Background

Rye (Secale cereale, 2n = 2x = 14, RR) belongs to the genus Secale in the Triticeae tribe of Poaceae, the grass family [1]. It is a good source of carbohydrates and a small amount of protein, potassium, B vitamins, lignans, ferulic acid, alkylpolysechenol, and prebiotics. Rye is mainly used for bread production and high-quality animal feed and pasture [2, 3]. In addition, Rye exhibits strong tolerance to abiotic stress, such as cold tolerance, drought resistance and soil impoverishment resistance, as well as biotic stress, including resistance to fungi and other pathogens [4, 5]. Recently, numerous studies have been devoted to studying Rye genome, which has laid the foundation for the promotion of breeding and targeted gene editing in Rye [6, 7].

Transcription factors (TFs), are DNA-binding proteins that specifically interact with cis-acting elements of the eukaryotic proteins [8]. As an essential class of regulators, transcription factors are involved in almost all biological processes in plants. When plants are subjected to biotic and abiotic stresses, transcription factors bind to the specific promoter regions of genes to activate or inhibit transcription of downstream target genes for defensive responses [9, 10]. Previous studies have demonstrated that heat shock transcription factors (HSFs) can regulate plant heat stress responses [11]. HSFs contain five basic functional and structural domains: N-terminal DNA binding domain (DBD), oligomerization structural domain (OD), nuclear localization signaling domain (NLS), nuclear export signaling domain (NES), and C-terminal transcriptional activation domain (CAD) [12, 13]. The oligomeric domain consists of two hydrophobic heptapeptide repeats, abbreviated as HR-A and HR-B. Heat shock transcription factors are divided into three subfamilies, A, B, and C, depending on the different number of amino acids inserted in HR-A and HR-B [13,14,15]. HSFA subfamily inserts a long amino acid chain between HR-A and HR-B, with an aromatic-rich hydrophobic acidic amino acid tail (AHA) at the C-terminus. AHA is essential for transcriptional activation function. On the other hand, the HSFB subfamily has no amino acid sequence between HR-A and HR-B, while members of HSFC subfamily have a short amino acid sequence inserted between HR-A and HR-B. Therefore, HSFA has a transcriptional activation function, while the opposite holds true for HSFB and HSFC. Under normal conditions, HSFs transcription factors are present as monomers in cytoplasm of plants and are inactive. When plants are exposed to abiotic stresses (e.g., heat stress), HSFs are active through OD binding to DNA-binding structural domains and AHA structural domains to form heat-kill transcription factor trimers. DBD structural domain will be recognized and bind to heat stress elements (HSEs) of the heat-kill protein promoter, which activates transcription of the heat-killed protein Hsp genes and enable plants to exhibit resistance to abiotic stresses [16,17,18,19].

Following the mining of genomic data, an increasing number of HSF gene families of plants such as A. thaliana [18], Populus L. [20], T. aestivum [21], Beta. vulgaris L. [22], and Gossypium spp. [23], O. sativa [24], H. vulgare [25]. and Z. mays [26] have been identified. Scharf first identified transcription factors HSFs associated with heat stress response elements by exposing cell cultures of tomatoes to heat stress. Furthermore, studies on HSFs have been carried out in a variety of plants [27]. In A. thaliana, overexpression of AtHSFB4 (AT1G46264.1) gene resulted in a shortened root length. Overexpression of AtHSFA2 improves plant salt tolerance, and promotes callus growth [28]. Overexpression of GmHSFA1 in soybean, and overexpression of SlHSFA1 in tomato enhances heat tolerance of transgenic plants [29, 30]. In addition, numerous studies have revealed that HSF family has a crucial regulatory function in various physiological aspects of the growth and development in other plants, such as stress response and plant phase transition. Currently, a number of HSF genes have been isolated and identified in different plants, which are still poorly understood in Rye [11, 31, 32].

In this study, based on the newly published Rye genome [7], we have identified 31 HSF genes and compared their gene structure, motif composition, chromosomal location, and gene duplication. To further study the developmental mechanisms among species, the HSF genes in Rye were compared with closely related genera to analyze the evolutionary distance and relationships. Finally, the expression of HSF genes (under hormones and stress treatment) was analyzed by qRT-PCR, it was found that there was a differential expression pattern of HSF genes in different tissues, which initially confirmed the biological function of HSF genes in Rye. This study provided a comprehensive analysis of the HSF gene family in Rye, which provided valuable information for screening important HSF genes in Rye under different development stages and stress treatment, and provided a theoretical basis for functional analysis of HSF gene family in other species.

Result

Identification of the HSF gene in Rye

A total of 31 ScHSF genes were identified according to different chromosomes and named between ScHSF1 and ScHSF31. The basic characteristics of ScHSF, such as molecular weight (MW), isoelectric point (pI), coding sequence length (CDS), and subcellular localization (http://cello.life.nctu.edu.tw/) were analyzed.

