Genome-Wide Identiﬁcation and Characterization of Hsf and Hsp Gene Families and Gene Expression Analysis under Heat Stress in Eggplant ( Solanum melongema L.)

: Under high temperature stress, a large number of proteins in plant cells will be denatured and inactivated. Meanwhile Hsfs and Hsps will be quickly induced to remove denatured proteins, so as to avoid programmed cell death, thus enhancing the thermotolerance of plants. Here, a comprehensive identiﬁcation and analysis of the Hsf and Hsp gene families in eggplant under heat stress was performed. A total of 24 Hsf -like genes and 117 Hsp -like genes were identiﬁed from the eggplant genome using the interolog from Arabidopsis. The gene structure and motif composition of Hsf and Hsp genes were relatively conserved in each subfamily in eggplant. RNA-seq data and qRT-PCR analysis showed that the expressions of most eggplant Hsf and Hsp genes were increased upon exposure to heat stress, especially in thermotolerant line. The comprehensive analysis indicated that different sets of SmHsps genes were involved downstream of particular SmHsfs genes. These results provided a basis for revealing the roles of SmHsps and SmHsp for thermotolerance in eggplant, which may potentially be useful for understanding the thermotolerance mechanism involving SmHsps and SmHsp in eggplant. data Q.P.; C.G., B.S. and


Introduction
Plants live in complex environments where multiple abiotic stresses, such as salt, drought and extreme temperature, may seriously restrict their growth and development [1]. As sessile organisms, plants cannot move to avoid these stresses and, thus, they have developed mechanisms, such as enhanced expression of tolerance-related genes, in response to heat stress [2,3]. To survive and acclimatize under adverse environment conditions, plants have established self-defense mechanisms during the course of long-term evolution. Previous studies have shown that under heat stress (HS), plant cells respond rapidly to high temperatures by inducing the expression of genes encoding heat shock proteins (Hsps), which are involved in preventing heat-related damage and confer plant thermotolerance in strawberry, walnut, barley and grapevines [4][5][6][7]. Many Hsps function as molecular chaperones in preventing protein misfolding and aggregation, consequently maintaining protein homeostasis in cells and inducing acquired thermotolerance in plants [8]. The expression of Hsps is controlled and regulated by specific types of transcription factors called heat shock factors (Hsfs), which normally exist as inactive proteins [9].
Currently, many plant Hsf and Hsp genes from various species have been isolated and comprehensively studied. Based on their approximate molecular weights and sequence homologies, Hsps are classified into five families, namely, the small Hsp (sHsp), Hsp60s, Hsp70s, Hsp90s and Hsp100s [10]. The expression of sHsp is positively correlated with thermostability [11]. As chaperones, Hsp60 proteins participate in the folding and aggregation of many proteins transported to organelles, such as chloroplasts and mitochondria [12]. Hsp70 chaperones, together with their co-chaperones, make up a set of prominent cellular machines that assist with a wide range of protein folding processes in almost all cellular compartments [13]. In Arabidopsis TU8 mutants, the downregulation of Hsp90 expression leads to mutants that are more sensitive to heat. In Arabidopsis thaliana seedlings, fungi producing Hsp90 inhibitors increase the expression of the Hsp101 and Hsp70 genes, resulting in the enhancement of plant heat resistance [14]. Arabidopsis has at least 21 Hsf s [15]. HsfA1a, HsfA1b and HsfA1d act as the main positive regulators of the heat shock response [16] and HsfA2 can enhance the thermotolerance of plants [17]. Above all, Hsfs and Hsps play crucial roles in plant thermotolerance. The Hsf and Hsp gene families have been extensively studied in the model plant Arabidopsis thaliana and in non-model plants, such as rice (Oryza sativa) [18], poplar (Populus trichocarpa) [19], maize (Zea mays) [20] and Chinese cabbage (Brassica rapa L. ssp. pekinensis) [21].
Eggplant (Solanum melongema L.) is an important economic solanaceous crop, ranking third, after potato and tomato. Eggplant is primarily cultivated in East Asia, South Asia, the Middle East and northern Africa. The optimal temperature for eggplant growth and development ranges from 22 • C to 30 • C. With global warming, the temperature in subtropical and tropical regions is often above 35 • C, resulting in serious heat injury in eggplant, including limited plant growth, reduced productivity and damaged quality [2]. Thermotolerance is an important agronomic trait for eggplants, but the molecular mechanisms of heat tolerance remain elusive. Hsfs and Hsps play core roles in the signal transduction pathways involved in plant response to heat stress. Due to the vital regulatory functions of Hsf and Hsp genes in plant responses to heat stress, Hsf and Hsp genes in eggplant under heat stress were studied. The eggplant genome was sequenced and assembled [22], enabling the characterization of the eggplant Hsf and Hsp families and their responses to heat stresses at the molecular level. Therefore, genome-wide identification of Hsf and Hsp genes in eggplant was conducted to infer their expansion and evolutionary history. RNA-seq data and quantitative real-time RT-PCR analyses were used to explore their expression difference in the thermotolerant line 05-4 and the thermosensitive line 05-1 as elicited by naturally increased temperature. The results provide a relatively complete profile of the Hsf and Hsp gene families in eggplant and elucidate their relationship with thermotolerance, which provides a foundation for further functional research on these genes in eggplant. Furthermore, these findings could potentially be useful for understanding the mechanism of thermotolerance mediated by Hsfs and Hsps in eggplant.

