# Source paper archive

- Source title: Comparative Analysis of Orthologous ANK-NLR-WRKY Proteins in Wheat Species: Evolutionary Divergence and Structural Characteristics
- Author(s): Venter, E., Mkhize, N. S.
- Source landing URL: https://www.biorxiv.org/content/10.64898/2026.11.18.767528v1
- PDF URL: https://www.biorxiv.org/content/10.64898/2026.11.18.767528v1.full.pdf
- PDF SHA-256: 5430ebb22d0297b4d1fe0f419f40f986819d91d026b6ac59942dcf8e6ad17852
- Converter: pymupdf4llm 0.0.27
- Limits: Automated PDF text extraction; reading order, equations, tables and formatting may be incomplete or incorrect. Images are omitted, and no OCR was performed. Markup and links are neutralized for safe downloading. Consult the original PDF for the authoritative version.

---

&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1 **Comparative Analysis of Orthologous ANK-NLR-WRKY Proteins in Wheat Species:**


2 **Evolutionary Divergence and Structural Characteristics**


3


4 Naledi Mkhize, Eduard Venter*


5


6 Department of Botany and Plant Biotechnology, University of Johannesburg, Auckland Park,


7 Johannesburg, 2006, South Africa


8 *Corresponding author: Eduard Venter


9 eduardv@uj.ac.za


10


11 ORCID ID:


12 Naledi Mkhize: 0000-0001-6210-281X


13 Eduard Venter: 0000-0002-1838-6453


14


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


15 **Abstract**


16 The plant nucleotide-binding leucine-rich repeat (NLR) immune receptors are known to play a crucial


17 role in plant host defence against a variety of biotic stresses. Several NLR proteins contain integrated


18 domains (NLR-ID) like ankyrin (ANK) and WRKY transcription factor domains and are important


19 evolutionary adaptations in plant immunity. They are responsible for the perception of effector


20 molecules from invading pathogens and pests and subsequent activation of plant immune responses.


21 However, the functional roles these ANK-NLR-WRKY proteins play in wheat ( _Triticum aestivum_ )


22 against various pathogens and pests remain poorly understood. Computational tools employed for


23 comparative sequence and phylogenetic analyses revealed that _TaAdnr1_ is highly conserved across


24 diploid ( _T_ . _urartu_ ), tetraploid ( _T_ . _dicoccoides_ ), and hexaploid ( _T_ . _aestivum_ ) wheat species, including its


25 well characterised orthologue YrU1 from _T. urartu_ . Three-dimensional (3D) protein structural


26 modelling using AlphaFold2 supported strong conservation in the multiple domains of these NLR


27 orthologues, more specifically within the integrated WRKY domain. This conservation may suggest a


28 strong evolutionary maintenance of TaADNR1. Furthermore, this may also suggest TaADNR1 reflects


29 specificity in effector recognition and regulation. Therefore, the integration of functional genomics


30 studies, sequence analysis, evolutionary analysis and structural modelling possesses the potential to


31 provide new insights into the pleiotropic roles of integrated domain NLRs in wheat.


32


33 **Keywords**


34 Plant Immune Receptors, Integrated Decoy Domain, NLR-ID, Cross Kingdom Immunity


35


2

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


36 **Introduction**


37 The WRKY transcription factor (TF) domain consists of the conserved amino acid sequence


38 WRKYGQK at its N-terminus, which is typically around 60 amino acids in length and in some instances


39 includes domain repetitions (Hussain et al., 2024). These domains are known to bind to specific W-box


40 elements ((T/C)TGAC(T/C)) in the promoter regions of target genes, thereby modulating their


41 transcription (Zhang et al., 2020; Zhang &amp; Wang, 2005). The WRKY proteins are classified into three


42 major groups based on the number of WRKY domain sequences and the features of their zinc-finger


43 motifs (Zhao et al., 2023). Group I consists of two WRKY domain sequences, with group II and group


44 III containing only one WRKY domain sequence. Group I and II WRKY proteins are characterised by


45 the presence of C2H2 zinc finger motifs while Group III WRKY proteins contains a C2HC zinc finger


46 motif (Zhao et al., 2023). WRKY proteins are further subdivided into various subgroups based on


47 phylogenetic analyses, which reveal distinct evolutionary pathways and functional specialisation among


48 them (Sinha et al., 2024). The presence of these structures and sequences enables the WRKYs to bind


49 to W-box cis-elements in the target gene promoter (Zhang &amp; Wang, 2005). These structural distinctions


50 within the WRKY family are critical for their varied DNA-binding specificities and diverse regulatory


51 functions across plant species (Sinha et al., 2024).


52 WRKY transcription factors interact with signalling pathways and coordinate a complex regulatory


53 network essential for plant defence and environmental adaptation (Villano et al., 2020; Wen et al., 2023).


54 These pathways are mediated by various phytohormones and secondary messengers, including salicylic


55 acid and jasmonic acid, which control the plant's defence mechanisms against pathogens and pests


56 (Zhao et al., 2023; Chen et al., 2019). They modulate the expression of diverse stress-related genes by


57 binding to their cognate W-box elements and initiate or repress gene expression to facilitate adaptive


58 responses. Furthermore, they also interact with other transcription factor families, such as those from


59 the bHLH, NAC, and MYB families, forming complex regulatory networks that regulate plant


60 responses (Finatto et al., 2018). This highlights the pleiotropic roles of WRKY proteins in orchestrating


61 the intricate molecular responses to diverse environmental settings and plant developmental stages


62 (Hussain et al., 2024; Zhang et al., 2020).


3

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


63 Plant nucleotide-binding leucine-rich repeat (NLR) proteins are a vital component of the plant innate


64 immune system for the recognition of pathogen or pest invasion and the subsequent activation of a rapid


65 and robust immune response (Césari, 2018). The architecture of plant NLR proteins consists of a


66 complex mechanism of defence and a comprehensive evolutionary divergence in both


67 monocotyledonous and dicotyledonous plants (Shao et al., 2016). These NLR proteins possess


68 additional domains that provides an evolutionary strategy to detect pathogen invasion. Such NLRs


69 consist of additional non-canonical domains that resemble host proteins that are targets of pathogen


70 effectors and are termed integrated decoy NLRs (ID-NLRs) (Ellis, 2016; Gu et al., 2025). This strategy


71 is a product of processes such as domain fusions, diversification and gene duplication that enable plants


72 to adapt to the evolving pathogens (Césari et al., 2014). Studies have reported that these additional


73 domains are implicated in the recognition of pathogen effectors either directly or indirectly. Following


74 recognition by these intricate IDs, a signalling cascade is activated which results in programmed cell


75 death (PCD) and systemic acquired resistance (SAR) (Marchal et al., 2022; Sarris et al., 2015). The


76 WRKY and zinc-finger BED domains are examples of the IDs that are frequently identified and studied


77 in plants (Kroj et al., 2016) with the diversification of these IDs a prominent evolutionary strategy for


78 NLR diversification (Ellis, 2016).


79 The WRKY domain-integrated NLR genes play a crucial role in wheat defence by enhancing


80 transcriptional regulation during pathogen and pest attacks. Plant NLRs with WRKYs suggest that


81 pathogen recognition and transcriptional reprogramming are interconnected as many WKRY


82 transcription factors have been implicated in these roles (Anbu et al., 2023; Jin et al., 2022). This enables


83 plants to produce a rapid and precisely regulated defence response within the host (Chen et al., 2019).


84 In wheat, the WRKY integrated NLR orthologs, _TaAdnr1_ and _YrU1_ were identified from _T. aestivum_


85 and _T. urartu_, respectively (Wang et al., 2020a; Nicolis &amp; Venter, 2018). Studies on _TaAdnr1_ and _YrU1_


86 represent defence mechanisms where the integrated WRKY domain likely mediates the transcriptional


87 activation of defence pathways upon pathogen or pest recognition. Knockdown of _TaAdnr1_ containing


88 a single WRKY integrated domain resulted in increased susceptibility to the aphid _Diuraphis noxia_


89 (Nicolis &amp; Venter, 2018), with the _YrU1_ WRKY linked to stripe rust resistance against _Puccinia_


4

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


90 _striiformis_ f. sp. _tritici_ ( _Pst_ ) (Wang et al., 2020). The unique structure of _YrU1_ was shown to aid in the


91 recognition of pathogenic effectors and thereby activating immune signalling (Wang et al., 2020a). This


92 functional integration of a transcription factor domain directly within an NLR sensor provides a


93 mechanism for the immediate transcriptional modulation upon effector recognition and subsequent


94 initiation of defence responses (Anbu et al., 2023).


