Genome-wide Scanning and Characterization of Sorghum bicolor L. Heat Shock Transcription Factors.

A genome-wide scanning of Sorghum bicolor resulted in the identification of 25 SbHsf genes. Phylogenetic analysis shows the ortholog genes that are clustered with only rice, representing a common ancestor. Promoter analysis revealed the identification of different cis-acting elements that are responsible for abiotic as well as biotic stresses. Hsf domains like DBD, NLS, NES, and AHA have been analyzed for their sequence similarity and functional characterization. Tissue specific expression patterns of Hsfs in different tissues like mature embryo, seedling, root, and panicle were studied using real-time PCR. While Hsfs4 and 22 are highly expressed in panicle, 4 and 9 are expressed in seedlings. Sorghum plants were exposed to different abiotic stress treatments but no expression of any Hsf was observed when seedlings were treated with ABA. High level expression of Hsf1 was noticed during high temperature as well as cold stresses, 4 and 6 during salt and 5, 6, 10, 13, 19, 23 and 25 during drought stress. This comprehensive analysis of SbHsf genes will provide an insight on how these genes are regulated in different tissues and also under different abiotic stresses and help to determine the functions of Hsfs during drought and temperature stress tolerance.

INTRODUCTION

High temperature and drought have adverse effects on water relations, photosynthesis and results in 50% crop reduction [1]. In response to heat stress, rapid accumulation of small heat shock proteins (Hsps) was observed in all eukaryotes and plants. Hsps act as molecular chaperones and prevent the aggregation and denaturation of proteins [2]. Heat shock transcription factors (Hsfs) transcriptionally regulate the Hsp genes. Plant Hsfs play a central role in the heat stress response. Tomato HsfA1, A2, and A3 confer heat stress tolerance when overexpressed [3-5]. LpHsfA1a and AtHsfA2 enhance thermotolerance upon overexpression but abolished when knocked-out or interfered [6, 7]. Transcription factor A2 has been found as a key regulator in response to many environmental stresses [8]. In Arabidopsis, overexpression of HsfA4a leads to decreased production of cytosolic H2O2 scavenging ascorbate peroxidase (APX) and it was hypothesized that Hsfs may act as H2O2 sensors in the plants [9]. HSFA1D, HSFA2, and HSFA3 act as key factors in regulating APX2 expression during diverse stress conditions [9]. Overexpression of AtHsfA1b-gusA in transgenic tomato plants led to the constitutive expression of Hsps, elevated levels of APX activity, with enhanced heat and chilling tolerance. Hsfs are also induced by other abiotic stresses likesalinity, temperature, cold, and metal [10]. Overexpression of OsHsfA2e and AtHsfA3 showed tolerance to salt stress [11, 12] but HsfA3 conferred enhanced thermotolerance and salt hypersensitivity in germination in Arabidopsis [13]. While HsfA1b (AtHsfA1b) gene is involved in chilling tolerance in tomato [14, 15], OsHsfA4a is involved in cadmium tolerance in rice and wheat [16]. Besides imparting abiotic stress tolerance, several heat shock factors are also involved with disease resistance and developmental activities. HsfB1 and HsfB2b are associated with pathogen resistance in Arabidopsis [17]. Further, HsfA9 was reported to be essential for embryogenesis and seed maturation in sunflower and Arabidopsis [18, 19]. Hsfs bind to the conserved cis-acting (5’-nGAAn-3’) heat shock elements (HSE) of the promoters. At least 3 HSE are required for better interaction with Hsf. Based on homology and conservation of domains, plant Hsfs are classified into three classes. When compared with fungi and animals, plants have many Hsf genes [20, 21]. Genome-wide screening of many plants resulted in the identification of 16 to 35 Hsfs depending on the species [22-24].

The Hsf gene family has not been characterized in Sorghum bicolor. But, functional and evolutionary relationship between organisms can be studied only when multiple sequences of these families are available for alignment and phylogenetic analysis. Therefore, an attempt has been made in the present study to identify, classify and to characterize Sorghum Hsf genes and predict their evolutionary relationship with Arabidopsis and Oryza. Further, it is also not known where and when these Hsf genes are expressed in Sorghum. Therefore, in the present investigation, tissue specific expression profiles of these Hsfs have been studied by carrying out quantitative real-time PCR under different abiotic stress treatments (by with-holding water for for 5-days for drought, by keeping at 4°C for 4 h for cold, by exposing to 40°C in a growth chamber for 4 h, by saturating potted plants with 150 mM NaCl and by collecting the tissue samples after 4 h treatment and by spraying 100 μM ABA and incubating the plants for 4 h for tissue collection). These results will be useful not only for studying the structure and function of SbHsfs but also for enhancing abiotic stress tolerance in this crop plant.

