Genome-wide identification of soybean WRKY transcription factors in response to salt stress

Members of the large family of WRKY transcription factors are involved in a wide range of developmental and physiological processes, most particularly in the plant response to biotic and abiotic stress. Here, an analysis of the soybean genome sequence allowed the identification of the full complement of 188 soybean WRKY genes. Phylogenetic analysis revealed that soybean WRKY genes were classified into three major groups (I, II, III), with the second group further categorized into five subgroups (IIa–IIe). The soybean WRKYs from each group shared similar gene structures and motif compositions. The location of the GmWRKYs was dispersed over all 20 soybean chromosomes. The whole genome duplication appeared to have contributed significantly to the expansion of the family. Expression analysis by RNA-seq indicated that in soybean root, 66 of the genes responded rapidly and transiently to the imposition of salt stress, all but one being up-regulated. While in aerial part, 49 GmWRKYs responded, all but two being down-regulated. RT-qPCR analysis showed that in the whole soybean plant, 66 GmWRKYs exhibited distinct expression patterns in response to salt stress, of which 12 showed no significant change, 35 were decreased, while 19 were induced. The data present here provide critical clues for further functional studies of WRKY gene in soybean salt tolerance. Electronic supplementary material The online version of this article (doi:10.1186/s40064-016-2647-x) contains supplementary material, which is available to authorized users.


Background
Soybean (Glycine max) is a global cash crop. Apart from its major contribution to human and animal nutrition, the seed provides a feedstock for biodiesel production and represents a significant raw material for a number of pharmaceutical and industrial processes (Phang et al. 2008;Wang et al. 2010). In recent years, the demand for soybean is increasing rapidly, so it attracted more and more attention to improve soybean agronomic traits, such as stress tolerance. Soybean productivity is greatly compromised by soil salt. However, during the long period of evolution, soybean has evolved complex strategies to survive salt stress. These strategies are originated from the changes of various aspects, such as the genome, gene expression, metabolism and physiology (Phang et al. 2008). To present, functionally reported salt tolerance related genes in soybean are mainly categorized into several classes, including ion transporter coding genes (e.g. GmHKT1, GmSALT3, GmNHX1, GmCAX1 and GmCHX1) (Chen et al. 2011;Guan et al. 2014;Li et al. 2006;Luo et al. 2005a;Qi et al. 2014), transcription factors (TFs) (e.g. GmNAC11/-20/-29, GmDREB1, GmMYB76/-92/-174/-177, GmbZIP44/-62/-78, GmWRKY27 and GmERF7) (Hao et al. 2011;Jin et al. 2010;Liao et al. 2008a, b;Wang et al. 2015;Zhai et al. 2013) and others genes (e.g. glutathione S-transferase gene GsGST, late embryogenesis abundant gene GmLEA, calcineurin B-like protein coding gene GmCBL1 and flavone synthase gene GmFNSII) (Ji et al. 2010;Lan et al. 2005;Li et al. 2012;Phang et al. 2008;Wang et al. 2011;Yan et al. 2014;Zhou et al. 2010).
