Genome-wide Identification of WRKY transcription factor family members in sorghum (Sorghum bicolor (L.) moench)

Abstract

WRKY transcription factors regulate diverse biological processes in plants, including abiotic and biotic stress responses, and constitute one of the largest transcription factor families in higher plants. Although the past decade has seen significant progress towards identifying and functionally characterizing WRKY genes in diverse species, little is known about the WRKY family in sorghum (Sorghum bicolor (L.) moench). Here we report the comprehensive identification of 94 putative WRKY transcription factors (SbWRKYs). The SbWRKYs were divided into three groups (I, II, and III), with those in group II further classified into five subgroups (IIa–IIe), based on their conserved domains and zinc finger motif types. WRKYs from the model plant Arabidopsis (Arabidopsis thaliana) were used for the phylogenetic analysis of all SbWRKY genes. Motif analysis showed that all SbWRKYs contained either one or two WRKY domains and that SbWRKYs within the same group had similar motif compositions. SbWRKY genes were located on all 10 sorghum chromosomes, and some gene clusters and two tandem duplications were detected. SbWRKY gene structure analysis showed that they contained 0–7 introns, with most SbWRKY genes consisting of two introns and three exons. Gene ontology (GO) annotation functionally categorized SbWRKYs under cellular components, molecular functions and biological processes. A cis-element analysis showed that all SbWRKYs contain at least one stress response-related cis-element. We exploited publicly available microarray datasets to analyze the expression profiles of 78 SbWRKY genes at different growth stages and in different tissues. The induction of SbWRKYs by different abiotic stresses hinted at their potential involvement in stress responses. qRT-PCR analysis revealed different expression patterns for SbWRKYs during drought stress. Functionally characterized WRKY genes in Arabidopsis and other species will provide clues for the functional characterization of putative orthologs in sorghum. Thus, the present study delivers a solid foundation for future functional studies of SbWRKY genes and their roles in the response to critical stresses such as drought.

Introduction

WRKY transcription factors (TFs), one of the largest TF families in plants, regulate various biological processes, including stress responses. WRKY proteins contain a conserved WRKYGQK motif at their N-terminus, along with a 60-amino-acid-long zinc finger motif at their C-terminus [1]. These two motifs are essential for the binding of WRKY TFs to the W-box cis-element [(T)TGAC(C/T)] located within the promoters of their target genes. WRKY proteins can be classified into three groups (I, II, and III) according to the number of WRKY domains and the type of zinc finger motif, i.e., C₂H₂ or C₂HC [2]. Group I members have two WRKY domains and the C₂H₂-type zinc finger. Group II members have only one WRKY domain and a C₂H₂ zinc finger motif and can be further classified into five subgroups (IIa, IIb, IIc, IId, and IIe) based on the sequence of the DNA-binding domain. Finally, group III members have one WRKY domain and a C₂HC-type zinc finger [3, 4].

Several studies have demonstrated that WRKY TFs regulate various biological processes and control gene expression via a combination of positive or negative regulation [5, 6]. WRKY TFs have been reported to be involved in responses to biotic stresses [7], developmental processes such as senescence, embryogenesis, and seed development, as well as abiotic stresses [8, 9]. For example, the wheat (Triticum aestivum) TaWRKY10 gene is considered to be a key regulator in salt and drought responses by regulating stress-responsive genes [10]. Heterologous expression of TaWRKY1 and TaWRKY33 enhanced drought and heat tolerance in Arabidopsis plants [11]. Also, Arabidopsis plants heterologously expressing the maize (Zea mays) ZmWRKY40 gene exhibited improved tolerance to drought [12]. Similarly, heterologous expression of TaWRKY13 in Arabidopsis increased root length and proline content, and reduced malondialdehyde content, thus improving salt stress tolerance [13]. Overexpression of WRKY13 in Arabidopsis enhanced cadmium tolerance of transgenic plants by inducing the expression of PLEIOTROPIC DRUG RESISTANCE 8 (PDR8), encoding an ATP-binding cassette transporter [14].

WRKY can however also act as a negative regulator of gene expression. Heterologous expression of the cotton (Gossypium hirsutum) GhWRKY33 gene reduced drought tolerance of transgenic Arabidopsis plants [15]. Likewise, heterologous expression of ZmWRKY17 impaired salt stress tolerance in transgenic Arabidopsis and lowered the abscisic acid (ABA) content by repressing ABA-dependent and stress-responsive genes [16].

Beyond the functional characterization of WRKY genes in Arabidopsis, the functions of many WRKY genes remain to be validated in non-model species. Indeed, few studies have tested the contribution of WKRY genes in their species of origin. Overexpression of TaWRKY2 in wheat enhanced tolerance to drought stress and increased yield [17]. Overexpression of OsWRKY11 enhanced tolerance to drought and heat in transgenic rice [18]. The grapevine (Vitis amurensis) VaWRKY12 gene enhanced cold tolerance in transgenic grapevine calli [19]. In wild sorghum (Sorghum propinquum), SpWRKY controls seed shattering but is unrelated to seed shattering genes selected during domestication, as it likely arose recently [20]. These studies confirm that WRKY TFs play important roles and suggest their potential use for crop improvement in terms of stress tolerance.

Since the identification of the first WRKY gene from sweet potato (Ipomoea batatas) [21], genome-wide analyses in different species have identified many WRKY genes, including 171 WRKY genes in wheat [22], 119 in maize [23], 103 in rice (Oryza sativa) [24], 71 in sesame (Sesamum indicum) [25], 70 in chickpea (Cicer arietinum) [26], and 79 in potato (Lycopersicum tuberosum) [27]. Sorghum is the fifth most important cereal crop in terms of production and dedicated arable land, and displays unique adaptations that allow it to withstand harsh conditions at different growth stages. Sorghum is also an excellent model for TF studies [28–30]. The availability of a complete genome assembly for sorghum now provides an opportunity for the genome-wide identification of SbWRKY genes. To gain insight into the roles of SbWRKYs in plant responses to stresses such as drought, we used a variety of approaches to identify and functionally characterize 94 putative members of the WRKY family in sorghum.

Material and methods

Identification of WRKY family genes in sorghum

We collected data from the following databases to identify putative SbWRKY genes. The Plant Transcription Factor Database version 4 (http://planttfdb.cbi.pku.edu.cn/) was used to download the amino acid sequences of sorghum WRKY proteins, and "WRKY" was used as a query to search against the Grassius Transcription Factor Database (https://grassius.org/grasstfdb.php). We used the WRKY domain ID (PF03106) to identify putative WRKY proteins encoded by the S. bicolor genome (v3.1) through the Joint Genome Institute (JGI) (https://phytozome.jgi.doe.gov/pz/portal.html#). We also used the keyword “WRKY” as a search query in the MOROCOSHI Sorghum Transcription Factor Database (http://sorghum.riken.jp/morokoshi/Home.html).

We employed CD-HIT suite (http://weizhong-lab.ucsd.edu/cdhit_suite/cgi-bin/index.cgi?cmd=cd-hit) to remove redundant and incomplete sequences, and Simple Modular Architecture Research Tool (SMART) (http://smart.embl-heidelberg.de/#) [31] to confirm that the sequences contained WRKY domain(s).

For all identified SbWRKY proteins, we obtained their predicted isoelectric point (pI) and molecular weight (MW) from the ExPASy proteomic server (http://web.expasy.org/protparam).

