Introduction

Environmental vagaries exert detrimental effects on plants including almost all phonological stages. Evolution of an intricate signaling system to sense and react to external stimuli is crucial for the survival and perpetuation of plants to the next generation (Takahashi and Shinozaki 2019). Signaling events encompass reception, interaction and reaction following a change in the environment, which remains under the control of regulatory molecules at the cellular and sub-cellular levels (Joshi et al. 2017). Transcription factors (TFs) are important constituents of plant signaling pathways that define plant responses against biotic and abiotic stimuli in addition to their response to internal signals which coordinate different interacting partners during developmental processes (Joshi et al. 2016). WRKY TFs are among the largest families of transcriptional regulators in plants (Bakshi and Oelmüller 2014). In recent years, several WRKY superfamily members have been reported in different plant species including Glycine max (197), Raphanus sativus (126), Oryza sativa (109), Populus (104), Pinus monticola (80), Arabidopsis thaliana (75), Sorghum bicolor (68), Citrullus lanatus (63), Carica papaya (52), Hordeum vulgare (45), Physcomitrella patens (38), Selaginella involucrate (35) and Ipomoea trifida (83) (Mangelsen et al. 2008; Liu and Ekramoddoullah 2009; Song et al. 2010; Pan and Jiang 2014; Fan et al. 2015; Karanja et al. 2017; Yang et al. 2018; Li et al. 2019). Binding of WRKY TFs with the (T)TGAC(C/T), i.e., W-box cis-element in the promoter of target genes induces gene expression to achieve cellular homeostasis (Bakshi and Oelmüller 2014; Machens et al. 2014). WRKY DNA-binding domain (DBD) is characterized by an invariant heptad WRKYGQK amino acid motif at their N-terminus and a zinc-binding motif at their C-terminus (Jiang et al. 2017). Considerable diversity has been evident in the structure and function of WRKY TFs from a growing body of research. Variations have been found in the number of WRKY genes and proteins per genome, number of introns, nucleotide sequence, among and within different evolutionary levels of the plant kingdom. WRKYs have gained attention owing to their involvement in diverse plant processes including growth, development, biotic and abiotic stress responses and plant innate immunity including microbe- or pathogen-associated molecular pattern-triggered immunity (MTI or PTI) and effector-triggered immunity (ETI) (Rushton et al. 2010; Phukan et al. 2016; Jiang et al. 2017). WRKYs can act as activators or repressors, in various homo- and heterodimer combinations, and form a TF net that contributes to various cytoplasmic and nuclear processes including signaling events from organelles or the cytoplasm to the nucleus (Bakshi and Oelmüller 2014). In this review, we summarize recent advances in our understanding of the WRKY TF family, with an emphasis on the plant stress response. We then elaborate on the WRKY-mediated regulatory networks that impart innate immunity to plants.

Structural characteristics and classification of functional domains

Despite having a highly conserved W-box, the varying binding abilities of WRKY TFs could be credited to variable numbers of DBDs and zinc-finger-like motifs. Based on these features, WRKY proteins are classified into four groups: I (two WRKY DBDs), II (single DBD with different C2–H2 zinc finger), III (single DBD with C2–HC zinc finger) and IV (incomplete WRKY domain without zinc finger) (Xie et al. 2005). Unlike group I, II and IV, group II is not monophyletic and is further divided into IIa, IIb, IIc, IId, and IIe based on additional conserved structural motifs besides the WRKY domain (Chen et al. 2012; Phukan et al. 2016). In lower plants, such as algae (Dunaliella bardawil), four WRKY TFs have been identified and all of them contained WRKYGEK sequence of amino acids at N-terminal, a difference due to Zn-finger at C-terminal. For instance, DbWRKY1 contained CX4CX23HXH, DbWRKY2 contained CX4CX21HXH while DbWRKY3 and DbWRKY4 contained CX4CX22HXH. Investigations on WRKY structures suggest that pre-WRKY structures (Pro-WRKY) must have a single domain as an origin. This might have doubled to give group I members of WRKY TFs, while a loss of N-terminal WRKY domain gave rise to group IIc members. Group IIc might have branched prophylactically to give rise to other subgroups of group II, and group III is the youngest and less diverse compared to other two groups (Song and Gao 2014; Wu et al. 2017a). The proposition that group II and III WRKY members should have evolved from group I gathers strong support from sequence similarity that the WRKY domains of group II and III members share with the C-terminal WRKY domain of group I WRKY TFs (Chen et al. 2019).