MATERIALS AND METHODS

Plant Materials and Stress Treatments

Sorghum bicolor variety cultivar Parbhani Moti, an improved desi variety was used for gene expression related experiments. Sorghum plants were grown in earthen pots containing 4.5 kg of black clay soil (Vertisol) under glass house conditions with 28/20°C day/night temperatures. Plants were maintained up to 28 days under well watered conditions and then used for different abiotic stress treatments. Drought stress was imposed by with holding the water supply for 5-days followed by leaf sample collection. For cold stress (low temperature) treatment, the plants were kept at 4°C in a refrigerator for 4 h and was used for tissue collection. For heat stress (high temperature) treatment, plants were kept at 40°C in a growth chamber and tissues were collected after 4 h of treatment. Salinity stress was induced by saturating the potted plants with 150 mM NaCl solution and leaf samples were collected after 24 h of treatment. For ABA stress, plant leaves were sprayed with 100 μM ABA solution and leaf sample was collected after 4 h. Different tissue samples like seedlings, leaf, flower, mature embryos, and roots were collected from different growth stages of Sorghum plants grown under normal growth conditions. For each sample, tissues were collected from three different plants grown under the same experimental condition (28/20°C day/night temperature), to provide biological replicates. Tissues were snap frozen immediately in liquid nitrogen and stored at -80°C until RNA extraction.

Identification and Localization of Hsfs in Sorghum Genome

Non-redundant nucleotide and amino acid sequences of Arabidopsis and rice Hsfs [25] were collected from TIGR and NCBI data bases. A total of 47 sequences were collected and each Hsf coding sequence (cds) was blasted against Sorghum bicolor genome in Gramene database by default settings. Gene sequences from the genome were retrieved using Edit plus (http://www.editplus.com/) and the sequences are subjected to Genscan (http://genes.mit.edu/GENSCAN.html) for coding sequences (cds) and amino acids. The redundant sequences which share the same chromosome location were eliminated and the remaining candidate genes were checked for Hsf DBD (DNA binding domain) in the Pfam database by employing SMART program [26], to identify coiled - coil structure and core of HR - A/B region. Sequences without the presence of DBD and coiled - coil regions have been eliminated.

Multiple Sequence Alignment

ClustalX2 [27] was used for multiple sequence alignment and domain prediction with default parameters. Bioedit (http://bioedit.software.informer.com/7.1/) and Genedoc (Free Software Foundation Inc.) were used for manually editing. For subcellular localization, WoLFPSORT [28], for finding out transmembrane helices TMHMM [29] and for gene characterization GSDS [30,(Gene Structure Display Server http://gsds.cbi.pku.edu.cn] were used. NLS and NES were predicted with the help of NLStradamus [31], Nucleo [32], and Net NES [33]. Conserved motif analysis was carried out using MEME [34].

Promoter Analysis

In silico promoter analysis was carried out using 1 kb sequence upstream to all the Sorghum Hsfs. Promoter sequences were retrieved from the genome using Edit plus. PLACE [35] and Plant Care [36] softwares were used to identify the cis-acting elements in the promoter sequences. The distribution of cis-elements in promoter regions were further identified using MEME tool [34].

Phylogenetic Tree

Phylogenetic tree was constructed by MEGA 5.1 using the N-J method with 1000 boot strap replicates [37] on the basis of amino acid sequences of Oryza sativa, Arabidopsis, and Sorghum. Gene duplication events were also investigated using phylogenetic tree based on the 70% similarity and 80% coverage of aligned sequences [38, 39].

RNA Isolation and qRT-PCR

The list of primers used for the qRT-PCR analysis is shown in the supplementary (Table ). Total RNA was extracted from control and treated tissues using MACHEREY-NAGEL kit according to the manufacturer’s instructions. A total of 2.5 μg RNA was reverse-transcribed into cDNA using SuperScript III First-Strand Synthesis Kit (Invitrogen) for qRT-PCR analysis. The cDNA was diluted into 1:12 with nuclease free water as template for qRT-PCR. The Bioline Master Mix (2X) was used to detect gene expression profile according to the manufacturer's recommendations on the RealPlex (Eppendrof). qRT-PCR was carried out in 96-well optical PCR plates, and the reaction was performed in a total volume of 10 μL containing 0.4 μM of each primer (1.5 μL), cDNA (1.0 μL) and Bioline Master Mix (2X) and nuclease free water was added upto 2.7 μL. qRT-PCR primers were designed using Primer3 software with GC content of 40-60%, Tm >50°C, primer length 20-25 nucleotides, with expected product size of 90-180 bp (Table ). The thermal cycles performed were as follows: 95°C for 10 min, followed by 40 cycles at 95°C for 15 s and 62°C for 1 min. Amplicon dissociation curves were recorded after cycle 40 by heating from 58 to 95°C with fluorescence measured within 20 min. Three technical replicates were used for each gene. Expression levels of the SbACP2, EIF4A, and S/T-PP genes were used as internal controls. The experiments were independently repeated three times, and the data from these experiments were averaged. Relative gene expression calculations were carried using Rest software [40].