Chromosome mapping and tandem duplications of SbWRKY genes

Information about the chromosomal locations of all identified SbWRKY genes was obtained using the Phytozome BioMart tool (https://phytozome.jgi.doe.gov/biomart/martview/), and the genes were mapped onto sorghum chromosomes by MapChart (v.2.32). Tandem duplications of SbWRKYs were based on the following criteria: genes within a 100-kbp region on an individual chromosome with a sequence similarity ≥ 70% [32]. We calculated sequence similarities using EMBOSS Water, which uses the Smith-Waterman local pairwise alignment algorithm (https://www.ebi.ac.uk/Tools/psa/emboss_water/).

Classification and phylogenetic analysis of SbWRKY genes

Arabidopsis WRKY amino acid sequences obtained from the Arabidopsis Information Resource (TAIR) (https://www.arabidopsis.org/)) together with our SbWRKY sequences were used to construct a phylogenetic tree and classify SbWRKY genes. We performed multiple sequence alignments with ClustalW for AtWRKY and SbWRKY protein sequences and constructed a phylogenetic tree using MEGA v7.0 (https://www.megasoftware.net/) and the p-distance model. Pairwise deletion and 1,000 bootstrap replicates were used for the neighbour-joining method [33]. Based on the phylogenetic tree of AtWRKY and SbWRKY sequences, SbWRKY genes were classified into different groups and subgroups. The identification of putative sorghum WRKY orthologs in Arabidopsis was based on sequence alignment data and the phylogenetic tree [27].

Gene structure analysis of SbWRKY genes

The exon-intron structure of SbWRKYs was analyzed using the Gene Structure Display Server (GSDS v. 2.0) (http://gsds.cbi.pku.edu.cn/) from the Center for Bioinformatics at Peking University [34]. The genomic sequence and coding sequence (CDS) of each SbWRKY gene were used to predict the exon-intron pattern.

Conserved motif distribution analysis of SbWRKY genes

Conserved motifs in the SbWRKY proteins were identified using Multiple Em for Motif Elicitation (MEME) (v. 4.9.0; http://meme.nbcr.net/meme/)) with the following parameters: maximum motif number: 20; site distributions: any number of repetitions; minimum and maximum width: 10 and 50, respectively [35].

Gene ontology annotation and analysis of cis-acting elements

The gene ontology (GO) annotation analysis for SbWRKY proteins was performed using the Blast2GO tool with default parameters [36]. We screened protein sequences using the Basic Local Alignment Tool for Proteins (BlastP), followed by functional analysis, including mapping and annotation. Moreover, we collected information related to the biological process, molecular function, and cellular component associated with each SbWRKY. We also analyzed the cis-acting elements of SbWRKY genes by extracting 1,500 bp of upstream region for all SbWRKY genes and running the sequences through the online website PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/).

Digital expression pattern analysis of SbWRKY genes

To survey SbWRKY expression profiles, Affymetrix transcriptomic array data for sorghum were obtained from the Sorghum Functional Genomics Database (http://structuralbiology.cau.edu.cn/sorghum/pattern.php). The Genevestigator platform was used to analyze the expression profiles of SbWRKYs under different environmental stresses (drought, salt, ABA, and heat) with different samples stored in the Genevestigator platform [37]. The SbWRKY expression profiles heatmaps were generated using a hierarchical clustering analysis tool on the Genevestigator platform.

Drought stress treatment and samples collection

Seeds of Sorghum bicolor genotype (SX44B) used in this study were provided by Professor Zhang Fuyao (Sorghum Institute, Shanxi Academy of Agricultural Sciences, Shanxi, China). The seeds were surface-sterilized with 75% (v/v) ethanol and 5% (v/v) sodium hypochlorite for 1–2 min, and then rinsed them three times with distilled water. Sterilized seeds were allowed to germinate on two layers of water-soaked paper and incubated at 25°C in darkness for 3 d. Seedlings were transferred to pots containing a soil mixture of vermiculite and peat moss in a 1:1 ratio. Seedlings were kept in a growth chamber with 60–70% relative humidity, 28°C/23°C day/night temperature cycles, and a 16-h light/8-h dark photoperiod. Seedlings were maintained under normal growth conditions for two weeks before exposure to drought treatment, seedlings were treated with PEG8000 20% [38]. The seedlings shoot samples were collected at 0, 3, 6, 12, and 24 h, all samples were frozen in liquid nitrogen, and then stored at –80°C until used for RNA extraction.

Gene expression analysis by quantitative Real-Time PCR (qRT-PCR)

RNAs were extracted from the samples using (Promega, China) according to the manufacturer’s specifications. The first-strand cDNAs were synthesized by reverse transcription of 100 μg total RNA which was generated using the Easy Script First-strand cDNA Synthesis SuperMix Kit (TransGen Biotech, China). Synthesized cDNA was diluted 1:10 with nuclease-free water for use in qRT-PCR. The expression levels of the genes were normalized to the sorghum housekeeping gene GLYCERALDEHYDE 3-PHOSPHATE DEHYDROGENASE (GAPDH) gene as an internal control. Gene-specific primers were designed using the National Center for Biotechnology Information (NCBI) Primer-BLAST tool and were synthesized by Sangon (Beijing, China). Primers used in this study are listed in Supplementary S7 File. qRT-PCR was carried out using TransStart Green qPCR SuperMix UDG (TransGen Biotech, China) following the manufacturer’s instructions, on a Bio-Rad CFX96TM real-time PCR detection system (Bio-Rad, USA). Reaction parameters for thermal cycling were as follows: 94°C for 10 min, 40 cycles of 94°C for 5 sec and 55°C for 15 sec, 72°C for 10 sec. We performed RT-qPCR on three biological replicates and used the 2^–ΔΔCt method for quantification [39].

Results

Identification of WRKY family members in sorghum

Taking advantage of the availability of a complete genome assembly for sorghum, we identified SbWRKY family members using the keyword “WRKY” and the WRKY domain consensus sequence (PF03106) as queries in different databases. In this study, the presence of the WRKYGQK or WRKYGQK-like conserved domain was the basic criterion for the inclusion of genes in the SbWRKY family. We initially identified 134, 94, 99, and 97 transcripts from the TFDB, Grassius, MOROCOSHI, and JGI databases, respectively. We then used CD-HIT and multiple sequence alignments to remove redundant SbWRKY protein sequences and confirmed the presence of the WRKY domain in the remaining sequences by running all proteins through the SMART database. We thus removed all redundant sequences and those with an incomplete WRKY domain. A total of 94 non-redundant SbWRKY sequences were identified, and the following gene and protein data were summarized in Table 1. We named SbWRKYs according to their physical positions along sorghum chromosomes, starting with the upper arm of chromosome 1 and moving down to the lower arm (SbWRKY1 to SbWRKY94), as described previously [1] (Table 1). We used this set of SbWRKY protein sequences for further characterization. SbWRKY proteins ranged from 110 to 1,584 amino acids, with an average length of 390 amino acids. Their predicted MW and pI values ranged from 12.23 to 78.24 and from 4.75 to 10.06, respectively.

Table 1. Characteristics of the identified SbWRKY genes.