Though WRKY domain is highly conserved, substitutions occur in amino acid sequence of WRKYGQK; for example, WRKYGKK in maize, banana, populus, mulberry and soybean (Eulgem et al. 2000; Zhou et al. 2008; Zhang et al. 2017), WRKYGRK in populus and banana, FWRKYGQK in populus, WRKYGEK in rice and banana. In banana, other variants are also reported, such as WRKYGNK and WRKYGHK. Several monocot crops have also been studied with regard to WRKY TFs. Distribution of WRKY genes across a genome is not uniform and varies even within chromosomes (Goel et al. 2016; Xu et al. 2016; Zhang et al. 2017; Chanwala et al. 2020). In rice, the WRKY family members have 19 variants of the WRKY domain where WRKYGEK and WRKYGKK are the two common variants shared by seven and five domains (Zhang and Wang 2005). The other variants include WRICGQK, WRMCGQK, WKKYGQK, WIKYGQK, WKRYGQK, WSKYEQK, and WRKYSEK (Zhang and Wang 2005). In a few WRKY proteins, the WRKY motifs are also replaced by WIKY, WRMC, WRIC, WKKY, WVKY, WKRY, WSKY patterns (Jiang et al. 2017). Further, in the primary WRKYGQK motif of DBD, the WRKY residues are replaced by WRRY, WSKY, WKRY, WVKY or WKKY (Jiang et al. 2017). In addition to the above, variants of metal-chelating zinc finger motifs, such as C–H4-5–X22-23–H–X–H and C–X5-8–C–X25-28–H–X1-2–C, are characteristics of WRKY proteins (Zhang and Feng 2014). Besides, HARF motif (RTGHARFRR (A/G) P) of unknown function was also reported in sub-group IId in Arabidopsis thaliana, Salvia miltiorrhiza and Phaseolus vulgaris (Li et al. 2015a; Wang et al. 2016). A Calmodulin (CaM)-binding domain (DxxVxKFKxVISLLxxxR) was also reported in Arabidopsis Group IId WRKYs, such as AtWRKY7 (Park et al. 2005).

WRKY TFs bind specifically through both the terminals of WRKY domains to W box-type cis-elements in target gene promoters and activate or inhibit their transcription (Kim et al. 2008; Yamasaki et al. 2013; Li et al. 2020a, b, c). W-box consists of a degenerated/core TGAC motif, a conserved GAC core which interacts with the WRKY proteins, and its flanking thymine and pyrimidine (C/T) residues are recognized by specific WRKY factors (Jiang et al. 2017; Chen et al. 2019). Further, functional W-box elements are mostly present in WRKY TFs that bind with their own promoters or other WRKY TFs for auto- or cross-regulation (Liu et al. 2021). WRKY-DBDs recognize the regulatory motif sequence and bind through an α-helix or a β-sheet with the major groove of hydrophobic W-box element allowing base-specific interaction between the protein and the DNA (Deeba et al. 2017). However, NtWRKY12 having WRKYGKK sequence binds with WK-box (required for the PR-1a induction by salicylic acid (SA) and bacterial elicitors), not with W-box (van Verk et al. 2008). Mutation from GKK to GQK or GEK is reported to impair the DNA-binding activity of NtWRKY12 (van Verk et al. 2008). Similarly, SUSIBA2 (HvWRKY46) binds to W-box and SURE (Sun et al. 2003), OsWRKY13 interacts with W-box and PRE4 (Xiao et al. 2013), AtWRKY70 binds to W-box and WT-box (Machens et al. 2014) and AtWRKY50 interacts with TGA2 and TGA5 TFs, and simultaneously bound to PR1 promoter (Hussain et al. 2018). Further, by swapping β4–β5 strands, OsWRKY45-DBD forms a homodimer which provides flexibility around strands (Cheng et al. 2019; Xu et al. 2018). The inhibitor and mutational analyses of WRKY domain structures demonstrate that Tyr, Trp and two Lys residues in the WRKYGQK sequence (Ciolkowski et al. 2008) and zinc finger structures (Cheng et al. 2019) are indispensable for DNA binding. Further, differences between amino acid heptad affect the structure of WRKY domain, thereby enabling the concerted and selective regulation of WRKY TF target genes (Chen et al. 2019).

The distribution of different groups of WRKY TFs is also reported to differ among different plant species, exemplified by the abundance of group I WRKY proteins in A. thaliana and O. sativa as compared to strawberry. Similarly, poplar genome consists of 50% of WRKY TFs that belong to group I. Group I members in mulberry family constitute only group Ia sub-group. Also, a distinct motif of GGDFDDNEPEAKRWKGE was found at C-terminal of all group Ia members and one member of the group IIb family (Baranwal et al. 2016). Tea also shows substitution, in residual protein sequence, of glutamine by lysine (Wu et al. 2016). Research on woody dicots like Populus has offered evidence for formation of WRKY III after duplication events that occurred during monocot-dicot divergence (Wang et al. 2015). The structural complexity of WRKY genes is also manifested in point of spliced domains with V-type or R-type introns, with a low diversity between monocots and dicots than among lower plant groups (Liang et al. 2017). For example, both in strawberry and wheat, number and type of introns varied from zero to twenty and V-type introns were specific to IIa and IIb type groups (Zhou et al. 2016; Ning et al. 2017). In dicots like pepper, group IIa family contains members having leucine zipper structure, that serves as elicitors that mediate binding of WRKY TFs to W-box (Cormack et al. 2002). In tomato, a divergence in group IIe and group III members was observed at WRKY domain sequence, and also in group III members at Zn-finger motifs (Huang et al. 2012). In addition to this, WRKY TFs are reported to have undergone positive or neutral selection as well as gene duplication and loss (Tang et al. 2013; Zhou et al. 2016). Presence of a large number of duplicated blocks suggests occurrence of duplications during evolution and genome expansion within the WRKY family members (Xu et al. 2016; Zhang et al. 2017). In gymnosperms like pine, introns show synteny with the angiosperms as well as, among different WRKY families, group IId + IIe shows expansion with the highest number of members and WRKY3 family with the least representation (Liu and Ekramoddoullah 2009). This observation contrasts with angiosperms where WRKY members of all the families are present, though the number differs with different genomes. Therefore, binding affinity of various WRKY domains to W-box and others seems to be ambiguous, requiring deeper investigations on the domains outside the DBD.