The product of a TF gene binds to a specific cis-regulatory sequence(s) in the promoter of its target gene. The WRKYs are among the largest class of plant TFs, and their promoter target (the W-box) has the sequence (T) (T)TGAC(C/T) (Rushton et al. 2010). WRKY TFs are recognized by the presence of a conserved DNA-binding region composed of about 60 residues (the "WRKY" domain), which harbors the WRKYGQK heptapeptide followed by a C 2 H 2 or C 2 HC zinc finger motif (Rushton et al. 2010). In some cases, the heptapeptide can take the form WRKYGKK or WRKYGEK (Rushton et al. 2010). WRKY TFs have been classified into three main groups: those possessing two heptapeptides are clustered into group I; both group I and II members harbor one C 2 H 2 type zinc finger motif, while the group III members feature a C 2 HC one. The large size of group II has been addressed by its division into five subgroups (IIa, IIb, IIc, IId and IIe), based on peptide sequence (Eulgem et al. 2000;Rushton et al. 2010).
Since the first reports of WRKY TFs (dePater et al. 1996;Ishiguro and Nakamura 1994;Rushton et al. 1995Rushton et al. , 1996, considerable progress has been made in revealing the function of WRKY TFs. They are elucidated to be involved in various developmental and physiological processes (Rushton et al. 2010), such as seed development (Lagace and Matton 2004;Luo et al. 2005b), seed dormancy and germination (Xie et al. 2005(Xie et al. , 2007Zentella et al. 2007), senescence (Miao et al. 2004;Robatzek and Somssich 2002;Ulker et al. 2007) trichome morphogenesis (Johnson et al. 2002), metabolic pathways (Kato et al. 2007;Sun et al. 2003) and plant development (Cai et al. 2014;Devaiah et al. 2007;Guo et al. 2015;Yu et al. 2013Yu et al. , 2016. The particularly prominent roles of WRKY in plant appear to be the modulation of response to biotic and abiotic stresses Eulgem and Somssich 2007;Pandey and Somssich 2009). In the root of the model plant Arabidopsis, 18 WRKY genes have been shown to be induced by exposure to salt stress (Jiang and Deyholos 2006); In rice, ten WRKY genes (of 13 analyzed) respond differentially to a range of abiotic stress treatment (Qiu et al. 2004), while in Brachypodium distachyon, over 60 % of a set of 86 WRKY genes assayed were up-regulated by heat and cold stress and over 50 % were down-regulated by salt, drought and/or oxidation stress (Wen et al. 2014). Among the soybean WRKY TFs, 25 out of 64 have been shown to be differentially expressed in response to at least one abiotic stress treatment (Zhou et al. 2008).
Base on the availability of the complete soybean genome sequence and several databases (PlantTFDB, SoyDB, SoyTFKB, NCBI and Phytozome), a previous study reported a genome-wide characterization of the WRKY family in soybean and a functional analysis of some genes involved in response to Phakopsora pachyrhizi (Bencke-Malato et al. 2014). However, in Phytozome (release v10), the new assembly (v2.0) replaces the Glyma1 assembly. The new database corrects several issues in pseudomolecule reconstruction in the Glyma1 assembly. According to the new database and a recent RNA-Seq result (Belamkar et al. 2014), Song et al. identified 176 GmWRKYs and analyzed their expression files in different tissues and in response to drought and salt stress (Song et al. 2016). However, beside Phytozome, there are many other soybean genome sequence databases. Furthermore, several other transcriptome experiment data sets of soybean under abiotic stress are provided by NCBI website. Here, integrating more databases and RNA-Seq results (Belamkar et al. 2014;Wei et al. 2015), we made a new genome-wide identification of soybean WRKYs and compared their response to salt stress in different tissues. In addition, we analyzed expression profiles of 66 GmWRKYs by quantitative RT-PCR (RT-qPCR). Our findings provide new clues for further investigation of WRKY gene in soybean salt tolerance.

