Identification of WRKY gene family and characterization of cold stress-responsive WRKY genes in eggplant

Plant treatment

In the present study, the seedlings of ‘sanyueqie’ eggplant variety were germinated and grown in a growth chamber at 28 °C with a 16/8 h light/dark photoperiod. The cotyledons of plants were inoculated with the cultures at the cotyledon stage and transferred to 22 °C to induce gene-silencing. When the fourth leaves were fully expanded, the seedlings were transferred into 4 °C for five days, followed by three days of recovery at 28 °C.

Identification of WRKY family in eggplant

To identify the WRKY proteins in eggplant, two methods were employed. We searched the eggplant genome database (http://eggplant.kazusa.or.jp/) for WRKY proteins based on the sequences of conserved WRKY domains in Arabidopsis (Dong, Chen & Chen, 2003). All peptide sequences of eggplant were download from the eggplant database. The Hidden Markov Model (HMM) of WRKY was used to BLASTP. The candidate WRKY transcription factors were obtained subsequently. The Pfam databases (http://pfam.xfam.org/) and SMART databases (http://smart.embl-heidelberg.de/) were used to validate the candidate WRKY proteins. ProtParam tool (https://web.expasy.org/protparam/) was used to predict the molecular weights (MWs) and isoelectric points (pIs) of WRKY proteins.

Phylogenetic analysis of SmWRKY proteins

To understand the evolutionary relationship between the eggplant and other plants, multiple sequence alignment was constructed using CLUSTAL W based on the conserved WRKY domain sequences of WRKY proteins from eggplant, tomato, Arabidopsis and rice (Thompson, Higgins & Gibson, 1994; Ross, Liu & She, 2007). To know the relationship among eggplant WRKY proteins, the phylogenetic tree of eggplant WRKY proteins was conducted with the conserved WRKY domain sequences of WRKY proteins from eggplant. Phylogenetic analysis was performed with MEGA v5.1 (Tamura et al., 2011; Hall, 2013). A phylogenetic tree was produced following the neighbor-joining method.

Intron-exon structure and conserved motifs analysis

The intron-exon structures of SmWRKY genes were determined by submitting a txt file of gene features with gff3 format to Gene Structure Displayer Server (http://gsds.cbi.pku.edu.cn/) (Hu et al., 2015) (Data S1). The conserved motifs in the WRKY proteins were predicted using the MutipleEm for Motif Elicitation (MEME) program (http://alternate.meme-suite.org/tools/meme). The parameters in this study were set as follows: 10 motifs should be found, minimum width was 15, maximum width was 50. Interpro Scan was used to annotate the ten motifs.

Virus-induced gene silencing (VIGS)

Differentially expressed WRKYs were identified from the RNA-seq data conducted previously. SmWRKY26 and SmWRKY32 were up-regulated in both cultivars after 4 °C treatment and the proteins were encoded by two genes with a high sequence similarity. We selected these two genes to performed a functional verification through VIGS. The fragments of SmWRKY26 and SmWRKY32 were amplified using the specific primers (Data S2), verified using VIGS tool in SGN (http://vigs.solgenomics.net/) and cloned to TRV (tobacco rattle virus) RNA2 vector (Fu et al., 2005; Huang et al., 2012a; Huang et al., 2012b). The empty vector TRV RNA2 was used as the control. The pTRV1, pTRV2, pTRV2-SmWRKY26 and pTRV2-SmWRKY32 were introduced into agrobacterium GV3101 with electroporation method, respectively. The mixture of Agrobacterium cultures containing pTRV1 and pTRV2 (1:1, v/v), pTRV1 and pTRV2-SmWRKY26 (1:1, v/v) and pTRV1 and pTRV2-SmWRKY32 (1:1, v/v) at OD₆₀₀ = 1 were incubated for 4 h in the dark at room temperature before inoculation. Three independent replicates were performed.