RESULTS

Identification and Localization of Hsfs

Screening of Sorghum genome resulted in the identification of 25 SbHsfs and are named according to their chromosomal locations (Table ). Hsfs are distributed on chromosomes 1, 2, 3, 4, 6, 9, and 10 and the number of Hsfs varied from chromosome to chromosome. Eight Hsfs were identified on chromosome 1, five on chromosome 2, four on 3 and 4, two on chromosome 10 and one on chromosomes 6 and 9 (Fig. ). WoLFPSORT was employed to identify subcellular localization of Hsfs and 12 of them are located in nucleus, 4 in cytoplasm, 8 in chloroplast and 1 in lysosome (Table ). Transmembrane helices were not observed in the Hsfs identified.

Sequence Analysis of SbHsfs

The length of the Hsf proteins varied from 143 to 561 amino acids, the molecular weights between 15.25 to 59.48 KDa and the pI from 4.7 to 10.10. Most of the SbHsf contain only 1 intron, 4 introns were noticed in SbHsf14, but no introns in SbHsf8, 10, 13, 17, and 25 (Table ). The multiple sequence alignment shows highly conserved DBD domains in Sorghum bicolor Hsfs (Fig. ). The N terminal DBD of Hsfs contains 3α and 4β folds, which is the specific location of HSE. The DBD is approximately 100 amino acids in length, but SbHsf2, 9 and 18 contain only 30 residues. HR-A/B domains in Hsfs are characterized by coiled - coil structures, which is the key feature containing Leu-Zipper protein interaction domains (Fig. ). SMART program was used to predict the DBD characteristic features of HR-A/B regions of Hsfs (Table ). NLS and NES are important for intracellular distribution of Hsfs between the nucleus and cytoplasm and was predicted by using cNLS, NLstradmus, and NET NES 1.1 tools. Most of the Hsfs contain two motifs of basic amino acids K/R. Previous comparisons from Arabidopsis, Oryza, and Zea mays show a wide range of NLS monopartite and bipartites found near C terminal of HR A/B regions of Hsfs. Only SbHsfs 2, 9, and 16 contain bipartite NLS (Table ). MEME tool was employed to explore motif distribution both in gene and promoter sequences. It supports the phylogenetic analysis and helps to determine conserved motifs which are species specific, class specific and group specific (Fig. ). The Sorghum Hsfs contain 30 highly conserved motifs with 5 to 43 residues in length (Fig. ) and the number of motifs vary from 4 to 12. The SbHsf 18, 19, 20, and 24 contain 4 conserved motifs. Out of these, 2 and 3 are DBD, 15, 16, and 19 are coiled coil structure, 21, 24, and 25 are NLS, 12, 13, and 14 are AHA and 23 is NES motif. MEME finds the NLS motifs in SbHsf 11, 12, 17, 18, 22, 23, and 24 which could not be detected by NLS software.

Table shows the conserved cis-acting element motifs present in promoter regions. Motifs 12, 14, and 16 have ABA responsive elements; 16 and 24 have TATA box 2; 16, 26, and 29 have TATA box 3; 21 has LTRE which are low temperature and cold responsive elements, 26 have Myb and

28 have Myc waterstress responsive elements (Fig. and Fig. ). The promoter elements like ABRE, ANAERO, ARF, DPBF, DRE, LTRE, MYB, and MYC responsive to ABA, drought, low temperature, and cold are commonly present in all the 25 Hsfs along with high temperature responsive elements. The Hsfs also contained pathogenesis and salt stress responsive cis-elements GT1GMSCAM4 and WBOXNTERF3 for wound response and WBOXANTNPR1 for salicylic acid signal response. The CGCGBOX cis-elements present in Hsfs are involved in multiple signal transduction and KST1 is involved in guard cell-specific gene expression and pollen specific elements associated with pollen and anther development in different stress conditions. SbHsfs 9 and 13 contain a maximum of 15 ABRE cis-elements and SbHsfs2, 4, and 21 contain a minimum of one ABRE cis-elements (Table ).

Phylogenetic Analysis

Phylogenetic tree was constructed by using MEGA 5.1, and neighbour joining method was employed for multiple sequence alignment of 22 Arabidopsis, 25 rice, and 25 Sorghum Hsfs. Based on the bootstrap values and phylogenetic relationship, they were classified into 3 major Hsf classes A, B, and C. Phylogenetic analysis of rice, Arabidopsis, and Sorghum depicts a close relationship of rice and Sorghum, both being members of poaceae. While 10 subgroups are present in class A, 4 are seen in B and the least in C. The contrasting feature of the phylogenetic analysis is in the number of Hsfs that varied among the subclass A in rice, Sorghum and Arabidopsis. For example A2 (five) subgroup is present in the species rice and Sorghum, it is absent in Arabidopsis. While A6, A7, and A8 subgroups could not be found in monocot species like rice and Sorghum, 2, 2, 3 have been detected respectively in Arabidopsis (Figs. and ). Among the four subclasses of B, B1 are absent in Sorghum, but one is detected in rice. Further, in class C, the genome of Arabidosis revealed only one Hsf, but four each could be identified in rice and Sorghum.