WRKY TF and plant stress response

Role in biotic stresses

Plants have evolved mechanisms to respond to a range of biotic stresses, such as pathogen attacks, that they encounter in their life cycle (Sun et al. 2015; Jha et al. 2020). The role of WRKY TFs as key regulators in plant immune response to a variety of biotic stresses has been extensively investigated (Singh et al. 2017; Cui et al. 2019). WRKYs regulate expression of defense-related genes through binding to a consensus cis-element referred to the “W-box” (TTGACT/C) in the promoter regions of these genes. As reviewed in Peng et al. (2018), PAMP (pathogen-associated molecular patterns)-triggered immunity (PTI) and effector-triggered immunity (ETI) constitute the two lines of plant immune response, with former involving recognition of molecular patterns of the pathogen (PAMPs) with pattern-recognition receptors (PRR) in host cells; whereas toxins/effectors are detected by plant resistance (R) proteins in ETI that often involves a hypersensitive response (HR). Receptor-like kinases (RLKs), one of the largest gene families in plants, form the major class of pathogen recognition receptors. WRKYs were reported to play pivotal roles in repressing or activating both kinds of plant defense responses through direct or indirect interaction with PAMPs/effector proteins or their regulation by mitogen-activated protein kinases (Phukan et al. 2016). Peng et al. (2008) demonstrated a role for a WRKY transcription factor OsWRKY62 in both PTI (basal) and ETI (mediated by Xa21) in rice. Transgenic plants overexpressing OsWRKY62.1 (one of the two splice variants) showed impairments in both PTI and ETI, with suppression of defense-related genes. This study established OsWRKY62 as a negative regulator of innate defense response in rice against bacterial blight. By contrast, OsWRKY67 acts as a positive regulator of PTI and ETI against two rice pathogens, Magnaporthe oryzae and Xanthomonas oryzae pv. oryzae (Xoo) (Vo et al. 2018).

In A. thaliana, Kim et al. (2008) described the action of WRKY38 and WRKY62 in reducing basal defense response to bacterial pathogen Pseudomonas syringae through transcriptional activation of some unknown regulatory players. Also, an antagonistic action of Histone Deacetylase 19 (HDA19) against WRKY38 and WRKY62 confirmed HAD19 as a positive regulator of defense response. An enhanced level of disease resistance was found in single (wrky38 and wrky62) as well double mutant (Kim et al. 2008). In response to pathogen infection (P. syringae) in Arabidopsis, the association of MAP kinase 4 (MPK4) and MKS1 with WRKY33 in the nucleus is disrupted following the activation (MPK4) and phosphorylation (MKS1) events. The released WRKY33 then represses phytoalexin deficient3 (PAD3) gene, thus impairing the synthesis of antimicrobial compound camalexin (Qiu et al. 2008). Earlier, Zheng et al. (2006) showed greater susceptibility of Arabidopsis overexpressing WRKY33 against necrotrophic fungal pathogens (Botrytis cinerea and Alternaria brassicicola). Previously in soybean, authors identified 174 WRKY genes from the genome sequence and overexpression of the three WRKYs GmWRKY 136, 53 and 86 in the soybean cyst nematode (SCN)-susceptible cultivar Williams 82 enhanced SCN resistance level up to 55% (Yang et al. 2017). No significant increase was observed in these transgenic lines following SA application. In wheat, RNA-Seq associated differential expression of two WRKY genes (TaWRKY49 and TaWRKY62) with high-temperature seedling-plant resistance to stripe rust (Puccinia striiformis f. sp. tritici). Gene silencing experiments further validated the RNA-Seq results, and TaWRKY49-silencing enhanced resistance while silencing of TaWRKY62 reduced resistance. Authors suggested negative and positive regulatory roles of TaWRKY49 and TaWRKY62, respectively, in High-Temperature Seedling-plant resistance to Pst (HTSP) through modulating SA, jasmonic acid (JA), ethylene and reactive oxygen species (ROS) pathways (Wang et al. 2016). Another example of negative regulation of plant immunity by WRKY includes CaWRKY40b, which regulates a set of immunity associated genes in pepper under Ralstonia solanacearum infection. Virus-induced gene silencing (VIGS) of CaWRKY40b and overexpression of CaWRKY40b-SRDX caused up- and down-regulations of positive and negative regulators, respectively; whereas CaWRKY40b-overexpression exerted an opposite role (Khan et al. 2018). Further, enhanced transcript levels of CaWRKY40 were reported to induce SA, JA and ethylene mediated pathways during R. solanacearum infection. Overexpression of CaWRKY40 regulates hypersensitive response (HR)-associated and pathogenesis-related genes, thus providing resistance to R. solanacearum (Dang et al. 2013). CaWRKY40b thus acts as a negative regulator by directly modulating immunity associated gene CaWRKY40. Further, CaWRKY6 positively regulates R. solanacearum resistance by activating CaWRKY40 (Cai et al. 2015). Isolation of SpWRKY1 from wild tomato (Solanum pimpinellifolium) following cloning and RT-PCR approaches, and its subsequent transformation into cultivated tomato (Solanum lycopersicum) caused an increase in resistance to Phytophthora infestans via mediating the regulation of abscisic acid (ABA) biosynthesis genes (Li et al. 2015b). In another attempt, tobacco plants transformed with SpWRKY1 showed lower contents of malondialdehyde, relative electrolyte leakage and higher antioxidant enzymes peroxidase (POD) and superoxide dismutase (SOD) and phenylalanine ammonia-lyase (PAL) activities, indicating plant’s resistance to Phytophthora nicotianae (Li et al. 2015c). Concomitantly, the expression of genes related to JA/SA and deference response also altered as a result of SpWRKY1 overexpression.