Identification and Classification of Hsf and Hsp Family Members in Eggplant
Published Arabidopsis Hsf and Hsp sequences [23] were retrieved and used as queries in BLAST searches against the eggplant genome database (http://eggplant.kazusa.or. jp/, accessed on 6 June 2021) to identify potential eggplant Hsfs and Hsps. All output genes identified according to Arabidopsis Hsf and Hsp sequences were collected and confirmed using Pfam (http://pfam.xfam.org/search, accessed on 6 June 2021) and SMART (http://smart.embl-heidelberg.de/, accessed on 6 June 2021). The isoelectric points and molecular weights were predicted using the Compute pI/Mw tool from ExPASy (http: //web.expasy.org/compute_pi, accessed on 6 June 2021).

Phylogenetic Analysis
Alignments of the full eggplant Hsf and Hsp proteins were performed using clustal X2.1 [24]. Phylogenetic trees were constructed using the neighbor-joining (NJ) method in MEGA (version 5.0) [25] with bootstrap values from 1000 replicates indicated at each node. To identify signature domains, the Hsf and Hsp protein sequences were compared with Arabidopsis and tomato. SmHsfs and SmHsps (sHsp, Hsp60s, Hsp70s, Hsp90s and Hsp100s) were named based on the subfamily classification and their phylogenetic relationships with the corresponding AtHsfs and AtHsps and gene names of eggplant sHsps were revised according to their molecular weights in the eggplant genome database based on Hirakawa et al. [22].

Gene Structures, Conserved Motifs and Protein Functional Network Analysis
The exon and intron structures were illustrated using the Gene Structure Display Server (GSDS, http://gsds.cbi.pku.edu.cn accessed on 6 June 2021) [26] by aligning the predicted cDNA sequences with their corresponding genomic DNA sequences. The conserved motifs in the encoded proteins were analyzed using the MEME online program (http://meme.sdsc.edu, v4.9.0, accessed on 6 June 2021) [27]. MEME was run locally with the following parameters: number of repetitions = any, maximum number of motifs = 20 and optimum motif width = 6-100 residues for Hsf, sHsp, Hsp60, Hsp70 and Hsp100. The STRING protein interaction database (http://string-db.org/, accessed on 6 June 2021) was used to analyze the interaction networks of Hsf and Hsp proteins in the highly specific protein and parameter selection model plant species Arabidopsis thaliana.

Plant Materials, Growth Conditions and Stress Treatments
In the present study, two inbred eggplant lines (selected by the Vegetable Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou, China), the thermotolerant line 05-4 and the thermosensitive line 05-1, were used. Eggplant seedlings were cultivated under 25/20 • C day/night conditions and a 16/8 h day/night photoperiod in a growth chamber until the four true leaves period for treatments. For the HS treatment, the seedlings of 05-1 and 05-4 with four leaves were directly placed in the 42 • C light incubator (RXZ-1000B3, Jiangnan Instrument Factory, Ningbo, China). For the heat treatment used for RNA-seq, the 3rd mature leaves of treated seedlings were collected at 0 and 6 h after HS treatment and 10 plants were used for each treatment. For the heat treatment used for the qRT-PCR, the 3rd mature leaves from two different lines were harvested at 0 and 6 h. The samples were harvested, immediately frozen in liquid nitrogen and stored at −80 • C for RNA extraction. Three biological replicates were performed and each replicate had 10 plants.