95 The role of integrated WRKY domains into NLR proteins is best studied for _Arabidopsis_ RRS1. The


96 integrated WRKY domain of RRS1 allows for the direct recognition of pathogen effectors such as


97 PopP2, which acetylates the integrated WRKY domain, leading to NLR activation and subsequent


98 immune responses (Sarris et al., 2015). This mechanism exemplifies an integrated decoy strategy, where


99 the WRKY domain within RRS1-R mimics a host target of PopP2, thereby converting effector-mediated


100 suppression of host defences into immune activation (Roux et al., 2015). The deletion of the WRKY


101 domain in RRS1 resulted in an auto-active allele, suggesting that this domain actively contributes to


102 maintaining the NLR complex in an inactive state prior to ligand detection (Ma et al., 2018). Upon


103 effector binding, this inhibitory effect is released and enables activation of the associated NLR and


104 subsequent initiation of defence responses (Li et al., 2019). This intricate regulatory mechanism defines


105 sophistication of plant innate immunity where NLR-WRKY fusions offer an alternative strategy for


106 detecting pathogen effectors that specifically target host transcription factors (Chen et al., 2019).


107 This evolutionary strategy suggests that integrated domains in plant NLRs, including WRKY and


108 others, are crucial for detecting various effectors and cofactors in plant immunity for activation of an


109 immune signal and ultimate defence response. This intricate mechanism enhances the specificity as well


110 as the efficacy of the plant immune response (Dangl et al., 2013). Furthermore, this integration strategy


111 allows for an efficient method for the NLR to directly perceive effector modifications of these host


112 targets and convert the pathogen's virulence strategy into an immune activation trigger (Adachi et al.,


113 2019). To further our understanding of these roles we investigated the predicted 3D architectures of


114 the different wheat ANK-NLR-WRKY and their evolutionary relationships, to evaluate the


115 possible similarities and differences between them, as well as the potential conserved resistance


116 mechanisms, using molecular biology computational tools.


5

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


117


118 **Material and Methods**


119 **Wheat ANK-NLR-WRKY protein (** _**T. aestivum**_ **,** _**T. dicoccoides**_ **and** _**T. urartu**_ **) sequence retrieval,**


120 **domain identification and alignment**


121 The full-length canonical protein sequences of TaADNR1 ( _T. aestivum_, AABBDD genome hexaploid


122 wheat), TRIDC5AG050380 ( _T. dicoccoides_, AABB genome tetraploid wheat) and YrU1 ( _T. urartu_, AA


123 genome diploid wheat and A-genome ancestor for tetra- and hexaploid wheat), were downloaded from


124 EnsemblPlants. The domain boundaries of the three proteins for the NB-ARC and WRKY domains


125 were predicted using InterProScan (Blum et al., 2026) searches against the plant family PF00931 for


126 the NB-ARC domain and PF03106 for the WRKY domain. The coordinates of the domain region of the


127 WRKY domains were extracted and validated by the presence of the conserved WRKYGQK motif. The


128 three full-length wheat ANK–NLR–WRKY proteins were aligned using MAFFT (v7) utilising the L

129 INS-i algorithm. The default gap opening and extension penalties from MAFFT were used and the


130 alignment was imported to the web-based ESPript 3.0 tool (https://espript.ibcp.fr/ESPript/ESPript/) for


131 visualisation. The parameters selected in the ESPript 3.0 tool included %MultAlin for generation of an


132 alignment with similar residues according to their chemical properties and a default Global score of 0.7


133 since the proteins are long with multiple domains. The generated ESPript alignment output showed


134 conserved residues based on the similarity thresholds. Finally, the produced ESPript alignment output


135 was used to annotate the different domains (ANK, NB-ARC and LRR domains) as well as the conserved


136 motifs (WRKYGQK and C2HC-type zinc-finger motif) inferred using the InterProScan annotations


137 from the BLAST search outputs.


138 The isolated WRKY domain sequences from the three wheat proteins and _Arabidopsis_ AtWRKY41 and


139 AtWRKY46, each consisting of 60 amino acids, were extracted according to the Pfam WRKY DNA

140 binding domain (PF03106) based on InterProScan annotations. These WRKY-only sequences for YrU1,


141 TRIDC5AG050380, and TaADNR1 were aligned using MAFFT (v 7) with default parameters and the


142 alignment was visualised using ESPript 3.0. They were chosen for further computational analyses as


6

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


143 the phylogenetic analysis indicated that they are the potential WRKY domain progenitors found in YrU1


144 and TaADNR1.


145


146 **Evolutionary relationships of ANK–NLR–WRKY proteins across grasses**


147 **Wheat NLR-WRKY protein sequence retrieval, alignment and construction of phylogenetic tree**


148 The full-length canonical protein sequences consisting of the NB-ARC and WRKY domain from wheat


149 and related grass species were obtained from EnsemblePlants and aligned using MAFFT with the L

150 INS-i algorithm, which is optimised for accurate alignment of protein sequences containing conserved


151 motifs and variable regions (Katoh &amp; Standley, 2013). ClipKIT v1.4.1 in smart-gap mode was used to


152 trim the MAFFT alignment to reduce alignment noise and poorly aligned regions and to retain


153 phylogenetically informative sites while removing columns dominated by gaps (Steenwyk et al., 2020).


154 The maximum-likelihood phylogenetic tree was constructed using IQ-TREE v2, branch support was


155 assessed by using 1 000 ultrafast bootstrap replicates, and the tree was rooted using the _Arabidopsis_


156 RRS1 sequences as outgroups.


157


158 **Phylogenetic analysis and identification of TaADNR1 WRKY transcription factor lineage**


159 **associated with WRKY domain integration**


160 The complete WRKY transcription factor proteome of _T. urartu_ was combined with the WRKY domain


161 sequences of YrU1, TRIDC5AG050380, and TaADNR1. _Arabidopsis_ WRKY1, WRKY40, and


162 WRKY55 were included as reference outgroups to anchor the tree relative to well-characterised WRKY


163 lineages. All the _T. urartu_ WRKY transcription factors that clustered with YrU1, TRIDC5AG050380,


164 and TaADNR1, were considered as possible candidate donor lineages for WRKY domain integration.


165 The candidate _T. urartu_ WRKY transcription factors identified from the clustering were annotated by


166 cross-referencing EnsemblPlants gene IDs with WRKY family classifications, _Arabidopsis_ orthologues


167 in the Plant Transcription Factor Database (PlantTFDB) as well as published functional studies.


7

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


168


169 **AlphaFold wheat (** _**T. aestivum**_ **,** _**T. dicoccoides**_ **and** _**T. urartu**_ **)** **ANK-NLR-WRKY** **structure**


170 **prediction and comparison**


171 The protein sequences of the different wheat ANK-NLR-WRKY proteins (TaADNR1, YrU1, and


172 TRIDC5AG050380) and _Arabidopsis_ WRKY proteins AtWRKY41 and AtWRKY46 were submitted to


173 AlphaFold2, v2.3.2, (Jumper et al., 2021) for structural prediction via the Latch.bio platform using the


174 default parameters. From the results of the prediction, the highest-ranked model (ranked_0.pdb) for


175 each protein was used for all the downstream analyses. The per-residue confidence scores (pLDDT)


176 were extracted from the B-factor column of the predicted PDB files using a custom Python script


177 (Supplementary data), utilising the NumPy library. The average pLDDT was calculated for the full

178 length of each of the protein, as well as the WRKY specific domain according to Supplementary Table


179 1 (Table S1). The residues were categorised by confidence percentage with very high (&gt;90%), high (70

180 90%), low (50-70%), and very low (&lt;50%).


181 For each of the proteins, the model structure with the highest rank was visualised, superimposed, and


182 further analysed using the UCSF ChimeraX (v1.8) software. The pairwise alignment of the structures


183 for the three wheat proteins, and TaADNR1 with AtWRKY41 and AtWRKY46, were performed using


184 the matchmaker command in ChimeraX. The model structures of YrU1 and TRIDC5AG050380 were


185 aligned to TaADNR1 as the reference. ChimeraX was also used to calculate the root-mean-square


186 deviation (RMSD, in Ångströms) and was reported for both the full-length superpositions, as well as


187 the isolated WRKY domains. The WRKY domains were isolated for each of the proteins following


188 superpositions by selecting the defined residue ranges and hiding the non-selected atoms for a focused


189 visual analysis.


190


8

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


191 **Results**


192 **Wheat ANK-NLR-WRKY protein (** _**T. aestivum**_ **,** _**T. dicoccoides**_ **and** _**T. urartu**_ **) sequence retrieval,**


193 **domain identification and alignment**


194 The domain architecture of the three ANK–NLR–WRKY proteins from wheat, TaADNR1 ( _T._


195 _aestivum_ ), TRIDC5AG050380 ( _T. dicoccoides_ ) and YrU1 ( _T. urartu_ ) revealed that they were highly


196 similar and conserved in the N-terminal ankyrin repeat (ANK) domain and the central NB-ARC


197 domain, with minor amino acid substitutions within the C-terminal leucine-rich repeat (LRR) region


198 and the terminal WRKY DNA-binding domain. The domain boundaries were defined using


199 InterProScan annotations (Table 1).


200


201 Table 1: The domain architecture of the different wheat ANK-NLR-WRKY proteins


202


203


204 The full-length sequence alignments shared 100% amino acid identity in all the ANK and NB-ARC


205 domains for the three proteins (Figure S1) strongly supporting the evolutionary conservation of these


206 regions. The LRR region shared 100% complete identity between TaADNR1 and TRIDC5AG050380.