Multiple sequence alignment and phylogenetic analysis
A multiple alignment of the WRKY sequences was performed using the ClustalW program implemented in MEGA v6.06 software package (http://www.megasoftware.net/) (Tamura et al. 2013). The sequences were also subjected to a phylogenetic analysis using the neighborjoining method; the resulting tree was based on 1000 bootstrap replicates, the p-distance model and pairwise deletion.

Gene structure and conserved motifs analysis
The exon-intron structure of each gene was derived by comparing its coding sequence with the corresponding genomic DNA sequence, using the GSDS program (http://gsds.cbi.pku.edu.cn/) (Hu et al. 2015). The online program MEME v4.10.1 (http://meme-suite.org/tools/ meme) was used to identify the conserved motifs present; the relevant parameters were: number of repetitions = any; maximum number of motifs = 16; optimum width of each motif = 6-70 residues.

Genomic location and gene duplication
Each WRKY gene was positioned in the genome by reference to the full genome sequence. The gene duplications in GmWRKY genes were identified based on the investigations described in previous study (Schmutz et al. 2010), and Circos software was used to provide a graphical representation of the position of homeologous chromosome segments (Krzywinski et al. 2009).

RNA-seq analysis
A transcriptomic analysis was based on archival RNAseq data collected from a set of salt stress experiments, mounted on the NCBI GEO database (Belamkar et al. 2014). Experiments GSM1377923, -24 and -25 represented three independent replicates of plants sampled before any exposure to salt; GSM1377935, -36 and -37 related to plants sampled after a 1 h exposure to the stress; GSM1377938, -39 and -40 after a 6 h exposure; and GSM1377941, -42 and -43 after a 12 h exposure. The other transcriptomic analysis was also based on archival RNA-seq data derived from the NCBI GEO database , experiment GSE57960 was related to plants sampled after a 12 h exposure to salt stress, the aerial part of plants was used for sequencing. The reads per kilobase of exon model per million mapped reads (RPKM) algorithm was used for normalization and mean normalized values were used for the analysis. The transcription response was given in the form of fold changes relative to the 0 h control. Cluster v3.0 software (University of Tokyo, Human Genome Center) was used to perform hierarchical clustering, which was visualized using Java TreeView software (Saldanha 2004). The relevant parameters were: similarity measurement: correlation (uncentered); linkage method: average linkage method.

Plant materials
Seed of cv. Williams 82 was germinated on a sheet of moist filter paper, and the seedlings were grown under a regime of 28/20 °C, 14 h photoperiod, light intensity 800 μmol m −2 s −1 and relative humidity 55 %. Two weeks old seedlings were exposed to 200 mM NaCl for either 0, 2, 6 or 24 h, after which the whole seedling was harvested, snap-frozen in liquid nitrogen and stored at −80 °C.

RT-qPCR analysis
Total RNA was extracted from the frozen plant material using the TRIzol reagent (Invitrogen, USA) following to the manufacturer's instructions. The resulting RNA was treated with RNase-free DNaseI (Promega, USA) to remove genomic DNA contamination, and the cDNA First strand was synthesized with 3 μg total RNA by TransScript One-step gDNA Removal and cDNA synthesis SuperMix (TransGen, China) following the manufacturer's protocol. The subsequent RT-qPCRs and data analysis were performed using a Bio-Rad Real-Time PCR detection system (Bio-Rad) based on the SYBR Green I master mix, as reported previously (Bustin et al. 2009;Seo et al. 2009). According to a previous study, GmELF1b was most stably expressed under salt stress, so it was used as a reference gene (Le et al. 2012). All reactions were carried out in triplicate, using samples harvested from independent plants. The relevant primer sequences are given in Additional file 1: Table S1.

Identification of WRKY genes in soybean
The WRKYs represent one of the largest families of plant TFs. The acquisition of full genome sequences has simplified the enumeration of WRKY copy number, so that it is now clear that there are 81 WRKY copies in tomato (Huang et al. 2012), 55 in cucumber (Ling et al. 2011), 104 in poplar (He et al. 2012, 59 in grapevine ), 116 in cotton (Dou et al. 2014) and 119 in maize (Wei et al. 2012). In a previous study, 182 putative WRKY gene models were identified (Bencke-Malato et al. 2014). In recent, phytozome updated the soybean assembly; the new assembly (v2.0) replaced the Glyma1 assembly. Therefore, we performed a comprehensive analysis of soybean WRKY sequences obtained from Phytozome, PlantTFDB and NCBI, and finally identified a total of 185 non-redundant putative WRKY genes. Compared with previous 182 WRKY genes, six genes GmWRKY49 (Glyma.05G203900), GmWRKY53 (Glyma.06G061900), GmWRKY72 (Glyma. 07G161100), GmWRKY108 (Glyma.10G171000), GmWR KY130 (Glyma.14G085500) and GmWRKY131 (Glyma. 14G100100) are novel ones, while three previous genes GmWRKY17 (Glyma06g06530), GmWRKY38 (Glyma 18g48460) and GmWRKY132 (Glyma14g11440) are considered obsolete according to the current version of the annotated genome. However, the three obsolete genes have been retained for further study. Thus, a total of 188 annotations of GmWRKYs were presented in this study (Table 1). All the 188 retrieved sequences were proved to contain WRKY domains using SMART analysis.
In previous genome-wide studies, the commonly accepted nomenclature for WRKY members was based on their location order on chromosomes (Dou et al. 2014;Ling et al. 2011). Identically, in the present study, GmWRKYs was designated from GmWRKY1 to GmWRKY188 based on their exact physical position from the top to the bottom on the soybean chromosomes 1-20 (Table 1). For genes producing more than one transcript, only the primary sequence was named. This nomenclature system was different from previous study   In silico mapping revealed that the WRKY genes were distributed over all 20 soybean chromosomes.