Quantative real time-polymerase chain reaction (qRT-PCR)

Specific primers were designed in the website (http://primer3.ut.ee/) (Data S2) (Koressaar & Remm, 2007). Total RNA was isolated from the tissue samples using the Trizol reagent according to the manufacturer’s instruction. Subsequently, RNA was reverse transcribed using the HiScript II Q Select RT SuperMix for qPCR kit (Vazyme, China). qRT-PCR was performed on an LightCycler 480 II (Roche, Switzerland) using the AceQ®qPCRSYBR® Green Master Mix (Vazyme, China). The PCR conditions consisted of denaturation at 95 °C for 5 min, followed by 40 cycles of denaturation at 95 °C for 10 s, annealing and extension at 60 °C for 30 s. Melt curve analysis was performed to determine the specificity of reactions. The transcript level was calculated according to the ΔΔCt method (Livak & Schmittgen, 2001). The eggplant Actin gene (Sme2.5_01462.1) was used as internal control. Three biological and technical repetitions were performed during the whole experiment.

Identification of WRKY protein family in eggplant

To identify the WRKY proteins in eggplant, BLASTP search was performed against the eggplant genome database (http://eggplant.kazusa.or.jp/) using the conserved WRKY domains in Arabidopsis. About 79 putative WRKY proteins were identified. Meanwhile, 68 putative WRKY proteins were identified using HMM (Hidden Markov Model) search program with a Hidden Markov Model (PF03106). By removing the redundant proteins and proteins without WRKY domain verified by Pfam and SMART programs, a total of 58 WRKY proteins were finally confirmed in eggplant. Obviously, the WRKY family of eggplant was smaller than those of other Solanaceae plants such as tomato (81 members) (Huang et al., 2012a; Huang et al., 2012b), pepper (Capsicum annuum L.) (61 members) (Cheng et al., 2016) and potato (82 members) (Liu et al., 2017a; Liu et al., 2017b).

The length of WRKY proteins in eggplant ranged from 134 to 1046 aa, with an average length of approximately 396 aa. The proteins with minimum length and maximum length were SmWRKY31 and SmWRKY47, respectively. Correspondingly, the molecular weights ranged from 15.75 kDa (SmWRKY31) to 116.54 kDa (SmWRKY47). The pIs of 58 SmWRKY proteins ranged from 5.13 to 9.72, among which 34 pIs<7 and 24 pIs>7 (Table 1) (Data S3).

Classification and phylogenetic analysis of SmWRKY proteins

Based on the number and amino acid sequences of the conserved WRKY domains, eggplant WRKY family can be classified into three main groups or seven subgroups, which is consistent with other plants. The group I contained 13 WRKY members which had two conserved WRKY domains, while 37 members were classed into group II (6 group II a, 7 group II b, 13 group II c, 8 group II d and 5 group II e). The remaining six WRKY proteins belonged to group III (Table 1).

To study the evolutionary relationships of WRKY proteins between eggplant and other plants, a phylogenetic tree was constructed based on the conserved sequences of WRKY domains from eggplant, tomato, Arabidopsis and rice. All the WRKY proteins were divided into seven subgroups (Fig. 1). Phylogenetic analysis indicated that WRKY proteins in eggplant and tomato were closely related, compared with those in Arabidopsis and rice, such as SlWRKY54 and SmWRKY1, SlWRKY53 and SmWRKY3, SlWRKY31 and SmWRKY26, SlWRKY33 and SmWRKY32, SlWRKY3 and SmWRKY8 and so on (Data S4).

To determine the phylogenetic relationships among the SmWRKY proteins, a phylogenetic tree was also constructed with WRKY conserved sequences using MEGAv5.1. As shown in Fig. 2, most of the SmWRKY members clustered in accordance with the classification, while SmWRKY2, SmWRKY4, SmWRKY10, SmWRKY11, SmWRKY41, SmWRKY42, SmWRKY47, SmWRKY54, SmWRKY55 and SmWRKY58 were not clustered into the corresponding branches.

