Introduction

The mitogen-activated protein kinase (MAPK) cascades consist of the three sequential protein kinases MAPKK kinase (MAPKKKs/MEKKs), MAPK kinase (MKKs), and MAPKs, which are highly conserved signal transduction pathways in all eukaryotes (Meng and Zhang 2013; Jonak et al. 2002). MEKKs are located in the upper reaches of the MAPK cascade pathway, which can be phosphorylated and activated by a series of upstream signals or receptor proteins. The activated MEKKs then transmit the signal by phosphorylating the serine/threonine residues in the S/T-xxxxx-S/T motif of the downstream MAPKKs. MAPKKs, which are located at the central position of the MAPK cascade signaling system, are a class of dual-specific protein kinases that activate MAPKs through phosphorylation of the conserved tyrosine and threonine residues in the TEY/TDY motif in the activation loop (T-loop) (Çakır and Kılıçkaya 2015; Rodriguez et al. 2010; Tena et al. 2001). The activated MAPKs phosphorylate the downstream transcription factors and other enzymes or signaling components that are related to numerous cellular processes, which include cell proliferation and differentiation (Jiménez et al. 2007; Xu and Zhang 2015), hormonal responses (Tena et al. 2001), and biotic and abiotic stress (Zhu 2002; Pitzschke et al. 2009b).

To date, several MAPK cascade components have been well characterized and studied in both model plants and crop species. In tobacco, the NPK1-NQK1/NtMEK1-NRK1 cascade regulated cytokinesis during meiosis and mitosis (Soyano et al. 2003), and the MAP kinase p45ntf4 was activated by hydration during pollen germination (Wilson et al. 1997). The AtMKK5-AtMPK6 cascade plays a vital role in primary root growth and stomatal response by responding to ABA (Li et al. 2017). AtMPK12 acted as a negative regulator in the auxin signaling pathway (Lee et al. 2009). The OsMKK4-OsMPK1-OsWRKY53 signaling cascade was positively involved in plant wounding (Yoo et al. 2014). In addition, some MAPK cascade genes played essential roles in response to abiotic stress. The AtMEKK1-AtMKK4/5-AtMPK3/6 cascade was involved in regulating biotic stress (Asai et al. 2002; Wang et al. 2007). The AtMEKK1-AtMKK1/AtMKK2-AtMPK4 cascade played a vital role in plant innate immunity, salt stress, and cold stress (Gao et al. 2008; Teige et al. 2004; Furuya et al. 2014; Kong et al. 2012b). Some MAPKs, such as AtMPK4 and OsMAPK3, negatively regulated biotic stress signals (Pitzschke et al. 2009a; Lee et al. 2011). Overexpression of SlMAPK1 improved the resistance of tomato to drought stress (Wang et al. 2018). ClMPK7, ClMPK20-1, and ClMPK9-4 in watermelon were significantly up-regulated after high-temperature treatment (Song et al. 2015).

With the completion of annotated genome databases, it is possible to systematically identify the MAPK cascade genes in plants. Genome-wide analysis revealed that the Arabidopsis thaliana genome contained 10 MKKs and 20 MPKs (Colcombet and Hirt 2008; Pitzschke et al. 2009a), whereas 8 MKKs and 17 MPKs were found in rice (Wankhede et al. 2013). MKK and MPK gene families were also analyzed in economic crops. Six MKKs and 14 MPKs were identified in cucumber (Wang et al. 2015), and five MKKs and 16 MPKs were found in tomato (Wu et al. 2014; Li et al. 2014). These studies suggested that the number of MAPKK and MAPK genes varied across species. Foxtail millet (Setaria italica), which is cultivated in Asia and Africa’s semi-arid regions, is well-known for its high nutritional value and strong adaptability to abiotic stress (Barton et al. 2009). In addition, the high-quality small genome, small stature, prolific seed production, and available genetic transformations have made foxtail millet a potential C4 model plant (Doust et al. 2009; Li and Brutnell 2011; Muthamilarasan et al. 2020). However, no systematic studies on the MAPK cascade genes in foxtail millet have been reported.