207 YrU1 exhibited 98.37% identity against both TaADNR1 and TRIDC5AG050380 proteins.


208 Contrastingly, six amino acid substitutions were identified in the LRR region of YrU1 compared to both


209 TaADNR1 and TRIDC5AG050380 that showed 100% identity in this region. The six amino acid


210 substitutions from the similar TaADNR1 and TRIDC5AG050380 compared to the YrU1 sequence were,


211 Ser1154Thr, Gly1199Arg, Ile1205Met, Thr1216Ala, Tyr1223Ala, and Ala1235Asp. The conserved


9

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


212 WRKY domain WRKYGQK motif as well as the C2HC-type zinc-finger motif was retained in all the


213 three proteins, which indicated that the DNA-binding functionality should be maintained. There were


214 some amino acid differences between TaADNR1 and TRIDC5AG050380 and YrU1 observed before


215 and after the WRKY domains.


216 High conservation was observed from the multiple sequence alignments of the isolated WRKY domains


217 of the three wheat proteins, especially on the WRKY core. TaADNR1 and TRIDC5AG050380 shared


218 100% identity and YrU1 indicated 83.33% identity compared to both proteins (Figure 1a). The


219 WRKYGQK motif was completely conserved in all three proteins which indicated preservation of


220 DNA-binding specificity. In contrast, there were nine amino acid substitutions observed in the regions


221 downstream of the WRKY motif in YrU1 compared to TaADNR1 and TRIDC5AG050380. These


222 substitutions included changes Thr14Ile, Met15Ile, Leu35Pro, Ala36Val, Asn45Thr, His54Tyr, and


223 Asn55His. There was only one amino acid substitution observed upstream of the WRKY motif at


224 position 1 (Asp1Glu). Importantly, there were no substitutions observed in WRKYGQK motif as well


225 as the conserved zinc-finger motif in all the three proteins (Figure 1a, Figure S1).


226


227 **Figure 1** : **A**, The WRKY domain-only multiple sequence alignments isolated from the ANK–NLR–
228 WRKY protein sequences of TaADNR1, TRIDC5AG050380 and YrU1 ( _T_ . _aestivum_, _T_ . _dicoccoides_
229 and _T_ . _urartu_ )The WRKY domain-only multiple sequence alignments of the isolated WRKY regions
230 from TaADNR1, AtWRKY41 and AtWRK46.Amino acid substitutions are present upstream and
231 downstream the WRKY core, including the zinc-finger motifs in AtWRKY41 and AtWRKY46. There
232 were no amino acid substitutions observed within the WRKY core for all the proteins.
233


10

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


234 The sequence alignment between TaADNR1 and AtWRKY41 shared 60% amino acid identity,


235 TaADNR1 and AtWRKY46 shared 61.67% amino acid identity, and AtWRKY41 and


236 AtWRKY46 shared a 66.67% identity (Figure 1b). The alignments of the isolated WRKY


237 domain of TaADNR1 and both the _Arabidopsis_ WRKY protein domains revealed that the


238 WRKY core was conserved. There were twenty-four amino acid substitutions for TaADNR1


239 against AtWRKY41, twenty-three between TaADNR1 and AtWRKY46, and twenty amino


240 acid substitutions between AtWRKY41and AtWRKY46. These substitutions were observed in


241 the region upstream and downstream the WRKY core. Notably, substitutions were also


242 observed in the region within the Zinc-finger motif between the three proteins.


243 There were sixteen amino acid substitutions between TaADNR1 and AtWRKY41 that showed


244 the most important polarity and charge changes, as well as broad physicochemical class


245 changes (Table S1). There were eight amino acid substitutions that retained the same broad


246 class changes. The sixteen amino acid substitutions that showed the most important polarity


247 and charge changes included Ser17Ala and Thr45Gly, changing from polar uncharged to non

248 polar, His19Phe and His28Phe, changing from polar-ionisable to non-polar. There were six


249 amino acid substitutions that exhibited changes from non-polar to polar uncharged and these


250 included Ala22Ser, Phe24Tyr, Val31Thr, Gly33Tyr, Leu48Thr and Val53Thr. Single amino acid


251 substitutions were observed for changes from non-polar to polar positively charged


252 (Met18Lys), polar uncharged to polar negatively charged (Ser44Asp), polar positively charged


253 to polar uncharged (Lys32Gln), polar ionisable to polar positively charged (His55Arg) and


254 polar negatively charged to polar uncharged (Glu57Thr).


255 The alignment between TaADNR1 and AtWRKY46 indicated fifteen amino acid substitutions


256 with the most important polarity and charge changes, and broad physicochemical class changes


257 (Table S2). There were eight amino acid substitutions that retained the same broad class


258 changes. The fifteen amino acid substitutions that indicated the most significant polarity and


11

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


259 charge changes included Thr37Val, changing from polar uncharged to non-polar, His28Phe and


260 His55Leu, changing from polar-ionisable to non-polar. There were three amino acid


261 substitutions that displayed substitutions from non-polar to polar uncharged and these included


262 Phe24Tyr, Val31Thr and Leu48Ser. Substitutions from non-polar to polar ionisable and non

263 polar to polar positively charged included Phe4His and Ile15His, and Met18Lys and Val53Lys,


264 respectively. Single amino acid substitutions were observed for changes from non-polar to


265 polar negatively charged (Gly33Asp), polar uncharged to polar negatively charged (Ser44Asp),


266 polar positively charged to polar uncharged (Lys32Gln), polar ionisable to polar uncharged


267 (His19Asn) and polar negatively charged to polar uncharged (Glu57Asn).


268 The AtWRKY41 and AtWRKY46 alignment showed fifteen amino acid substitutions that were


269 non-identical with significant polarity and charge changes (Table S3). These included


270 Ser22Ala, changing from polar uncharged to non-polar and Ala17Ser, Phe19Asn and


271 Gly45Thr, changing from non-polar to polar uncharged. There were three amino acid


272 substitutions that displayed substitutions from non-polar to polar ionisable and these included


273 Phe4His, Leu15His and Phe28His. Single amino acid substitutions were observed for changes


274 from polar uncharged to polar positively charged (Thr53Lys) and polar positively charged to


275 non-polar (Arg55Leu).


276


277 **Phylogenetic relationships between NLR-WRKY proteins across grasses and placement of the**


278 **integrated WRKY domain**


279 The evolutionary relationships of NLR-WRKY proteins across monocot grass species included _T_ .


280 _aestivum_ (TaADNR1), _T_ . _dicoccoides_ (TRIDC5AG050380), _T. urartu_ (YrU1), and homologues from


281 _Aegilops_, _Oryza_, and _Brachypodium_ with the resulting topology indicating TaADNR1,


282 TRIDC5AG050380 and YrU1 forming a strongly supported (100 %) monophyletic clade (Figure 2).


283 The WRKY-only phylogenetic tree was constructed using the complete _T. urartu_ WRKY transcription


12

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


284 factor proteome isolated from EnsemblPlants, the WRKY domains derived from YrU1,


285 TRIDC5AG050380, and TaADNR1 full-length sequences, and the _Arabidopsis_ WRKY1, WRKY40,


286 and WRKY55 as outgroups (Figure 3). This was to identify the likely WRKY transcription factor


287 lineage that contributed the integrated WRKY domain into the wheat NRL protein. The WRKY domains


288 from YrU1, TRIDC5AG050380, and TaADNR1 clustered together in a well-supported clade (≥ 99 %)


289 with YrU1 basal to TRIDC5AG050380, and TaADNR1 (Figure 3). This confirmed the shared


290 evolutionary origin of these proteins with YrU1 as ancestral to TRIDC5AG050380 and TaADNR1. This


291 clade grouped well with five _T. urartu_ WRKY transcription factors that included WRKY41 and


292 WRKY46 suggesting that these WRKY’s represent the closest relatives of the WRKY donor lineage. A


293 BLAST search against _Arabidopsis_ WRKY proteins from the Plant Transcription Factor Database


294 (PlantTFDB) and The Arabidopsis Information Resource (TAIR) confirmed the six _T. urartu_ WRKY


295 transcription factors were similar to the WRKY41- and WRKY46-like proteins (Figure S2, Table S4).


296 Four of the candidates (TuG1812G0100004483.01, TuG1812G0300002773.01,


297 TuG1812S0003165700.01, TuG1812G0200001696.01) showed similarity to WRKY41-like protein,


298 with TuG1812G0500002157.01 exhibiting similarity to the WRKY46-like protein. These WRKY41

299 and WRKY46-like sequences showed strong sequence similarity (E-values ≤ 10⁻³¹, Table S4), high


300 bootstrap support (76–100 %, Figure 3), and clustered tightly with the integrated WRKY domains to


301 support a close evolutionary relationship.