Gene Name	Gene Locus ID	Chromosome Location	Gene Start	Gene End	pI	MW	Conserved Heptapeptide	Zinc Finger Type	Domain Number	Group	Protein Length (aa)
SbWRKY1	Sobic.001G006600.1	Chr01	661532	669834	6.10	42.50	WRKYGQK	C2H2	1	lle	392
SbWRKY2	Sobic.001G055400.1	Chr01	4145000	4147281	9.42	46.08	WRKYGQK	C2H2	1	lle	434
SbWRKY3	Sobic.001G083000.1	Chr01	6408709	6412370	6.76	45.19	WRKYGQK	C2H2	2	l	424
SbWRKY4	Sobic.001G084000.1	Chr01	6506824	6511079	6.2	34.37	WRKYGQK	C2H2	1	llc	331
SbWRKY5	Sobic.001G095500.1	Chr01	7340447	7343822	10.05	43.61	WRKYGQK	C2H2	1	lld	406
SbWRKY6	Sobic.001G148000.1	Chr01	11929277	11931246	8.89	15.12	WRKYGQK	C2H2	1	NG	141
SbWRKY7	Sobic.001G162100.1	Chr01	13349344	13352564	9.78	37.88	WRKYGQK	C2H2	1	lld	352
SbWRKY8	Sobic.001G282400.1	Chr01	55413200	55415329	5.97	46.31	WRKYGQK	C2H2	1	lle	427
SbWRKY9	Sobic.001G332500.1	Chr01	62096233	62103315	7.34	52.62	WRKYGQK	C2HC	2	l	498
SbWRKY10	Sobic.001G381300.1	Chr01	66930715	66931988	5.60	33.95	WRKYGQK	C2HC	1	lll	328
SbWRKY11	Sobic.001G389000.1	Chr01	67559616	67561361	8.71	24.20	WRKYGEK	C2H2	1	lll	225
SbWRKY12	Sobic.002G008600.2	Chr02	791610	795938	9.71	29.42	WRKYGQK	C2H2	1	llc	272
SbWRKY13	Sobic.002G128400.2	Chr02	17755739	17757847	5.75	29.09	WRKYGQK	C2H2	1	lll	275
SbWRKY14	Sobic.002G168300.3	Chr02	52688090	52696535	6.92	17.74	WRKYGSK	C2HC	2	lll	1584
SbWRKY15	Sobic.002G174200.1	Chr02	55134836	55136822	5.67	32.89	WRKYGQK	C2HC	1	lll	310
SbWRKY16	Sobic.002G174300.1	Chr02	55188598	55190855	5.99	34.60	WRKYGQK	C2HC	1	lll	328
SbWRKY17	Sobic.002G202700.1	Chr02	59258884	59260622	6.46	34.72	WRKYGQK	C2H2	1	lla	327
SbWRKY18	Sobic.002G202800.1	Chr02	59275303	59277597	6.72	31.77	WRKYGQK	C2H2	1	lla	295
SbWRKY19	Sobic.002G242500.1	Chr02	63158788	63162729	6.94	64.54	WRKYGQK	C2H2	1	l	602
SbWRKY20	Sobic.002G355000.1	Chr02	71806701	71813179	6.24	65.56	WRKYGQK	C2H2	2	l	611
SbWRKY21	Sobic.002G418500.1	Chr02	76629585	76632298	9.17	37.05	WRKYGQK	C2HC	1	lll	354
SbWRKY22	Sobic.003G000600.1	Chr03	65678	68474	8.62	59.48	WRKYGQK	C2H2	1	llb	570
SbWRKY23	Sobic.003G037400.1	Chr03	3517021	3519078	6.07	26.65	WRKYGKK	C2H2	1	llc	260
SbWRKY24	Sobic.003G037500.1	Chr03	3520321	3522720	7.27	56.13	WRKYGQK	C2H2	1	llb	548
SbWRKY25	Sobic.003G040800.1	Chr03	3787973	3794918	9.75	30.73	WRKYGQK	C2H2	1	llc	295
SbWRKY26	Sobic.003G138400.1	Chr03	13400561	13405829	6.77	60.58	WRKYGQK	C2H2	1	llb	582
SbWRKY27	Sobic.003G199400.1	Chr03	52648841	52650568	5.57	23.75	WRKYGKK	C2H2	1	llc	216
SbWRKY28	Sobic.003G200700.1	Chr03	52845063	52846563	9.17	33.94	WRKYGQK	C2H2	1	llb	332
SbWRKY29	Sobic.003G226600.1	Chr03	56293002	56295461	8.7	44.22	WRKYGQK	C2H2	1	lle	413
SbWRKY30	Sobic.003G227300.1	Chr03	56472263	56479136	6.62	38.87	WRKYGQK	C2H2	1	llc	361
SbWRKY31	Sobic.003G242800.2	Chr03	58197072	58201064	6.45	35.97	WRKYGQK	C2HC	1	lll	346
SbWRKY32	Sobic.003G248400.1	Chr03	58718991	58722963	7.32	42.36	WRKYGQK	C2H2	1	llc	410
SbWRKY33	Sobic.003G276000.1	Chr03	61248940	61250384	8.64	23.95	WRKYGKK	C2H2	1	llc	225
SbWRKY34	Sobic.003G285500.1	Chr03	61880213	61882652	5.27	32.23	WRKYGQK	C2H2	1	lle	319
SbWRKY35	Sobic.003G287200.1	Chr03	62028638	62032127	7.08	26.67	WRKYGQK	C2H2	1	llc	246
SbWRKY36	Sobic.003G296300.1	Chr03	62835017	62837081	4.75	33.69	WRKYGQK	C2H2	1	lle	310
SbWRKY37	Sobic.003G337500.1	Chr03	66057329	66064893	5.85	26.64	WRKYGQK	C2HC	1	lll	236
SbWRKY38	Sobic.003G337600.1	Chr03	66070688	66076206	5.98	29.56	WRKYGQK	C2HC	1	lll	264
SbWRKY39	Sobic.003G337700.1	Chr03	66081053	66083714	5.87	40.92	WRKYGQK	C2HC	1	lll	377
SbWRKY40	Sobic.003G337800.1	Chr03	66089525	66092348	5.17	36.43	WRKYGQK	C2HC	1	lll	333
SbWRKY41	Sobic.003G337900.1	Chr03	66104667	66107290	5.91	29.63	WRKYGQK	C2HC	1	lll	277
SbWRKY42	Sobic.003G341100.1	Chr03	66392943	66395715	6.32	59.35	WRKYGQK	C2H2	2	l	556
SbWRKY43	Sobic.003G353000.1	Chr03	67192266	67198503	6.13	39.68	WRKYGQK	C2H2	1	NG	377
SbWRKY44	Sobic.003G444000.1	Chr03	74207255	74210127	4.76	38.85	WRKYGQK	C2H2	1	llc	354
SbWRKY45	Sobic.004G065900.1	Chr04	5352084	5353418	8.45	38.95	WRKYGQK	C2H2	1	lla	364
SbWRKY46	Sobic.004G117600.1	Chr04	12353902	12355528	5.81	39.57	WRKYGQK	C2H2	1	lle	375