Gene Duplication Events

Two paralogs participated out of 25 Sorghum Hsfs in regional duplications within the chromosomes. These paralogs evolved from their common ancestral genes through gene duplication events. While no segmental duplication events were observed in Sorghum 8 and 7 were recorded in maize and rice respectively out of nine paralogs. Maize and rice Hsf family is expanding with large number of segmental duplications (Fig. ).

Transcript Profiling of SbHsfs in Different Tissues

SbHsf genes displayed differential expression in different tissues (Fig. ). Out of four major tissues (mature embryo, seedling, root, and panicle), panicle showed higher levels of Hsf abundance than the mature embryos. No Hsfs were up- or down-regulated in the case of mature embryos (Fig. ). While in seedling Hsfs4, 9 are highly expressed, 13 and 22 are moderately expressed. In the case of roots, only 4 and 13 are well expressed. Moderate expression levels were also recorded in Hsfs5, 6, 21, 23, and 25 in roots. On the otherhand, Hsfs4 and 22 are highly expressed, Hsf1, 3, 5, 9, 10, 13, 16, 19, 23, and 25 recorded moderate transcript levels in panicle tissues (Fig. ).

Abiotic Stress Induced Expression of Hsfs

All Hsfs displayed a differential expression in response to various abiotic stresses (Fig. ). Among the five treatments (ABA, cold, heat, salt, and drought), drought stress induced higher transcript abundance than the other treatments. ABA, did not enhance the levels of Hsfs except in Hsf23, where only minor increase was noticed. Expression was significantly upregulated in HSf1, 15, 19, and 25 under cold stress (Fig. ). Moderate levels of expression was observed in Hsfs 2, 3, 4, 5, 6, 8, 10, 13, 16, 21, 23, and 24. During heat stress, Hsf1 was highly expressed, and moderate expression were displayed in Hsfs 6, 9, 13, 21 etc. During salt stress, Hsfs4, 6, 13, 16, 21, and 23 were up-regulated. In contrast, many Hsfs like 1, 5, 6, 10, 13, 18, 22, 23, and 25 were upregulated during drought stress (Fig. ).

DISCUSSION

Sequence Analysis

Hsfs have been identified in several plants [21-24, 41-43] but not in S. bicolor which is often exposed to salt, drought, and temperature stresses. Genetic variability for drought tolerance exists in Sorghum [44] but the effects of high temperature and water stresses on reproductive biology and seed-set needs further investigations and identification of candidate genes for breeding programs aimed at crop improvement. While eight Hsfs are distributed on chromosome 1, no Hsfs could be detected on 5, 7 and 8. In the case of Arabidopsis, maize and rice, Hsfs are spread all over the chromosomes but chromosomes 11 and 12 lack them [47, 25]. Like rice and maize, S. bicolor has also the same number of Hsfs, which reflects that Hsfs are conserved during the process of evolution [47, 25]. The theoretical pI of Hsfs range between 4.7 to 10.10, which indicates that they contain both acidic and basic proteins. Hsfs 2, 9 and 18 contain 30 residue-length DBD, which may occur due to deletions in DBD regions of α and 4 β-helices and due to genetic diversity in SbHsfs. Class A requires AHA motifs for their functioning, but SbHsf14 and 20 lack such motifs. SbHsfs18 and 24 belong to class C but do not contain AHA motifs. They may bind to other classes of A and C Hsf types and form hetero oligomers and start their function [25]. In silico survey of the putative cis-elements of the Sorghum Hsfs showed the presence of HSE, ABA responsive elements, ARR, Anaero, CACTT, low temperature responsive elements (LTRE), pollen specific cis-regulatory (AGAAA) and desiccation responsive elements. This indicates that Hsfs are not only expressed during high temperature but also during other environmental stresses. The presence of HSE cis-elements in the promoter regions is correlated with the expression of Hsf genes under high temperature stress in Arabidopsis, rice, maize, and wheat [47-50]. Bate and Twell [51] observed that transcriptional activation of late pollen gene (lat52) is controlled by a pollen-specific cis-regulatory elements AGAAA and TCCACCATA to attain high gene expression levels associated with pollen maturation. Promoter analysis of the endo-β-mannanase gene demonstrated pollen-specific cis-acting elements POLLEN1LELAT52 (AGAAA) which are associated with anther and pollen development [52]. In the present study also, such AGAAA elements were detected in the promoter regions of Hsfs indicating that these Hsfs may be involved in anther and pollen development in Sorghum. Promoter analysis of the KST1 gene, (an inward rectifying potassium channel) revealed a sequence motif TAAAG and the involvement of these elements suggests a role for Dof transcription factors in guard cell-specific gene expression and stomatal conductivity [53]. Such TAAAG elements have been observed in our promoter analysis, raising scope for speculation of Hsf promoters in K+ influx and guard cell movement. Hsfs are not only expressed during abiotic stress, but also biotic stress since their promoter regions contain potential cis-elements such as WBOXNTERF3 and WBOXATNPR1 which are responsive to biotic stresses like wound, pathogen, and salicylic acid [54, 55]. While ERF3 gene is activated by wounding in tobacco [55], the disease resistance regulatory protein NPR1 has been found to be required to activate AtWRKY18 [56]. Detecting ABA and salicylic acid response elements in the promoter regions of Hsfs provide valuable clues on the underlying regulatory mechanisms of Hsfs that may further lead to development of plants with biotic and abiotic resistance.