RNA Extraction and Quantitative Real-Time PCR Analysis
Total RNA was extracted using a TransZol Plant kit (TransGen Biotech/TransBionovo, Beijing, China) and the cDNA was synthesized according to the manufacturer's instructions (Takara, Dalian, China). Primers with amplicon lengths of 80-150 bp were designed using Primer5 software. All primer sequences are listed in Table S14. Real-time qRT-PCR was conducted on a Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) using the SYBR Premix Ex Taq kit (TaKaRa, Dalian, China) according to the manufacturer's instructions. The 10 µL reaction system contained 5 µL of SYBR Green Supermix (2×), 4 µL of cDNA template (30 ng/µL) and 0.5 µL of each primer (10 µM). The qRT-PCR reaction was performed using the following parameters: pre-denaturation at 95 • C for 30 s, followed by 39 cycles of denaturation at 95 • C for 5 s, annealing at 60 • C for 15 s and extension at 72 • C for 15 s. The fluorescent signal was measured at the end of each cycle and the melting curve analysis was performed by heating the PCR product from 65 • C to 90 • C to verify the specificity of the primers. Three independent biological replicates were performed and the qPCR of each replicate was performed in triplicate. The relative expression levels of eggplant Hsf and Hsp genes were calculated using the 2 − Ct method [28]. The SmEF1a genes were used as internal controls.

Genome-Wide Identification and Analysis of Hsf and Hsp Gene Family Members in Eggplant
To search for Hsf and Hsp genes in eggplant, we used the conserved Hsf and Hsp domain consensus sequences of several proteins as BLASTP queries against the eggplant Horticulturae 2021, 7, 149 4 of 15 genome database (http://eggplant.kazusa.or.jp/, accessed on 6 June 2021). In addition, homology searches using identified protein sequences of Arabidopsis thaliana were performed. After automated database searching and a manual review, 24 and 117 genes were identified as members of the Hsf and Hsp families in eggplant, respectively, whose classification and naming were based on the rules of the Hsp gene families from Arabidopsis and tomato, including sHsp, Hsp60, Hsp70, Hsp90 and Hsp100. The Hsf and Hsp gene families in eggplant were relatively large compared with those in Arabidopsis and those in tomato and rice, respectively. The numbers of identified genes in the Hsf, sHsp, Hsp60, Hsp70, Hsp90 and Hsp100 families of eggplant were 24, 39, 21, 30, 17 and 10, respectively (Table 1). As shown in Supplementary Materials Table S1, the amino acid lengths for Hsfs ranged from 111 (SmHsfA1c) to 496 (SmHsfA1b), with deduced molecular weights from 12.2 kDa to 54.8 kDa and the predicted isoelectric points of Hsfs were divergent, ranging from 4.60 (SmHsfA3) to 9.64 (SmHsfA1d). The length of sHsp proteins ranged from 87 (Sm10.2-sHsp) to244 amino acids (Sm27.2-sHsp) and the predicted molecular weights were between 10.2 kDa (Sm10.2-sHsp) and 27.2 kDa (Sm10.2-sHsp). In addition, the predicted pI-values of sHsp proteins ranged from 4.56 (Sm10.2-sHsp) to 10.49 (Sm12.7-sHsp). The amino acids lengths were consistent with the molecular weights of Hsp60s. The amino acid number and molecular weight for Smcpn60-4 was the highest, while that for SmCpn60-7.3 was the lowest and the predicted pI-values ranged from 5.26 (SmCpn60-a1) to 10.29 (SmCpn60-7.3). The deduced length of the Hsp70 proteins ranged from 85 (SmmtHsc70-3) to 914 (SmHsp70-18) amino acids and the highest-and lowest-molecular-weight SmHsp70s were SmHsp70-18 (103.1 kDa) and SmHsp70-5 (11.5 kDa), respectively, while the pI values ranged from 4.52 (SmmtHsc70-3) to . The length of Hsp90 proteins ranged from 137 (SmHsp90-4.4) to 782 (SmHsp90-6) amino acids, the predicted molecular weights of Hsp90s were between 16.2 kDa (SmHsp90-4.4) and 89.6 kDa (SmHsp90-7.1) and the predicted isoelectric points ranged from 4.78 (SmHsp90-5) to 9.55 (SmHsp90-2.1). The longest amino acids lengths and highest molecular weights in Hsp100s were SmHsp100-ClpB1, with 979 amino acids and 110.2 kDa, respectively. In contrast, the smallest was SmHsp100-ClpC3; the predicted isoelectric points ranged from 5.38 (SmHsp100-ClpB3) to 9.07 (SmHsp100-ClpC1) and these proteins were distributed from the alkaline to acidic.