SbWRKY47	Sobic.004G138400.2	Chr04	38911863	38913035	9.78	32.03	WRKYGQK	C2H2	1	lld	299
SbWRKY48	Sobic.004G271800.1	Chr04	61583724	61588753	5.67	51.61	WRKYGQK	C2H2	1	lle	497
SbWRKY49	Sobic.004G298400.1	Chr04	63778460	63782424	8.43	26.10	WRKYGQK	C2H2	1	llc	238
SbWRKY50	Sobic.004G312200.1	Chr04	64900147	64902898	5.09	61.31	WRKYGQK	C2H2	1	llb	578
SbWRKY51	Sobic.005G013400.1	Chr05	1211003	1212487	5.69	36.49	WRKYGEK	C2HC	1	lll	334
SbWRKY52	Sobic.005G013500.1	Chr05	1224392	1225924	6.31	38.95	WRKYGEK	C2HC	1	lll	361
SbWRKY53	Sobic.005G013600.2	Chr05	1235265	1241922	8.85	25.24	WRKYGEK	C2HC	1	lll	227
SbWRKY54	Sobic.005G013800.1	Chr05	1245144	1246547	5.88	30.67	WRKYGQK	C2HC	1	lll	271
SbWRKY55	Sobic.005G014000.1	Chr05	1261347	1263314	6.27	33.04	WRKYGQK	C2HC	1	lll	289
SbWRKY56	Sobic.005G014200.1	Chr05	1302596	1304021	6	30.80	WRKYGQK	C2HC	1	lll	272
SbWRKY57	Sobic.005G117400.2	Chr05	51657060	51660442	9.18	24.24	WRKYGQK	C2H2	1	llc	225
SbWRKY58	Sobic.006G051700.1	Chr06	38157163	38162677	5.97	12.23	WRKYGQK	C2HC	1	lll	110
SbWRKY59	Sobic.006G115700.1	Chr06	48356715	48362695	6.77	78.24	WRKYGQK	C2H2	2	l	740
SbWRKY60	Sobic.006G166300.1	Chr06	52377828	52383758	8.66	29.10	WRKYGQK	C2H2	1	llc	269
SbWRKY61	Sobic.006G201000.1	Chr06	55241047	55245365	7.01	55.79	WRKYGQK	C2H2	1	lle	532
SbWRKY62	Sobic.006G206000.1	Chr06	55572908	55574447	10.05	33.13	WRKYGQK	C2H2	1	lld	315
SbWRKY63	Sobic.007G077466.1	Chr07	8936871	8940145	7.7	49.41	WRKYGQK	C2H2	1	llc	455
SbWRKY64	Sobic.007G085300.1	Chr07	10754655	10756113	9.87	32.97	WRKYGQK	C2H2	1	lld	318
SbWRKY65	Sobic.007G111600.1	Chr07	41587770	41594141	6.05	61.78	WRKYGQK	C2H2	2	l	569
SbWRKY66	Sobic.007G118301.1	Chr07	51192936	51194468	6.48	35.62	WRKYGQK	C2HC	1	lll	340
SbWRKY67	Sobic.007G217700.3	Chr07	64590499	64600679	6.89	74.52	WRKYGQK	C2H2	2	l	685
SbWRKY68	Sobic.008G028600.1	Chr08	2541872	2544263	5.35	36.20	WRKYGQK	C2HC	1	lll	341
SbWRKY69	Sobic.008G029000.2	Chr08	2604929	2606770	6.72	33.68	WRKYGQK	C2HC	1	lll	294
SbWRKY70	Sobic.008G029200.1	Chr08	2612734	2614762	5.99	30.33	WRKYGQK	C2HC	1	lll	267
SbWRKY71	Sobic.008G029400.1	Chr08	2625199	2632242	8.94	25.19	WRKYGEK	C2HC	1	lll	225
SbWRKY72	Sobic.008G060300.1	Chr08	6462595	6464519	5.15	35.55	WRKYGQK	C2HC	1	lll	334
SbWRKY73	Sobic.008G107500.1	Chr08	50603514	50609969	7.71	52.42	WRKYGQK	C2H2	2	l	496
SbWRKY74	Sobic.008G153600.1	Chr08	58584243	58587317	10.06	39.29	WRKYGQK	C2H2	1	lld	371
SbWRKY75	Sobic.008G174100.1	Chr08	60851165	60856614	8.89	13.13	WRKSGQR	C2HC	1	lll	1163
SbWRKY76	Sobic.009G034800.1	Chr09	3182138	3188861	5.79	58.46	WRKYGQK	C2H2	1	llb	567
SbWRKY77	Sobic.009G068900.1	Chr09	7567284	7568764	6.96	22.09	WRKYGKK	C2H2	1	llc	206
SbWRKY78	Sobic.009G092100.1	Chr09	20912936	20928927	6.39	40.83	WRKYGQK	C2HC	1	III	378
SbWRKY79	Sobic.009G100500.1	Chr09	39826393	39831148	7.69	54.06	WRKYGQK	C2H2	2	l	517
SbWRKY80	Sobic.009G171600.1	Chr09	52696414	52699630	8.28	68.54	WRKYGQK	C2H2	2	l	649
SbWRKY81	Sobic.009G174300.1	Chr09	52963804	52966801	6.25	26.62	WRKYGEK	C2HC	1	lll	235
SbWRKY82	Sobic.009G206800.1	Chr09	55463596	55464452	9.2	25.32	WRKYGQK	C2H2	1	llc	241
SbWRKY83	Sobic.009G212800.1	Chr09	55836119	55837874	6.08	22.71	WRKYGKK	C2H2	1	llc	219
SbWRKY84	Sobic.009G234100.1	Chr09	57342319	57345920	6.38	44.44	WRKYGQK	C2H2	1	llc	424
SbWRKY85	Sobic.009G234900.1	Chr09	57418453	57422639	6.52	65.84	WRKYGQK	C2H2	1	llb	631
SbWRKY86	Sobic.009G238200.1	Chr09	57628850	57630763	5.81	29.08	WRKYGQK	C2HC	1	lll	272
SbWRKY87	Sobic.009G247300.2	Chr09	58261832	58265060	8.22	28.36	WRKYGQK	C2H2	1	llc	262
SbWRKY88	Sb09g029810	Chr09	58261882	58265547	6.07	36.12	WRKYGQK	C2H2	1	llc	343
SbWRKY89	Sobic.009G247700.1	Chr09	58309106	58310986	6.05	38.89	WRKYGQK	C2H2	1	lle	364
SbWRKY90	Sobic.010G035300.2	Chr10	2858141	2865169	6.79	65.39	WRKYGQK	C2H2	1	llb	625
SbWRKY91	Sobic.010G045700.1	Chr10	3566332	3570889	6.11	40.52	WRKYGQK	C2HC	1	lll	378
SbWRKY92	Sobic.010G148600.2	Chr10	42567454	42569300	6.01	41.69	WRKYGQK	C2H2	1	lle	385
SbWRKY93	Sobic.010G148800.1	Chr10	42829856	42831874	7.58	37.43	WRKYGQK	C2H2	1	lle	350
SbWRKY94	Sobic.010G209100.1	Chr10	55261687	55263735	9.64	37.82	WRKYGQK	C2H2	1	lla	348