302


303 **Figure 2** : The phylogenetic relationships of the wheat NLR-WRKY proteins


13

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


304


305 **Figure 3** : The WRKY-only phylogenetic analysis for the identification of the likely donor lineage of
306 the integrated WRKY domain
307

308


309 **AlphaFold2 structure predictions, comparison and analysis of wheat** **ANK-NLR-WRKY** **full-**


310 **length proteins and WRKY domain**


311 The AlphaFold2 structure predictions generated quality structural models for the three full-length ANK

312 NLR-WRKY proteins (Figure 4, Table S5). The full-length model for TaADNR1 and


313 TRIDC5AG050380 showed a high average pLDDT scores of 71.91 and 71.64 (Figure 4A and 4C),


314 respectively. YrU1 indicated a moderate average pLDDT score of 68.32 (Figure 4B). The average


14

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


315 pLDDT scores for the WRKY specific domain of all three proteins was high with 82.27, 81.61 and


316 78.48 for TaADNR1, TRIDC5AG050380 and YrU1, respectively. Notably, both TaADNR1 and


317 TRIDC5AG050380 WRKY domains exhibited pLDDT scores above 81.


318


319 **Figure 4** : Alphafold2 predicted structural alignment of full-length ANK-NLR-WRKY proteins


320 The structural superposition revealed a high degree of conservation across the three proteins (Figure 4,


321 Table S5). The three ANK-NLR-WRKY full-length proteins showed high similarity with pairwise


322 RMSD values below 0.75 Å (Figure 4A, Table S5). The structural superposition between TaADNR1


323 and YrU1 indicated a low RMSD of 0.746 (Figure 4B, Table S5), while TaADNR1 and


324 TRIDC5AG050380 exhibited the lowest RMSD of 0.424 (Figure 4B, Table S5). The conservation was


325 even more pronounced within the WRKY domain. Notably, the WRKY domains of TaADNR1 and


326 TRIDC5AG050380 were nearly indistinguishable, with an RMSD of only 0.130 Å (Figure 4C, Table


327 S5). The WRKY domain of the more distantly related YrU1 also maintained very high similarity to


328 TaADNR1 RMSD 0.560 Å (Figure 4B, Table S5). A similar trend was observed in the high-confidence


329 residue (70-90 %) score, with the full-length proteins indicating reduced low-confidence scores of 53.6


330 %, 53.9 % and 52.1 % for TaADNR1, TRIDC5AG050380 and YrU1, respectively (Figure 6A-C, Figure


331 S2, Table S5).


15

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


332 Compared to the full-length protein, the WRKY specific domain exhibited an increased high-confidence


333 residue score, with higher percentage scores ranging between 76.7-81.7 % (Figure 5, Table S5).


334 Contrastingly, the WRKY-only domain of TaADNR1 and TRIDC5AG050380 both exhibited an


335 increased high-confidence percentage (80 and 81 %, respectively) compared to YrU1 (76.7 %) (Table


336 S5). A reduced low-confidence percentage was similar in both TaADNR1 and TRIDC5AG050380 (6.7


337 %) compared to the increased low-confidence percentage in YrU1 (18.3 %). Therefore, both TaADNR1


338 and TRIDC5AG050380 exhibited good low-confidence as well as good high-confidence percentages


339 in the WRKY domain, confirming the high structural prediction for this region. This highlights the high


340 degree of structural similarity and conservation between the two protein domains from the different


341 _Triticum_ species.


342


343 **Figure 5** : The structural alignment of the isolated WRKY domains of the wheat ( _T_ . _aestivum_, _T_ . _urartu_
344 and _T_ . _dicoccoides_ ) ANK-NLR-WRKY proteins


16

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


345 **AlphaFold2 structure predictions, comparison and analysis of TaADNR1 and AtWRKY41 and**


346 **AtWRKy46**


347 The AlphaFold predicted structure comparisons of the TaADNR1 isolated WRKY integrated region as


348 well as the _Arabidopsis_ AtWRKY41 and AtWRKY46 isolated WRKY region exhibited a high degree


349 of structural conservation between the three proteins regardless of the moderate sequence divergence


350 (Figure 6, Table S6). The average pLDDT for TaADNR1, AtWRKY41 and AtWRKY46 were 82.27,


351 93.71 and 92.34, respectively (Table S6). This indicated a high confidence in the prediction of the


352 structure with significantly high confidence for both AtWRKY41 and AtWRKY46. The AtWRKY41


353 protein exhibited a significantly larger high-confidence residue percentage of 88.3% for 53 residues


354 (&gt;90%), while TaADNR1 exhibited a high-confidence residue percentage of 80% for 48 residues (70

355 90%). The TaADNR1 and AtWRKY41 structural superpositions of the isolated WRKY region indicated


356 a near-identical fold with a RMSD of 0.360 Å, while TaADNR1 and AtWRKY46 exhibited a RMSD


357 of 0.378 Å. This observation demonstrated high similarity in the predicted three-dimensional structure


358 of the compared WRKY regions.


359


360 **Figure 6** : The structural alignment of the isolated WRKY domains of the wheat ( _T. aestivum_ ) TaADNR1
361 WRKY domains against the _A. thaliana_ AtWRKY41 and AtWRKY46 proteins. The protein structures
362 are coloured by chain with TaADNR1 WRKY domain coloured blue, AtWRKY41 coloured white and


17

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


363 AtWRKY46 coloured yellow. **A** and **B**, Pairwise alignments of TaADNR1 against AtWRKY41 and
364 AtWRKY46, respectively. **C**, Superposition of all three isolated WRKY domains from the three
365 proteins. The near-identical overlap in **A**, with a RMSD of 0.360 Å highlights the high degree of
366 structural conservation with significant differences between TaADNR1 and AtWRKY41 (RMSD &lt; 0.5
367 Å). This emphasizes that these domain regions adopt strong three-dimensional structural folds
368 regardless of the differences in the amino acid substitutions between the two protein regions.
369


370 **Discussion**


371 This is the first study that provides a comparative structural analysis of three ANK-NLR-WRKY


372 immune receptor proteins across the wheat diploid, tetraploid and hexaploid species. The analysis of


373 both the full-length and the isolated WRKY domain alignments show how these ANK–NLR–WRKY


374 proteins have evolved in wheat and its relatives. The high degree of similarity and conservation of both


375 the ANK and NB-ARC domains in TaADNR1, TRIDC5AG050380 and YrU1 reflects a strong


376 indication that the structure of these domains are essential for functioning of these immune proteins and


377 do not seem to change significantly over the evolutionary course of time (Mosavi et al., 2004; Steele et


378 al., 2019). This further supports that alteration or disruption of these domains will compromise the


379 functioning of the NLR protein as they hold important roles including NLR activation and downstream


380 defence responses (Takken et al., 2006; Tameling et al., 2006; Wang et al., 2020b).


381


382 **Multiple sequence alignments reveal high conservation in the wheat ANK-NLR-WRKYs and**


383 **WRKY specific domains**


384 The small sequence variations that are present in both the LRR and WRKY domains suggest that these


385 domains are important in enabling the toleration of small changes that may contribute to various


386 adaptations for the functioning of the protein. The LRR domain was recently categorised further,


387 consisting of the highly conserved segment (HCS) and the highly variable segment (HVS) (Liu et al.,


388 2022). YrU1 shows variations within the HVS of its LRR domain compared to both TaADNR1 and


389 TRIDC5AG050380. Such variations are attributed to shaping the functional specificity of the NLR in


390 recognising specific pathogens under positive selection pressures. In contrast, these variations allow the


18

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


391 LRR domain to harbour dual functions, including the auto-inhibition of the NLR as well as effector


392 recognition (Anbu et al., 2023).


393 The wheat WRKY-only alignment shows strict conservation of the WRKYGQK motif as well as the


394 zinc-finger-like motif across all three proteins and confirms the preservation of the core DNA-binding


395 functionality of these proteins. These conserved motifs indicate that they all belong to the group III


396 class of WRKY transcription factors, with the WRKY domains consisting of the C2-HC zinc finger


397 motif and a sequence pattern C-X5-8-C-X25-28-H-X1-2-C (Grzechowiak, 2014). Specifically, the


398 conserved pattern observed for the three proteins is C-X7-C-X23-H-X1-C. These motifs are directly


399 involved in the interaction with the W-box elements in target promoters and are crucial for


400 transcriptional regulation of plant defence-related genes (Bakshi and Oelmüller, 2014). In _Arabidopsis_,


401 the acetylation of the lysine residues within the WRKYGQK domain disrupted the functional activity


402 of RRS1 as well as the RPS4/RRS1 complex for W-box DNA interaction (Sarris et al., 2015). Therefore,


403 the evolutionary conservation of the WRKY core and the zinc-finger motif is crucial for maintaining


404 the high affinity DNA-binding activity.