Phylogenetic and Sequence Structure Analysis of Hsf and Hsp Proteins in Eggplant
To evaluate the evolutionary relationship of the eggplant Hsf and Hsp proteins, a phylogenetic analysis of each family was performed based on the full-length amino acid sequences from Arabidopsis, eggplant and tomato and each family could be classified into different subfamilies. The SmHsf family contained three subfamilies: type A (18 genes), type B (5 genes) and type C (1 gene). Based on the phylogenetic tree, class HsfA had the maximum number of subclasses among the three classes and was closer to tomato Hsf proteins, which coincided with the botanical classification (Table S2). A total of 39 sHsp genes could be grouped into 12 distinct subfamilies, containing 6 groups of cytosolic sHsp genes, C-I, C-II, C-III, C-IV, C-V and C-VI and 2 groups of mitochondrial sHsp genes, MT I and MT II. Notably, the C-I sHsp group in the eggplant genome was large, containing 24 genes, compared with 6 in Arabidopsis (Table S3). The Hsp60 family was divided into 4 subfamilies, including cytosol-localized Cpn60 (12 genes), mitochondrion-localized Hsp60 (4 genes) and chloroplast-localized Cpn60-a (2 genes) and Cpn60-b (3 genes) ( Table S4). The Hsp70 family contains genes encoding 19 cytosolic Hsp70s, 4 binding proteins (BIPs, Hsp70 homologs in the ER), 3 mitochondrial Hsp70s (mtHsc70s) and 2 chloroplastid Hsp70s (cpHsc70s) (Table S5). Seventeen Hsp90 family genes could be divided into cytoplasm (Cyt), mitochondrial (MT), endoplasmic reticulum (ER) and chloroplast, containing 8, 3, 2 and 1 proteins, respectively (Table S6). The Hsp100 family can be classified into ClpB, C, D and X classes as follows: 3 ClpB proteins (designated as B1, B2 and B3), 4 ClpC proteins (C1, C2, C3 and C4), 1 ClpD protein (D1) and 2 ClpX proteins (X1 and X2) (Table S7).

Structure of Hsf and Hsp Genes and Conserved Motifs of Hsf and Hsp Proteins in Eggplant
To obtained further insights into the structural diversity of Hsf and Hsp genes in eggplant, we used the Multiple Expectation maximization for Motif Elicitation (MEME) [27] to predict the conserved motifs shared among the related proteins within these families. In each family, 20 putative motifs were identified. The details of these motifs are listed in Tables S8-S13. Most of the closely related members in the phylogenetic tree shared common motif compositions.
The exon/intron structures of eggplant Hsf and Hsp members were analyzed based on their coding sequences and the corresponding genome sequences. The eggplant Hsfs shared highly conserved exon/intron structures with 0-3 intron phases ( Figure 1A). The intron phases were remarkably well conserved among family members. Most of the eggplant sHsps did not contain introns and only a few had 1-3 introns ( Figure 1B). Interestingly, in the Hsp60 family, two members, SmCpn60-8 and