Structural analysis of WRKY variants

The WRKY domain consists of a WRKY amino acid sequence at the N-terminal and a zinc-finger structure that is either Cx4–5Cx22–23HxH or Cx7Cx23HxC at the C-terminal. Some variants in WRKY domain were reported in the earlier studies. For example, the core sequence of WRKY domain was replaced by WRRY, WSKY, WKRY, WVKY or WKKY (Eulgem et al., 2000; Xie et al., 2005; Wei et al., 2016). The amino acids of GQK were substituted by GKK or EIG. Interestingly, WHKF and WDKF were found in eggplant WRKY proteins and GEK replaced GQK following the WRKY sequence (Fig. 3). Furthermore, it existed that the sequence of the zinc-finger motif at the C-terminal was Cx23RxH (SmWRKY58) or Cx23QxH (SmWRKY42) in eggplant WRKY proteins (Fig. 3). In SmWRKY11 protein, the sequence between the core domain (WRKYGQK) and the first cysteine of zinc-finger was two amino acids less than those normal sequences (Fig. 3). SmWRKY42 and SmWRKY58 were lack of the first histidine of conserved zinc-finger (Fig. 3). It may result in the failure of zinc-finger formation.

Intron-exon structure and conserved motifs of SmWRKY genes

To understand the intron-exon structure of SmWRKY genes, we submitted the gff3 file of SmWRKY genes to Gene Structure Display Server (http://gsds.cbi.pku.edu.cn/) (Data S1). As shown in Fig. 4, the scope of intron number in eggplant WRKY family was from 0 (SmWRKY10) to 11 (SmWRKY9), with an average number of 4.4. In rice and cassava, the number of introns varied from 0 to 8 and 1 to 5, respectively (Xie et al., 2005; Wei et al., 2016). Group I contained 3-5 introns with the exception of SmWRKY9 (11 introns). Group II a contained 3-5 introns, except for SmWRKY38 (2 introns). Group II b contained 2–5 introns. Although several WRKY proteins were not clustered together in group II c, most of WRKYs contained 2 introns. The number of intron in group II d WRKYs was almost 2, excluding SmWRKY10 (0 intron) and SmWRKY20 (5 introns). All of the group II e SmWRKYs consisted of 2 introns. The majority of group III WRKY genes also included 2 introns.

To further know the structure of eggplant WRKYs and their phylogenetic relationships, conserved motifs were analyzed using MEME tool and annotated with InterPro Scan. As we set, ten motifs were found in the sequences (Fig. 5). The first three motifs were annotated as WRKY DNA-binding domain. All the SmWRKY proteins contained two WRKY motifs at least. These results suggested that the features were conserved in the WRKY family identified from eggplant. Notably, almost the members in each group contained the same motifs. However, Yang et al. (2019) mentioned that the function of family members largely varied, even if these members were from the same clade. Therefore, these WRKY proteins classified into the same group may differentially function. Although SmWRKY18 belonged to group II c, it had the similar conserved motifs with the group II d.

The differentially expressed SmWRKYs in response to cold stress

Eggplant is sensitive to cold stress and often undergoes the cold stress during its growth and development, especially in the early spring (Zhou et al., 2018a; Zhou et al., 2018b). Therefore, it is important to explore the mechanism for cold tolerance and how to improve this tolerance. Previous studies have shown that WRKY transcription factors regulated the plant defense against cold stress (He et al., 2016; Zhang et al., 2016). In order to study the function of WRKY gene family in eggplant response to cold stress, the leaf samples from a cold-tolerant genotype and a cold-sensitive genotype were collected with the three biological replicates when the seedlings were treated for twelve hours. Then the differentially expressed WRKY genes in eggplant cultivars to cold stress were investigated from RNA-seq data which has been not published yet (Data S5). As shown in Fig. 6, twenty SmWRKY genes were identified in each cultivar. Interestingly, sixteen WRKY genes were common in both cultivars. Meanwhile, the sixteen WRKY genes showed similar expression pattern in these two cultivars when exposed to cold stress. Only SmWRKY1 and SmWRKY8 were blocked in the sensitive cultivar (Fig. 6A). SmWRKY4, SmWRKY7, SmWRKY29 and SmWRKY39 were also down-regulated in the tolerant cultivar except these two WRKY genes, (Fig. 6B). The transcript level of the rest of WRKY genes in both cultivars were enhanced. These results indicated that WRKY transcription factors could respond to cold stress and constituted a complicated network to regulate the cold tolerance in eggplant.

