR Green Super Mix (Bio-Rad Labs, Hercules, CA, USA) was employed in conjunction with a qPCR (detection system (CFX96 Touch-TM RT PCR detection system). Gene-specific primers were designed using the Oligo Calculator (http://mcb.berkeley.edu/labs/krantz/tools/oligocalc.html/) (Kibbe 2007) and the NCBI Primer BLAST algorithm (https://www.ncbi.nlm.nih.gov/tools/primer-blast/) was used to validate gene-specific primers (Ye et al. 2012). T-test was employed for statistical analysis of expression data.

Targeting of PIN auxin efflux carriers by AGC kinases

The PPI network was assembled with data from the STRING database (https://string-db.org/), and the proteins were subsequently grouped into three distinct clusters using the k-means clustering technique. The protein structures were then predicted using an online server known as iTASSER (https://zhanggroup.org/I-TASSER/). To analyse and compare the protein sequences, we employed the MEGA alignment software (ver. 7, https://megasoftware.net/). In this analysis, the consensus regions or motifs were identified and prominently marked using black boxes.

Identification and sequence analysis of AGC gene family in Z. mays

The process of identifying biological sequences typically encompasses four stages. Initially, a reference sequence is identified, which in our study, involved selecting a member of the AGC Gene family from TAIR. This reference sequence is subsequently employed as a query for conducting a BLAST search against the plant sequence of interest, which, in our case, was Z. mays. The obtained results are then scrutinised for reproducibility and validated by examining a specific signature domain within the plant-specific protein sequence.

A total of 39 reference protein sequences from A. thaliana were retrieved from the TAIR database. Using these sequences, homologues in Z. mays were identified and retrieved from NCBI, resulting in the recognition of 39 homologues for further analysis. All sequences underwent a thorough re-examination to ensure that the conserved Pkinase domain is present. This domain was also discussed in the context of the analysis of AGC protein kinases within this cereal crop, i.e. ZmAGC members.

Phylogenetic analysis of the AGC gene family in Z. mays

Phylogenetic analysis is a powerful tool for studying the evolutionary relationships of genes or proteins between different species. Using the online IQ-TREE tool, a maximum likelihood parsimony tree was constructed and seven AGC subfamilies were generated. A total of 78 protein sequences were used in the phylogenetic analysis. Phylogenetic analysis of the AGC gene family in Z. mays was performed using Z. mays (ZmAGC) and A. thaliana (AtAGC) proteins. AGC gene sequences from Z. mays were collected and aligned to begin the phylogenetic analysis of AGC genes in Z. mays. Similarly, protein sequences identified in various groups all fall into the same category. In our study, we predicted the existence of several subfamilies within the AGC gene family in Z. mays. Specifically, we identified and counted the following subfamilies and their members: PDK (four members); AGC1 (34 members); AGC2 (eight members); AGC2-associated (four members); NDR (16 members); IRE (eight members); and S6K (four members). By comparing the subfamilies of Z. mays with A. thaliana, we observed a significant expansion in Z. mays. The expansion was most pronounced in subfamilies such as PDK, AGC1, AGC2, and the related AGC2, NDR, IRE and S6K, where the number of members in Z. mays was significantly greater than in A. thaliana. These differences in subfamily size between Z. mays and A. thaliana suggest different evolutionary trajectories or species-specific adaptations within the AGC gene family (Fig. 1).

Fig. 1.

Phylogenetic tree of AGCs from A. thaliana, and Z. mays. Each coloured range in the tree represents the main species which are Z. mays and the reference specie which is A. thaliana in each sub-family of AGCs. The PDK family is illustrated as an outgroup.

Physiochemical properties of AGC gene family in Z. mays

Table 1 shows the physiochemical properties of 39 AGC proteins. Our analysis found proteins with higher amino acid numbers and molecular weights. Specifically, ZmIRE-H1, ZmIRE3 and ZmIRE were among the proteins that exhibited these higher values. Theoretical predictions of the isoelectric points (pI) of the proteins gave an average value of 7.62. Furthermore, the proteins were predicted to be located in the nucleus, cytoplasm, plasma membrane and extracellular spaces of the cell, indicating their likely association with activities in the nucleus, cytoplasm, plasma membran, and extracellular spaces. Table 1 contains information on various other properties of the proteins.