Among the 31 ScHSF proteins, ScHSF20 protein had the least amino acids (226), while ScHSF23 protein had the most amino acids (520). The molecular weight of the protein ranged from 24.64 kDa (ScHSF20) to 57.32 kDa (ScHSF23), and the pI ranged from 4.80 (ScHSF29) to 10.24 (ScHSF20), with an average of 6.14. All ScHSF proteins contained HSF_DNA-bind domains. Subcellular localization results showed that all ScHSF proteins were located in the nucleus, while four were located in the peroxisome, three were located in the cytoplasm, and one (ScHSF13) was located in the extracellular. The number of HSF genes in Rye (31) was higher than that in A. tauschii Coss (19), A. thaliana (21), Z. mays (25), and O. sativa (26) but lower than that in H. vulgare (32) and T. aestivum (82) (Additional file 1: Table S1).

Multiple sequence alignment, phylogenetic analysis, and classification of ScHSF proteins

To determine the phylogenetic relationships among Rye HSF proteins, a phylogenetic tree of Rye (31 ScHSFs) and Arabidopsis (22 AtHSFs) was constructed using MEGA 7.0 software. The 31 ScHSF proteins were divided into seven branches (groups 1–7) in the phylogenetic tree according to the previously proposed Cenci and Rouard classification method and topology [33]. Consensus exists with the taxa of HSF proteins in Arabidopsis, indicating that these HSF genes remained stable during the evolutionary process.

Among seven subfamilies, subfamily C had the largest number of members (9 ScHSFs), and subfamilies A and A1 had the fewest members (only 1 ScHSF). All members were usually concentrated in the three subfamilies A2, B and C. Comparison with the phylogenetic tree of Rye revealed that some of the ScHSFs clustered closely with AtHSFs (bootstrap support ≥ 70), suggesting that these proteins in Rye and A.thaliana might be homologous and have similar biological functions (Fig. 1a; Additional files 1 and 2: Table S1 and Fig. S1).

Fig. 1
figure 1

The evolutionary relationship and sequence alignment of the rye HSF proteins. a A phylogenetic tree of Rye and Arabidopsis thaliana HSF proteins showing that HSF proteins were divided into seven subfamilies. Rye proteins were denoted in red, whereas A. thaliana proteins were indicated in black. b Multiple sequence alignment of the DBD domains among the seven subfamilies of the ScHSF protein family

Full size image

Previous studies have shown that all HSF proteins contain DBD-conserved domain and are highly conserved. The conserved domain contains three α-helical bundles (α1-α3) and four reverse parallel β-folds (β1-β4), consisting of about 100 amino acid residues (Fig. 1b). However, there are varying degrees of insertions or deletions in these proteins. For example, in subfamily C, ScHSF21 had an 8 amino acid sequence inserted between α3 and β3, and ScHSF18 had an 11 amino acid sequence missing between α3 and β3. Overall, DBD domains in Arabidopsis and Rye were highly conserved, indicating that DBD structural domains were established early in plants.

Conserved motifs and gene structure analysis of ScHSF genes

The structural and taxonomic diversity of HSF genes in Rye were explored by comparing genomic DNA sequences. By comparing the localization and number of exon–intron structures, 31 ScHSF genes were found to have different numbers of exons, ranging from 1 to 4. Moreover, five ScHSF genes had only one exon, while most ScHSF genes (20, ~ 64.5%) contained 2 exons, and all genes contained HSF-DNA binding sites (Additional file 3: Figure S2). In addition to this, ScHSF1 and ScHSF14 genes belonged to A2 subfamily, which have the same intron and exon structure with 4 exons and 3 introns, the highest number of HSF genes (Additional file 3: Figure S2a, b). It was worth mentioning that ScHSF21 had a very large intron structure. In general, ScHSF genes of the same subfamily had similar gene structures. Subfamily A2 showed greater structural differences in the number of introns. Therefore, it could be speculated that they might have more biological functions.

To further evaluate the structural diversity of ScHSF genes, the motifs of ScHSF genes were analyzed using online motif software. Ten different motifs were identified in ScHSF proteins (motif 1 to motif 10). Motif 1, 2, 4, and 5 were usually located together, indicating that these four motifs were closely associated with SPL proteins. Notably, ScHSF genes of the same subfamily usually had similar motif compositions. For example, subfamily C contained motifs 1, 2, 3, 4, and 5, while subfamily A2 contained all the same motifs (motifs 1–9). Furthermore, it was found that some motifs were located at specific positions. For instance, motif 2 was always located at the beginning of the motif, while the motifs located at the end were different. Motif 5 was always between motif 1 and motif 2 (Additional file 3: Figure S2c). Overall, these results indicated that genes from the same subfamily had similar genetic composition and structure tended to cluster together, which was consistent with the population classification of the phylogenetic tree.