Role in abiotic stress

A growing body of literature supports the involvement of WRKY transcription factors in the regulation of abiotic stress response in plants. Meeting the challenge of feeding nine billion by 2050 necessitates a sustainable increase in crop yields in an environment that is increasingly challenged by drought and salinity (Tester and Langridge 2010; Singh et al. 2015; Wani et al. 2018; Jha et al. 2019). A variety of biotic and abiotic factors have been reported to induce WRKY TFs, such as OsWRKY45, and Arabidopsis transformed with OsWRKY45 had enhanced tolerance to drought via regulating stomatal closure and stress-related genes (Qiu and Yu 2009). An acquired drought tolerance (adt) mutant having improved drought tolerance was isolated in Arabidopsis after surveying 43 WRKY-associated T-DNA insertion lines. The insertion in adt caused an elevation in WRKY57 expression level. The drought tolerance of adt could be credited to enhanced ABA levels. Further, the binding of WRKY57 with stress-responsive genes (RD29A, NCED3) via the core W-box sequence was confirmed with chromatin immunoprecipitation (ChIP), PCR and quantitative real-time (RT–PCR) analysis (Jiang et al. 2012). In wheat, overexpression of TaWRKY146 having homology with the AtWRKY46 showed hypersensitivity towards salt and drought stress in Arabidopsis (Ma et al. 2017). Significant up-regulation of WRKY146 in wheat leaves and roots was found under osmotic stress. Further, transgenic Arabidopsis overexpressing TaWRKY146 showed improved drought tolerance with elevated contents of proline and soluble sugar and lesser malondialdehyde (MDA) content, besides induced stomatal closure and reduced transpiration rate. Moreover, Arabidopsis plants overexpressing WRKY46 from Fagopyrum tataricum (FtWRKY46) showed improved tolerance against salt stress by scavenging ROS (Lv et al. 2020). In contrast, overexpression of AtWRKY46 in Arabidopsis (OV46) resulted in hypersensitivity to drought and salt stress, achieved by modulating the expression of QUA-QUINE STARCH (QQS) (Ding et al. 2014). Wan et al. (2018) identified a set of 53 WRKY genes from the transcriptome data in Caragana intermedia, and functional validation of the two WRKY genes CiWRKY75–1 and CiWRKY40–4 was evident with enhanced drought tolerance of transgenic Arabidopsis. Genome-wide analysis of 97 WRKY genes in Pennisetum glaucum demonstrated differential expression patterns in different tissues under drought and salinity stress conditions (Chanwala et al. 2020). Out of these PgWRKY33, PgWRKY62 and PgWRKY65 were found to be regulated under both dehydration and salinity stress responses. It was also shown earlier that transformation of tobacco with SpWRKY1 increases tolerance against drought and salt stress (Li et al. 2015b). Given the highly up-regulated expression under drought stress, the transformation of WRKY3 TF gene, isolated from horse gram (Macrotyloma uniflorum), was performed in groundnut and consequently, improved drought tolerance of the transgenic lines was evident by less accumulation of MDA, more proline and higher antioxidant activities (Kiranmai et al. 2018). Other recent examples where overexpression of WRKY TF caused an improvement in drought tolerance of plants including wheat WRKY1, WRKY2 and WRKY33 into Arabidopsis (He et al. 2016; Gao et al. 2018) and Malus baccata WRKY2 and WRKY3 into tobacco (Han et al. 2018a, b).

Currently, salt-deteriorated soils form 20% of the irrigated land and 2% of the dry land, and half of the total arable land is likely to be impacted by salinity by 2050 (Kaashyap et al. 2017). Under osmotic/salt stress, an interplay between WRKY46 and “ABA signaling” and “auxin homeostasis” was elucidated in the regulation of lateral root (LR) development in Arabidopsis (Ding et al. 2015; Li et al. 2020a, b, c). LR development was hampered in a mutant (wrky46) in contrast to WRKY46 overexpressing lines that had enhanced LR. Exogenous application of the mutants with α-naphthaleneacetic acid (NAA)/ indole-3-acetic acid (IAA) rescued the LR phenotypes, whereas 2,3,5-triiodobenzoic acid (TIBA) inhibited the LR development in both WT and OV46. WRKY46 was also shown to affect auxin conjugation in lateral roots. Also, the action of WRKY46 on downstream ABI4 was illustrated through the genetic analysis of the double mutants (wrky46 abi4). Recently, Wu and colleagues (2017b) found significant differences in expression levels of PeWRKY83 in Moso bamboo (Phyllostachys edulis) during abiotic stresses (drought, salinity and ABA treatment). Further, authors defined the functional role of PeWRKY83 by transforming it into Arabidopsis, and the transgenic Arabidopsis plants showed enhanced tolerance to salt stress, with higher germination rates, higher content of proline, less electrolyte leakage and lower MDA. Transgenic plants showed reduced sensitivity to exogenous ABA and PeWRKY83 positively regulated ABA-related genes under salt stress. A new WRKY TF DgWRKY5 was isolated recently from Chrysanthemum and the expression level of the TF showed up-regulation in response to various stresses (salt, ABA and H2O2) (Liang et al. 2017). Further, the transformation of Chrysanthemum with DgWRKY5 elucidated improved salt tolerance of the transgenic plants consequent upon DgWRKY5 overexpression, with contents of proline, soluble sugars and proteins accumulating proportionate to salt stress. Differential expression of GmWRKY49 was obtained in salt-tolerant and salt-susceptible soybean genotypes and authors further performed functional characterization of GmWRKY49 by overexpressing it into soybean seedlings and Arabidopsis (Xu et al. 2018). Enhanced tolerance to salt stress was evidenced in both soybean seedlings and transgenic Arabidopsis, with improvements in germination rate, survival rate, root length, in addition to proline content. Similar studies where overexpression of WRKY genes improved salt tolerance of plants include Dendranthema grandiflorum DgWRKY2 and DgWRKY4 into chrysanthemum (Wang et al. 2018; He et al. 2018) and so forth. In contrast, salt sensitivity was promoted following overexpression of WRKY genes like CmWRKY17 (Chrysanthemum) and PcWRKY33 (Polygonum cuspidatum) in Arabidopsis, resulting in down-regulation of stress-related genes and reduction in other parameters like antioxidant enzymatic activities, and proline content (Li et al. 2015c; Bao et al. 2018).