In this study, we characterized 16 MPK and 11 MKK genes at the genome-wide level in foxtail millet. Subsequently, we analyzed the detailed information of the identified SiMPK and SiMKK genes, which included the gene structures, conserved motifs, physicochemical properties, chromosomal locations, and phylogenetic relationships. Finally, we evaluated the expression patterns of the SiMPK and SiMKK genes under phytohormone treatment and abiotic stress using qRT-PCR. Our research lays a foundation for future studies of the roles of SiMPK and SiMKK genes in response to hormone and abiotic stress. Our data presented here also provide clues to understanding the crosstalk between the SiMPKs and SiMKKs in foxtail millet and other plant species.

Materials and methods

Identification of family genes of SiMPKs and SiMKKs in foxtail millet

The reported gene sequences of MPK and MKK genes in Arabidopsis thaliana and rice were downloaded from the Arabidopsis genome database (https://www.arabidopsis.org/) and Rice Genome Annotation Project (http://rice.uga.edu/). These gene sequences were submitted to the plant genome database Phytozome (https://phytozome-next.jgi.doe.gov/) for BLAST using default parameters (e-value < 1.0) (Goodstein et al. 2012), which resulted in candidate gene sequences for SiMPK and SiMKK. To further test the presence of intact protein structures of these candidate genes, amino acid sequences of SiMPKs and SiMKKs were submitted to the Pfam database (https://www.sanger.ac.uk/) for identification (Mistry et al. 2021), and genes without complete structural domains were removed. Finally, members of the SiMPK and SiMKK gene families were identified. Isoelectric point (PI) and the molecular weight (Mw) of SiMPKs and SiMKKs in foxtail millet were analyzed using the bioinformatics online analysis software ExPASY (https://web.expasy.org/protparam/) (Wilkins et al. 1999).

Phylogenetic tree and chromosomal location of SiMPK and SiMKK genes in foxtail millet

The amino acid sequences of MPKs and MKKs of Arabidopsis thaliana, rice, and foxtail millet were aligned homologously by ClustalX software (Thompson et al. 2002) using default parameters. Phylogenetic trees of the MPK and MKK members of Arabidopsis, rice, and foxtail millet were constructed using the neighbor-joining (NJ) method of MEGA 7.0 software (Parameters set as: Poisson model and pairwise deletion, Bootstrap 1000 repetitions) (Kumar et al. 2016). We used Gene Location Visualize in Tbtools to map the chromosome locations of SiMPK and SiMKK genes based on their gene numbers and GFF files of foxtail millet genome annotations (Chen et al. 2020a).

Analysis of gene structure and conserved motifs of SiMPKs and SiMKKs in foxtail millet

The amino acid sequences of SiMPK and SiMKK members were submitted to the Multiple Em for Motif Elicitation (MEME) online program (https://meme-suite.org/meme/index.html) to predict their conserved protein motifs (Parameters: number of repetitions = any, maximum number of motifs = 10) (Bailey et al. 2009). The coding sequences and genome sequences of SiMPK and SiMKK genes were downloaded from Phytozome, and the gene structures of SiMPKs and SiMKKs were analyzed using the Gene Structure Display Serve online program (http://gsds.gao-lab.org/index.php) (Hu et al. 2015).

In silico transcript profiling analysis of foxtail millet SiMPKs and SiMKKs in different tissue

To explore the spatial and temporal expression patterns of SiMPK and SiMKK genes in foxtail millet, transcription data in different tissue parts and growth stages of foxtail millet were downloaded from the Phytozome Datasets GeneAtlas v1 Tissue Sample library, which included seedlings (5 d), germ shoots (6 d), leaves (14 d), panicles (7 d), and roots (10 d). Gene expression was shown by the FPKM (Fragments per kilobase of the exon model per million mapped) value. The downloaded data were imported into the Heatmap program of TBtools to map the expression profiles of SiMPKs and SiMKKs in different tissues and different developmental stages of foxtail millet (Chen et al. 2020a).