Expression Patterns of Eggplant Hsf and Hsp Genes
To examine the heat response for Hsfs and Hsps in eggplant, an RNA sequencing profile (data not shown) in leaves of thermosensitive line 05-1 and thermotolerant line 05-4, at 0 and 6 h after HS treatment, was used. Hsf and Hsp genes were selected according to annotations and their expression profiles were analyzed. We analyzed the transcription levels of 18 Hsf, 25 sHsp, 6 Hsp60, 18 Hsp70, 11 Hsp90 and 6 Hsp100 genes in the leaves. As shown in Figure 3, for the thermosensitive line 05-1, 16 genes (89%) of the Hsf family were upregulated and two members, SmHsfA4e and SmHsfB3, were downregulated under HS conditions, which were more than in the thermotolerant line 05-4, in which 17 Hsf genes (94%) were upregulated by HS and only SmHsfA8 was downregulated. In contrast to line 05-1, SmHsfA4e and SmHsfB3 were strongly induced in treated 05-4 leaves. In the leaves of the thermotolerant line 05-4, among the upregulated members, the expression levels of most A (A1a, A1b, A3, A4a, A4b, A4d, A4e, A5, A6a and A6b), B1 and B2a were higher than those of other members under HS.

Expression Patterns of Eggplant Hsf and Hsp Genes
To examine the heat response for Hsfs and Hsps in eggplant, an RNA sequencing profile (data not shown) in leaves of thermosensitive line 05-1 and thermotolerant line 05-4, at 0 and 6 h after HS treatment, was used. Hsf and Hsp genes were selected according to annotations and their expression profiles were analyzed. We analyzed the transcription levels of 18 Hsf, 25 sHsp, 6 Hsp60, 18 Hsp70, 11 Hsp90 and 6 Hsp100 genes in the leaves. As shown in Figure 3, for the thermosensitive line 05-1, 16 genes (89%) of the Hsf family were upregulated and two members, SmHsfA4e and SmHsfB3, were downregulated under HS conditions, which were more than in the thermotolerant line 05-4, in which 17 Hsf genes (94%) were upregulated by HS and only SmHsfA8 was downregulated. In contrast to line 05-1, SmHsfA4e and SmHsfB3 were strongly induced in treated 05-4 leaves. In the leaves of the thermotolerant line 05-4, among the upregulated members, the expression levels of most A (A1a, A1b, A3, A4a, A4b, A4d, A4e, A5, A6a and A6b), B1 and B2a were higher than those of other members under HS.  A strong response to HS in all of the 25 sHsp genes from both lines (05-1 and 05-4) was observed, in which a majority of these genes were upregulated and only Sm15.6-sHsp and Sm15.8-sHsp were downregulated. After high-temperature treatment, the expression of SmHsp60-1, SmHsp60-2, SmCpn60-a2 and SmCpn60-b3 was increased in the two inbred lines, while SmHsp60-4 and SmCpn60-b3 was increased in 05-4 and no significant difference could be observed in 05-1. Among the 18 Hsp70 genes, the expression was remarkably changed in response to heat treatment in the thermosensitive 05-1 and thermotolerant A strong response to HS in all of the 25 sHsp genes from both lines (05-1 and 05-4) was observed, in which a majority of these genes were upregulated and only Sm15.6-sHsp and Sm15.8-sHsp were downregulated. After high-temperature treatment, the expression of SmHsp60-1, SmHsp60-2, SmCpn60-a2 and SmCpn60-b3 was increased in the two inbred lines, while SmHsp60-4 and SmCpn60-b3 was increased in 05-4 and no significant difference could be observed in 05-1. Among the 18 Hsp70 genes, the expression was remarkably changed in response to heat treatment in the thermosensitive 05-1 and thermotolerant 05-4 leaves and these genes were upregulated in both plants. Among these upregulated genes, the expression quantity of SmHsp70-1, SmcpHsp70-2 and SmHsp70-BIP4 was higher in 05-4, compared with 05-1. However, SmHsp70-14 and SmHsp70-15 were increased in the thermotolerant line, but decreased in the thermosensitive line. Considering the Hsp90 genes, the expression levels of most genes (SmHsp90-1.3, SmHsp90-1.4, SmHsp90-1.5, SmHsp90-2.3, SmHsp90-5, SmHsp90-6, SmHsp9-7.1 and SmHsp90-7.2) were increased and only SmHsp90-2.2 and SmHsp90-4.3 were downregulated in the two lines. Among the upregulated Hsp90 genes, gene expression levels of six genes in 05-4 were obviously higher than in 05-1. After heat treatment, SmHsp100-ClpB1, SmHsp100-ClpB2, SmHsp100-ClpB3, SmHsp100-ClpD1 and SmHsp100-ClpX2 expressions in the two lines were significantly increased. Among these genes, SmHsp100-ClpB1, SmHsp100-ClpB3 and SmHsp100-ClpX2 showed higher expression in 05-4 than in 05-1, but SmHsp100-ClpB2 was more abundant in the thermosensitive line.