405 Nine out of the ten amino acid substitutions identified in YrU1 compared with both TaADNR1 and


406 TRIDC5AG050380 occur mostly downstream of the WRKY core and involve changes in amino acid


407 physicochemical properties. For example, computational analysis suggests that amino acid substitutions


408 from threonine to isoleucine (Thr14Ile) and threonine to methionine (Thr18Met), in YrU1 to TaADNR1


409 and TRIDC5AG050380, represents non-conservative substitutions that shift from a polar uncharged


410 residue to a non-polar hydrophobic residue. These types of replacements may destabilise a protein’s


411 structure because the introduction of a hydrophobic residue is unfavourable in aqueous environments,


412 while a polar residue could form stable hydrogen bonds with water (Wolynes, 1995). Ultimately, these


413 substitutions may potentially present an increasing flexibility or hydrogen-binding capacity (Abuzaid


414 et al., 2024). Similarly, substitutions Ala36Val and Met15Ile preserves hydrophobicity as they are non

415 polar. However, the loss of the larger sulphur-containing side chain may impact amino acid local


416 packing in TaADNR1 and TRIDC5AG050380. The Leu35Pro substitution replaces a flexible


417 hydrophobic residue with a more rigid amino acid which may potentially influence the local secondary


19

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


418 structure of the domain (Pahuja et al., 2007). Further substitutions such as Tyr19His, His4Tyr and


419 Asn55His, affect the amino acid’s aromatic and charged properties. This has implication in influencing


420 the electrostatic interactions or pH-dependent binding properties. These changes do not necessarily


421 affect the residues that are directly involved in DNA binding and zinc-finger conformations. Indeed, the


422 Alphafold2 modeling supports very limited changes between the WRKY domains of these three


423 proteins. This suggests that the WRKY domain maintains its transcriptional activity and presents


424 regulatory differences or specificity in its interactions (Phukan et al., 2016). Therefore, YrU1 seems to


425 present a diverged ancestral form that is still functional, while TaADNR1 and TRIDC5AG050380 show


426 complete similarity in identity in their WRKY domains. In contrast, YrU1 may display regulatory


427 differences that may contribute to species-specific immune responses, while TaADNR1 and


428 TRIDC5AG050380 may show broad spectrum species recognition in the immune responses (Wang eet


429 al., 2020a). These results indicate that functional activity should be intact. Therefore, the small


430 variations observed in the YrU1 WRKY sequence suggests that they may not negatively affect the


431 overall functioning of the WRKY and may suggest possible differences in gene regulation rather than


432 complete functional divergence.


433 The amino acid sequence alignment between TaADNR1, AtWRKY41 and AtWRKY46 shows


434 conservation of the WRKYGQK core motif, including the zinc-finger-like motif between wheat and


435 _Arabidopsis_ . This observation supports the preservation of the DNA-binding property (Brand et al.,


436 2013; Kim et al., 2024). The substitutions with potential impact on WRKY functioning between


437 TaAdnr1 and AtWRKY41 included His28Phe, Lys32Gln and Gly33Tyr. These substitutions were


438 observed within the zinc-finger core and include changes such as loss of a positive charge (Lys32Gln)


439 and a change from a flexible to a bulkier aromatic amino acid (Gly33Tyr). Furthermore, substitutions


440 outside both the WRKY core and zinc-finger motif included Met18Lys and His55Arg that present


441 introduction of a positive charge, Ser44Asp and Glu57Thr that present local charge creation and


442 removal, and lastly, Ser17Arg and Thr45Gly that represents polar to non-polar changes that can remove


443 hydrogen bonding. Potential impactful substitutions between TaADNR1 and WRKY46 within the


444 WRKY core included Gly33Asp, His28Phe and Lys22Gln. These substitutions may present changes


20

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


445 such as loss of ionisation. Substitutions outside both the WRKY core and zinc-finger motif included


446 Val53Lys and Met18Lys that introduces a positive charge, Phe4His and Ile15His that introduces pH


447 sensitivity, Ser44Asp and Glu57Ans that may present electrostatic changes, and lastly, Thr37Val that


448 represents polar to non-polar change. There was only one potentially impactful amino acid substitution


449 between AtWRKY41 and AtWRKY46 within the zinc-finger motif where an ionisable amino acid


450 (Phe28His) was introduced. The other substitutions observed outside the zinc finger motif included


451 Arg55Leu and Thr53Lys that introduced loss of a positive charge and addition of pH sensitivity,


452 respectively. Furthermore, substitutions in Thr53Lys represented an introduction of a positive charge,


453 while Ser22Ala introduced a direct polar-to-non-polar change, and Phe19Asn and Gly45Thr introduced


454 hydrophobic to polar changes that may possibly alter the hydrogen-bonding surfaces. Therefore, these


455 substitutions are suggested to introduce changes in the local hydration, hydrogen bonding, electrostatic


456 potential or packing, without necessarily affecting the folding of the WRKY domain, DNA binding,


457 effector recognition and function of the NLR protein (Kim et al., 2024; Pandey et al., 2018; Mukhi et


458 al., 2021).


459


460 **Evolutionary relationships of NLR–WRKY proteins across grass species**


461 The integration of WRKY transcription factor domains into plant NLR’s is a significant evolutionary


462 event in plant immunity as well as other important plant processes in response to environmental stimuli


463 (Mukhtar, 2013). In the present study, we further investigated the evolutionary history of the TaADNR1


464 from the hexaploid wheat and the origin of the WRKY domain that was integrated into this ANK-NLR

465 WRKY proteins. The NLR–WRKY phylogenetic tree constructed using NLR’s from different grass


466 species, which consist of the WRKY integrated domain, indicated that TaADNR1, TRIDC5AG050380


467 and YrU1 (from _T. aestivum_, _T. urartu_ and _T. dicoccoides_, respectively) cluster together and form a


468 robust monophyletic clade. This observation supports its origin from a common ancestral gene. The


469 high conservation displayed in the domain architecture in these proteins across the diploid, tetraploid,


470 and hexaploid wheat indicates that the WRKY integration event occurred before polyploidisation of


471 wheat from a distant ancestor and has since been retained. This conservation suggests that the


21

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


472 evolutionary integration of the WRKY domain provides an added advantage in the functioning of the


473 NLR, which is consistent with the proposed integrated decoy model for perception of pathogen derived


474 effector molecules (Césari et al., 2014).


475 Initial studies on TaADNR1 indicated that the most probable WRKY domain donor was the


476 TaWRKY50 transcription factor from group III (Nicolis and Venter, 2018). At the time of this study the


477 _T. urartu_ and _T. dicoccoides_ genomes were not fully available for comparison when the _T. aestivum_ and


478 related grass species’ proteomes were queried. The divergence, 79.7%, over the 69 amino acid WRKY


479 domain raised the question why such a large amount of change took place in such a short timespan


480 (Nicolis and Venter, 2018). The additional information generated through the extended phylogenetic


481 analysis in this study updates this to recognise that the WRKY domain found on TaADNR1,


482 TRIDC5AG050380 and YrU1 originated from a common ancestor and increases our understanding of


483 the evolutionary history of TaADNR1. Similar to Wang et al. (2020), we postulated that the donor of


484 the integrated WRKY domain either contributed to _T. urartu_ or possibly _T. dicoccoides_ . The


485 phylogenetic tree containing WRKY transcription factors from _T_ . _aestivum_, _T_ . _dicoccoides_ and _T_ . _urartu_


486 showed that TaADNR1, TRIDC5AG050380, and YrU1 clustered specifically with WRKYs from _T_ .


487 _urartu_ . Resolving the tree to focus on _T. urartu_ WRKY transcription factors supported a shared an


488 evolutionary origin for the integrated WRKY domains of TaADNR1, TRIDC5AG050380 and YrU1,


489 indicating that TaWRKY41 from _T. urartu_ as the closest extant relative of the donor lineage. This


490 narrows the donor contribution to already have occurred in _T. urartu_ with subsequent contribution to _T._


491 _dicoccoides_ and _T. aestivum_ .


492 The six _T. urartu_ WRKY candidates (TuG1812G0100004483.01, TuG1812G0300002773.01,


493 TuG1812S0003165700.01, TuG1812G0200001696.01) shared high homology to a WRKY41-like


494 transcription factor and one candidate (TuG1812G0500002157.01) showed homology to WRKY46

495 like transcription factor similarity to _Arabidopsis_ RRS1. Their high sequence similarity with the


496 integrated domains and close phylogenetic clustering further supported the evolutionary relationship


497 with the donor lineage. WRKY41 and WRKY46 are both known regulators of defence signalling,


498 salicylic acid-mediated immunity, and stress responses, making them biologically plausible targets for


22

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


499 pathogen effector manipulation (Higashi et al., 2008; Hu et al., 2011; Verk et al., 2011; Ding et al.,


500 2015). Indeed, the AtWRKY41 integrated onto RRS1 recognises AvrRps4 from _Pseudomonas_ _syringae_


501 in _Arabidopsis_ (Sarris et al., 2016). This effector trapping leads to the initiation of a defence response


502 against the invading pathogen, which suggests a similar role for integrated WRKY domain on


503 TaADNR1, YRU1 and TRIDC5AG050380.


504 Collectively, these findings support a model in which a WRKY41-like transcription factor was


505 integrated into an ancestral NLR gene in a diploid wheat progenitor, giving rise to the ANK–NLR–


506 WRKY lineage conserved across wheat evolution. This work advances our understanding of the


507 molecular origins of wheat immune receptors and provides a framework for exploring the functional


508 consequences of WRKY integration in resistance to aphid and fungal pathogens.