The phylogenetic tree revealed that proportion of the three Hsf classes differed considerably among the three species. While class A contained the large number of Hsfs, class B accounted for small number, and class C the minimum. Hsfs with three distinct classes A, B, and C appeared to be more in number in majority of angiosperms except in Medicago truncatula (class C absent), when compared to lower plants that contain classes like A and B as in the case of Picea abies, Selaginella moellendorffii, Physcomitrella patens, Chlorella sp. NC64 etc. [21]. Differences in different subgroups of A4, A9, B1 and B2 were observed between rice and a relatively temperature and drought tolerant S. bicolor, which is a C4 plant. Subgroup B1 is absent in Sorghum while it is present in rice. Perhaps these differences in different subclasses of A and B play critical roles during various types of abiotic stresses and developmental activities in these two contrasting plants. However, such an assumption needs to be validated experimentally. In plants, gene duplication events play an important role in evolution [57]. In polyploidization, gene duplicates accumulate [58] and these processes involve several transcription factors [59]. Recently, Song et al. [24] observed duplication events in the expansion of Hsf genes in Chinese cabbage. These observations clearly indicate that Hsf transcription factor family contributed to polyploidy [24, 59]. In the present study, segmental gene duplication events could not be observed in Sorghum unlike that of maize and rice [47, 25].

Transcript Analysis in Different Tissues and During Different Abiotic Stress Conditions

The expression patterns of different Hsf genes may differ depending on the plant species [21]. Yamaguchi-Shinozaki and Shinozaki [60] have shown that transcription factors interact with each other. It appears that each of the Hsf genes respond differentially to different abiotic stresses and developmental stages. Several transcriptome studies show that Hsf transcription plays significant roles in response to abiotic stress [23, 24, 48]. This type of unique expression patterns of Hsf transcripts were observed in response to both abiotic stresses and developmental stages also [9, 21, 43]. The varied patterns of Hsf expressions in different tissues may relate to the differences in cis-acting elements present in different promoter sequences. In the present study, Hsfs that are expressed during one type of abiotic stress, did not up-regulate when exposed to the other type of stresses, the exception being Hsf1 for cold and drought, and Hsf6 for salt and drought. Cross-talk exists between abiotic stress signal and plant growth and the expression of different transcription factor gene families [24, 41, 45, 60] indicating that these Hsfs play critical roles in maintaining drought and temperature stress tolerance and also play a vital role during development [21, 41, 46].

Six out of 21 Hsfs in Arabidopsis and 8 and 9 out of 25 in Oryza and Sorghum were induced by heat stress respectively [61, 62]. In many plants, intron-mediated enhancement (IME) of gene expression was noticed as in the case of Alcohol dehydrogenase 1, and Bronze 1 as reported by Callis et al. [63], Shrunken 1 in maize [64] and Phosphoribosylanthranilate transferase 1 in Arabidopsis [65]. Introns increased the transcription initiation and mRNA levels in these cases [66]. While in rice, intron mediated enhanced gene expression was observed, in Sorghum, exceptions were noticed in SbHsf08, 10, 13 and 25. These Hsfs in S. bicolor showed elevated expression levels without any intron. Intriguingly, SbHsf14 contains 4 introns but displayed lower expressions during stress. This infers that IME gene expression may vary depending upon the Hsf present in a specific species. OsHsfA2d, which is duplicated with OsHsfA2c, has two introns in place of one in the original gene A2c and OsHsfB2b/OsHsfB2c. This OsHsfB2b/OsHsfB2c has 2 introns and exhibited more expression during heat stress and considerably higher expression in almost all the other abiotic stresses and during seed development [62]. In S. bicolor, regional duplicated gene pair SbHsf02/SbHsf08 has no introns instead of 2 in the original gene SbHsf02, Hsf08 expressed abundantly in all the tissues and during all stress treatments. On the other hand, SbHsf10/SbHsf11 has 2 introns, but not expressed during all stresses.