pI, isoelectric point; MW, molecular weight; aa, amino acid.

Chromosome mapping of the SbWRKY genes and tandem duplication analysis

MapChart was used to determine the locations of the 94 SbWRKY genes, and they were distributed on all 10 sorghum chromosomes (Chr) (Fig 1). Chr 3 counted the highest number of SbWRKYs with 23 genes, corresponding to 24.5% of the entire gene family, followed by 14 genes on Chr 9, 11 genes on Chr 1, and 10 genes on Chr2. Chromosomes 6, 7, and 10 had the fewest number of genes, with only five SbWRKYs each. The remaining SbWRKY genes were located on Chr4 (six genes), Chr5 (seven genes), and Chr8 (eight genes).

Fig 1. Distribution of 94 SbWRKY genes on sorghum chromosomes.

Chr01-Chr10 above the colored bars indicates chromosome (Chr) numbers. The physical location of each SbWRKY gene is shown, and the gene name is indicated on the right side of each bar as SbWRKY#. Red boxes indicate tandem duplications, and green lines denote gene clusters.

Tandem repeats were identified based on previously reported criteria [40]; namely, two or more genes should be located within a 100-kbp window and display a sequence similarity of at least 70%. Gene cluster events were observed on six chromosomes. Specifically, there were five clusters on Chr3; two each on Chr2 and Chr9; and one cluster each on Chr1, Chr5, and Chr8. No clusters were found on Chr4, Chr6, Chr7, and Chr10. High-density clusters were detected on Chr5 and Chr3, and identified tandem repeats on two chromosomes: two group III genes (SbWRKY51 and SbWRKY52) on Chr5 and two group-IIc genes (SbWRKY87 and SbWRKY88) on Chr9 (Fig 1). These tandem-duplicated genes clustered together in the phylogenetic tree within their respective clades, and the sequence similarity metrics for the gene pairs are provided in S1 File.

Classification and phylogenetic analysis of SbWRKYs

To investigate the evolution of SbWRKY family members, an unrooted phylogenetic tree was constructed based on multiple sequence alignment between full-length protein sequences of 65 AtWRKYs and 94 SbWRKYs, using the neighbour-joining method in MEGA7.0 (Fig 2). The constructed phylogenetic tree was used to classify the SbWRKYs into three major groups (I, II, and III), according to the classification in Arabidopsis [1] (Fig 2). Of the 11 SbWRKYs in group I, 10 had two WRKYGQK motifs and two C₂H₂-type zinc finger motifs (C-X_3-4-C-X_22-23-H-X₁-H), corresponding to two full WRKY domains. Although the protein encoded by SbWRKY19 had only one WRKY domain, it belonged to group I on the phylogenetic tree. Fifty protein sequences with one WRKY domain and the C₂H₂-type zinc finger motif (C-X_4-5-C-X₂₃-H-X₁-H) were classified into group II. This group was further divided into five subgroups, IIa, IIb, IIc, IId, and IIe, with 4, 8, 20, 6, and 12 members, respectively. Group III contained 31 members with one WRKY domain and the C₂HC-type zinc finger motif (C-X₇-C-X₂₃-H-X₁-C). SbWRKY14 was unique in that it comprised of two WRKY domains with the C₂HC-type zinc finger motif (C-X₇-C-X₂₃-H-X₁-C). Thus, it had features associated with both group I and group III WRKYs, but was classified into group III based on its position in the phylogenetic tree. SbWRKY6 and SbWRKY43 did not belong to any group (Fig 2 and Table 1). Group II was the largest group and accounted for 53.2% of all putative SbWRKYs, which is similar to reports in wheat, soybean (Glycine max), and pepper (Capsicum annuum). Overall, the classification of SbWRKYs confirms their diversification, which suggests that different family members may have varied functions.

Fig 2. Phylogenetic tree of WRKY members in sorghum and Arabidopsis.

SbWRKY and AtWRKY protein sequences were aligned with ClustalW, and a phylogenetic tree was constructed with MEGA7.0 using the neighbour-joining method and 1,000 bootstrap replicates. The members were divided into groups I, II, and III, and group II was further divided into subgroups IIa, IIb, IIc, Ild, and Ile.

The highly conserved heptapeptide motif WRKYGQK was present in 81 SbWRKY proteins. We observed several heptapeptide variants in the remaining 13 proteins as follows: WRKYGEK in six proteins (SbWRKY11, SbWRKY51, SbWRKY52, SbWRKY53, SbWRKY71, and SbWRKY81); WRKYGKK in five proteins (SbWRKY23, SbWRKY27, SbWRKY33, SbWRKY77, and SbWRKY83); WRKYGSK in SbWRKY14; and WRKSGQR in SbWRKY75. Among the 94 identified SbWRKYs, only 11 had two WRKY domains, whereas the remaining members had one WRKY domain. SbWRKY protein sequences, genomic sequences, and CDS are provided in S2, S3, and S4 Files, respectively. AtWRKY protein sequences are provided in S5 File.

SbWRKY gene structure analysis

To obtain additional clues about the evolution of SbWRKY family members and their specific features, SbWRKY exon-intron structures were analyzed. The intron number of SbWRKY genes ranged from zero to seven, whereas their size varied. Among the 94 SbWRKY genes identified, 10 contained two exons (and one intron), 58 had three exons (two introns), eight had four exons (three introns), eight had five exons (four introns), five had six exons (five introns); the SbWRKY14 gene had eight exons and seven introns (Fig 3). Four genes (SbWRKY51, SbWRKY64, SbWRKY66, and SbWRKY88) lacked introns.

Fig 3. Exon–intron structures of SbWRKY genes.

We used the Gene Structure Display Server (GSDS) for gene structure analysis and constructed the phylogenetic tree using MEGA v7.0. In the gene diagrams, blue bars indicate upstream and downstream UTRs, yellow bars indicate coding sequences (CDS), and black lines indicate introns.

The exon–intron distribution patterns showed some similarities in terms of their numbers and positions within the same group. However, there were also differences within groups. For instance, all genes in group II had zero to five introns, all SbWRKY genes in subgroup IIb had five introns except SbWRKY22 and SbWRKY50, which had four, SbWRKY24 had two, and SbWRKY28 had one intron. Intron numbers in group III ranged from zero to seven, with SbWRKY14 being the only gene with seven introns among all SbWRKYs. These results indicate that there is considerable structural variation among SbWRKYs, which may correspond to functional diversification between closely related members (Fig 3).

Motif composition analysis of SbWRKYs

MEME (version 4.11) was used to analyze all SbWRKY protein sequences for conserved motifs, resulting in the identification of 20 distinct conserved motifs, ranging from 6 to 50 amino acids in length (Fig 4). Motifs 1, 2, 3, and 4 corresponded to the WRKY domain located at the C-terminus of sorghum WRKY proteins. Most SbWRKY members within the same group or subgroup shared a similar motif composition. Motifs 12, 15, and 20 were unique to group III. Motifs 7, 10, 11, and 18 were unique to group IIb, and motif 14 was exclusively detected in group I. All group I members had two WRKY domains except for SbWRKY19, as mentioned earlier, suggesting it may have lost its N-terminal WRKY domain. Motif 13 was unique to group IIe. Examples of motifs shared by different groups included motif 8, shared by groups IIe and IId, and motif 5, shared by groups I and IIc. Although SbWRKY6 and SbWRKY43 clustered with group III, they are not associated with any group; they contained motifs 1, 2, 6, 3, and 17 (Fig 4). Groups IIe and IId were two close subgroups in the phylogenetic tree, and the vast majority of their members contained motifs 8, 2, 4, 3, and 17. Both subgroups had a similar domain arrangement, which may be indicative of functional similarity. Some motifs occurred only in a few SbWRKY members, such as motifs 12 and 20, which were only present in SbWRKY14, SbWRKY75, and SbWRKY58.

Fig 4. Motif analysis of SbWRKYs.

(a) The distribution of 20 conserved motifs identified by MEME in the different groups of SbWRKYs. (b) Each motif is indicated by a different color. (c) Sequence logos for motifs 1–20.