Validation of Hsf and Hsp Gene Expression Levels by qRT-PCR
To verify the accuracy of the transcriptome sequencing, the expressions of 12 randomly selected genes were validated using quantitative real-time RT-PCR (qRT-PCR). The results showed that the expression pattern of each tested gene was similar to that of the transcriptome sequencing and the increase rate of all these Hsf and Hsp genes in the thermotolerant line 05-4 were significantly higher than those in thermosensitive line 05-1 (Figure 4). 05-4 leaves and these genes were upregulated in both plants. Among these upregulated genes, the expression quantity of SmHsp70-1, SmcpHsp70-2 and SmHsp70-BIP4 was higher in 05-4, compared with 05-1. However, SmHsp70-14 and SmHsp70-15 were increased in the thermotolerant line, but decreased in the thermosensitive line. Considering the Hsp90 genes, the expression levels of most genes (SmHsp90-1.3, SmHsp90-1.4, SmHsp90-1.5,  SmHsp90-2.3, SmHsp90-5, SmHsp90-6, SmHsp9-7.1 and SmHsp90-7.2) were increased and only SmHsp90-2.2 and SmHsp90-4.3 were downregulated in the two lines. Among the upregulated Hsp90 genes, gene expression levels of six genes in 05-4 were obviously higher than in 05-1. After heat treatment, SmHsp100-ClpB1, SmHsp100-ClpB2, SmHsp100-ClpB3, SmHsp100-ClpD1 and SmHsp100-ClpX2 expressions in the two lines were significantly increased. Among these genes, SmHsp100-ClpB1, SmHsp100-ClpB3 and SmHsp100-ClpX2 showed higher expression in 05-4 than in 05-1, but SmHsp100-ClpB2 was more abundant in the thermosensitive line.

Validation of Hsf and Hsp Gene Expression Levels by qRT-PCR
To verify the accuracy of the transcriptome sequencing, the expressions of 12 randomly selected genes were validated using quantitative real-time RT-PCR (qRT-PCR). The results showed that the expression pattern of each tested gene was similar to that of the transcriptome sequencing and the increase rate of all these Hsf and Hsp genes in the thermotolerant line 05-4 were significantly higher than those in thermosensitive line 05-1 (Figure 4).

Discussion
Many studies have suggested that Hsfs and Hsps play central roles in plant developmental and defense processes [29,30]. Benefiting from genome availability, the functions of the Hsf and Hsp family genes have been characterized in many plants. Although Hsfs and Hsps exist in all living organisms, their numbers vary in different plants. There are 22 Hsfs in Arabidopsis, 25 Hsfs in rice [18], 30 Hsfs in maize [20], 25 Hsfs in pepper [31] and 52 Hsfs in soybean [32]. Compared to the 27 sHsp genes in Arabidopsis [33], there are 35, 51 and 27 sHsp genes in pepper [31], soybean [34] and Chinese cabbage [35], respectively. Previous studies have identified 18 Hsp70 genes in Arabidopsis and 32 genes in rice [36]. The grapevine genome contains at least seven genes encoding members of the Hsp90