Class A HSFs have been characterized in more detail than class B and C HSFs in plants. In Arabidopsis, expression of HsfA2 was high among the class A HSFs under high temperature and light stresses [8]. In rice, the expression of all OsHsfA2 genes increased by heat stress except for A2b, which is actually a duplicated gene with A2e [62]. In Sorghum, 5 members of HsfA2 genes have been noticed in contrast to 6 in rice, and are also highly induced during drought, salt, heat, and cold stresses. HsfB1 is absent in Sorghum and Oryza but present in Arabidopsis. Though HsfB1 is heat inducible, its overexpression did not lead to thermotolerance in Arabidopsis [14, 61]. On the other hand, in tomato, HsfB1 is a transcription co-activator functioning along with HsfA1 and hypothesized as a heat shock induced factor essential for maintenance and restoration of house keeping gene transcription during stress [67]. OsHsfB2a, B2b and B2c were induced by heat stress but expressed in developing seeds. In Sorghum, Hsf B2 was not induced under any stress but observed in panicles. Double knock-out mutants for AtHsfB1 and B2 displayed normal fertility and thermotolerance as compared with single knock-out mutants in Arabidopsis [17]. In S. bicolor, B3 was highly expressed in panicle and early seedling stage during droght but not in rice. On the other hand, B4 and class C Hsfs are moderately induced under all stress conditions. Thus, several differences exist among different classes of Hsfs between water loving rice and relatively drought tolerant S. bicolor.

In conclusion, 25 SbHsfs genes were identified in the genome of S. bicolor. Such a systematic analysis of Hsfs help us in finding out the functions of Hsf signaling pathways associated with different abiotic stress conditions and also growth and development. The diverse expression patterns suggest that these genes may perform different physiological functions depending on the type of tissue and its needs. Some SbHsfs were constitutively expressed, while others exhibited a distinct expression pattern in different tissues and under diverse abiotic stress treatments, implying that SbHsfs genes have functional diversity. This study provides the first step towards the future studies of Hsf protein functions and enhancing drought or thermotolerance stress and also the association of SbHsf genes under diverse environmental conditions.

Supplementary material is available on the publisher’s web site along with the published article.

Table 1.

List of Sorghum Hsf proteins. The identified Hsf proteins are listed according to their chromosome location. Hsf proteins are designated according to their locus id, protein sequence (AA) length, annotations, chromosomal locations their molecu-lar weight (Mw), isoelectric point (pi), protein localization, and introns.

S. No.	Gene Name	Locus	AA Length	Annotation	Chromosome Location	Mw   (Da)	pI	Localization	Intron
1	SbHsf01	Sb03g06630	467	RHsf 7	1	51515.94	5.48	Cytoplasm	2
2	SbHsf02	Sb03g12370	371	RHsf 8/Hsf 3	1	42186.32	4.96	Cytoplasm	2
3	SbHsf03	Sb03g53340	371	RHsf 4	1	40678.33	4.94	Nucleus	1
4	SbHsf04	Sb03g63750	477	-	1	52598.68	4.92	Nucleus	1
5	SbHsf05	Sb10g28340	328	RHsf 6	1	37535.13	5.00	Nucleus	1
6	SbHsf06	Sb3g02990	383	Putative Hsf sp 17	1	43217.74	5.58	Nucleus	2
7	SbHsf07	Sb3g63350	313	-	1	35158.51	7.23	Nucleus	1
8	SbHsf08	Sb01g042370	415	RHsf 8/ Hsf 3	1	46456.79	4.91	Nucleus	0
9	SbHsf09	Sb03g25120	302	RHsf 12 / Hsf 5	2	33727.63	6.78	Nucleus	1
10	SbHsf10	Sb08g36700	334	-	2	34544.70	9.71	Nucleus	0
11	SbHsf11	Sb09g28200	482	-	2	51240.71	10.10	Chloroplast	2
12	SbHsf12	Sb01g35790	561	-	2	59489.99	7.61	Chloroplast	1
13	SbHsf13	Sb02g004370	372	RHsf 5	2	41766.57	4.70	Nucleus	0
14	SbHsf14	Sb01g39020	456	Putative Hsf 8	3	49714.46	6.73	Chloroplast	4
15	SbHsf15	Sb01g53220	421	RHsf 11/Hsf 8	3	46415.83	9.60	Chloroplast	1
16	SbHsf16	Sb01g54550	434	RHsf 9	3	48351.37	5.13	Nucleus	1
17	SbHsf17	Sb03g028470	365	RHsf 13/Put. Hsf 1	3	39232.33	6.05	Lysosome	0
18	SbHsf18	Sb02g13800	347	-	4	37301.03	9.63	Chloroplast	2
19	SbHsf19	Sb02g29340	143	-	4	15257.53	8.07	Chloroplast	2
20	SbHsf20	Sb02g32590	176	-	4	19217.10	4.78	Chloroplast	1
21	SbHsf21	Sb4g13980	404	Putative Hsf sp 17	4	44957.02	5.34	Nucleus	1
22	SbHsf22	Sb04g48030	439	RHsf 1	6	46314.56	5.52	Chloroplast	1
23	SbHsf23	Sb09g026440	476	RHsf 10/ Hsf sp 17	9	52621.30	5.05	Nucleus	2
24	SbHsf24	Sb06g35960	279	-	10	29070.51	6.98	Cytoplasm	1
25	SbHsf25	Sb06g36930	439	-	10	47365.84	4.85	Cytoplasm	0

Table 2

Functional domains and motifs of Sorghum bicolor Hsfs.