Chromosomal spread and gene duplication in ScHSF genes

According to the newly published Rye genome database, HSF genes (31) were distributed on seven chromosomes (Chr), and each HSF gene was named based on its physical position on chromosomes. Chr5 contained the most ScHSF genes (11, ~ 35.5%), followed by Chr7 (8, ~ 25.8%). Chr1 had only 1 gene (~ 4.76%), while Chr4 had no distribution of ScHSF genes. Notably, nearly half of HSF genes were located at the bottom of chromosomes (Fig. 2a).

Fig. 2
figure 2

The chromosomal distribution and synteny blocks of HSF genes in Rye. a Distribution of the 31 ScHSF genes on different chromosomes. The scale represented the length of chromosomes, whereas the green bars indicated chromosomes. The chromosome number is displayed on the left side of each green bar. b Analysis of interchromosomal fragment duplication of HSF genes in the Rye genome. The colored lines represented all synthetic blocks and the red lines specifically indicated the duplicated pairs among the 31 ScHSF genes

Full size image

Gene duplication events, including tandem repeat events and segmental duplications, play an essential role in gene amplification and the generation of new functions, [34]. Chromosomal regions with a range of 200 kb containing two or more genes are defined as tandem repeat events [35]. Accordingly, a duplication event analysis of HSF genes in Rye was performed to explore the evolutionary conservation of gene family. The results showed that there were no tandem duplication events in Rye genome, but two pairs of duplicated fragments were present (Fig. 2b, Additional file 4: Table S2). Four homologs in HSF genes indicated an evolutionary relationship between these genes. The highest number of ScHSF was found in LG5 (n = 2), followed by LG2 and LG7 (n = 1), and all genes were linked within subfamily B, suggesting that some ScHSF genes might be accompanied by fragment replications. These replication events were the main drivers of new functions of ScHSF genes during evolution.

Evolutionary analysis of the ScHSF genes and HSF genes of several different species

A dicotyledon (A. thaliana) and five monocotyledons (H. vulgare, O. sativa, Z. mays, T. aestivum and A. thuschii Coss) were selected to analyze the evolution of HSF in Rye. 31of ScHSF genes identified were compared with 10 conserved motifs from 6 other plants. ScHSF genes were not evenly distributed in the phylogenetic tree. As shown in Fig. 3a, the ScHSF proteins tended to gather with the HSF proteins of T. aestivum and H. vulgare, suggesting that they are more closely related (Fig. 3a). These genes, from the same subfamily, tended to have the same themes and tended to cluster together. Remarkably, almost all HSF genes from these seven plants contained motifs 1, 2, 3, 6, and 7 (Fig. 3, Additional file 5: Table S3). Subfamily A2 contained the most motifs and showed a diversity of expression. In addition, all genes except HSF in wheat began with motif 3. And in subfamily A1, A2, A3, and C, motif 8 was always distributed at the end of the pattern. In summary, ScHSF genes in subfamily A3 had high homology with barley HSF gene clusters, while most of the HSF genes in other groups had homology with wheat HSF gene clusters, indicating a closer distance in evolution and with similar potential functions.

Fig. 3
figure 3

Phylogenetic relationships and motif compositions of HSF proteins of seven different plant species (Rye, A. thaliana, H. vulgare, O. sativa, Z. mays, T. aestivum, and A. tauschii Coss). a An unrooted phylogenetic tree was constructed using the neighbor-joining method as implemented by Geneious R11. b Distribution of the conserved motifs in HSF proteins. Ten differently colored boxes represent different motifs and their position in each HSF protein (Table S3)

Full size image

A homology map between Rye and six representative species was constructed to explore the phylogenetic mechanisms of HSF genes in Rye. These species included one dicot (A. thaliana) and five monocots (H. vulgare, O. sativa, Z. mays, T. aestivum, and A. tauschii Coss). A total of 31 ScHSF genes were colinear with those of A. tauschii Coss (19), A. thaliana (21), Z. mays (25) and O. sativa (26), H. vulgare (32), and T. aestivum (82). The number of homologous pairs among the other six species (A. thaliana, H. vulgare, O. sativa, Z. mays, A. tauschii Coss, and T. aestivum) was 1, 19, 19,21, 26, and 69, respectively (Fig. 3, Table S3).