Silencing SmWRKY26 and SmWRKY32 increased the sensitivity to cold

VIGS was an effective tool to study the gene functions (Zhang et al., 2018; Xu et al., 2019). VIGS mediated by TRV has been successfully employed in the functional analysis of eggplant genes (Zhou et al., 2018a; Zhou et al., 2018b). We conducted a functional validation for SmWRKY26 and SmWRKY32 using VIGS. The fragments of SmWRKY26 and SmWRKY32 were cloned into the TRV silencing vector. Subsequently, these vectors were introduced into the eggplant. Using the online VIGS tool, we found that the transcript level of SmWRKY22 was likely affected. To detect the efficiency of silencing, the transcript level of these two WRKY genes was determined with qRT-PCR. As shown in Fig. 7A, the transcript level of SmWRKY26 and SmWRKY32 in the leaves with target gene silenced displayed a reduction of approximately 80%, compared with the leaves inoculated with the control bacterium. However, the expression of SmWRKY22 showed no alter in any one kind of the silenced seedlings. The seedlings were transferred into 4 °C for 5 days when the fourth leaf was fully expanded. The gene-silenced seedlings showed more severe wilting and shrink, while the control seedlings presented a little chlorosis (Fig. 7B). Whereafter, the leaf chilling injury index (LCI) was calculated after treatment. The LCI of SmWRKY26- and SmWRKY32- silenced seedlings were significantly higher than that of control seedlings (Fig. 7C). The phenotypes and index upon cold stress suggested that SmWRKY26 and SmWRKY32 positively regulated the tolerance to cold stress in eggplant.

Identification of WRKY proteins in eggplant

Characterization of gene families from eggplant genome is more and more important and imperative due to its importance for understanding the roles in plant responses to environmental stresses. In this study, we have identified 58 WRKY genes through searching the sequenced genome (Table 1). The WRKY gene family of eggplant was smaller than those of other plants such as the representative of Solanaceae plant tomato (81 WRKY genes). The WRKY members identified in this study was fewer than that reported in the PlantTFDB website (http://planttfdb.cbi.pku.edu.cn/family.php?sp=Sme&fam=WRKY#family_intro). 65 eggplant WRKY proteins were recorded in the plant transcription factor database. However, nine proteins either contained no WRKY feature structures, or had incomplete WRKY domains. Moreover, another two WRKY proteins were found in our study. 74 hits were obtained when “WRKY” was set as the keyword in eggplant genome. Among these, 55 hits were consistent with the WRKY proteins in our study. It lacked three WRKY proteins (Sme2.5_06310.1_g00004.1, Sme2.5_04517.1_g00003.1 and Sme2.5_00196.1_g00004.1) which were found in our study. The rest 19 predicted proteins were misannotated and listed in Data S6.

Interestingly, it was larger than that reported in Yang’s study (50 SmWRKY genes) (Yang et al., 2015), which might be due to that Yang identified the WRKY genes from transcriptome data.