Gene	Group	DBD	NLS	NES
SbHsf01	A2	119-212	327 (ASRKRRRPIG)	384 (LENLALNI)
SbHsf02	A9	3-43	155 (DGNRKRRFQAL)	94 (LLMQQLLV)
SbHsf03	A2	55-148	143 (RTIKRRRPPS)	333 (VELLSLGL)
SbHsf04	A1	1-88	199 (ANKKRRLPKQ)	-
SbHsf05	A2	8-101	207 (ISKKRRRPID)	-
SbHsf06	A2	36-129	232 (ISKKRRRRIV)	165 (LLMTEVVKL)
SbHsf07	B4	44-137	280 (DGKKRRAQQV)	-
SbHsf08	A9	1-83	252 (DGNRKRRFQAL)	191 (LLMQQLVDL)
SbHsf09	B4	3-33	280 (GKKKKRAHQD)	-
SbHsf10	B4	87-181	-	313 (LALEGADLSLTV)
SbHsf11	B4	200-293	-	461 (LALEGADLSLTV)
SbHsf12	B2	107-232	-	90 (FFLVLLLLL)
SbHsf13	A2	155-288	246 (ISKKRRRRID)	-
SbHsf14	A10	10-103	227 (KNIKRRRASK)	-
SbHsf15	C	107-200	382 (PAPGKRRRIG)	366 (VVLRAML)
SbHsf16	A4	23-115	199 (HGKKRRLPIP)	166 (LEDKLIFL)
SbHsf17	C	63-135	-	11 (LHTELALGLL)
SbHsf18	C	2-36	-	-
SbHsf19	A4	140-233	-	113 (LVYDALLVL)
SbHsf20	A3	9-102	-	23 (MLLEPKLEDEDV)
SbHsf21	A5	88-203	137 (FHKKRRLPG)	97 (VSQIEDLERRV)
SbHsf22	B3	47-140	-	422 (LDVLTLSV)
SbHsf23	A4	30-123	-	279 (MELALVSL
SbHsf24	C	49-142	-	179 (MLAFLLKVV)
SbHsf25	A10	24-117	307 (AGRKRRLLD	336 (VLAFEELAL)

Number in brackets indicates the position of the putative localization signal (NLS), nuclear export signal (NES) and DNA Binding Domains (DBD).

Table 3

Conserved cis-acting elements of Sorghum bicolor Hsfs. MEME motifs, cis elements, signal sequence and their functional roles.

S. No.	Motif	Cis Elements	Seq (signal)	Functions
1	12, 14, 16	ABA	ACGTG	Etiolation-induced expression (erd1)
2	17	Anaero 2	AGCAGC	Fermentative pathway
3	2,3,23, 24	ARR	NGATT	Response regulator
4	3, 25	CAAT	CAAT	Promoter of legumin gene
5	27,29,30	CACTT	CACT	Promote phosphoenolpyruvate carboxylase
6	6,17	CGC Box	VCGCGB	Ca++/calmodulin binding
7	1,3,21,24,25,26,27,28,29,30	DOF	AAAG	DNA binding proteins and carbon metabolism
8	2,12,28	DPBF	ACACNNG	ABA and embryo-specification
9	2,26,29	GATA	GATA	Chlorophyll a/b binding
10	6,13	GCC CORE	GCCGCC	G box high level constitution expression
11	9,25,27	GT1	GRWAAW	SA inducible
12	20,23,24,26	GTGA	GTGA	Late pollen gene g10, pectate lyase
13	5	HEXA	CCGTCG	Histone H4
14	3	I BOX CORE	GATAAG	Light regulated
15	21	LTRE	CCGACA	Low temperature and Cold
16	26	MYB	CNGTTR	Water stress
17	28	MYCONSES	CANNTG	erd1 (etiolation responsive to dehydration)
18	5,6	PAL BOX	CCGTCC	Phenylalanine ammonia-lyase
19	8,16,23	POLASI GI	AATAAA	Poly adenylation
S. No.	Motif	Cis Elements	Seq (signal)	Functions
20	4,20	POLASI G2	AATTAAA	Poly adenylation rice amylase
21	16,23,29	POLASI G3	AATAAT	Poly adenylation
22	25,27,28,29	POLLEN	AGAAA	Endo beta mannose, anther and pollen Development
23	21	PRE CONSES	SCGAYNRNN	Plastid responsive and light
24	3,16,29	ROOT MOTIF	ATATT	Promotes rol D
25	2	RYREPEATLE	CATGCAT	GLYCININE, ABA res., embryogenesis
26	11,13,14	SORLIP1	GCCAC	Phytochrome A, root development
27	5,12,24	SORLIP2AT	GGGCC	Light inducible
28	15	SORLREPSAT	TGTATATAT	Phytochrome A
29	2	SPH CORE	TCCATGCAT	Viviporous 1, seed specific development
30	1,12,14,20,23	SURE	GAGAC	Sulfur transporter
31	27	TAAAGSTKSTK1	TAAAG	Controlling guard cells and K+ influx
32	16,29	TATA2	TATAAAT	Accurate initiation for phaseolin
33	16,26,29	TATA 3	TATTAAT	Accurate initiation
34	8,16	TATA 4	TATATAA	Accurate inhibition G
35	8,16,23,26,29	TATA 5	TTATTT	lutamine synthase (non photo syn) ?
36	30	WBOXATNPR1	TTGAC	Response to SA signal
37	2, 11, 30	WBOXNTERF3	TGACY	Response to wound signal
38	2, 12, 30	WRKY	TGAC	Repressor for gibberellin signaling