509


510 **AlphaFold2 structure predictions of wheat NLR-WRKY proteins and WRKY-specific domains**


511 The AlphaFold2 structure predictions generated high-quality structural models for the three wheat


512 ANK-NLR-WRKY proteins. The full-length models exhibited moderate average pLDDT scores (68.32

513 71.91), a characteristic feature of large, multi-domain intracellular immune receptors which often


514 contain flexible linkers. Shorter proteins generally display higher pLDDT scores for good predictions,


515 and these scores tend to decrease with increasing amino acid length (Ilzhöfer et al., 2022). Prediction


516 confidence, as indicated by pLDDT scores, tends to decline for proteins longer than approximately 1


517 000 amino acids (Abbas et al., 2023). However, protein sequences with approximately 1 500 aa report


518 moderate to good accuracy in the predictions as well as good confidence in the backbone fold. These


519 scores represent the characteristic nature of large multidomain immune receptors that are often known


520 to consist of flexible linkers between multiple domains in a protein (Jumper et al., 2021; Huang et al.,


521 2023; Hetmann et al., 2024). Therefore, for both TaADNR1 and TRIDC5AG050380 with similar amino


522 acid lengths, the pLDDT scores of 71.91 and 71.64 indicate a moderate to high confidence in the


523 accuracy of the predicted structures. The lower pLDDT score in YrU1 may be attributed to slightly


524 longer amino acid length. These observations are consistent with the fact that there is flexibility in the


23

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


525 NLR conformation during activation or inactivation, in the regions where the interacting domains link


526 together (Ohto, 2022).


527 The isolated WRKY domain regions of the proteins were predicted with significantly higher average


528 pLDDT scores (78.48-82.27). The average pLDDT scores for the WRKY domain specific region of all


529 three proteins was high with 82.27, 81.61 and 78.48 for TaADNR1, TRIDC5AG050380 and YrU1,


530 respectively. Although the scores were not above 90, which may be comparable to experimental


531 resolutions, these high scores, especially for TaADNR1 and TRIDC5AG050380, could indicate a


532 reasonable structural prediction. The high-confidence prediction of the WRKY domains supports that


533 its role in DNA binding is conserved and intact for different NLRs with minimal differences that may


534 regulate distinct signalling mechanisms (Sarris et al., 2016; Anbu et al., 2023).


535 The integrated WRKY domain of TaADNR1 exhibits a highly conserved 3D conformation compared


536 with the _Arabidopsis_ WRKY41 and WRKY46 domains. The average pLDDT scores for the WRKY


537 domain specific region of all three proteins was exceptionally high with 82.27, 93.71 and 92.34 for


538 TaADNR1, AtWRKY41 and AtWRKY46, respectively. Notably, both AtWRKY41 and AtWRKY46


539 exhibited high-confidence residue percentages with 88.3% (53 residues at &gt; 90%) and 78.7 for (48


540 residues at &gt; 90%), respectively. These observations indicate that both the AtWRKY41 and AtWRKY46


541 WRKY domain region may be comparable to experimental resolutions. The low RMSD values of 0.360


542 Å for AtWRKY41 and 0.360 Å for AtWRKY46 confirm a highly conserved conformation. Despite the


543 moderate degree of amino acid substitutions between the three proteins, the overall 3D conformation


544 was retained. The AtWRKY41 and AtWRKY46 proteins have been previously reported in multiple


545 studies for their contribution in plant defence against various pathogens (Pandey et al., 2018). The study


546 on the structural analysis and conservation of the WRKY domain of the _Solanum lycopersicum_


547 SlWRKY3 and SlWRKY4 revealed high similarity in the structural superposition, with RMSD values


548 less than 0.25 Å and several amino acid substitutions (Aamir et al., 2017). These observations may


549 suggest the possibility that TaADNR1 and both the _Arabidopsis_ AtWRKY41 and AtWRKY46 may


550 share the functional ability to recognise pathogen or pest effectors during defence signalling, while


551 exhibiting minor amino acid substitutions. Taken together, the TaADNR1 WRKY integrated domain


24

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


552 shares structural conservation across the examined integrated WRKYs from _T. dicoccoides_ and _T._


553 _urartu_, including closely related _A. thaliana_ WRKY proteins that play a role in plant defence.


554 Pairwise superposition of the WRKY domains revealed a high degree of conservation across the three


555 wheat orthologs. The folds of the full-length proteins were highly similar, with all pairwise RMSD


556 values below 0.75 Å .This suggests that these proteins not only have an identical topology but also share


557 similarities in atomic bond lengths, bond angles, and overall spatial arrangements of their atoms. This


558 high degree of overall structural conservation is notable given the evolutionary divergence between the


559 diploid and polyploid wheat species and suggests strong maintenance of function on the overall NLR

560 WRKY protein. This high conservation level further supports conservation of function when


561 considering that the NLR family is the largest family within the wheat genome with extensive


562 diversification of its members (Andersen et al., 2020; Walkowiak et al., 2020; Jumper et al., 2021).


563 A high level of conservation was observed when comparing the isolated WRKY domains in the three


564 proteins. The WRKY domains of TaADNR1 from the hexaploid wheat and its tetraploid progenitor


565 TRIDC5AG050380 were found to be nearly indistinguishable with an RMSD of 0.130 Å. This


566 observation suggests that the 3D fold of this domain has been strictly preserved after the


567 polyploidisation events that gave rise to hexaploid wheat. The extreme structural conservation further


568 validates the strong preservation in the DNA-binding role in the immune signalling mechanism (Aamir


569 et al., 2017). The WRKY domain of the more distant diploid progenitor YrU1 also maintained very high


570 similarity to TaADNR1 (RMSD 0.560 Å) even though this domain is positioned differently in the full

571 length sequence (residues 1429–1488 for YrU1 vs. 1333-1392 for TaADNR1). This level of


572 conservation supports that the WRKY domain is under high evolutionary conservation and


573 maintenance, suggesting that this is a key functional domain that plays a critical role in immune


574 signalling in wheat species (Sarris et al., 2015). Such a strong conservation of the observed domains is


575 remarkable for NLR proteins which are known for evolving rapidly for pathogen recognition (Jacob et


576 al., 2013; Wang et al., 2020b). However, this high level of conservation may result in the functioning


577 of the WRKY domain being highly susceptible to disruption of its structure and mutations in this


25

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


578 domain’s core residues could negatively impact the relay of an immune signal and affect the


579 evolutionary function (Nicolis and Venter, 2018; Wang et al., 2020a).


580


581 **Acknowledgements**


582 This work was funded by public grants from the Winter Cereal Trust (WCT/W/2007/04


583 Induced systemic resistance in wheat).


584


585 **References**


586 Aamir M, Singh VK, Meena M, Upadhyay RS, Gupta VK and Singh, S. Structural and functional


587 insights into WRKY3 and WRKY4 transcription factors to unravel the WRKY–DNA (W-Box) complex


588 interaction in tomato ( _Solanum lycopersicum_ L.). A computational approach. Front Plant Sci. 2017; 8:


589 819–819. https://doi.org/10.3389/fpls.2017.00819


590 Abbas UL, Chen J, Shao Q. Assessing fairness of AlphaFold2 prediction of protein 3D structures.


591 Proceedings of the 14th ACM International Conference on Bioinformatics, Computational Biology, and


592 Health Informatics. ACM. 2023; 1–10. https://doi.org/10.1145/3584371.3612943


593 Abuzaid O, Idris AB, Yılmaz S, Idris EB, Idris LB, Hassan MA. Prediction of the most deleterious non

594 synonymous SNPs in the human IL1B gene: evidence from bioinformatics analyses. BMC Genom Data.


595 2024; 25. https://doi.org/10.1186/s12863-024-01233-x


596 Adachi H, Derevnina L, Kamoun S. NLR singletons, pairs, and networks: evolution, assembly,


597 and regulation of the intracellular immunoreceptor circuitry of plants. Curr Opin Plant Biol.


598 2019; 50: 121–131. https://doi.org/10.1016/j.pbi.2019.04.007


599 Anbu S, Swart V, van der Berg N. Unmasking the invaders: NLR-mal function in plant defense. Front


600 Plant Sci. 2023; 14: 13074. https://doi.org/10.3389/fpls.2023.1307294


26

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


601 Andersen EJ, Nepal M, Purintun JM, Nelson D, Mermigka G, Sarris PF. Wheat disease resistance genes


602 and their diversification through integrated domain fusions', Front Genet. 2020; 11: 898.