Homology analysis of these six plants revealed at least one pair of genes homologous to ScHSF, such as ScHSF25 with AET4Gv20678400.5/ Zm00001d032923_T002/ AT3G22830.1/ HORVU4Hr1G073650.2/ Os10t0419300-01/ TraesARI4B01G298300.1, indicating that these homologous genes were highly conserved and might have existed prior to the ancestral divergence. Accordingly, it was speculated that they might have played a crucial role in the evolution of HSF gene family in Rye. Interestingly, In H. vulgare, O. sativa, A. tauschii Coss, Z. mays, and T. aestivum, gene pairs were found to be collinear with eight ScHSF genes (ScHSF6, ScHSF7, ScHSF8, ScHSF12, ScHSF13, ScHSF24, ScHSF25, and ScHSF27) (Fig. 4, Additional file 6: Table S4). These homologous gene pairs could have been formed by gene replication during the differentiation of dicotyledons and monocotyledons.

Fig. 4
figure 4

Analysis of the HSF genes between rye and six representative plant species (A. thaliana, H. vulgare, O. sativa, Z. mays, T. aestivum, and A. tauschii Coss). Gray lines in the background indicate the neighboring blocks in genomes of rye and other plants, whereas red lines highlight the syntenic rye HSF gene pairs

Full size image

Analysis of cis-acting elements in ScHSF promoters

The promoter regions of ScHSFs were analyzed to provide ideas for tissue-specific expression of genes and stress response patterns. The cis-acting elements in promoter could be divided into four categories: light-responsive, hormone- and stress-responsive, plant growth- and development-related elements. Individual ScHSF gene in Rye covered most of phytohormone response elements, including abscisic acid response elements (ABRE). MeJA hormone response elements (CGTCA-motif and TGACG-motif). In addition, cis-regulatory elements were associated with low temperature, drought, anaerobic conditions, and other defenses found in all ScHSF genes (Fig. 5, Additional file 7: Table S5).

Fig. 5
figure 5

The distribution of cis-acting elements in promoters of ScHSF gene family members

Full size image

All genes contained light (G-BOX), drought (MYC), and stress (STRE) elements, while 93.3% of ScHSF genes contained MeJA and ABA response elements. The promoters of ScHSF1, ScHSF8, ScHSF22, and ScHSF26 contained growth hormone -, ethylene -, SA-, and gibberellin- reaction elements. All ScHSF promoters except the ScHSF3 promoter contained drought-related element as-1 (Fig. 5, Additional file 7: Table S5). These results suggested that certain cis-acting elements might be involved in regulating the expression of different tissues, such as seeds and meristematic tissues. Furthermore, we speculated that ScHSF genes might be involved in tissue developmental processes and in response to various abiotic stresses.

Protein–protein interaction network analysis of ScHSF family members

The combination of promoter cis-elements and transcription factors can regulate the precise initiation and efficiency of transcription. The PlantTFDB was used to explore the potential TFs that binding to ScHSF promoter. The results showed that ScHSF18 and ScHSF15 had the most and least transcription factors, respectively (Fig. 6a). Meanwhile, All ScHSF genes were regulated by a large number of ERF transcription factors. Studies have shown that ERFs can regulate the expression of target genes JA-based target genes and defense against Boea chinensis in Arabidopsis thaliana, suggesting that ScHSF might indirectly participate in regulation of JA synthesis against pathogens. Meanwhile, it has been reported that OsERF3 acts as a central switch that enables plant metabolism to respond appropriately to insects [36]. Therefore, it was speculated that ScHSF and participating in JA response played a potential role in insecticide traits through ERF regulation [37, 38].

Fig. 6
figure 6

Predicted interactions between rye HSF proteins. a Regulation network between ScHSFs and potential TFs. Red boxes represent ScHSFs genes and different colored ovals represent different TFs. Yellow shows ERF, gray shows HSF, blue shows bZIP, green shows SPL, pink shows MYB. b Prediction of the protein– protein interaction network among 31 ScHSFs

Full size image

To better understand regulatory relationships of HSF genes in Rye. We performed mutual protein prediction of HSF genes in Rye based on the most homologous wheat species. As shown in Fig. 6b, interactions existed among the nine ScHSF members. Moreover, ScHSF5 and ScHSF13 could interact with each other. Interestingly, ScHSF3 and ScHSF11 could interact with each other, along with ScHSF1, ScHSF2, ScHSF4, ScHSF16, and ScHSF25. Overexpression of AtHSFA2 (homologous ScHSF1, ScHSF2, ScHSF16, and ScHSF25) significantly increased basal and acquired heat tolerance in Arabidopsis plants [39]. ScHSF3, ScHSF5, ScHSF11, and ScHSF13, were identified as homologs proteins of Arabidopsis HSFB2b, a protein involved in plant resistance to pathogens [40].