Table 4

Conserved cis-acting elements present in promoter of Sorghum Hsfs.

Gene	Cis Acting Elements
	ABRECTAL (MACGYGB)	ANAERO (AAACAAA)	ARF (TGTCTC)	CGCGBOXAT (VCGCGB)	CURE (GTAC)	DPBF (ACACNNG)	DRE (RCCGAC)	GT1GM SCAM4 (GAAAAA)	LTRE (CCGAAA)	MYB (WAACCA/YAACKG/ CNGTTR)	MYC (CATGTG/CANNTG)	POLLEN1LE LAT52 (AGAAA)	TAAAGSTKST1 (TAAAG)	WBOXNT ERF3 (TGACY)	WBOXAT NPR1 (TTGAC)
SbHsf01	5	1	2	6	10	8	1	4	2	15	38	9	3	7	6
SbHsf02	1	2	1	0	8	0	0	0	2	16	8	4	3	5	3
SbHsf03	3	0	0	2	6	1	1	1	2	4	4	2	0	0	0
SbHsf04	1	2	1	10	20	6	4	3	4	23	24	8	7	8	3
SbHsf05	8	0	3	4	16	5	3	3	4	27	40	6	3	7	6
SbHsf06	9	9	0	26	10	4	2	4	7	30	38	9	10	5	4
SbHsf07	4	5	1	0	22	3	1	4	0	18	22	4	7	13	6
SbHsf08	4	7	0	6	6	3	2	3	5	16	24	4	5	3	6
SbHsf09	15	2	0	24	6	7	5	4	7	18	16	9	3	4	3
SbHsf10	3	3	2	0	12	4	0	1	0	7	14	7	6	3	0
SbHsf11	9	2	0	0	8	2	1	3	0	9	18	11	3	3	2
SbHsf12	7	4	2	4	14	1	0	5	0	23	10	5	1	3	2
SbHsf13	15	6	0	24	6	5	4	5	7	21	18	10	6	7	2
SbHsf14	3	3	2	4	12	3	0	0	1	29	22	11	4	7	1
SbHsf15	9	2	1	12	4	2	0	3	0	12	8	3	5	2	1
SbHsf16	2	1	0	10	0	0	0	0	0	10	0	2	0	2	2
SbHsf17	6	3	0	12	0	4	0	1	1	3	12	1	1	2	1
SbHsf18	12	3	0	14	4	1	3	1	3	10	4	0	0	3	1
SbHsf19	6	0	0	0	14	2	2	7	3	30	8	5	5	12	16
SbHsf20	2	6	1	10	8	4	4	10	6	25	36	16	6	9	7
SbHsf21	1	2	0	0	4	3	2	7	2	22	12	12	2	4	8
SbHsf22	4	0	0	4	8	3	0	5	4	14	10	12	2	6	2
SbHsf23	9	6	1	44	6	4	12	6	12	23	16	7	1	8	3
SbHsf24	8	4	3	42	12	4	3	5	7	35	30	13	3	9	6
SbHsf25	6	6	1	8	6	2	1	1	2	15	14	7	5	2	2

ABRECTAL: Response to ABA, ANAERO: Anaerobic conditions, ARF: ABA and auxin responsive, CGCGBOX: Multiple signal transduction, CURE: Cu and oxygen responsive, , DPBF: ABA, DRE: Dehydration responsive elements, GT1GMSAM4: Salt and pathogenesis related, LTRE: Low temperature and cold responsive, MYB: responsive to drought and ABA, MYC: Response to drought, cold and ABA, POLLEN: pollen and anther development, TKST1: Guard cell-specific gene expression, WBOXNTERF3: Wound signal and WBOXATNPR1: Salicylic acid responsive