Classification of WRKY genes in soybean
As described previously, WRKY family is typically categorized into three main groups defined by the number of WRKY domains present and the configuration of their zinc finger (Rushton et al. 2010). The 188 soybean WRKY genes were also categorized into the three main groups (Additional file 3: Fig. S1). The group I members numbered 32 (GmWRKY65 and -72 harbored a single N-terminal WRKY domain); there were 130 sequences assigned to group II, sub-divided into subgroup IIa (14 members), IIb (33 members), IIc (42 members), IId (21 members) and IIe (20 members); the remaining 26 sequences belonged to group III (Table 1; Fig. 1). In Arabidopsis and poplar, group I houses the largest number of WRKYs, while in rice, group III is the largest (He et al. 2012). However, the largest group in soybean is group II, implying that this group had experienced more gene duplications during the evolutionary course.
Although most of the sequences harbored the well conserved WRKYGQK motif, variants were present in 24 of the sequences: WRKYGKK in 11, WRKYGEK in three, WKKYGQK in two, and WRKYGKR, WRKYEDK, WIKYGQK and WHQYGLK each in one. Strikingly, A WRKYGQK-like stretch was lacking in GmWRKY20, -49, -105 and -130, while the group I members WRKY65 and -72 had both lost their N terminal WRKYGQK-like stretch (Table 1; Additional file 3: Fig. S1). The largest number of variants belonged to group IIc, 11 out of 24. The WRKYGQK sequence was highly conserved in subgroups IIb, IId and IIe, as well as in the C terminal WRKY domain of group I members (Table 1; Additional file 3: Fig. S1). This is consistent with the implication that this group experienced more gene duplications. There was also some variation in the zinc finger motif (including its complete absence) in 11 of the sequences (WRKY6, -42, -52, -65, -72, -79, -94, -106, -112, -139 and -165) (Additional file 3: Fig. S1).

Gene duplication of soybean WRKY genes
Duplication events contribute not only to functional redundancy, but also generate functional novelty (Moore and Purugganan 2005). The modern soybean genome has undergone two whole genome duplication (WGD) events, the first, associated with the evolution of the legume clade occurred ~59 million years ago (Lavin et al. 2005), while the second, which was responsible for the creation of the Glycine genus, occurred ~13 million years ago (Schmutz et al. 2010). To investigate whether the expansion of GmWRKY genes had primarily happened during both WGD events, we mapped the GmWRKYs to the duplicated blocks (Fig. 2). Consistent with previous study (Schmutz et al. 2010), the blocks between chromosomes involved more than just two chromosomes. Of 188 GmWRKYs, 180 (95.7 %) genes were located in the blocks (the exceptions were GmWRKY46, (Fig. 2), indicating that WGD was the primary reason for the expansion of GmWRKYs (Fig. 2).
Besides WGD, tandem duplication event is the other approach for gene expansion. Precise mapping analysis showed the presence of 14 adjacent genes possibly due to tandem duplication ( Fig. 2; Additional file 4: Fig. S2a). These 14 WRKY genes were localized in 6 distinct tandem duplicate gene clusters, with four clusters containing two tandem genes (GmWRKY120/123, GmWRKY121/122, GmWRKY131/132 and GmWRKY151/152) and two clusters possessing three ones (GmWRKY108/109/110 and GmWRKY136/137/138). All the 14 tandem duplicated WRKY genes were mapped onto the duplicated blocks, implying that local duplications occurred earlier than the WGD.