Gene ontology annotation and analysis of cis-acting elements

Gene ontology (GO) annotations of 94 SbWRKY proteins were analyzed using the Blast2GO tool. The SbWRKY target genes were categorized into different functional groups under three main categories including biological processes, molecular functions, and cellular components (Fig 5). Under the larger umbrella of biological processes, most SbWRKYs were identified as being involved in the regulation of cellular processes, biosynthetic processes and different metabolic processes, as well as response to different stimuli, signalling, cell communication, and responses to other organisms, chemicals and stress. The molecular functions of SbWRKYs were associated mostly with DNA-binding, DNA-binding transcription factor activity, catalytic activity, acting on a protein, and hydrolase activity. The cellular component of this protein family included organelle and intracellular organelle (Fig 5). In addition, all SbWRKY proteins were predicted to be localized in the nucleus.

Fig 5. Gene ontology analysis of identified SbWRKYs.

The enrichment analysis shows the involvement of SbWRKY in biological processes, molecular functions, and cellular components.

Cis-acting elements within promoters are the binding sites through which transcriptional regulation is enacted. We therefore, extracted the 1.5-kbp promoter regions upstream of all SbWRKY genes from the sorghum genome assembly to identify cis-acting elements using the online tool PlantCARE. Thus, various cis-acting regulatory elements were found in all SbWRKY genes promoter regions. Featuring prominently in our list of cis-elements were stress-responsive elements, including MBS (MYB transcription factor binding site involved in drought inducibility), LTR (low-temperature responsive element), ARE (anaerobic induction responsive element), TC-rich repeats (defense-responsive and stress-responsive elements, WUN-motif (wound-responsive elements), and GC-motif (anoxic specific inducibility element). Phytohormone-responsive elements: ABA-responsive element (ABRE), methyl jasmonate (MeJA) responsive element (TGACG-motif and CGTCA-motif), auxin-responsive elements (AuxRR-core and TGA-element), salicylic acid-responsive element (TCA-element), and gibberellin-responsive element (GARE-motif). Multiple light-responsive elements were present in the promoters of SbWRKY genes, including Sp1, TCT-motif, GT1-motif, GATA-motif, GA-motif, BoX II, and G-box, as well as elements associated with development, including the CAT-box (element related to meristem expression), and the o2-site (metabolism regulation). The promoter related and binding sites elements were found included TATA-box, CAAT-box, A-box, HD-Zip, and W-box (a classic WRKY DNA-binding motif). Many unknown functions were detected included AAGAA-motif, DRE core, and MYB. The most common cis-acting regulatory elements in the SbWRKY promoter regions were TGACG-motif, ABRE, CGTCA-motif, CAAT-motif, MYB, TATA-box, and G-box. We identified W-box elements in the promoters of 38 SbWRKY genes. All SbWRKY genes contained at least one stress-responsive element along with other cis-elements, reflecting their potential functional variation (S6 File).

Digital expression analysis of SbWRKY genes at different growth stages and in different tissues

As a preliminary survey of the potential roles of SbWRKY genes during sorghum growth and development, the temporal and spatial expression profiles of SbWRKY genes were investigated using microarray data available from SorghumFDB. We used the Genevestigator platform for the analysis, and present the results as heatmaps. The sorghum accessions represented in the microarray datasets included R159, Fremont, Atlas, PI455230, PI152611, and AR2400 [41]. The microarray datasets comprised 37 samples representing leaves, roots, shoots, stems (pith and rind), and internodes. Seventy-eight of the 94 SbWRKY genes were represented in the microarray data and displayed distinct expression patterns across all tested tissues. For example, all 78 genes were expressed in leaves, 76 in roots, 77 in shoots and pith, 76 in internodes, and 75 in rind. Thirty-two of the 78 genes exhibited high expression levels in at least one tissue (Fig 6A). The number of genes with high expression levels (>65% expression) varied between tissues, although roots showed the highest number of highly expressed SbWRKY genes with 17 members, followed by nine in pith, seven in leaves, six in rind, five in internodes and four in shoots. The most highly expressed SbWRKY genes were SbWRKY19, SbWRKY83, SbWRKY45, SbWRKY79, SbWRKY5, SbWRKY42, SbWRKY73, SbWRKY22, SbWRKY34, SbWRKY72, SbWRKY25, and SbWRKY70 in different tissues (Fig 6A). Notably, SbWRKY72 was highly expressed in all tissues. By contrast, SbWRKY70 expression was only detected in leaves and shoot. Clustering analysis of SbWRKY expression patterns grouped rind and pith, this is consistent with their biological features.

Fig 6. Heatmaps of SbWRKY gene expression.

SbWRKY expression levels in different tissues (a) and at different growth stages (b). (c) Hierarchical clustering of SWRKY gene expression patterns under different environmental conditions, including drought, ABA, heat, salt and combination stress.

SbWRKY expression patterns at different growth and developmental stages (seedling, stem elongation, flowering, boot, and dough stage) were also analyzed (Fig 6B). SbWRKY genes were expressed differently (up- or down-regulated) at all stages, and those with high expression at different stages included SbWRKY74, SbWRKY75, SbWRKY19, SbWRKY5, SbWRKY45, SbWRKY79, SbWRKY25, and SbWRKY72. SbWRKY expression was slightly higher during the seedling, flowering, and dough stages, suggesting that they may be involved in stress responses during sensitive developmental stages to improve plant tolerance (Fig 6B).

Hierarchical clustering analysis of the expression patterns of 78 SbWRKY genes under different environmental stresses was performed in Genevestigator. Two major clusters were obtained, which divided SbWRKY genes into two groups. The first major cluster consisted of highly expressed genes, including SbWRKY45, SbWRKY79, SbWRKY83, and SbWRKY16, under various abiotic stress conditions such as drought, salt and ABA (Fig 6C). The second major cluster contained several sub-clusters of SbWRKY genes with different expression patterns, i.e., up- or downregulated at least 2.5-fold (in absolute terms), in response to drought, salt, ABA in different sorghum tissues (Fig 6C). Several genes were found to have a stable expression level across different tissues and may therefore be considered constitutively expressed. We hypothesize that other SbWRKY genes with low expression levels may work cooperatively with other proteins throughout plant development.

qRT-PCR expression analysis of SbWRKY genes in response to drought stress

Previous studies proved that WRKY genes involved in plant responses to drought stress in several crops such as maize, wheat, and rice [42]. To investigate the role of SbWRKY genes in drought responses in sorghum, we selected five genes (SbWRKY45, SbWRKY72, SbWRKY74, SbWRKY75, and SbWRKY79) for expression analysis by qRT-PCR in the shoot of sorghum seedlings subjected to drought stress. qRT-PCR results revealed that transcripts levels for these five SbWRKY genes were remarkably increased under drought stress at different time points (Fig 7) suggesting that these genes may function in this process. The relative expression of our selected SbWRKY genes peaked at different time points. SbWRKY75 exhibited the highest expression level, with an 89-fold change in expression after 6 h of drought stress, with a final decline after 24 h. The peak expression of SbWRKY74 occurred 12 h after the onset of drought stress, whereas SbWRKY45 and SbWRKY79 expression peaked 24 h into drought stress (Fig 7). Although the peak expression of SbWRKY72 and SbWRKY75 occurred at 6 h, the expression of SbWRKY72 gradually increased to this level, then decreased gradually at later time points. SbWRKY79 showed the least induction during drought stress relative to the other genes tested here. Overall, the expression pattern of these selected SbWRKY genes under drought stress conditions suggests that different SbWRKY genes may play an essential role in drought stress tolerance.