Expression patterns of ScHSF genes in different plant organs

To further evaluate the potential function of ScHSF genes, a total of 15 genes in seven subfamilies were selected and the expression of these representative genes in four plant organs (root, stem, leaf, and flower) was analyzed by qRT-PCR. ScHSF genes showed different expression patterns in roots, stems, leaves, and flowers, suggesting that these genes contribute to diverse regulatory roles (P < 0.05). All genes were expressed in different tissues; two genes (ScHSF3, and ScHSF9) had the highest expression level in fruits; seven genes (ScHSF1, ScHSF4, ScHSF13, ScHSF15, ScHSF22, ScHSF26, and ScHSF28) had the highest expression level in roots, while six genes (ScHSF8, ScHSF18, ScHSF21, ScHSF23, ScHSF24, and ScHSF31) had the highest expression level in flowers (Fig. 7a). Most genes from the same subfamily had similar expression patterns, suggesting that these genes might have similar functions. By analyzing the expression of ScHSF genes in different tissues, it was obvious that all HSF genes were least expressed in leaves. We could speculate that HSF genes might be relevant to the development of stems, roots, and flowers in plants (Fig. 7b).

Fig. 7
figure 7

Tissue-specific gene expression of the 15 ScHSF genes and gene expression levels during various fruit development stages. a Expression patterns of the 15 ScHSF genes in flower, leaf, root, stem, and fruit tissues were analyzed using qRT-PCR. Error bars represent the stand errors with three replications. Lowercase letters indicate significant differences among treatments (α = 0.05, LSD). b Positive number = positive correlation; negative number = negative correlation. Red numbers indicate a significant correlation at the 0.05 level. c Expression patterns of the 15 ScHSF genes at different fruit developmental stages were analyzed using qRT-PCR (7 DPA, 14 DPA, 21 DPA, 28 DPA, and 35 DPA). Error bars represent the stand errors with three replications. Lowercase letters indicate significant differences among treatments (α = 0.05, LSD). d Positive number = positive correlation; negative number = negative correlation. Red numbers indicate a significant correlation at the 0.05 level

Full size image

The ScHSFs may also regulate the fruit development of Rye, thus affecting its nutritional composition and development rate [41, 42]. Therefore, 15 HSF genes were analyzed for expression level at five different post-anthesis stages (7d, 14d, 21d, 28d, 35d) after anthesis to identify genes that could potentially regulate Rye fruiting-related genes (P < 0.05). Most ScHSF genes displayed different expression patterns at five stages of fruit development. In the fruit of Rye, the expression of five genes (ScHSF4, ScHSF15, ScHSF18, ScHSF22, and ScHSF31) were significantly increased with fruit development, while the expression of most genes (ScHSF3, ScHSF13, ScHSF23, ScHSF26, and ScHSF28) decreased with fruit development (Fig. 7c). The expression of most genes showed a down-regulation trend in fruit expression with increasing time, and it could be speculated that HSF genes showed negative regulation in fruits (P < 0.05) (Fig. 7d). This also demonstrated that HSF genes played an essential role in fruit development and provided a theoretical basis for the studying of the nutritional value of Rye.

Expression patterns of ScHSF genes under various treatments

To determine whether the expression of ScHSF genes was restricted by different abiotic stresses, a representative expression of 15 ScHSF genes were expressed under six abiotic stresses. The results showed that some ScHSF genes exhibited significant up-regulated and down-regulated expression patterns under different stress treatments. Most of ScHSF genes also displayed significant differences in diverse tissues with the treatment period (P < 0.05). For example, most HSF genes were induced by cold stress in stems and leaves, whereas most genes were induced to express in roots under heat stress. Notably, ScHSF3, ScHSF9, ScHSF13, ScHSF15, ScHSF21, ScHSF23, ScHSF28, and ScHSF31 showed opposite patterns of highest expression levels in leaves and stems, compared to roots under cold and heat stress. Under flooding stress, expression level of ScHSF1, ScHSF3, and ScHSF31 were most significantly up-regulated and mostly concentrated at 4 h treatment time. Under drought stress, ScHSF1, ScHSF8, ScHSF18, ScHSF22, ScHSF24, ScHSF26, and ScHSF28 were significantly up-regulated in leaves. Meanwhile, most of the genes also showed a significant up-regulation under UV and NaCl stresses (Fig. 8). It should be emphasized that ScHSF1 and ScHSF3 were significantly highly expressed under all six different stress treatments and it could be further investigated as a potential candidate gene for stress management (P < 0.05).

Fig. 8
figure 8

Expression analysis of the 15 ScHSF genes in three tissues (roots, stems, and leaves) at the seedling stage under different abiotic stresses (UV radiation, flooding, PEG, NaCl, heat, and cold treatments). a Expression analysis of the 15 ScHSF genes was performed using qRT-PCR. Error bars represent the stand errors with three replications and the lowercase letter above the bar indicates a significant difference (α = 0.05, LSD) among the treatments. b Positive numbers = positive correlations; negative numbers = negative correlations. Red numbers indicate a significant correlation at the 0.05 level

Full size image

In addition, the expression patterns of ScHSF genes under ABA, IAA, and GA3 were used to further explore the functions of genes. The genes exhibited different expression patterns under different hormone treatments (P < 0.05) (Fig. 9). Under ABA treatment, most genes showed an up-regulation trend, while ScHSF15, ScHSF24, ScHSF26, and ScHSF31 showed a down-regulation trend. Moreover, ScHSF9 showed the highest expression level under IAA treatment, while only ScHSF5 expression level was down-regulated. Expression level of ScHSF22 and ScHSF28 were the highest under IAA treatment.