Heat stress is defined as an increase in air temperature above a threshold level (Teixeira et al. 2013). In tropical and subtropical areas, threshold levels of heat stress are above 32–35 °C (Bita and Gerats 2013), while in temperate crops, temperature above 25 °C is considered as heat stress (Wahid et al. 2007). Implications of heat stress on plant growth and function depend upon the intensity, duration of exposure, and the degree of the elevated temperature (Sita et al. 2017). In response to heat stress, a variety of WRKY genes enable plants to adapt temperature levels beyond “optimal tolerance range” (Jiang et al. 2017). In Arabidopsis, research groups have shown participation of WRKY 25,WRKY26, WRKY33, and WRKY39 in plant response to heat stress (Li et al. 2009, 2010,2011). Previous study has shown the interaction of B-cell lymphoma2 (Bcl-2)-associated athanogene 7 (BAG7) and WRKY29 is required for heat tolerance in Arabidopsis (Li et al. 2017a, b). It was reported that CaWRKY40 is induced by heat shock and its overexpression enhanced heat shock tolerance in tobacco, while its silencing impaired thermotolerance in pepper (Dang et al. 2013). Further, it was observed that CaWRKY6 binds and activates CaWRKY40 promoter and thus acts as a positive regulator of heat stress tolerance (Cai et al. 2015). Other important plant WRKY TFs that confer combined tolerance to drought and heat include TaWRKY1 and TaWRKY33 from wheat (He et al. 2016), OsWRKY11 from rice (Wu et al. 2009) etc. It was also shown earlier in Arabidopsis that WRKY47 directly regulates Xyloglucan Endo Transglucosylase Hydrolases17 (XTH17) and Extensin‐Like Protein (ELP), thus regulating aluminium tolerance (Li et al. 2020a, b, c).

Signalling of WRKY TFs and regulation by plant hormones

As described in earlier sections, WRKY TFs play an essential role in the adaptation of crop plants to both biotic and abiotic stresses, and the adaptation results from an interplay between WRKYs and a variety of plant hormones. Empirical evidence suggests that exogenous application of different plant hormones altered expression of WRKY genes upon exposure to multiple abiotic stress including drought, salt and H2O2 (Yan et al. 2014). Under stressed conditions, WRKY TF may work independently or in coordination. Synergistic action of different WRKY TFs and enormous crosstalk with plant hormones and stress tolerance pathways demonstrates their differential expression. For instance, the synergistic interaction between ABA-inducible OsWRKY51 and OsWRKY71 genes inhibits gibberellic acid (GA) signalling in the aleurone cells of rice seeds (Xie et al. 2006; Hwang et al. 2016). Likewise, AtWRKYs 18, 40 and 60 are shown to participate in signaling pathways that are mediated by plant hormones SA, JA and ABA (Chen et al. 2010). One of the mechanisms of plant signaling involved receptor stimuli interaction and downstream signaling cascade that leads to expression of TFs and amplification of target genes and the corresponding response (Joshi et al. 2016). One of the major signalling pathways underlying both biotic and abiotic stresses is mediated through MAP kinase signaling cascade. It was demonstrated recently that ABA signalling repressor OsWRKY29 suppresses expression of OsABF1 and OsVP1, thus leading to a repressed seed dormancy (Zhou et al. 2020).

WRKY signalling in biotic stress

With regard to biotic stress, receptor R genes having NBS LRR domain activate downstream MAP kinase signaling cascades and WRKY18/22/29/30/33/53 genes, thus serving against P. syringe, Magnaporthe grisea, and B. cinerea (Fig. 1) (Sheen et al, 2002; Chujo et al 2014). Upon infection to B. cinerea, MAPK3/6 phosphorylates WRKY33 and phosphorylated WRKY33, in turn, activates PAD3, causing an increase in resistance through phytoalexin biosynthesis (Fig. 1) (Mao et al. 2007; 2011). In contrast, a decline was observed in resistance level as a result of overexpression of WRKY38 and WRKY62. Both genes, induced in a Non-expresser of Pathogenesis-Related genes 1 (NPR1)-dependent manner by SA or by virulent P. syringae, work additively as negative regulators of plant’s basal defense (Kim et al. 2008). The NO signalling pathway mediated by AtWRKY27 improves defense against R. solanacearum (Mukhtar et al. 2008). Several defense-related genes and regulatory NPR1 genes have statistically enriched WRKY TF binding sites (W box elements) in their promoter region, such as pathogen inducible ICS1 gene, suggesting either autoregulation or cross-regulation by other WRKY proteins (Dong et al. 2003).