Gene structure and conserved motifs of GmWRKYs
Gene structural diversity may reflect the evolution of multigene families (Hu et al. 2010). In order to look into the structural diversity of GmWRKY genes, we first   Fig. 1 The distribution of WRKY genes on the 20 chromosomes of soybean constructed a phylogenetic tree based on the full-length GmWRKY polypeptide sequences, and they were also categorized into seven subfamiles as above (Additional file 4: Fig. S2a). From the tree, we could find that each clade consists of two to four genes, which well matched the two WGD events and confirmed that the expansion of GmWRKY happened during both WGD events. We then analyzed the exon-intron organization in the coding sequences of each soybean WRKY genes HD-ZIP genes (Additional file 4: Fig. S2b    group IIa harbored three (7/14) or four (5/14) (Additional file 4: Fig. S2b). In contrast, the gene structure appeared to be more variable in groups IIb, the number of introns in this group varied from one to six (Additional file 4: Fig.  S2b). In Populus, although there were only eight members in group IIb, the numbers of introns varied from three to six (He et al. 2012). These results indicated that WRKY genes in different species were relatively conserved during the evolution. Furthermore, genes shared similar exonintron organization within the same subgroup, while they were strikingly distinct in the gene structure among different groups, suggesting that they were not only conserved, but diverged during the evolution.
To better understand the conservation and diversification of WRKY genes in soybean, putative motifs of GmWRKYs were predicted using MEME software and finally 16 distinct motifs were identified (Additional file 4: Fig. S2c). As expected, most of the closely related members in the phylogenetic tree shared common motif compositions, suggesting that the WRKY proteins within the same subfamily might be of similar functions. However, like putative motifs predicted in ZmWRKYs (Gao et al. 2014), the biological significance of most of the putative motifs in GmWRKYs was also unclear because they did not have homologs when searching against Pfam (http://pfam.sanger.ac.uk/search) and SMART (Simple Modular Architecture Research Tool) databases. The same phenomenon also existed in Populus NAC and HD-ZIP proteins (Hu et al. 2010(Hu et al. , 2012. According to previous study, WRKY proteins harbor typical WRKY domains and zinc-finger motifs (Eulgem et al. 2000;Rushton et al. 2010). Here, motif 1, 2 and 9 comprised the WRKY domain, motif 3 and 10 were the partial zinc-finger motifs followed motif 2 and 9 (Additional file 5: Table S3). The product size of the group I and subgroup IIb genes was larger than that of members of the other groups (Table 1), consistent with their harboring a greater number of motifs (Additional file 4: Fig. S2c). In contrast, although subgroup IIc possessed the largest number of members, they harbored the least number of motifs (one to three). Even though the C-terminal regions of GmWRKYs were highly divergent, we could also identify several conserved motifs which were present in GmWRKYs from specific subgroups, for example, motifs 3, 5 and 6 in group I, motif 14 in subgroup IIb, and motif 13 in subgroups IIa and IIb (Additional file 4: Fig. S2c). Whether these motifs play functional roles remained to be further elucidated.

Expression profiles of GmWRKYs in response to salt stress
The WRKY gene family is heavily implicated in the plant response to abiotic stress , as indicated by a number of microarray-based transcriptomic data sets. Several studies have reported the influence of abiotic stress on WRKY genes based on these data sets (Dou et al. 2014;Satapathy et al. 2014;Wei et al. 2012). In soybean (cv. Kefeng No. 1), the response of a set of 64 WRKY genes following the plant's exposure to salt stress has been described (Zhou et al. 2008), but this number represents only about one-third of the total WRKY genes. In order to give insight to the function of GmWRKYs in plant response to salt tolerance, we analyzed the soybean (cv. Williams 82) gene expression profiles under salt stress (Belamkar et al. 2014). Finally, 66 of the 188 GmWRKY genes were transcriptionally regulated under salt stress ( Fig. 3a; Additional file 6: Table S4). 65 genes were up-regulated, with only WRKY71 being down-regulated (Fig. 3a). The response of WRKY was typically quite rapid (Eulgem et al. 2000), most notably in the case of GmWRKY20,, which responded by an at least five fold rise in transcript abundance after a 1 h exposure to the stress. In some cases, the response was transient: some examples were the genes WRKY44, -51, -54, -78, -81, -85, -102 and -107, for which transcript abundance peaked after a 6 h exposure and then fell away ( Fig. 3a; Additional file 6: Table S4). The most responsive gene (WRKY134) belonged to subgroup IIa, which was increased to ~226 fold after a 6 h exposure ( Fig. 3a; Additional file 6: Table S4). By contrast, the expression of GmWRKY71, a member of group IId was downregulated in response to salt stress (Fig. 3a).
In soybean, a notable number of responsive genes belonged to subgroup IIb (18 of 33), although their level of induction by salt treatment was only modest (Fig. 3b). In addition, most of subgroup IIc (14/42) and group III (14/26) members were significantly induced and these genes tended to be dramatically up-regulated after a 6 h exposure (Fig. 3c, d). In Arabidopsis root, the 18 salt induced members belong to group I (4/18), II (11/18) and III (3/18), respectively (Jiang and Deyholos 2006). These data indicated that either in soybean or Arabidopsis, group II members made major contribution in salt response, suggesting that WRKY functions in response to salt stress in different organisms appeared to be conserved during evolution.
The material used in the above RNA-seq was soybean root which was not able to represent other parts. We then analyzed the other RNA-seq data which were derived from the aerial part of soybean plants (cv. SuiNong 28) . A total of 49 GmWRKY genes were transcriptionally regulated under salt stress ( Fig. 3e; Additional file 7: Table S5). 47 genes were downregulated, with only two (WRKY155 and WRKY183) being up-regulated (Fig. 3e). These results were quite different with the above RNA-seq analysis, indicating that GmWRKY genes showed distinct response profiles in different tissues.