Fig. 9
figure 9

Expression analysis of the 15 ScHSF genes in fruits under different hormones (ABA, IAA, and GA3). a Expression analysis of the 15 ScHSF genes was performed using qRT-PCR. Error bars represent the stance error of three replicates, with lowercase letters above the error bars indicating significant differences among the treatments (α = 0.05, LSD). b Positive numbers = positive correlation; negative numbers = negative correlation. Red numbers indicate a significant correlation at the 0.05 level

Full size image

Discussion

ScHSF gene structure and evolutionary analyses

During the growth and development of Rye, HSF, a transcription factor involved in various stress responses, including high-temperature stress, salt stress, drought stress, and oxidation stress, played a crucial role [31, 32]. In recent years, the rapid development of metagenomics has resulted in the identification and characterization of HSF genes in many plants, including O. sativa, A. thaliana, Z. mays, Poplar, tomatoes, T. aestivum. However, the study of ScHSF family was still poorly understood to date.

In this study, 31 HSF genes were identified in Rye with HSF proteins ranging from 260 to 520 amino acids in length (Additional file 1: Table S1). A comparative genomic analysis of gene structure revealed that all HSF genes contained different numbers of introns, ranging from 0 to 4. All of the encoded proteins showed complex and variable structures, while variability might be attributable to gene duplication during evolution. The introns, as a part of plant evolution, might not only increase the length of genes as well as the frequency of recombination between genes and played a major role in the regulatory roles [43]. In contrast, genes without introns have no advantages during the evolution of the species and delay regulatory responses [44,45,46,47]. Therefore, many ScHSF members respond rapidly when subjected to stress treatments. The same subfamily has a similar number of motif compositions and introns, which allows us to speculated that they might share a common evolutionary origin and molecular functions. This approach can also be used to predict the function of unknown proteins.

Based on the conserved structural domains of Arabidopsis, they were divided into seven subfamilies. Each group contains at least one HSF gene from Arabidopsis and Rye, suggesting that these genes have not been missing during evolution and might have some biologically important functions (Fig. 1a). Gene amplification is the main driver for the generation of new functional genes during evolution, which is divided into two types: segmental duplications and tandem replication. Compared to segmental replication [48], tandem duplication events reprensent a larger proportion of plant genomes, with an approximately 10% incidence in Arabidopsis and rice [49, 50]. We found more HSF proteins in Rye compared with A. tauschii Coss (19), A. thaliana (21), Z. mays (25), and O. sativa (26). This suggested a possibility of more gene duplication events happened in Rye, which could also lead to the production of new functional genes in plants to adapt to harsh environment [51]. Based on physical location, 31 ScHSF genes were unevenly distributed on 7 chromosomes of Rye (Fig. 2a). Homology analysis of the HSF gene in Rye showed that no tandem duplicate gene pairs were discovered. Nevertheless, two pairs of fragment duplicates were identified (Fig. 2b). The homologous genes on different chromosomes of Rye might have promoted the evolution of ScHSF genes, resulting in a higher number of HSF genes in Rye than in other monocotyledons (A. tauschii Coss, Z. mays, and O. sativa).

To further speculate on phylogenetic developmental mechanisms of HSF genes, six comparative syngeneic maps of Rye connections with one dicotyledon and five monocotyledons were constructed. As we can see from Fig. 3, HSF genes from both Rye and different plants were classified into seven taxa. ScHSF genes from subfamily A3 showed higher homology with barely HSF genes clusters, whereas most of HSF genes from other groups were clustered with wheat. Interestingly, there was at least one pair of co-linear genes between ScHSF25 and AET4Gv20678400.5/ Zm00001d032923_T002/ AT3G22830.1/ HORVU4Hr1G073650.2/ Os10t0419300-01/ TraesARI4B01G298300.1, which might provide a theoretical basis for understanding whether they shared a common ancestor. Analysis of orthologous genes also illustrated that ScHSFs had the highest number of homologous gene pairs with wheat, indicating a higher level of homology among them. In addition, by analyzing the motif composition of HSF genes in plants, we found that HSF genes contained 10 motifs, with different subfamilies containing similar motifs. Moreover, HSF genes contained almost all motifs of the A2 subfamily. These results also reaffirmed that HSF genes in Rye were more closely related to wheat and might have a common ancestor.