Plant material and experimental treatment

In this experiment, “Jingu 21” was used as the experimental material, which were provided by Prof. Xingchun Wang’s lab at Shanxi Agricultural University. The seeds of “Jingu 21” were cultivated in a 1:1 vermiculite and soil mixture of nutrient soil at 26/23°C, 50,000 lx light (16 h/8 h period), and 30–50% relative humidity in greenhouse. After seed germination, seedlings with uniform growth were selected and transferred to nursery pots with three plants in each pot. Seedlings at the double-leaf stage were ready for the subsequent treatment. For hormonal stress and for drought and salt stress, seedlings were washed and transferred to a liquid 1/2 MS medium for culture. To acclimatize the plants to the hydroponic environment, hormonal stress treatments were performed after 3 d of hydroponics (the nutrient solution was changed once a day). Specific stress types included 100 µM ABA, 100 µM BR, 1 Mm GA3, 100 µM melatonin, 100 µM MeJA, 10 mM SA, 10 µM IAA, 75 µM 6-BA, 10 nM NAA (Wu et al. 2014), 10% PEG6000, and (150 mM/200 mM) NaCl. For extreme temperature stress, double-leaf stage seedlings were grown in soil for another 3 d, and then they were subjected to low (4 °C) and high temperature (40 °C day/32°C night) in constant temperature plant incubators. Three biological replicates were used in each set of experimental treatments. Samples (approximately 100 ng) that were harvested after 24 h of treatment were frozen instantaneously with liquid nitrogen and stored in a -80℃ refrigerator.

RNA extraction and reverse transcription, qRT-PCR expression analysis

Total RNA of foxtail millet leaves was extracted using TransZolTM UP Plus RNA Kit (Transgen Biotech, China, Beijing), according to the instructions for use. The synthesized cDNAs were stored at -20 °C for subsequent experiments. The primers used in the experiments (Supplementary Table 1) were synthesized commercially, with the SiActin (Seita.8G043100) gene as an internal reference. cDNA was diluted to 300ng/µL as the template for qRT-PCR analysis. The reaction system with 10 µL, which included 5 µl of TransStart Tip Green qPCR SuperMix, 1 µL of template, 0.4 µL of upstream and downstream primers, and 3.2µL of each ddH2O was set. The reaction procedure of qRT-PCR contained two steps: 94 ℃ for 30 s, 94 ℃ for 15 s, and 60 ℃ for 30 s (40 cycles). The 2−ΔΔCT method was used to analyze the qRT-PCR results, and a t-test was used to analyze any significant differences (Livak and Schmittgen 2001).

Results

Identification of SiMPK and SiMKK gene families in foxtail millet

Following a BLASTP search in the foxtail millet Phytozome database using the protein sequences of the MPK and MKK genes from Arabidopsis and rice, the candidate loci of SiMPKs and SiMKKs were obtained. After a Pfam search with the protein kinase domains (PF00069.28) to exclude the loci without any kinase domain, 16 SiMPK and 11 SiMKK genes were identified in foxtail millet (Table 1). The open reading frame (ORF) of the 16 SiMPKs ranged from 1044 to 1836 bp, and the resulting protein molecular weights ranged from 39.13 kDa to 69.60 kDa. SiMPK3 had the minimum Mv with 347 amino acids, and SiMPK10 had the maximum with 611 amino acids, which was approximately a two-fold difference. The range of protein isoelectric points for SiMPKs varied from 5.46 to 9.34. The SiMKKs family consisted of 11 genes with ORF lengths that ranged from 996 to 1572 bp. The protein length of the SiMKK members was from 331 to 523 amino acids, which resulted in molecular weights from 34.69 kDa to 58.52 kDa. The corresponding isoelectric point of the 11 SiMKKs ranged from 5.47 to 10.02.