Fig. 1
figure 1

Schematic diagram depicting the role of WRKY transcription factors under various biotic elicitors. WRKY TFs play an essential role in regulatory pathways of non pathogenic rhizobacteria and necrotic pathogens. During pathogen infection WRKY regulate downstream signalling pathways modulated by either salicylic acid dependent, or jasmonic acid dependent or nitric oxide dependent or by interacting with other WRKY genes. These WRKY regulated defense pathways has significant role in plant survival either by providing systemic acquired resistance or delaying senescence

Full size image

Out of 72 WRKY genes reported in Arabidopsis, 49 genes have been reported to be differentially regulated by P. syringae or SA treatment (Dong et al. 2003). Research suggests that TFs like WRKY3, WRKY4 and WRKY33 positively regulate the SA-mediated plant resistance pathways to necrotrophic fungal infection, such as B. cinerea (Zheng et al. 2006; Lai et al. 2008). Similarly, WRKY33, WRKY46, WRKY53 and WRKY70 positively regulate SA-mediated defense pathway against biotrophic bacterial pathogen, such as P. syringae (Zheng et al. 2006; Hu et al. 2012). Two alleles of OSWRKY45, both regulated by WRKY13, showed a contrasting response to Xoo (Qiu et al. 2007). OsWRKY45-1 overexpressing plants showed increased susceptibility in contrast to OsWRKY45-2-overexpressing plants that had enhanced resistance to Xoo and X. oryzae pv. oryzicola (Xoc) (Tao et al. 2009). Further, OsWRKY45-1 exhibited control to SA and JA levels, while OsWRKY45-2 only JA levels. Recently it was found that in rice, VQ13 (JA-responsive valine-glutamine (VQ)-motif-containing protein) activates MPK6 and WRKY45, and positively regulates JA-induced resistance to Xoo (Uji et al. 2019). AtWRKY18 and AtWRKY40 suppress Jasmonate-Zim Domain (JAZ) repressors which repress JA-signaling, causing suppression of several defense-related genes, such as FMO1, PAD3, and CYP71A13. All these eventually help Trichoderma root colonization (Brotman et al. 2013). As shown in Fig. 1, CaWRKY27 acts as a positive regulator in tobacco resistance responses to R. solanacearum infection through modulation of SA, JA, ethylene, and NO-mediated signaling pathways (Dang et al. 2014). A recent study revealed that HbWRKY83 is a positive regulator of JA-, ethylene-, and wound-responsive genes in the laticiferous cells of rubber tree (Kang et al. 2020).

Fu and Dong (2013) found that SA or benzothiadiazole (BTH)-induced systemic acquired resistance (SAR) can be achieved by transcriptional regulation of NPR1 genes. Various WRKY TFs were identified as key regulators of NPR1-dependent or -independent SAR, such as AtWRKY18, AtWRKY58, and AtWRKY70 (Wang et al. 2006), OsWRKY03 and OsWRKY71 (Liu et al. 2005, 2007) and OsWRKY45 (Shimono et al. 2007). Recently, wheat transgenic lines overexpressing barley HvWRKY6 and HvWRKY70 showed improved resistance against Puccinia striiformis f. sp. tritici pathotype CYR32 and Blumeria graminis f. sp. tritici pathotype E20 (Li et al. 2020a, b, c).