Phylogenetic tree and variant analysis of eggplant WRKY

The phylogenetic tree among the eggplant WRKY proteins has shown their relatively conserved evolutionary relationships (Fig. 2). As shown in Fig. 2, some WRKY proteins were not clustered into the corresponding branches. These WRKY proteins were identified as WRKY variants through analyzing the amino acid sequence of these proteins (Fig. 3). For example, the C-terminal WRKY domain of WRKY2 protein had the sequence similarity with group III WRKY proteins and the last amino acid in the zinc-finger of the N-terminal WRKY domain is cysteine. These results suggested that the substitutions, addition and subtraction of amino acids of WRKY domain might be responsible for the changes in clustering. Furthermore, these WRKY variants probably lose the DNA-binding activity and the zinc-finger motif can not be formed, which lead to new molecular activities, novel biology functions and a non-functional protein. We found this phenomenon in our earlier study performed in soybean (Yang et al., 2017a; Yang et al., 2017b). The phylogenetic relationship between eggplant and other plants confirmed that WRKY proteins in eggplant contained the same subgroups with other reported plants (Data S4). In addition, eggplant WRKY proteins had a highly close relationship to tomato WRKY proteins, on account of being the Solanaceae.

Intron-exon structure and conserved motifs in eggplant WRKY

Analysis of the distribution of intron and exon is a better way to understand the phylogenetic relationship, because intron-exon structural diversity is an important part in the evolution of gene families (Wei et al., 2016). The WRKY genes in group I, group II a and group II b have more introns, whereas a vast majority of the rest WRKY genes have only two introns. It seems that eggplant WRKY proteins originate from the first group WRKY. The conserved motifs of eggplant WRKY proteins were identified using MEME tool and InterPro Scan. All the eggplant WRKY proteins contained at least two WRKY motifs with the exception of SmWRKY41. Unlike SmWRKY4, another WRKY variant in which the Cx22HxH replaced for Cx23HxH, the core sequence of the WRKY domain in SmWRKY41 was replaced by WRKYGQN. This specific variation was potentially responsible for one WRKY motif in SmWRKY41. Motif 10 was mainly included in group II d proteins except SmWRKY10 and SmWRKY11 (Fig. 5). Using InterPro Scan, motif 10 was identified as Zn-cluster domain (IPR018872) which was found associated with the WRKY domain (Babu et al., 2006). The types of conserved motifs contained in SmWRKY10 and SmWRKY11 more closely resembled those of group II e proteins, which might be one of the reasons for their clustering (Fig. 5).

SmWRKY genes in response to cold stress

Cold stress is an important factor affecting the growth and development of eggplant. To a certain extent, the response of WRKY genes to cold stress reflects the roles of WRKY genes in cold tolerance of plants (He et al., 2016). For example, Zhou et al. (2008) found that GmWRKY21 was induced by cold treatment and the transgenic plants over-expressing GmWRKY21 exhibited enhanced tolerance. In this study, we identified 24 differentially expressed SmWRKY genes in the two investigated materials from the RNA-seq data (Fig. 6). Four eggplant WRKY genes were specially differentially expressed in any one cultivar. This indicated that the differentially expression of eight WRKY genes might lead to difference in cold tolerance. In the common sixteen genes, WRKY26 and WRKY32 which shared a high sequence-similarity displayed up-regulated under cold stress. We conducted a VIGS experiment to verify the function of these two genes in response to cold. The different phenotypes and the chilling injury index between the silenced and the control plants illuminated that SmWRKY26 and SmWRKY32 could positively regulate the response to cold stress (Fig. 7).

Zhou et al. (2016) identified that several genes in response against Verticillium dahlia were annotated as WRKY genes, such as Unigene21521, Unigene11753, Unigene6502 and Unigene20920. SmWRKY26 and SmWRKY32 were the orthologous of Unigene20920 and Unigene6502 in cultivated eggplant, respectively. In addition, the proteins encoded by WRKY26 and WRKY32 had a 93% and 85% degree of similarity with SlWRKY31 and SlWRKY33 in tomato, respectively. SlWRKY31 and SlWRKY33 were reported as SlWRKY33A and SlWRKY33B to function in disease resistance and heat tolerance in tomato (Zhou et al., 2015). These results implied that orthologous could perform variant functions in different plants.