Table 1 Characteristics of predicted MPKs and MKKs in foxtail millet
Full size table

Phylogenetic tree and chromosomal location analysis of SiMPKs and SiMKKs in foxtail millet

To reveal and to classify the evolutionary relationships of the SiMPK and SiMKK genes in foxtail millet, unrooted phylogenetic trees were created by sequence alignment of the 16 SiMPKs and 11 SiMKKs with homologous genes in Arabidopsis and rice using the MEGA7.0 neighbor-joining (NJ) method. The phylogenetic tree showed that the SiMPK genes were divided into four groups that labelled A, B, C, and D (Fig. 1a). Two members were assigned to group A (SiMAPK2 and SiMAPK5), two members belonged to group B (SiMPK1 and SiMPK3), and two members were placed in group C (SiMPK4 and SiMPK6). The remaining eight members were classified as group D. Similarly, the 11 SiMKK genes were also classified into four groups that labelled A, B, C, and D (Fig. 1b). Group A contained SiMKK1, SiMKK5, and SiMKK9, group B had SiMKK2 and SiMKK10, group C contained SiMKK3, SiMKK4, and SiMKK11, and the remainder were members of group D. The evolutionary tree revealed that MPKs/MKKs in foxtail millet and rice were near to each other evolutionarily, which resulted in more direct homologous gene pairs (e.g., OsMPK11/SiMPK5, OsMKK1/SiMKK1, etc.), but they were somewhat far from MPKs/MKKs in Arabidopsis, which is a characteristic that could be related to plant evolution (Fig. 1b).

Fig. 1
figure 1

Phylogenetic relationships of MPK and MKK proteins inArabidopsis, rice, and foxtail millet. Using MEGA7.0 software and neighbor-joining (NJ) method, Arabidopsis thaliana, rice, and foxtail millet protein sequences were constructed as unrooted phylogenetic trees. Bootstrap values of 1000 repeats were calculated on each branch. (a) phylogenetic tree for MPK and (b) phylogenetic tree MKK

Full size image

Next, the chromosome distributions of 16 SiMPKs and 11 SiMKKs were analyzed by using the Gene Location Visualize of Tbtools (Fig. 2). Our results revealed that these genes existed individually or in sub-groups on chromosomes, rather than being distributed randomly. Based on Holub’s definition of the gene cluster as ≥ 4 genes within 200 kb on a chromosome, no gene clusters were found in either the SiMAPK and SiMAPKK gene families (Holub 2001). SiMPKs and SiMKKs were distributed on 7 of the 9 chromosomes, and none of these genes were found on chromosome 2 and 7. More interestingly, SiMKK2 and SiMKK10 were linked tightly on chromosome 1, and they showed > 98% similarity in amino acid sequence, which suggested that gene duplication may have occurred.

Fig. 2
figure 2

Chromosome distribution ofSiMPKandSiMKKgenes in foxtail millet.SiMPK and SiMKK genes were mapped on nine chromosomes of foxtail millet using TBtools software. The scale on the left represents chromosome length

Full size image

Analysis of the conserved motifs and gene structure in SiMPK and SiMKK families of foxtail millet

Using amino acid sequence alignment, we found that all the SiMPKs and SiMKKs have conserved activation loops of protein kinase (Supplementary Fig. 1). SiMPKs had an evolutionarily conserved T-X-Y motif, which is a typical phosphorylation site required for MAPK activation. The SiMPKs in groups A, B, and C contained the conserved TEY motifs, whereas group D contained the conserved TDY motif. SiMPK5 was an exception that had the MEY motif instead of the TEY motif. The MEY motif was identified in tomato (Kong et al. 2012a) and Brachypodium (Chen et al. 2012a), but not in rice and Arabidopsis. In addition, we discovered that the SiMPKs in A and B groups all had a CD domain conserved as (LH)DXXDE(P), which is the anchoring site of MKKs, while no such domain was found in groups C and D (Supplementary Fig. 1a). Alignment of amino acid sequences revealed that all 11 SiMKKs had the conserved D(L/I/V)K active site and the VGTxxxYMSPER motif. The activation-loop motif (S/T-X5-S/T) was present in the SiMKKs in groups A and B, but not in the other groups (Supplementary Fig. 1b).