603 https://doi.org/10.3389/fgene.2020.00898


604 Bakshi M, Oelmüller R. WRKY transcription factors jack of many trades in plants. Plant Signal Behav.


605 2014; 9: e27700. https://doi.org/10.4161/psb.27700


606 Blum M, Hobbs E, Florentino L, Bateman A. InterProScan 6: a modern large-scale protein function


607 annotation pipeline. Bioinform Adv. 2026; 6: vbag141... https://doi.org/10.1093/bioadv/vbag141


608 Brand LH, Fischer NM, Harter K, Kohlbacher O, Wanke D. Elucidating the evolutionary conserved


609 DNA-binding specificities of WRKY transcription factors by molecular dynamics and in vitro binding


610 assays. Nucl Acids Res. 2013; 41: 9764-9778... https://doi.org/10.1093/nar/gkt732


611 Césari S. Multiple strategies for pathogen perception by plant immune receptors. New Phyt. 2018; 218:


612 17-24... https://doi.org/10.1111/nph.14877


613 Césari S, Bernoux M, Moncuquet P, Kroj T, Dodds PN. A novel conserved mechanism for plant NLR


614 protein pairs: the “integrated decoy” hypothesis. Front Plant Sci. &#91;2014; 5: 606. https://doi.org/10&#93;(https://doi.org/10.3389/fpls.2014.00606)


615 3389/fpls 2014 00606


616 Chen X, Li C, Wang H, Guo Z. WRKY transcription factors: evolution binding and action.


617 Phytopath Res. 2019; 1: 13. https://doi.org/10 1186/s42483-019-0022-x


618 Dangl JL, Horváth D, Staskawicz BJ. Pivoting the plant immune system from dissection to


619 deployment. Sci. 2013; 341: 746–751. https://doi.org/10 1126/science 1236011


620 Ding ZJ, Yan JY, Li CX, Li G, Wu Y, Zheng SJ. Transcription factor WRKY46 modulates the


621 development of _Arabidopsis_ lateral roots in osmotic/salt stress conditions via regulation of ABA


622 signaling and auxin homeostasis. Plant J. 2015; 84: 56–69. https://doi.org/10 1111/tpj 12958


623 Ellis JG. Integrated decoys and effector traps: how to catch a plant pathogen. BMC Biol. 2016;


624 14: 13. https://doi.org/10 1186/s12915-016-0235-8


27

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


625 Eulgem T, Rushton PJ, Robatzek S, Somssich IE. The WRKY superfamily of plant transcription factors


626 Trends Plant Sci. 2000; 5: 199–206... https://doi.org/10.1016/s1360-1385(00)01600-9


627 Finatto T, Viana VE, Woyann LG, Busanello C, da Maia LC, de Oliveira AC. Can WRKY


628 transcription factors help plants to overcome environmental challenges? Genet Mol Biol. 2018;


629 41: 533–544. https://doi.org/10 1590/1678-4685-gmb-2017-0232


630 Grzechowiak M. The role of WRKY transcription factors in plants. BioTechnologia (Pozn). 2014; 95:


631 215–233. https://doi.org/10 5114/bta 2014 54797


632 Gu Q, Liu S, He Z, Meng X, Deng Y. NLRs in plant immunity: structural insights and molecular


633 mechanisms. Crop Des. 2025; 4: 100103–100103. https://doi.org/10 1016/j cropd 2025 100103


634 Hetmann M, Parigger L, Sirelkhatim H, Stern AC, Krassnigg A, Gruber K, et al. Folding the human


635 proteome using BioNeMo: A fused dataset of structural models for machine learning purposes. Sci Data.


636 2024; 11: 591–591. https://doi.org/10 1038/s41597-024-03403-z


637 Higashi K, Ishiga Y, Inagaki Y, Toyoda K, Shiraishi T, Ichinose Y. Modulation of defense signal


638 transduction by flagellin-induced WRKY41 transcription factor in _Arabidopsis thaliana_ . Mol Genet


639 Genom. 2008; 279: 303–312. https://doi.org/10.1007/s00438-007-0315-0


640 Hu YR, Dong Q, Yu D. _Arabidopsis_ WRKY46 coordinates with WRKY70 and WRKY53 in basal


641 resistance against pathogen _Pseudomonas_ _syringae_ . Plant Sci. 2011 185-186: 288–297.


642 https://doi.org/10.1016/j.plantsci.2011.12.003


643 Huang B, Kong L, Wang C, Ju F, Zhang Q, Zhu J, et al. Protein structure prediction: challenges advances


644 and the shift of research paradigms. Genom Proteom Bioinform. 2023; 21: 913–925.


645 https://doi.org/10.1016/j.gpb.2022.11.014


646 Hussain A, Qayyum A, Farooq S, Almutairi SM, Rasheed RA, Qadir M, et al. Pepper immunity


647 against _Ralstonia_ _solanacearum_ is positively regulated by CaWRKY3 through modulation of


28

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


648 different WRKY transcription factors. BMC Plant Biol. 2024; 24: 5.


649 https://doi.org/10.1186/s12870-024-05143-z


650 Ilzhöfer D, Heinzinger M, Rost B. SETH predicts nuances of residue disorder from protein embeddings.


651 Front Bioinform. 2022; 2: 1019597. https://doi.org/10.3389/fbinf.2022.1019597


652 Jacob F, Vernaldi S, Maekawa T. Evolution and conservation of plant NLR functions. Front Immunol.


653 2013; 4: 297–297.. https://doi.org/10.3389/fimmu.2013.00297


654 Jin Y, Liu H, Gu T, Xing L, Han G, Ma P, et al. PM2b a CC-NBS-LRR protein interacts with


655 TaWRKY76-D to regulate powdery mildew resistance in common wheat. Front Plant Sci.


656 2022; 13: 973065. https://doi.org/10.3389/fpls.2022.973065


657 Jumper J, Taki K, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein


658 &#91;structure prediction with AlphaFold. Nature. 2021; 596: 583–589. https://doi.org/10.1038/s41586-021-&#93;(https://doi.org/10.1038/s41586-021-03819-2)


659 03819-2


660 Katoh K, Standley DM. MAFFT Multiple Sequence Alignment Software version 7: improvements in


661 performance and usability. Mol Biol Evol. 2013; 30:772-780. https://doi.org/10.1093/molbev/mst010


662 Kim H, Kim J, Choi DS, Kim MS, Deslandes L, Jayaraman J, et al. Molecular basis for the interference


663 of the _Arabidopsis_ WRKY54-mediated immune response by two sequence-unrelated bacterial effectors.


664 Plant J. 2024; 118: 839–855. https://doi.org/10.1111/tpj.16639


665 Krattinger SG, Keller B. Molecular genetics and evolution of disease resistance in cereals. New Phyt.


666 2016; 212: 320–332. https://doi.org/10.1111/nph.14097


667 Kroj T, Chanclud E, Michel-Romiti C, Grand X, Morel JB. Integration of decoy domains


668 derived from protein targets of pathogen effectors into plant immune receptors is widespread.


669 New Phyt. 2016; 210: 618-626. https://doi.org/10.1111/nph.13869


670 Levy AA, Feldman M. Evolution and origin of bread wheat. Plant Cell. 2022 34: 2549–2567.


671 https://doi.org/10 1093/plcell/koac130


29

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


672 Li J, Huang H, Zhu M, Huang S, Zhang W, Dinesh‐Kumar SP, et al. A plant immune receptor


673 adopts a two-step recognition mechanism to enhance viral effector perception. Mol Plant. 2019;


674 12: 248–262. https://doi.org/10.1016/j.molp.2019.01.005


675 Liu Z, Ren ZY, Yan L, Li F. DeepLRR: An online webserver for leucine-rich-repeat containing protein


676 characterization based on deep learning. Plants. 2022; 11: 136 https://doi.org/10.3390/plants11010136


677 Ma Y, Guo H, Hu L, Martinez PP, Moschou PN, Cevik V, et al. Distinct modes of derepression


678 of an _Arabidopsis_ immune receptor complex by two different bacterial effectors. Proc Natl


679 Acad Sci USA. 2018; 115: 10218–10227. https://doi.org/10.1073/pnas.1811858115


680 Marchal C, Michalopoulou VA, Zou Z, Çevik V, Sarris PF. Show me your ID: NLR immune


681 receptors with integrated domains in plants. Essays Biochem. 2022; 66: 527–539


682 https://doi.org/10.1042/ebc20210084


683 Mosavi LK, Cammett TJ, Desrosiers DC, Peng Z. The ankyrin repeat as molecular architecture for


684 protein recognition. Prot Sci. 2004; 13: 1435–1448. https://doi.org/10.1110/ps.03554604


685 Mukhtar MS. Engineering NLR immune receptors for broad-spectrum disease resistance. Trends Plant


686 Sci. 2013; 18: 469–472. https://doi.org/10.1016/j.tplants.2013.08.005


687 Mukhi N, Brown H, Gorenkin D, Ding P, Bentham AR, Stevenson CE, et al. Perception of structurally


688 distinct effectors by the integrated WRKY domain of a plant immune receptor. Proc Natl Acad Sci USA.