Gene structures and motif analysis of AGC gene family in Z. mays

The intron and exon structures of ZmAGC members were analysed using the Gene Structure Display Server 2.0 to acquire a better knowledge of their structural properties. The study found that the number of introns in genes can vary (0–24). The ZmPHOT2 gene has the most introns, whereas ZmAGC2.2, ZmAGC2.3 and ZmAGC2.4 have no introns. Thus, the AGC1, AGC2 and related AGC and S6K gene families contain fewer intron and exon regions. Fig. 2b clearly shows that the site distribution of un-gapped motifs was chosen to be around 10. Motifs 1 and 2 were the most conserved across the entire ZmAGC kinase family. The motif logos was useful in predicting the functional motifs involved in AGC genes found in Z. mays. The site distribution of un-gapped motifs was chosen to be around 10. The motifs play an important role in performing their distinct function, and different colours are used in the figure to differentiate all of the conserved motifs as well as motif consensus.

Fig. 2.

Gene structure and motif analysis. (a) An online server Gene Structure Display Server 2.0 is used in which introns, exons, CDS, upstream and downstream sequences were emphasised. Coding sequence (CDS) regions are presented by purple bars, introns are represented by grey lines and upstream and downstream sequences are highlighted by red bars. The intron phase is shown by numbering 0, 1, 2. (b) Graphical representation of 10 conserved motifs found in all the protein sequences and their homologues. Each motif is represented by a distinct colour to differentiate it from the others. The presence and arrangement of these motifs are likely important for the function or structure of the proteins.

ZmAGC members domain analysis and protein-protein interaction analysis

Domain analysis of Z. mays revealed that members of the AGC gene family possessed conserved domains such as Pkinase, Pkinase_C and Per-Arnt-Sim (PAS). Pkinase or Pkinase_C was found in almost all proteins, whereas the PAS domain was found in only two proteins; (1) ZmHOT1; and (2) ZmHOT2. Protein–protein interactions revealed that members of the AGC gene family interacted with each other as well as with other proteins in the cell. ZmHOT1 and ZmHOT2 showed interaction with a discrete set of cellular proteins, while they showed weak interaction with the set of proteins that are members of the AGC gene family and other proteins with which they interact (Fig. 3).

Fig. 3.

Domain analysis and protein-protein interaction: (a) Protein domains of ZmAGCs were retrieved from NCBI CDD decamped. Pkinase is the most conserved domain. (b) Potein–protein interaction network was developed using the STRING database illustrating the most and least network of interacting proteins.

Promoter region and regulatory elements analysis of ZmAGC genes

The 2000-bp promoter sequences of 39 ZmAGC genes were examined for cis-regulatory elements. The analysis revealed 63 different types of regulatory elements, with the G-box being the most common. TGACG motif, AuxRR core, box-4, chs-CMA2a, 3-Af1 binding site, TC-rich repeats, AE-box, TATC-box, TCA, BOX-III and GAP-box were also common. Some elements were present in smaller numbers, such as B. HD Zip 1, GATT motif, GC motif, AT-rich element, WUN motif, LAMP elements, O2 site and I-Box. These results suggest that cis-regulatory elements play an important role in gene expression and shed light on the regulation of gene expression in Z. mays. Heat stress is mediated by TGACG motif, Sp1, CAAT box, G box, box 4, GCN4_Motif, AuxRR core, AT-rich element, TCA element, ARE, CGTCA motif and chs-CMA2a (Fig. 4).

Fig. 4.

(a) Cis-acting regulatory elements of AGCs gene family in Z. mays are presented according to its identified members. Furthermore, they are also represented by their ligands. (b) Frequency of cis-elements in promoter regions of AGCs.

Gene expression analysis of ZmAGC members under heat stress

In this study, Next-Generation Sequencing (NGS) data analysis was employed to investigate the genetic response of Z. mays kernels under heat stress conditions. The research focused on identifying heat-related genes and unravelling the underlying biological pathways that have a significant impact on heat stress tolerance also potentially providing valuable insights for crop improvement strategies in the face of climate change. The expression profile of ZmAGC genes in Z. mays under heat stress (Fig. 5). The results showed that ZmNDR5 and ZmIRE3 were significantly upregulated (with at least a 2-fold change) in response to heat stress. ZmNDR4, on the other hand, showed a slight upregulation, but the change was not statistically significant. Furthermore, genes such as ZmAGC2.2, ZmPID, ZmKIPK, ZmIRE, ZmS6K1 and other members of this gene family did not show significant changes in expression, suggesting that they were not affected by heat stress. In contrast, ZmNDR3, ZmAGC1.8, ZmAGC1.9 and ZmAGC1.5 showed significant downregulation in response to heat stress with a fold change of 0.5 or less. Similarly, ZmNDR8, ZmAtPK5 and ZmAtPK7 showed slight downregulation, albeit not as pronounced. Taken together, these findings highlight the diverse response of ZmAGC genes to heat stress in Z. mays, with some genes significantly upregulated, others unchanged, and still others significantly downregulated.

Fig. 5.