Fig 7. Relative expression of selected SbWRKY genes in response to drought stress.

Relative expression levels of five SbWRKY genes under drought stress. Genes expression was analyzed by RT-qPCR, with the 0 h sample used as untreated control (expression = 1). Error bars represent standard errors; data were calculated using the 2^–ΔΔCt method.

Discussion

WRKY TFs are key regulators of many processes in plants, including responses to abiotic and biotic stresses. In both model and non-model plants, considerable progress has been made towards identifying and functionally characterizing WRKY TFs, and many WRKY genes have been found to promote stress tolerance [42]. The completed genome assembly of sorghum now makes it possible to perform a genome-wide analysis of the SbWRKY gene family. We identified a total of 94 SbWRKY genes in S. bicolor, which is slightly higher than in other species; for example, there are 75 WRKY genes in Arabidopsis [1], 71 in sesame [25], 79 in potato, 85 in cassava (Manihot essculenta) [43], and 59 in grapevine [44]. In contrast, sorghum contains fewer WRKY genes than maize, soybean, or rice [23, 45, 46]. The present findings in sorghum, an important cereal crop and a model plant for drought tolerance, add to the recent identification of WRKY genes in a variety of plant species, including chickpea [26], Chinese jujube (Ziziphus jujube) [47], sugar beet (Beta vulgaris) [48], coffee (Coffea arabica) [49], pepper [50], eggplant (Solanum melongena) [51], Asian legume crops [52], sweet potato (Ipomoea batatas) [53], and pearl millet (Pennisetum glaucum) [54].

SbWRKY genes are distributed on all 10 sorghum chromosomes. Previous studies mapped many traits, such as stay-green phenotypes, lodging tolerance, pre-flowering drought tolerance, and yield-related components, to various chromosomal locations in sorghum [55, 56, 57]. Several SbWRKY genes map within these candidate regions; for example, SbWRKY26 is located on chromosome 3, close to a mapping interval for stay-green and pre-flowering drought tolerance Quantitative Trait Loci (QTL) in sorghum. Likewise, SbWRKY67 maps to chromosome 7, in a genomic region associated with a lodging tolerance QTL, whereas SbWRKY59 maps to the same region associated with pre-flowering drought tolerance, flowering time, and stay-green QTLs. These results suggest that several SbWRKY genes might control or contribute to stay-green, pre-flowering drought tolerance, or other traits related to stresses tolerance.

Gene duplication events affect genome expansion, family size, and the distribution of genes on chromosomes. Distinct types of duplication events, such as tandem and segmental duplication, differ in terms of the resulting number of gene copies and their distribution, and these factors are important for functional prediction. Tandem duplication of chromosome regions can also give rise to a cluster of family members, with subsequent structural and functional divergence over time leading to the expansion and evolution of the gene family [58]. Our analysis identified two tandem duplications of SbWRKY genes, which is a smaller number than that reported for potato [27], rice [24], wheat [22], or Arabidopsis [59]. Despite these differences, tandem duplications may have shaped the evolution of the SbWRKY gene family in sorghum.

Both tandem duplications and segmental duplications have played essential roles in the evolution and diversification of the WRKY gene family in plant species [60]. These evolutionary events were assigned to three types of gene duplication: tandem, segmental and whole-genome duplications [61]. Duplication events are significant for WRKY diversification, as duplicated WRKY genes may acquire new functions. In this study, we focused on tandem duplication events. We hypothesize that tandem duplications have played an important role in the evolution and diversification of WRKY genes in sorghum, although the expansion of this gene family likely arose mainly through other events. Tandem duplications were a critical but recent gene duplication contributor in Arabidopsis [62]. Previous studies have shown that gene duplication greatly accounts for new genes [63]. Gene duplication may result in sub-functionalization: for example, expansion of the functions among the wheat WRKY gene family members has occurred through tandem duplication and whole-genome duplication [64]. Tandem duplications and gene clusters have been previously described for multiple WRKY genes in rice within the same intergenic region [65].

The presence of the conserved WRKY domain, which binds to the W-box motif in the promoters of WRKY target genes, is the most essential characteristic of the WRKY family [66, 67]. We performed a classification of SbWRKY genes here according to the approach used in other crop species, based on phylogenetic tree topology. We also adopted the divisions of WRKY family members into the same groups and subgroups described in Arabidopsis: groups I, II, and III according to the number of WRKY domains and the type of zinc finger motif, along with further subdivision of group II into subgroups IIa, IIb, IIc, IId, and IIe [1]. In this study, 11, 50, and 31 SbWRKY genes were classified into groups I, II, and III, respectively; two genes did not belong to any group. Group II was the largest with 50 members and accounted for 53.2% of all SbWRKY genes. These results are consistent with WRKY group sizes in sesame, Arabidopsis, and sugar beet [25, 1, 46]. Among the subgroups, subgroup IIc was the largest with 20 SbWRKY genes, or 40% of the genes assigned to group II (Fig 2), which is similar to results in sugar beet [48], Arabidopsis [1], and soybean [68].

Whereas most SbWRKY proteins contained the conserved WRKYGQK motif, other heptapeptide variants were identified in 13 SbWRKY proteins (SbWRKY11, SbWRKY14, SbWRKY23, SbWRKY27, SbWRKY33, SbWRKY51, SbWRKY52, SbWRKY53, SbWRKY71, SbWRKY75, SbWRKY77, SbWRKY81, and SbWRKY83; Table 1). Similar variations in the heptapeptide motif were also identified in potato [27]. Several examples indicate that these differences may affect the binding ability of WRKY TFs to the W-box element. For example, two soybean WRKYs with the WRKYGKK motif variant were unable to bind to the W-box element [66]. In tobacco (Nicotiana tabacum), NtWRKY12 with the WRKYGKK motif bound to TTTTCCAC instead of the W-box consensus sequence (TTGACT/C) [69]. Therefore, further investigation is warranted to identify the preferred DNA-binding sequences associated with different WRKYGQK-like motifs.

Group-specific patterns were detected in their exon–intron structure. Indeed, SbWRKY genes within the same group had similar exon–intron patterns. The number of introns in SbWRKY genes ranged from 0 to 7, which is similar to that reported in chickpea [26]. Four SbWRKY genes lacked an intron, indicating that intron loss may have occurred during the evolution of this gene family; a similar observation was reported in the rice WRKY gene family [46]. Intron-less genes within plant gene families may imply their close relationship. Intron-less genes is not a specific feature of the SbWRKY gene family, as they have been reported in other gene families like GRAS-domain TFs, F-box TFs [70], small auxin-up RNAs [71], and DEAD-box RNA helicases [72]. Intron-less genes may arise from one of three major mechanisms: retroposition (the integration of a sequence derived from RNA into the genome), duplication of existing intron-less genes, and horizontal gene transfer [73].

The variations in intron sizes within and between SbWRKY groups may have resulted from duplication, inversion, and/or fusion events [74]. These results were similar to findings in wheat [22], carrot (Daucus carota) [74], and cassava [43]. Overall, the diversification of the exon-intron pattern will provide important clues about the evolution of SbWRKY genes.