689 2021; 118: e2113996118. https://doi.org/10.1073/pnas.2113996118


690 Nicolis V, Venter E. Silencing of a unique integrated domain nucleotide-binding leucine-rich repeat


691 gene in wheat abolishes _Diuraphis noxia_ resistance. Mol Plant-Micr Int. 2018; 31: 940–950.


692 https://doi.org/10.1094/mpmi-11-17-0262-r


693 Ohto U. Activation and regulation mechanisms of NOD-like receptors based on structural biology. Front


694 Immunol. 2022; 13. https://doi.org/10.3389/fimmu.2022.953530


30

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


695 Pahuja KB, Madhusudhan MS, Adkar BV, Chakrabarti P, Ramakrishnan C, Šali A, et al. Stereochemical


696 criteria for prediction of the effects of proline mutations on protein stability. PLoS Comp Biol. 2007; 3.


697 https://doi.org/10.1371/journal.pcbi.0030241


698 Pandey B, Grover A, Sharma P. Molecular dynamics simulations revealed structural differences among


699 WRKY domain-DNA interaction in barley ( _Hordeum vulgare_ ). BMC Genom. 2018; 19: 132.


700 https://doi.org/10.1186/s12864-018-4506-3


701 Phukan UJ, Jeena GS, Shukla RK. WRKY transcription factors: molecular regulation and stress


702 responses in plants. Front Plant Sci. 2016; 7: 760–760. https://doi.org/10.3389/fpls.2016.00760


703 Roux CL, Huet G, Jauneau A, Camborde L, Trémousaygue D, Kraut A, et al. 2015. A receptor


704 pair with an integrated decoy converts pathogen disabling of transcription factors to immunity. Cell.


705 2016; 161: 1074–1088. https://doi.org/10.1016/j.cell.2015.04.025


706 Sarris PF, Cevik V, Dagdas G, Jones JDG, Krasileva KV. Comparative analysis of plant immune


707 receptor architectures uncovers host proteins likely targeted by pathogens. BMC Biol. 2016; 14.


708 https://doi.org/10.1186/s12915-016-0228-7


709 Sarris PF, Duxbury Z, Huh SU, Ma Y, Segonzac C, Sklenář J, et al. A plant immune receptor detects


710 pathogen effectors that target WRKY transcription factors. Cell. 2015; 161: 1089–1100.


711 https://doi.org/10.1016/j.cell.2015.04 024


712 Shao ZQ, Xue JY, Wu P, Zhang YM, Wu Y, Hang YY, et al. Large-scale analyses of angiosperm


713 nucleotide-binding site-leucine-rich repeat genes reveal three anciently diverged classes with


714 distinct evolutionary patterns. Plant Physiol. 2016; 170: 2095–2109.


715 https://doi.org/10.1104/pp.15.01487


716 Sinha K, Jana S, Pramanik P, Bera B. Selection on synonymous codon usage in soybean


717 ( _Glycine max_ &#91;) WRKY genes. Sci Rep. 2024; 14: 26530. https://doi.org/10.1038/s41598-024-&#93;(https://doi.org/10.1038/s41598-024-77156-5)


718 77156-5


31

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


719 Steele JFC, Hughes RK, Banfield MJ. Structural and biochemical studies of an NB-ARC domain from


720 a plant NLR immune receptor. PLoS ONE. 2019; 14. https://doi.org/10.1371/journal.pone.0221226


721 Steenwyk JL, Buida TJ, Li Y, Shen X-X, Rokas A. ClipKIT: a multiple sequence alignment trimming


722 software for accurate phylogenomic inference. PLoS Biol. 2020; 18: e3001007.


723 https://doi.org/10.1371/journal.pbio.3001007


724 Takken FLW, Albrecht M, Tameling WIL. Resistance proteins: molecular switches of plant defence.


725 Curr Opin Plant Biol. 2006; 9: 383–390. https://doi.org/10.1016/j.pbi.2006.05.009


726 Tameling WIL, Vossen JH, Albrecht M, Lengauer T, Berden JA, Haring MA, et al. Mutations in the


727 NB-ARC domain of I-2 that impair ATP hydrolysis cause autoactivation. Plant Physiol. 2006; 140


728 1233–1245. https://doi.org/10.1104/pp.105.073510


729 Verk MCV, Bol JF, Linthorst HJM, WRKY transcription factors involved in activation of SA


730 biosynthesis genes. BMC Plant Biol. 2011; 11. https://doi.org/10.1186/1471-2229-11-89


731 Villano C, Esposito SD, Amelia V, Garramone R, Alioto D, Zoina A, et al. WRKY genes family


732 study reveals tissue-specific and stress-responsive TFs in wild potato species. Sci Rep. 2020;


733 10: 7196. https://doi.org/10.1038/s41598-020-63823-w


734 Walkowiak S, Gao L, Monat C, Haberer G, Kassa MT, Brinton J, et al. Multiple wheat genomes reveal


735 &#91;global variation in modern breeding. Nature. 2020; 588: 277–283. https://doi.org/10.1038/s41586-020-&#93;(https://doi.org/10.1038/s41586-020-2961-x)


736 2961-x


737 Wang H, Zou S, Li Y. An ankyrin-repeat and WRKY-domain-containing immune receptor confers stripe


738 rust resistance in wheat. Nat Comm. 2020; 11: 1353. https://doi.org/10.1038/s41467-020-15139-6


739 Wang J, Chen T, Han M, Qian L, Li J, Wu M, et al. Plant NLR immune receptor Tm-22 activation


740 requires NB-ARC domain-mediated self-association of CC domain. PLoS Path. 2020; 16.


741 https://doi.org/10 1371/journal.ppat.1008475


32

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


742 Wen F, Wu X, Zhang L, Xiao J, Li T, Jia M. Molecular cloning and characterization of


743 WRKY12 a pathogen induced WRKY transcription factor from _Akebia_ _trifoliata_ . Genes. 2023;


744 14: 1015–1015. https://doi.org/10.3390/genes14051015


745 Wolynes PG. Biomolecular folding in vacuo!!!(?). Proc Natl Acad Sci USA. 1995; 92: 2426–2427.


746 https://doi.org/10.1073/pnas.92.7.2426


747 Zhang W, Gao T, Li P, Tian C, Song A, Jiang J, et al. Chrysanthemum CmWRKY53 negatively


748 regulates the resistance of chrysanthemum to the aphid _Macrosiphoniella_ _sanborni_ . Hort Res.


749 2020; 7. https://doi.org/10.1038/s41438-020-0334-0


750 Zhang Y, Wang L. The WRKY transcription factor superfamily: its origin in eukaryotes and


751 expansion in plants. BMC Evo Biol. 2005; 5: 1. https://doi.org/10.1186/1471-2148-5-1


752 Zhao J, Kang Z. Fighting wheat rusts in China: a look back and into the future. Phytopath Res.


753 2023; 5: 6. https://doi.org/10.1186/s42483-023-00159-z


754


33

Mkhize &amp; Venter


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767528; this version posted November 19, 2026. The copyright holder for this&#93;(https://doi.org/10.64898/2026.11.18.767528)

preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
&#91;perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


755 **Supplementary information**


756 **Table S1** : The amino acid sequence comparison of the WRKY domain region of TaADNR1 and
757 AtWRKY41. The table lists residue substitutions and their physicochemical classification identities.
758 The substitutions are classified based on the changes in polarity and charge between the residues.
759
760 **Table S2** : The amino acid sequence comparison of the WRKY domain region of TaADNR1 and
761 AtWRKY46. The table lists residue substitutions and their physicochemical classification identities.
762 The substitutions are classified based on the changes in polarity and charge between the residues.
763
764 **Table S3** : The amino acid sequence comparison of the WRKY domain region of AtWRKY41 and
765 AtWRKY46. The table lists residue substitutions and their physicochemical classification identities.
766 The substitutions are classified based on the changes in polarity and charge between the residues.
767
768 **Table S4** : The BLAST &#91;™&#93; search annotations of the six identified _T. urartu_ WRKY transcription factors
769
770 **Table S5** : The structural predictions and comparison metrics of the different wheat ANK-NLR-WRKY
771 proteins
772
773 **Table S6** : The structural predictions and comparison metrics of the _T. aestivum_ TaADNR1 protein
774 WRKY-only domain against the _A. thaliana_ AtWRKY41 and AtWRKY46 proteins. The average
775 pLDDT score and high confidence percentages for the three isolated WRKY domains for each protein
776 are indicated, including the RMSD measured in Ångstrom (Å) with reference to the TaADNR1 isolated
777 WRKY domain. The key insight into the overall confidence of the predictions is also indicated.
778


779 **Figure S1** : The multiple sequence alignment of the full-length ANK–NLR–WRKY proteins from _T._
780 _urartu_ (YrU1), _T. dicoccoides_ (TRIDC5AG050380), and _T. aestivum_ (TaADNR1).


781 **Figure S2** : The high- and low-confidence percentages of the full-length and WRKY-only proteins from
782 the AlphaFold2 structure predictions of TaADNR1, TRIDC5AG050380 and YrU1


34

Mkhize &amp; Venter