Investigation of GmWRKY gene expressions by RT-qPCR
The transcriptional profiles we analyzed above could provide clues for revealing the function of GmWRKYs in plant response to salt tolerance. However, the material used in RNA-seq was soybean root or aerial part which was not able to represent the entire plant. To shed light on the expression profiles of GmWRKY genes, 2 weeks old soybean seedlings (cv. Williams 82) were exposed to  IId  IIb  IIe  IIb  III  I  IIb  IIb  IIa  IIc  IIe  I  IIe  III  IIb  I  IIb  I  I  III  IIc  III  IIc  III  IIa  IIa  IIa  IIb  IIb  IIc  IIb  IIc  IIc  IIc  III  III  IIc  IIb  III  III  IIb  IIe  IIc  I  III  IIc  IIc  IIb  III  IIb  III  III  IIa  IIb  IIc  IIa  IIb  IIe  IIa  IIb  III  IIc  I  IIc  IIb 6h  24h  0h  2h  6h  24h   0h  2h  6h  24h  0 h  2 h  6 h  24h  0h  2h  6h  24h  0h  2h  6h  24h  0h  2h  6h  24h  0h  2h  6h  24h  0h  2h  6h  24h   0h  2h  6h  24h  0h  2h  6h  24h   0h  2h  6h  24h  0h  2h  6h  24h   0h  2h  6h  24h  0 h  2 h  6 h  2 4h  0h  2h  6h  24h Relative expression ** ** ** Fig. 4 RT-qPCR-based transcription profiling of 66 WRKY genes. a Unaffected WRKY genes. b Salt-inhibited WRKY genes. c Salt-inducible WRKY genes. Error bars represent SD (n = 3). (t test, *P < 0.05; **P < 0.01) 200 mM NaCl for 0, 2, 6 or 24 h, respectively, and then the total RNA of the whole plant was isolated used for RT-qPCR analysis. 66 GmWRKY genes were tested and exhibited distinct expression patterns in response to salt stress, of which 12 showed no significant change (Fig. 4a), 35 were decreased (Fig. 4b), while 19 were induced (Fig. 4c). GmWRKY38, -120 and -185 were substantially decreased, especially GmWRKY120. In contrast, GmWRKY20, -89, -114 and -142 were remarkably induced. These expression patterns were different with the above RNA-seq analysis, indicating that GmWRKY genes showed distinct response profiles in the whole plants compared to different tissues. The response of WRKY to abiotic stresses was generally rapid and transient (Eulgem et al. 2000). Likely, most of the GmWRKY genes responded rapidly, their expressions were decreased (31/35) or induced (16/19) after only a 2 h exposure (Fig. 4b, c). In addition, the response of GmWRKYs was transient, such as GmWRKY36,c).
In tomato and cucumber, most WRKYs are up-regulated by salt stress (Huang et al. 2012;Ling et al. 2011). In contrast, the majority of Brachypodium distachyon WRKYs are down-regulated by the stress (Wen et al. 2014). In soybean, most WRKYs were up-regulated in root, while down-regulated in the aerial part by the stress (Fig. 4). Species differences presumably reflected a major degree of functional divergence in the WRKY gene family.

Conclusion
The present study has taken a genome-wide view of the soybean WRKY gene family, and characterized their transcriptional response to salt stress. An analysis of their phylogeny, chromosomal location, gene structure and content of conserved motifs has allowed the genes to be classified into the standard set of groups. The expansion in copy number of the GmWRKYs has occurred largely as a result of the two well recognized ancient whole genome duplication events. To date, only three GmWRKY genes have been functionally investigated (Jiang and Deyholos 2009), leaving unknown the function of the remaining more than 180. The responsiveness to salt stress of about one-third of the GmWRKY complement confirms the potential of gene manipulation within this gene family as means of improving the salt tolerance of important crop species.