WRKY signalling in abiotic stress

In plants, ABA is a key regulator of both biotic and abiotic stress responses and its biosynthesis and accumulation are also enhanced under stress (Xiong et al. 2002). Abiotic stress tolerance signaling works through both ABA-dependent and ABA-independent pathways. Different WRKY TFs function co-ordinately or independently to the ABA signaling pathway. Thus, overexpression of AtWRKY21/33/40/57 and AtWRKY70 improves osmotic stress tolerance in an ABA-dependent manner; while AtWRKY25, ATWRKY45 and TaWRKY93 work via ABA-independent pathways. ATWRKY40 represses the expression of ABA-responsive genes (Rushton et al. 2012). Under abiotic-stressed scenario, ABA binds to PYR⁄PYL⁄RCAR protein phosphatase 2C-ABA complex and the chloroplast envelope located ABAR–ABA complex in the nucleus. These two complexes remove AtWRKY40 from the nucleus and depress ABA-responsive genes including ABI4, ABI5, ABF4, MYB2, DREB1a, DREB2b, RAB18 and AtWRKY60 (Fig. 2) (Rushton et al. 2012). Research in rice on the regulation of WKRY TFs under abiotic stress and plant hormone revealed that a set of WRKY genes is induced following abiotic stress whereas plant hormones act on another set of genes. Authors noted that majority of genes were regulated by both abiotic stress and plant hormones [ABA, IAA, GA3, Methyl jasmonate (MeJA) and SA] with extensive occurrence of crosstalks between both (Ramamoorthy et al. 2008). Analysis of GsWRKY20-overexpressing lines in Arabidopsis led authors to find an improved response to ABA, which reduced water loss and enhanced tolerance to drought stress via stomatal closure and reduced stomatal density (Luo et al. 2013). Further, multiple WRKY genes are deemed to be regulated by ABA-mediated responses (Fig. 2). ABA induces expression of different WRKY genes in rice, such as OsWRKY11, OsWRKY71, OsWRKY72 and OsWRKY77, and it may reduce expression or negatively regulate OsWRKY24 and OsWRKY45 genes. Further, OsWRKY51 and OsWRKY71 mediate the cross-talk between GA and ABA signaling. In rice, ABA-inducible nature of these genes is evident; however, these genes remain GA-repressible (Xie et al. 2006). OsWRKY24, OsWRKY51 and OsWRKY71 negatively regulate GA signaling pathway (Xie et al. 2005; Zhang et al. 2009). The transgenic lines overexpressing OsWRKY45-1 and OsWRKY45-2 registered their contrasting response to ABA. OsWRKY45-2 overexpressing lines showed increased ABA sensitivity and reduced salt stress tolerance which is phenotypically contrasting to OsWRKY45-2-suppressing lines (Tao et al. 2011). Overexpression of different WRKY genes could lead to salt tolerance by following different tolerance mechanisms. The enhanced expression of TaWRKY93 upon exogenous application of NaCl or abscisic acid improved salt tolerance via both ABA-dependent and -independent pathways and also by enhancing osmotic stress tolerance (Fig. 2) (Qin et al. 2015). Osmotic stress tolerance has been regulated by AtWRKY70 and AtWRKY54 with both working cooperatively and negatively regulating stomata closure (Li et al. 2013). The study established the role of both WRKY genes for improving drought tolerance in plants. Under salt stress, AtWRKY25 and AtWRKY33 have improved salt stress tolerance through SOS-pathway-independent manner. However, AtWRKY33 showed salt tolerance through ABA-dependent signaling pathway (Jiang and Deyholos 2009). Recently, it was demonstrated that Arabidopsis plants overexpressing IbWRKY2 (Zhu et al. 2020) and HbWRKY83 (Kang et al. 2020) were tolerant against salinity and drought stress via improved ROS scavenging system. Under heat stress, AtWRKY25 regulated expression of heat-inducible and oxidative stress-inducible genes containing W box at their promoter region, independent of the SA pathway. Overexpression of WRKY25 increased the expression of HsfA2, HsfB1, HsfB2a and Hsp101, implying a role in HsfB2a and HsfB1 pathway (Li et al. 2009). Cold stress tolerance is primarily regulated through cold-responsive (COR) regulon (Fig. 2). Overexpressing WRKY46 in cucumber conferred cold tolerance by modulating cold signalling pathway in an ABA-dependent manner (Zhang et al. 2016). AtWRKY34 regulates CBF and GM20 regulates ICE1 gene. Further, COR regulon regulates AtWRKY6/187/22/30/32 and AtWRKY40 genes (Banerjee and Roychoudhury 2015). Generally, abiotic stress signal activates or works through MAPK signaling pathways.

Fig. 2
figure 2

Schematic diagram depicting the signalling of WRKY transcription factors under different abiotic stresses. Various WRKY members plays different roles in regulation of abiotic stress regulatory pathways in different plants and tissues indicates their essential role in plant stress tolerance. During abiotic stress regulation, response occurs by complex cross-talk between phytohormones, such as, ABA, SA, MeJA, and ETH, and individual pathway is regulated by specific WRKY genes further modulting different physiological and biological processes

Full size image

More recently, a global analysis of sesame (Sesamum indicum L.) WRKY TF identified 71 genes, of which 26 were drought-responsive while 33 genes responded to waterlogging (Li et al. 2017b). The gene WRKY1 of Poncirus trifoliate and Citrus grandis was analyzed under cold and drought stresses, and the results showed that both genes showed induced expression in response to drought while cold could evoke response only from PtrWRKY1 (Şahin-Çevik and Moore 2013). Similarly, overexpression of various WRKY TFs led to tolerance to various abiotic stress, such as OsWRKY89-overexpressing lines in rice, showed enhanced tolerance to ultraviolet B irradiation (Wang et al. 2007), OsWRKY11 under the control of the heat shock protein 101 (HSP101) promoter enhanced heat and drought tolerance in rice (Wu et al. 2009). In summary, recent evidence in plant species supports multiple roles for WRKY genes against a variety of abiotic stresses.

WRKY-mediated crosstalk between abiotic and biotic stress responses

Recent evidence supports the involvement of WRKY TF in mediating response and adaptation to biotic and abiotic stresses. This crosstalk phenomenon is exemplified by the action of several WRKY TFs, such as OsWRKY13, in rice that alters plant response to both biotic (Xoo and M. grisea) and abiotic stresses (Qiu et al. 2007; 2008). OsWRKY13 is possibly regulated by both OsWRKY45-1 and OsWRKY45-2, the two alleles identified, respectively, from japonica and indica rice. Previous research has established OsWRKY45 and SlWRKY8 as the positive regulators of PR gene expression, resistance to P. syringae pv. tomato DC3000 (PsPto), tolerance to drought and salt stress and ABA sensitivity (Qiu and Yu 2009; Gao et al. 2020). Subsequent experiments elucidated that OsWRKY45-1, a negative regulator of ABA signaling, plays no role in salt tolerance in rice. In contrast, OsWRKY45-2 is a positive regulator of ABA signaling, and a negative regulator of salt stress tolerance (Tao et al. 2011). Similarly, constitutive expression of grape WRKY3 gene in Arabidopsis provided insights into its role in defining plant response to biotic (Golovinomyces cichoracearum, B. cinerea) and abiotic stresses (drought and salinity) (Guo et al. 2018). Research by Lee et al. (2018) established OsWRKY11 as a positive regulator of rice–Xoo interaction and drought stress in rice through activating expressions of deference- (Chitinase 2) and drought-(RAB21) related genes. Improved resistance to PsPto and drought/salt stress in tomato was achieved by overexpression of SlWRKY39 (Sun et al. 2015). The higher tolerance level of transgenic plants could be ascribed to the abundance of PR (SlPR1, SlPR1a1) and stress-related genes (SlRD22, SlDREB2A), and lower and higher contents of MDA and proline, respectively. Taken together, WRKY TFs participate in various cross regulatory networks related to defense response and their interaction provides insight into their coregulated functional dynamics of signaling response during different biological processes.