Discussion
Many studies have suggested that Hsfs and Hsps play central roles in plant developmental and defense processes [29,30]. Benefiting from genome availability, the functions of the Hsf and Hsp family genes have been characterized in many plants. Although Hsfs and Hsps exist in all living organisms, their numbers vary in different plants. There are 22 Hsfs in Arabidopsis, 25 Hsfs in rice [18], 30 Hsfs in maize [20], 25 Hsfs in pepper [31] and 52 Hsfs in soybean [32]. Compared to the 27 sHsp genes in Arabidopsis [33], there are 35, 51 and 27 sHsp genes in pepper [31], soybean [34] and Chinese cabbage [35], respectively. Previous studies have identified 18 Hsp70 genes in Arabidopsis and 32 genes in rice [36]. The grapevine genome contains at least seven genes encoding members of the Hsp90 super family [37].  reported 28 Hsf, 37 sHsp, 28 Hsp60, 20 Hsp70 and 5 Hsp100 genes in the poplar genome [19]. However, with the limited investigations into the molecular mechanism of heat tolerance, little is known about the Hsf family in eggplant.
In the present study, we identified 24 Hsf genes, 39 sHsp genes, 21 Hsp60 genes, 30 Hsp70 genes, 17 Hsp90 genes and 10 Hsp100 genes based on the eggplant genome (Table 1). Although the total number of Hsf and Hsp genes was similar to that of Arabidopsis [18,[38][39][40], rice [18,41] and tomato [42], the members of some specific Hsf and Hsp subclasses in eggplant were different from the other three species. Two members were identified that belonged to subclass HsfC2 in rice, while no eggplant Hsf members were classified into subclass HsfC2 and the same events were also observed in Arabidopsis thaliana [18] and pepper [31]. Rice is the model plant use for the monocot lineage and we inferred that the gene duplications led to the unique HsfC2 subclass in monocot species [17,42], which was the most marked difference between monocots and eudicots. In contrast, similar to tomato and Arabidopsis thaliana [18,43], eggplant also has members that were partitioned into the HsfA6 subclass, but no rice Hsf members were classified into subclass HsfA6 [44]. This finding suggested that Hsf genes were doubled and gained new functions during the evolution of the eggplant genome. Another interesting observation was that the subclass HsfA9 had 1 member in eggplant, compared with 4 members in pepper [31] and Eucalyptos grandis (Myrtaceae) contained at least 17 closely related HsfA9-encoding genes [17], suggesting a gene loss event during the evolutionary process of eggplant. However, there were two HsfA4 subclass genes in eggplant, more than in pepper CaHsfA4, which showed that some Hsfs might have the similar functions, as in maize [20]. The reasons for the increase in the HsfA9 genes need further investigation.
The phylogenetic analysis revealed that eggplant Hsf and Hsp members were more closely related to those from tomato than to those from Arabidopsis, which was consistent with the fact that both eggplant and tomato are members of the Solanaceae family [45]. Based on the previous analysis of the evolution of Hsfs and Hsps in Chinese cabbage [21,35], rice [46] and soybean [47], Hsf and Hsp genes essentially cover all the subfamilies and are relatively stable and conserved in the evolutionary process of eggplant and most of the Hsf and Hsp gene families were closely related to the evolutionary species.
Divergences in coding regions, particularly those that change the function of the gene, reflect amino acid altering substitutions and/or alterations in exon-intron structure [19]. The differences in intron and exon structure play important roles in the evolution of family genes. Structural analyses showed that the eggplant Hsf genes contained 0-7 introns and there were significant differences in the intron length; similar results were also obtained in cucumber [48], rice [49] and chickpea [50], but this result was different from that of pepper [31], for which all members have one intron. The number of introns of the Hsp gene family members in eggplant also showed differences, similar to the results of previous studies on poplar sHsp, Hsp60, Hsp70 and Hsp100 [19]. Qiao (2015), researching the pear Hsf and , researching the pepper Hsp20, showed a lack of conserved motifs among all the family genes and none of these genes contained the whole sequence, consistent with the eggplant Hsfs and Hsps in the present study [31,51]. We speculated that the deletion of introns and domains leads to structural changes during evolution, leading to functional diversity in Hsf and Hsp genes in eggplant; however, this theory needs experimental confirmation.
Hsfs, as transcriptional activators of Hsps, cooperate with Hsps to form a network responding to various stresses. These factors play a broad role in the tolerance to multiple environmental stress treatments apart from heat stress [52,53]. The comprehensive analysis of the expression for individual Hsf and Hsp members under HS was necessary for further functional analyses in plant thermotolerance [23,54]. The present study showed that most members of the eggplant sHsp, Hsp60, Hsp70, Hsp90 and Hsp100 families were induced by HS treatment in lines 05-1 and 05-4 and only a few members were significantly downregulated. Several studies have indicated that the expression and accumulation of heat shock proteins and heat shock transcription factors can enhance the thermostability of tomato [55], wheat [56] and rice [57]. Hsfs are activated under HS conditions and subsequently bind the HSE elements of the promoters of the Hsp genes to regulate the expression of downstream Hsp genes [17]. The accumulation of the Hsps effectively reduces the damage from HS and enhances thermotolerance by binding denaturing proteins and preventing them from irreversible aggregation [58]. Thus far, only sHsp has been shown to play a major role in improving plant thermotolerance in the form of molecular chaperones and cell membrane stabilizing factors [59]. However, the specific mechanisms of other Hsp genes are less well established. Previous studies have shown that the response of plants to high temperature was a quantitative trait controlled by multiple genes; some normal genes were closed and some stress tolerance-related genes were induced under high-temperature stress, thus altering plant morphogenesis, physiological functions and biochemical and molecular structures, which in turn influenced the growth of plants [60]. In addition, heat shock proteins are different from other stress proteins and have their own unique characteristics. In the present study, Hsps (sHsp, Hsp60, Hsp70, Hsp90 and Hsp100) showed species diversity, universal distribution and instantaneous response and structural conservation. For example, the synthesis of heat shock protein was fast, beginning between the first few minutes and tens of minutes and the expression lasted for up to several hours, occasionally continuing for 12 or more hours (Figure 4). Similar results were also observed in poplar [19] and grape [61].
In Arabidopsis, there are four members of the HsfA1 family, A1a, A1b, A1c and A1d [62]. Studies have shown that HsfA1a can directly sense heat stress and become activated and the same treatments also induced the binding to Hsp18.2 and Hsp70 promoters, as examined by chromatin immunoprecipitation [63]. Overexpressing HsfA1a enhances diverse stress tolerance by promoting stress-induced Hsp18.2 and Hsp70 gene expression [64]. In addition, AtHsfA1 was also related to drought stress [65] and programmed cell death [66]. Thus, in eggplant, HsfA1 may also play a similar function to AtHsfA1 and simultaneously communicate with Hsps. Increasing evidence suggests that Hsp is one of the most important heat stress proteins regulated by Hsf and is the material basis of the response of plant cells to high temperature damage [67][68][69]. Once exposed to high temperature, most of Hsf and Hsp genes in eggplant were induced to express rapidly and the expression level of these genes in the thermotolerant line was much higher than that in the thermosensitive line. Therefore, the Hsf-Hsp involved protein degradation pathway is also the main pathway of eggplant response to high temperature stress and may play an important role in the production of heat-tolerance in eggplant. The results provide a foundation for further functional research of these genes in eggplant, which could potentially be useful for elucidating the mechanism of thermotolerance in eggplant, even in other solanaceous plants.