Next, MEME analyses were performed to get further insight into the functional motifs of both gene families. A total of 10 different conserved motifs were detected in SiMPKs. Among them, motifs 6, 9, and 10 were found only in group D, but the other seven motifs were found in all four groups (Fig. 3a). Motif analysis revealed that motifs 1, 2, 4, and 5 were conserved in all SiMKKs (Fig. 3b). Meanwhile, specific motifs were found in different groups. For example, motif 6 was identified only in groups A and B, and motif 7 was detected in group D specifically. These unique motifs may indicate their specific gene functions (Zhang et al. 2020).

Fig. 3
figure 3

Schematic diagram of the conserved motifs of amino acids of (a) SiMPKs and (b) SiMKKs. A total of 10 motifs of SiMPKs and SiMKKs were identified by the Meme program. The Grey solid lines represent SiMPK and SiMKK genes and their lengths.The boxes with different colors represent 10 different motifs and their positions in each gene sequence. The results of SiMPK and SiMKK motifs correspond to the evolutionary tree by TBtools software, and the two family motifs correspond to the same legend

Full size image

Analysis of exon/intron organization was then performed to understand the gene structures of the SiMPKs and SiMKKs. All the SiMPK genes contained introns that varied in number, location, and length. The SiMPK members in Group D contained the most, usually 9–11, but Group C had the least number of introns, 2–3. (Fig. 4a). The gene structures of SiMKKs showed that the exon-intron organizations were the same in the classified groups. The SiMKK members in group A all had seven introns, and the members in group B all had eight introns. The SiMKKs in groups C and D had no introns, except for SiMKK6, which had one intron (Fig. 4b). Above all, both the conserved motif analysis and the gene structures validated the phylogenetic tree-based evolutionary relationship.

Fig. 4
figure 4

Gene structures of (a)SiMPKand (b)SiMKKin foxtail millet. The exon-intron structure of SiMPK and SiMKK genes was predicted by GSDS 2.0. The yellow boxes represent the gene coding region (CDS), the black lines represent introns, and the blue boxes represented untranslated Regions (UTR)

Full size image

In silico temporal and spatial transcript profiling of SiMPK and SiMKK genes

Numerous studies showed that the MAPK cascade genes were involved in various life processes in different organisms. To further study the roles of SiMPK and SiMKK genes in the growth and development of foxtail millet, the expression levels of SiMPK and SiMKK genes in diverse tissues were collected from the RNA-seq database, which included etiolated seedlings (5 d), germ shoots (6 d), leaves (14 d), panicles (7 d), and roots (10 d). The heatmap results revealed that the expression levels of SiMPK and SiMKK genes varied in different tissues and developmental stages (Fig. 5). In the SiMPK family, SiMPK5, SiMPK6, SiMPK7, SiMPK9, SiMPK15, and SiMPK16 genes were expressed constitutively, and the SiMPK1, SiMPK4, SiMPK8, SiMPK13, and SiMPK14 genes had expression patterns that were highly tissue-specific in all the tested tissues. SiMPK2 and SiMPK10 showed relatively higher expression levels than others in shoot and root tissues. For the 11 SiMKK genes, the expression levels of SiMKK1, SiMKK3, and SiMKK5 varied in all the tissues, whereas SiMKK4, SiMKK6, SiMKK8, and SiMKK10 were constitutive with lower expression. The SiMKK genes also showed developmental, stage-specific, expression patterns. For example, SiMKK7 was highly expressed in all the leaves at different stages, but SiMKK2 and SiMKK11 were only found in particular leaves at specific stages. These results suggested that the expression profiles of the MAPK cascade genes depended on their particular functions rather than their sequence similarities and evolutionary relationships and further more efforts are needed to determine their biological functions.