Among 20 identified functional motifs in SbWRKY proteins, motifs 1, 2, 3, and 4 corresponded to WRKY domains containing zinc finger domains that are present in most SbWRKY members. Motif 8 represented the nuclear localization signal (NLS), mainly distributed in subgroups IId and IIe. As described previously, members of subgroup IId possess an NLS and a conserved calmodulin-binding domain. Interestingly, three members of subgroup IId (SbWRKY74, SbWRKY62, and SbWRKY64) contained the conserved HARF motif (RTGHARFRR [A/G] P), which was also identified in Arabidopsis and poplar (Populus trichocarpa) WRKY subgroup IId, although the function of this motif is unknown [75, 1]. Some motifs were located nearby the WRKY domain, for instance, motifs 6, 9, and 17. Just as phylogenetic analysis divided SbWRKY genes into groups I-III and subgroups IIa-e, the presence or absence of shared motifs between SbWRKY proteins followed the same general separation into groups, consistent with previous studies [76–79]. Indeed, each of these motifs occurs in most subgroups, and each subgroup can be distinguished based on the motifs they present. Group-specific motifs might be involved in responses to a given biological process [80] and may provide clues about their potential function. The functions of other motifs identified by MEME are yet to be elucidated.

The promoter regions of 94 SbWRKY genes exhibited various conserved cis-acting regulatory elements involved in various functions, such as abiotic and biotic stress responses (MBS, LTR, ARE, TC-rich repeat, and GC-motif) and phytohormone regulation (ABRE, TCA-element, TGA-element, TGACG-motif, CGTCA-element, AuxRR-core, and GARE-motif). The presence of many cis-acting elements mediating responses to environmental stress and phytohormones indicates their involvement in different biological processes. Regulation of WRKY expression may occur via binding of a WRKY TF to W-box or by the binding of another TF to a different cis-element along WRKY promoters [81]. The wheat TaWRKY2 and TaWRKY19 exert their regulation of gene expression by binding to the promoter regions of target genes when overexpressed in Arabidopsis [82]. WRKY TFs may regulate the expression of their encoding genes by binding to their promoter or may regulate other WRKY TFs by cross-regulation [83]. Consistent with this hypothesis, the promoters of 38 SbWRKY genes had one or more W-boxes.

The essential roles of WRKY TFs in plant growth, development, and stress tolerance are supported by WRKY gene expression data from several species. From extensive studies in the model plant Arabidopsis, many AtWRKYs have been functionally characterized. Therefore, identifying the closest Arabidopsis homologue(s) of individual SbWRKYs may provide a hint as to their potential functions. For example, the highly expressed SbWRKY72 gene is most closely related to the Arabidopsis AtWRKY70 and AtWRKY54, which have been reported to modulate osmotic stress tolerance by regulating stomatal aperture [84].

SbWRKY79, SbWRKY80, and SbWRKY42 are putative orthologs of AtWRKY25, AtWRKY26, and AtWRKY33, which regulate heat shock proteins and the heat-induced ethylene-dependent response [85]. AtWRKY53 and AtWRKY70 belong to group III and both play important roles in leaf senescence [86]. SbWRKY72 and SbWRKY75 also belong to group III and are highly expressed in leaves. In addition, AtWRKY70 plays a critical role in osmotic stress signalling and plant defense responses in Arabidopsis [87]. SbWRKY19 and SbWRKY73, members of group I, are expressed in root tissues, although their putative Arabidopsis counterpart AtWRKY34 is involved in regulating gene expression during tapetum formation [88]. Several group II-a WRKY genes have documented roles in biotic stress responses in Arabidopsis [42]. Moreover, AtWRKY8, AtWRKY50 and AtWRKY57 are involved in phytohormones signalling and pathogen responses [89]. The highly expressed sorghum gene SbWRKY45 is the putative ortholog of AtWRKY18, AtWRKY40, and AtWRKY60, which are involved in abscisic acid signalling and abiotic stress [90]. Additionally, SbWRKY45 is the putative ortholog of maize ZmWRKY40, which confers drought resistance when expressed in transgenic Arabidopsis [12]. A putative ortholog of SbWRKY8 is Arabidopsis WRKY57, which can improve drought tolerance through elevated abscisic acid levels [52].

Based on the available sorghum transcriptomic data, the present analysis of SbWRKY genes revealed their different expression patterns at different growth stages and in different tissues. As indicated by the above examples, the known functions of putative Arabidopsis WRKY orthologs, together with expression data, will guide future functional analyses of SbWRKY genes, with a focus on their roles in responses to environmental stress. Highly expressed SbWRKY genes identified during our digital expression analysis of published sorghum microarrays were validated by qRT-PCR, which also provided general clues on SbWRKY responses to drought stress. We tested five genes that belonged to group I (SbWRKY79), IIa (SbWRKY45), IId (SbWRKY74), and III (SbWRKY72 and SbWRKY75). All selected genes were upregulated in response to drought. WRKY genes from the same groups were reported to play important roles in other plants. For example, the expression of sweet potato group I members ItfWRKY66, ItfWRKY69, and ItfWRKY80 was induced in response to drought, cold and salt stresses [53]. Furthermore, AtWRKY25 overexpression improved heat and salt stress tolerance in Arabidopsis [91]. Its putative ortholog SbWRKY79 was induced by drought as shown by RT-qPCR. Several putative SbWRKYs orthologs have been confirmed to be involved in drought stress tolerance in other crops. For example, SbWRKY45 was orthologous to maize ZmWRKY40, which is itself involved in drought stress tolerance [12].

Since we only tested a small fraction of SbWRKY genes, other SbWRKY genes may also be involved in drought stress responses. Therefore, further investigation into SbWRKY expression under other abiotic stress conditions (cold, salinity, and heat) is necessary. Our results provide several promising SbWRKY candidates for these future studies.

Conclusion

This study identified 94 WRKY genes in sorghum, and the following analyses were performed: characterization and classification, gene structure analysis, chromosome mapping, and conserved motif analysis. SbWRKY gene expression profiles indicated that SbWRKY genes may be important in different tissues and at different developmental stages. Several SbWRKY genes displayed tissue-specific expression. Besides, several SbWRKY genes were highly expressed in response to environmental stresses. qRT-PCR analysis revealed several SbWRKY genes induced by drought stress. In Arabidopsis, many AtWRKYs regulate abiotic and biotic stress responses, and the available information about specific WRKY members will facilitate functional validation and characterization of their putative orthologs in sorghum. Overall, the present findings provide a foundation for future functional analyses of SbWRKY genes in response to abiotic and biotic stress in sorghum.

Supporting information

S1 File. Detailed information on sequence similarity of putative paralogous pairs for tandem duplications.

(DOCX)

S2 File. SbWRKY protein sequences.

(DOCX)

S3 File. SbWRKY genomic sequences.

(DOCX)

S4 File. SbWRKY CDS sequences.

(DOCX)

S5 File. AtWRKY protein sequences.

(DOCX)

S6 File. Cis-acting elements in SbWRKY genes.

(XLSX)

S7 File. SbWRKY specific primers used for RT-qPCR.

(DOCX)

Data Availability

All relevant data are within the manuscript and its Supporting Information files.

Funding Statement

This project was supported by The National Key Research and Development Program of China (2016YFD0100605), the Innovation Academy for Seed Design, Chinese Academy of Sciences; special exchange programme of the Chinese academy of sciences (category A).The authors sincerely thank the Center for Agricultural Resources Research, Institute of Genetics and Developmental Biology (CAS) for providing the facilities‛.