Other roles of WRKY TFs

In addition to stress response, WRKY TFs role in a range of important processes including plant development are well documented (Chen et al. 2017). The SPF1 (Sweet Potato Factor1), first identified WRKY protein, participated in tuberous root development in Ipomoea batatas (Ishiguro and Nakamura 1994). Further, WRKY proteins ABF1 and ABF2, isolated from Avena fatua were found to regulate seed germination (Rushton et al. 1995). Besides, WRKY1, WRKY2, and WRKY3 were reported to regulate ribosomal protein gene expression in Petroselinum crispum (Rushton et al. 1996). WRKY42, WRKY45, WRKY75, WRKY6 from Arabidopsis and WRKY74, WRKY80 from rice were involved in plant nutrient (i.e., phosphorus, boron and iron) utilization (Chen et al. 2017). AtWRKY6, AtWRKY53 and OsWRKY45 regulate leaf senescense (Miao and Zentgraf 2010). Moreover, AtWRKY12, AtWRKY13, AtWRKY71 and OsWRKY11 were reported to be involved in floral development in angiosperms (Cai et al. 2014; Li et al. 2016; Yu et al. 2016). A recent study in Arabidopsis elucidated down-regulation of a set of genes associated with senescence (SAG12, SAG13 and SAG29) and chlorophyll degradation genes (PPH, PAO, NYE1/SGR1, NYC1, NOL) resulting from the overexpression of CiWRKY40–4 caused delayed leaf senescence in the transgenic plants (Wan et al. 2018).

WRKY’s participation in plant reproductive development is evident from severe fertility defects resulting from a compromise in WRKY-mediated regulatory network. In Arabidopsis, gene expression patterns in mature pollen grains, inferred from publically available data, qRT-PCR assay and genetic analysis, led authors to assume that WRKY2, WRKY34 and VQ20, suppress the expression of MYBs (MYB97, MYB101 and MYB120) during male gametogenesis. Further, binding of the WRKY2/WRKY34 with MYB97 via the promoter containing W-box was confirmed (Lei et al. 2018). Prioritization of defense response overgrowth, as shown by OsWRKY70 in mitigating insect herbivory in rice, can often yield unintended consequences (Li et al. 2015a). The authors highlighted the profound implications that these trade-offs may have, in breeding resistant plants.

Implications for future crop improvement

In recent years, a plethora of information has been generated in the wake of growing research interest on TFs, both in model and non-model crops. More comprehensive information is warranted on the activities of WRKY TFs in the plant genome, their communication with other biomolecules and sorting complexities of upstream elements that regulate them. It is further being traversed by modern tools of omics, analysis of complementarity with (cis-) promoter, systemics and several useful tools of bioinformatics. Different WRKY genes/TFs behave in a versatile manner even to an extent that homologs behave differently in different backgrounds (Cai et al. 2014). A set of WRKY TFs is known to regulate multiple genes with opposite effects (Liu et al. 2015) or could be induced and repressed depending on the external stimuli (Huang et al. 2016; Yan et al. 2016; Ding et al. 2013). In some cases, WRKY TFs upregulate both useful genes as well as unwelcomed genes (Wang et al. 2014). So, it is important to characterize them, elucidate their function and select the key TF translating gene for crop improvement. This deluge of information will be immensely useful when translated in crop improvement for the benefit the farmers and the mankind. However, practical implications of WRKY for crop improvement face a set of challenges.

Evidence suggests that WRKY TFs not only participate in plant growth and development, but also show complex regulatory mechanisms and networks involved in external abiotic stresses. A variety of roles of WRKYs have been elucidated on parts of the plant like root, leaf, etc.; however, evidence based on the whole plant is currently lacking. Considerable possibility exists that inferences gained from in vitro experiments might yield uncertain outcomes when applied for plant improvement in a different environment. Nothwithstanding this, there has been plenty of crop improvement illustrations recently based on transgenics and genomics-assisted breeding. Some of the recent examples are listed in Table 1. Growing body of literature on WRKY TFs advocates their usefulness for genetic characterization of a given crop population and evolutionary studies at different levels as a function of WRKY TFs may vary as a consequence of environmental stress and in the process of adaptation (Yan et al. 2016). In conclusion, considerable scope exists for understanding WRKY TFs with respect to their functionality, evolution and intra/inter communication with other TFs in cooperative and antagonistic manner, noncoding RNAs and epigenetic modifications to find a common link in the complex signal pathways to facilitate improvement of agricultural crop yield and quality.

Table 1 WRKY TF playing crucial roles in plant reproductive development
Full size table