Fig. 5
figure 5

Expression profiles of (a)SiMPKand (b)SiMKKgenes in different tissues and developmental stages of foxtail millet. Genes expression data were downloaded from the Phytozome Datasets GeneAtlas v1 Tissue Sample database, which included etiolated seedlings (5 d), germ shoots (6 d), leaves (14 d), panicles (7d), roots (10 d), and shoots (7 d). The gene expression maps of SiMPKs and SiMKKs were drawn using the Heatmap program in TBtools

Full size image

Expression pattern of SiMPK and SiMKK genes in response to phytohormones

Hormones affect plant physiological and biochemical responses through a variety of signaling pathways, which include MAPK cascades (Chen et al. 2020b). Gene expression patterns are frequently used to predict gene function. To explore the gene expression patterns of SiMPKs and SiMKKs in foxtail millet after exogenous hormone treatment, we harvested the samples after 24 h of treatment with nine different phytohormones (ABA, BR, GA3, MT, MeJa, SA, IAA, 6-BA, and NAA). We determined their expression patterns using qRT-PCR technology (Supplementary Fig. 2).

Several SiMPK and SiMKK genes responded rapidly to specific hormones, with more SiMPKs involved than SiMKKs (Figs. 6 and 7). For example, 15 out of 16 SiMPKs (except for SiMPK3) were induced by different hormones, but only 6 out of 11 SiMKK genes were induced. The transcriptions of SiMKK6 and SiMKK8 were too low to be detected by qRT-PCR in the samples. Both up-regulated and down-regulated genes were detected in some treatments. When treated with ABA, the expression levels of nine SiMPK genes (SiMPK1, SiMKP2, SiMPK5, SiMPK6, SiMPK7, SiMPK10, SiMPK11, SiMPK12, and SiMPK13) and four SiMKK genes (SiMKK2, SiMKK4, SiMKK10, and SiMKK11) changed significantly, and all of them were up-regulated. With the BR treatment, only four genes, which included SiMPK7, SiMKP15, SiMKK2, and SiMKK4, were up-regulated. After being treated with GA3, five SiMPKs (SiMPK3, SiMPK5, SiMPK8, SiMPK13, and SiMPK14) showed no significant response, and the other SiMPK genes (except for SiMPK16) showed increased expression, whereas SiMKK2 and SiMKK4 in the SiMKK family showed significant differences in transcription levels. Additionally, three SiMPKs (SiMPK5, SiMPK6, and SiMPK16) and two SiMKKs (SiMKK3 and SiMKK11) were up-regulated after MT treatment. After 24 h of MeJA treatment, eleven SiMPK genes and two SiMKKs were up-regulated significantly, while SiMPK16 showed decreased expression. Seven SiMPKs and five SiMKKs were induced by 6-BA treatment, and none of them were down-regulated. Unlike other phytohormones, all the SiMPK and SiMKK genes induced by SA were down-regulated, except for SiMPK5 and SiMKK4. Similar expression patterns were observed after exogenous IAA and NAA treatment, where SiMPK5 and SiMPK15 showed relatively higher increases than others. Taken together, the SiMPK and SiMKK genes exhibited specific expression patterns in response to various hormone stress. SiMPK5 and SiMKK4 responded to the most hormones, which suggested they played significant roles in the phytohormone signaling pathways.

Fig. 6
figure 6

Analysis of expression patterns ofSiMPKs under hormonal stress. Transcript levels of the SiMPK gene familiy under hormonal stress were analyzed using qRT-PCR. Hormones included 100 µM ABA, 100 µM BR, 1 mM GA3, 100 µM MT, 100 µM MeJA, 10 mM SA, 10 µM IAA, 75 µM 6-BA, and 10 nM NAA. Three biological replicates and three technical replicates were set up for all samples. Each bar represents the mean ± SE normalized to SiActin (Seita.8G043100). Asterisks denote significant differences using a t-test (* for p < 0.05 and ** for p < 0.01)

Full size image
Fig. 7
figure 7

Analysis of expression patterns ofSiMKKsunder hormonal stress. Transcript levels of SiMKK gene families under hormonal stress were analyzed using qRT-PCR. Hormones included 100 µM ABA, 100 µM BR, 1 mM GA3, 100 µM MT, 100 µM MeJA, 10 mM SA, 10 µM IAA, 75 µM 6-BA, and 10 nM NAA. Each bar represents the mean ± SE normalized to SiActin (Seita.8G043100). Asterisks denote significant differences using a t-test (* for p < 0.05 and ** for p < 0.01)