Gene expression profiles of ZmAGCs are presented under the heat stress and heat map was generated using Tbtools and in addition to this, scaling is done between −2.50 and 3.00. Red, upregulated genes; blue, downregulated genes.

The gene transcript quantity of nine AGC genes in Z. mays kernels was determined using qPCR. Based on the NGS analysis the qPCR-based quantification of nine genes which include the upregulatory genes, downregulatory genes and genes that were identified as the PIN auxin efflux carriers were included like ZmAGC1.5, ZmIRE3, ZmNDR5, ZmNDR3, ZmAGC1.9, ZmPHOT1, ZmKIPK, ZmAGC1.8 and ZmPID was performed in response to heat. ZmAGC1.5 was found to be significantly expressed with heat stress, and genes such as ZmIRE3 and ZmNDR5 were strongly upregulated, while ZmNDR3 was significantly downregulated (Fig. 6).

Fig. 6.

Real-time qPCR analysis was conducted to investigate the expression of ZmAGCs in Z. mays leaves under heat stress conditions. The coloured bar graphs represent the mean expression levels ± s.e. based on three independent replicates. *, statistical significance (t-test) at the 95% confidence level; **, significance at the 99% confidence level.

Targeting of PIN auxin efflux carriers by AGC kinases in Z. mays

In the realm of vascular plants, the AGC1 to AGC4 kinases represent distinct subfamilies within the AGC kinase group. Interestingly, the evolutionary origins of all AGC kinases found in terrestrial plants can be traced back to PHOTOTROPIN2-like proteins originally present in algae. Fig. 7a illustrates the intricate network of protein–protein interactions within the PIN auxin factor proteins, further categorised into three distinct clusters that collectively house the Z. mays PIN auxin proteins. Within the vibrant Red Cluster, we observe the presence of AGC1 proteins, including D6PKL1, D6PKL2, AGC1-7 and AGC1-5. Shifting our attention to the Green Cluster, which belongs to AGC3, we find the protein members PID and PID2. Lastly, nestled in the Blue Cluster of AGC4, we encounter the Phot1 and Phot2 protein members. In Fig. 7b, we are presented with a detailed view of AGC1 kinase D6PKL1, allowing us to explore the intricate structure of this protein. Notably, the structure has been thoughtfully constructed using the Itasser server, and the consensus motif bases have been thoughtfully highlighted to enhance our understanding. The protein structure is colour-coded for clarity, with the red segments representing the kinase domain, while the green, blue and cyan regions denote the positions of the consensus motif bases. This comprehensive visualisation provides valuable insights into the key elements of this important protein. Fig. 7c, Notably, a consensus in the form of specific motifs has been identified within the sequences of all AGC family members, with common two-base amino acid pairs shared among some. These motifs, including DHP, YLH, and FVGT, along with pairs such as FF, PP, PF, DF and KP, provide intriguing insights into the functional diversity and evolutionary history of AGC kinases in Z. mays and beyond. In the case of Z. mays, two phototropin (phot) blue light receptor kinases are members of the AGC4 subfamily, playing pivotal roles in various light-responsive processes. Notably, they are deeply involved in regulating phototropic stem bending. For instance, Z. mays phot1 serves as a substantial adverse modulator of ATP-binding module B auxin transporters reduces auxin transport yet allows for transverse axis mobility during phototropic growth. Conversely, within Z. mays, the AGC1 kinase family, comprising 13 members, has diversified to acquire a range of regulatory functions in terrestrial plants. Remarkably, despite their evolutionary lineage from PHOTOTROPIN2-like proteins, these AGC1 kinases have retained their positions at the plasma membrane. Here, they play critical roles in auxin transportation apical development, enlargement of cells, and other activities. AGC3 kinases, on the other hand, have evolved to fulfil roles in controlling the polarity of PIN proteins, representing a critical adaptation to meet the developmental and differentiation needs of multicellular organisms. Meanwhile, AGC2 kinases have primarily been associated with nuclear functions and do not appear to be directly involved in phosphoregulatory processes.

Fig. 7.

(a) Protein-protein interaction network has been categorised into three distinct clusters: the red and blue clusters encompass AGC1 members; while the green cluster emcompasses the AGC2 and AGC3 members. AGC4 members are represented within the blue cluster. (b) We present a model of the AGC1 kinase D6PKL1, showcasing the protein’s structure. The consensus motif bases have been thoughtfully highlighted in the structure, which was constructed using the Itasser server. (c) In the protein sequence alignment of the insertion loop of selected Z. mays AGC kinases, various sequence features have been marked. Notably, the sequence variation within the DHP motif may hold functional significance, particularly for D6PKL1, KIPK and D6PKL2, which are proposed as AGC1 targets.