Expression patterns and function prediction of ScHSFs

The analysis of gene expression is often used as an essential step in providing useful clues for functional prediction [52]. In this study, the expression patterns of 15 genes, which were represented in seven subfamilies, were selected and explored in different tissues and at different developmental stages. The results showed that most of HSF genes were significantly expressed (more than a twofold difference). For instance, most genes were significantly up-regulated in stems and leaves under cold treatment, while all genes were significantly up-regulated under UV and drought treatment (Fig. 8). This explained the high adaptability of Rye crop in alpine or arid areas. Most of HSF genes were significantly up-regulated in response to stress in these six treatments, and the genes expression was mainly in leaves and stems. However, the expression of most of HSF were highest in roots, suggesting that roots played a key function under drought conditions. Notably, both ScHSF1 and ScHSF3 were expressed in response to all six stresses. It could be further validated as a potential candidate gene for improving crop breeding.

Previous studies have shown that HSF genes were mainly involved in several environmental stress responses, such as high-temperature, salt, drought, and oxidation stress [11, 31, 32]. ClassA HSFs are major regulators of heat stress and could induce the expression of resistance genes [53]. The high expression levels of most HSF genes under different stress treatments also suggested that HSF genes played a significant role in roots and leaves. For instance, Tahmina reports that overexpression of the AtHSFB4 (AT1G46264.1) gene in Arabidopsis resulted in a shortened root length [28]. ScHSF23, the corresponding homologous gene, was up-regulated in roots for stress-responsive expressions under heat treatment and drought conditions. Based on the fact that homologous genes with similar structures may have similar functions, it is speculated that ScHSF23 may be related to root growth and development. Overexpression of AtHSFA2 in A. thaliana increases stress tolerance, and enhanced callus growth [49, 54]. The ScHSF1, ScHSF 2, ScHSF 14, ScHSF 15, ScHSF 16, ScHSF 25, and ScHSF 31 belong to A2 subfamily, which had high similarity to AtHSFA2.. Meanwhile, we found that A2 subfamily had the most abundant motifs, ScHSF1 and ScHSF15 also showed significant up-regulation in stress treatment (Fig. 8). Finally, two genes from each subfamily were screened for qRT-PCR analysis and verification of functional traits. The results showed that these genes were significantly up-regulated in different tissues during stress treatment, suggesting that these genes might respond to stress through different tissues. Interestingly, these up-regulated genes were not only expressed in roots but also dominantly in leaves and stems (Fig. 8). Thus, we speculated that this was likely due to complex protein interactions that coordinated the expression of multiple genes through a network of feedback mechanisms [55].

Conclusion

Altogether, identification and systematic analysis of HSF genes in Rye showed that the 31 ScHSF genes were unevenly distributed on 7 chromosomes that were classified into seven subfamilies. By comprehensively analyzing gene structures and conserved motifs of 31 putative ScHSF genes, we found that motifs and gene structures of the same family were similar and might have the same biological functions. Furthermore, fragments and tandem repeats were the main drivers of novel functions in ScHSF family. Fragment repeats might have more substantial contributions to the evolution of Rye HSF genes. Overall, we performed a preliminary analysis of the structure of HSF gene family in Rye and further detailed its expression pattern. The results indicated that ScHSF gene family played a critical role not only in stem and flower development but also in hormonal and abiotic stress response during Rye development.

Materials and methods

Gene identification

The whole Rye genome was downloaded from the Ensembl website (http://ensemblgenomes.org). HSF gene family members were obtained based on two BLASTp approaches (PFAM and SMART) [56,57,58]. Firstly, all possible HSF proteins were identified using BLASTp (score value ≥ 100, e value ≤ 1e-10) with reference to the trihelix protein sequence of Arabidopsis. Secondly, the PFAM protein family database (http://pfam.sanger.ac.uk) was used to produce a Hidden Markov Model (HMM)with HSF domains, and then an HMM model cutoff value of 0.01 in HMMER 3.0 was applied to compare HSF protein sequences of Rye (http://plants.ensembl.org/hmmer/index.html). The availability of the HSF core sequence was confirmed using PFAM and the SMART program (http://smart.emblheidelberg.de). A total of thirty-one HSF genes were identified and then used as initial sequences to confirm HSF proteins (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastp&PAGE_TYPE=BlastSear-ch&LINK_LOC=blasthome) with BLASTp. Finally, several characteristics of the HSF genes, such as the sequence length, isoelectric point (pI), molecular weight (MW), and subcellular localization, were identified using ExPasy. A 2000 bp sequence upstream of the start codon (ATG) of the ScHSF gene was extracted from Rye genome using TBtools, followed by an analysis of the cis-acting elements using PlantCare (http://bioinformatics.psb.ugent.be/webtools/plantcare/html). Finally, TFs were predicted through PlantTFDB and shown by using Cytoscape [59, 60].