Full size image

Expression pattern of SiMPK and SiMKK genes under abiotic stress

To investigate the roles of SiMPK and SiMKK genes under abiotic stress, we conducted qRT-PCR analysis to examine their expression levels in response to four abiotic treatments (cold, heat, salt, and PEG) (Supplementary Fig. 2). Almost all SiMPK and SiMKK genes exhibited differential expression levels after the four abiotic challenges (Figs. 8 and 9). When we subjected the young seedlings to cold, all the responsive eleven SiMPKs and five SiMKKs were up-regulated. For the heat treatment, all six SiMPK genes (except for SiMPK11) were significantly down-regulated, and no significant changes were obtained in the SiMKK families under heat stress. The expression levels of SiMPK and SiMKK genes changed irregularly following PEG treatment, with eight SiMPKs (SiMPK1, SiMPK3, SiMPK4, SiMPK5, SiMPK6, SiMPK9, SiMPK10, and SiMPK13) and SiMKK1 up-regulated, and two SiMPKs (SiMPK11 and SiMPK14) and SiMKK4 were down-regulated. We used two concentrations of NaCl (150 mM and 200 mM) to simulate salt stress, and similar expression patterns of the two gene families were obtained. Nine SiMPKs and four SiMKKs with significantly changed transcription levels were detected under both concentrations. Overall, our findings revealed that SiMPKs and SiMKKs responded to a variety of abiotic stress, which provides a basis for further studies on the mechanism of SiMPK and SiMKK genes under abiotic stress.

Fig. 8
figure 8

Analysis of expression patterns ofSiMPKsunder abiotic stresses. Transcript levels of SiMKKs gene families under abiotic stresses were analyzed using qRT-PCR. Abiotic stresses included 10% PEG6000, (150 mM/200 mM) NaCl, cold (4 °C) and heat (40 °C day/32°C night) stress. Three biological replicates and three technical replicates were set up for all samples. Each bar represents the mean ± SE normalized to SiActin (Seita.8G043100). Asterisks denote significant differences using a t-test (* for p < 0.05 and ** for p < 0.01)

Full size image
Fig. 9
figure 9

Analysis of expression patterns ofSiMKKsunder abiotic stresses. Transcript levels of SiMKKs gene families under abiotic stresses were analyzed using qRT-PCR. Abiotic stresses included 10% PEG6000, (150 mM/200 mM) NaCl, cold (4 °C) and heat (40 °C day/32°C night) stress. Three biological replicates and three technical replicates were set up for all samples. Each bar represents the mean ± SE normalized to SiActin (Seita.8G043100). Asterisks denote significant differences using a t-test (* for p < 0.05 and ** for p < 0.01)

Full size image

Discussion

Abiotic stress can cause a reduction in crop yield or even extinction; therefore, understanding how plants respond to adversity is critical for global food security (Chen et al. 2012b). The external stress received by upstream receptors is transmitted to the downstream stress resistance genes through several signal transduction pathways (Genot et al. 2017; Bari and Jones 2009). The MAPK cascade, which comprises many MAPKKK-MAPKK-MAPK modules, is an essential signaling pathway that transmits and amplifies the external signal through sequential phosphorylation (Rodriguez et al. 2010; Kong et al. 2013; Hamel et al. 2006). Genome-wide analyses of the MAPK cascade genes have been done in several plant species, which provides a crucial basis for subsequent functional characterization. However, little information on the MAPK cascade genes was available for foxtail millet. In this study, we systematically identified the MPK and MKK genes in foxtail millet, defined their fundamental characteristics, and analyzed the specific expression patterns in different tissues as well as under hormonal and abiotic stress.