Conclusions
In the present study, 24 Hsf genes and 117 Hsp genes (including sHsp, Hsp60, Hsp70 and Hsp100) were identified from the eggplant genome. The phylogeny, gene structure, expression profiles and heat stress responses of these genes were analyzed. The total number of Hsf and Hsp genes of eggplant was similar to that of Arabidopsis, tomato and rice, covering all the subfamilies, and the gene structure and motif composition were relatively stable and conserved in the evolutionary process. SmHsf genes, as key transcriptional activators of Hsp genes, regulated different subfamilies of Hsps in eggplant. Most of Hsf and Hsp genes are highly induced by HS in eggplant leaves, which indicated these genes participate in the response to heat stress. The expression levels of these genes in the thermotolerant line were enhanced significantly higher than those of in the thermosensitive line under HS in eggplant, which may be the main reason for strong thermotolerance in thermotolerant eggplant. According to the above results, it is expected to evaluate the thermotolerance of different eggplant resources by analyzing the expression change of specific Hsf and Hsp genes under HS and even the thermotolerance of other solanaceae species resources. The present study was undertaken to establish a solid foundation for functional research on the eggplant Hsf and Hsp gene families and broaden our understanding of the mechanism of thermotolerance mediated by Hsf and Hsp genes in solanaceous plants.