HSF gene structure

Multiple protein sequence alignments based on the domain sequences in characterized HSF proteins of A. thaliana were created by ClustalW with default settings. The deduced amino acid sequences of HSF domains of different subfamilies were manually regulated using GeneDoc software and Mega 7.0. Furthermore, Gene Structure DiHSFay Server (http://gsds.cbi.pku.edu.cn) online program was applied to analyze the exon–intron structure of HSF genes. MEME Online Applications (http://meme.nbcr.net/meme/intro.html) were then employed to identify the protein sequences by adjusting the optimum motif width to 6 ~ 200 and the maximum number of motifs to 10.

Chromosomal distribution and gene duplication events

All ScHSF genes were mapped to locations on different Rye chromosomes by using physical location information and handled using the Circos program [61]. The multiple collinear scanning toolkits (MCScanX) were then used, with default parameters, to analyze replication events of ScHSF genes [62]. Finally, the HSF genes homology between Rye and six other plants (O. sativa, Z. mays, H. vulgare, A. thaliana, T. aestivum, and A. tauschii Coss) was measured using Dual Synteny Plotter (https://github.com/CJ-Chen/TBtools).

Phylogenetic analysis and classification of the ScHSF family

With regard to the classification of AtHSFs, all identified ScHSF proteins were first clustered into diverse groups. Next, a neighbor-joining (NJ) tree was built using Jukes-Cantor model in MEGA 7.0 [63]. The phylogenetic tree was then generated, with a bootstrap value of 1000, and was assigned by gene R11 and BLOSUM62 cost matrix. Moreover, we generated a multi-species phylogenetic evolutionary tree that included all HSF protein sequences from Rye as well as six others plants species (O. sativa, Z. mays, H. vulgare, A. thaliana, T. aestivum, and A. tauschii Coss). Notably, all protein sequences were downloaded from the UniProt database (https://www.uniprot.org). A protein–protein interaction analysis was performed on the STRING database (http://string-db.org) using ScHSFs as the queries and T. aestivum proteins as references. Promoter cis-acting elements were predicted by both PlantCare and PlantTFDB [60, 64].

Plant materials, growth conditions, and different abiotic stress in Rye

The rye seed used in the experiment was provided by Fan Yu from Guizhou University. Wei ning Rye is the variety we used. Rye plants were cultivated in pots containing a mixture of soil and vermiculite (1:1) in a growth room. The growth room was maintained at a temperature regime of 25 °C during the 16-h daytime period and 20 °C during the 8-h nighttime period. The 3arelative humidity in the growth room was set at 75%. After 21 days of growth, stress treatment was initiated. Fruit sampling was conducted when the first seed setting occurred, and subsequent samples were collected every other week for five consecutive harvests.After planting, leaves, roots, stems, grains, anthers, and styles were collected from five individual plants under the same growth environment. The samples were immediately stored in liquid nitrogen at -80℃ until further analysis. The expression pattern of 31 HSF genes under different stresses was explored. Specifically, the abiotic stress treatments, including salt treatment (5% sodium chloride), water immersion (full plant), drought treatment (30% PEG 6000), UV radiation (70W/cm2, 220 V, 30W), high temperature treatment (40℃), and low temperature treatment (4℃), were applied at the seedling stage (after 21 days). Each stress treatment was replicated five times, and qRT-PCR analysis was performed after sampling at 1 h, 4 h, and 12 h, respectively. Hormone stress treatments were applied at seedling stage (after 21 days) with package expansions of ABA (100 μmol/L), IAA (100 μmol/L), and GA3 (100 μmol/L). Grain samples were collected at 7D, 14D, 21D, 28D, and 35D, and each treatment was replicated five times, respectively.

Total RNA extraction, cDNA reverse transcription, and qRT-PCR analysis

Total RNA was extracted using a plant RNA extraction kit (Vazyme Biotech) following the manufacturer's instructions. A cDNA library was constructed through reverse transcription of 1 mg RNA samples using 5 × HiScript® Reverse Transcriptase (vazymes) and 4 × gDNA (vazymes) kits in accordance with the manufacturer's protocol. The expression of some representative genes was then analyzed by qRT-PCR, with at least three biological replicates. The primers used were designed by Beacon Designer 7 (Additional file 8: Table S6). Relative mRNA expression was normalized to the actin gene (GADPH) mRNA expression as internal control and was calculated using the delta-delta Ct (2−ΔΔCt) method [65].

Statistical analyses

JMP6.0 (SAS Institute) was used to perform analysis of variance (ANOVA) tests; multiple comparison tests of ANOVA results were performed using the least significant difference (LSD) method at two difference significance levels p < 0.05* and p < 0.01**. Finally, histograms were generated using Origin version 8.0.