We identified 16 SiMPK and 11 SiMKK genes in foxtail millet based on the MAPK cascade genes in Arabidopsis and rice. Both SiMPKs and SiMKKs were classified into four groups by phylogenetic analysis, which was supported by the conserved motif analysis and exon/intron organization. The same classification with other plant species suggested their evolutionary conservation in different organisms.We also discovered that the MAPK cascade genes changed during the evolution of foxtail millet. Compared with other species, there were certain differences in quantity (Supplementary Table 2). For example, in Arabidopsis and rice, there was only one gene in the B group of MKKs, while two genes in foxtail millet, SiMKK2 (Seita.1G307500) and SiMKK10 (Seita.1G307400). We also checked the copy number of SiMKK2 and SiMKK10 in other species (Supplementary Fig. 3). There was one copy maize and sorghum, while two copies in Setaria viridis and Brachypodium. In Setaria italica and Setaria viridis, the two copy genes were closely connected on chromosome 1, and the amino acid sequence homology approaches 98%. The double-copy genes in Brachypodium genomes were located on different chromosomes, and have an amino acid sequence similarity of more than 90%. As a result, we speculated that gene duplication may have occurred between SiMKK2 and SiMKK10. Gene duplication is an essential mechanism for providing genetic material for functional evolution of genes. This may provide evidence for a gene duplication event during the evolution of the MAPK cascade in Setaria grasses. Moreover, we found that the SiMPK5 contained an M-E-Y motif in the protein kinases activation loop, which is different from the conserved T-X-Y domain in other SiMAPKs (Supplementary Fig. 1). SiMPK5 reacted to the majority of hormones in nine hormone treatments, indicating that it plays an essential role in phytohormone signal transduction pathways. In rice MAPKs, SA stimulated the expression of OsMPK17-1 and OsMPK17-2 (Singh and Jwa 2013), but in foxtail millet, SA only induced the expression of SiMPK8 (the homolog of OsMPK17-1) to regulate the defensive response. In silico, RNA-seq data showed that most SiMPKs and SiMKKs were expressed constitutively in all the tested tissues, which suggested their essential roles in maintaining basic life processes. Some genes showed significant spatiotemporal expression characteristics. For example, SiMPK8 showed variable transcription levels in almost all tissues, and SiMPK13 was highly expressed in leaves, which exhibited their tissue-specific function.

Plant hormones are involved in complex signaling pathways that regulate plant growth and development as well as play a vital role in responses to biotic and abiotic stress (Bari and Jones 2009; Zhang et al. 2012; Bennetzen et al. 2012; Liu et al. 2015). The MAPK cascade pathway responds to hormonal signals and coordinates with the hormonal signal pathway to regulate responses to abiotic stress (Jagodzik et al. 2018). For example, the MEKK1-MKK2-MPK4/MPK6 pathway in Arabidopsis participated in the tolerance of salt stress and cold stress (Teige et al. 2004). The MEKK6-MAPK5 cascade pathway in rice was involved in root development in response to salt stress (Schmidt et al. 2013). In this study, we analyzed the expression patterns of SiMPK and SiMKK genes under nine different phytohormones and four abiotic stresses. Our results indicated that all the SiMPK and SiMKK genes were involved in the hormone and abiotic treatment. (Singh and Jwa 2013)Previous studies have shown that AtMAPK1/2/7/14 was induced by ABA and mediated drought signaling in Arabidopsis (Danquah et al. 2015; Matsuoka et al. 2018). In foxtail millet, SiMPK4/6, which belongs to the same subfamily as AtMAPK1/2/7/14, responded positively to ABA and 10% PEG stress and was up-regulated. Therefore, we hypothesize that SiMPK and SiMKK genes may be in a crosstalk between hormones and abiotic stress. The complicated interactions between different MAPK cascade components maintain their high-fidelity and specific response to certain signals. For example, AtMKK1/AtMKK2 interacted with AtMPK4 to form the AtMKK1/AtMKK2-AtMPK4 cascade, which was involved in regulating low temperature and salt stress in Arabidopsis (Teige et al. 2004; Lee et al. 2008; Popescu et al. 2009; Kong et al. 2012b). In our study, SiMKK1 and SiMPK5, which are homologs of AtMKK1/AtMKK2 and AtMPK4, respectively, were up-regulated in expression under both salt and cold stress. As a result, we speculated that SiMKK1-SiMPK5 may have protein-protein interactions. In conclusion, our findings provide a foundation for further investigation of the regulatory network of the MAPK cascade in response to hormonal stress or abiotic stress for various biological functions.