NLR immune receptors and diverse types of non-NLR proteins control race-specific resistance in Triticeae

: Recent progress in large-scale sequencing, genomics, and rapid gene isolation techniques has accelerated the identification of race-specific resistance (R) genes and their corresponding avirulence (Avr) genes in wheat, barley, rye, and their wild relatives. Here, we describe the growing repertoire of identified R and Avr genes with special emphasis on novel R gene architectures, revealing that there is a large diversity of proteins encoded by race-specific resistance genes that extends beyond the canonical nucleotide-binding domain leucine-rich repeat proteins. Immune receptors with unique domain architectures controlling race-specific resistance possibly reveal novel aspects on the biology of host-pathogen interactions. We conclude that the polyploid cereal genomes have a large evolutionary potential to generate diverse types of resistance genes. Abstract Recent progress in large-scale sequencing, genomics, and rapid gene isolation techniques has accelerated the identification of race-specific resistance ( R ) genes and their corresponding avirulence ( Avr ) genes in wheat, barley, rye, and their wild relatives. Here, we describe the growing repertoire of identified R and Avr genes with special emphasis on novel R gene architectures, revealing that there is a large diversity of proteins encoded by race-specific resistance genes that extends beyond the canonical nucleotide-binding domain leucine-rich repeat proteins. Immune receptors with unique domain architectures controlling race-specific resistance possibly reveal novel aspects on the biology of host – pathogen interactions. We conclude that the polyploid cereal genomes have a large evolutionary potential to generate diverse types of resistance genes. recognition by the kinase domain, heterocomplex might undergo conformation changes, leading activation of the kinase activity and disease resistance. on the identification of two major yellow rust resis- tance genes encoding proteins containing a zinc-finger BED domain followed by canonical NB-ARC and LRR domains, demonstrating the involvement of ID-NLRs in race-specific resistance in wheat. This paper reports on the identification of the yellow rust resistance gene YrU1 that encodes an NLR protein with ID domains at both protein termini: an N-terminal ankyrin-repeat and a C-terminal WRKY domain, which could be involved in effector recognition. describes Sr60 that encodes a TKP protein to confer resistance against wheat stem rust. Authors show that Sr60 is a partial race-specific resistance gene that, upon pathogen infection, is upre- gulated, which moreover up-regulates several pathogenesis-related the first of a dominant gene-for-gene resistance controlled by a cell surface immune receptor, a WAK pro- tein, that detects the corresponding AvrStb6 leading to a strong resistance response not associated with HR, and completely blocking the infection progression. on the cloning of a race-specific resistance gene to leaf rust encoding a membrane-bound protein with multiple ankyrin domains. Lr14a is structurally similar to non-selective cation chan- nels.The cloning of Lr14a is one of the examples how gene cloning is supported by high-quality genome assemblies. Together with [56] it describes a novel protein family involved in race-specific resistance in wheat, which forms the basis to exploit this type of proteins in resis- tance breeding. the first wheat stem rust Avr gene, which directly recognised by the corresponding NLR receptor. This finding provides tools for molecular surveillance and early detection of virulent races that can assist in pathogen-informed breeding strategies.


Introduction
Genetic analysis of plant disease resistance against adapted pathogens has revealed two distinct, main forms of resistance in response to pathogen infection. Race-specific resistance provides mostly complete resistance to some races of a pathogen species and is controlled by single resistance (R) genes. Resistance only occurs in the presence of an R gene and the corresponding pathogen avirulence (Avr) gene [1]. In contrast, nonerace-specific resistance provides mostly partial, quantitative resistance (QR) to all races of a pathogen species and is independent of specific avirulence genes. QR delays disease development, and several quantitative trait loci act additively to confer resistance [2]. Some of the genes underlying these quantitative trait loci may make major contributions to QR. This is the case for the wheat genes Lr67/Yr46 [3] and Lr34/Yr18/Pm38 [4] that provide partial, QR against leaf rust, stripe rust, stem rust, and powdery mildew or Fhb7 against Fusarium [5]. In addition to race-specific and QR genes, there are susceptibility genes, like the barley Mlo gene, where loss-of-function was found to confer recessively inherited resistance to virtually all the barley powdery mildew isolates [6].
Different types of plant immune receptors recognize pathogen-derived molecules initiating differential defense responses which then converge into common signaling pathways [7]. Pattern recognition receptors, PRRs, (either receptor-like kinases or receptor-like proteins) recognize conserved pathogen-associated (or microbial-associated) molecular patterns, triggering pattern-triggered immunity (PTI) to induce defense reactions against nonadapted pathogens. Residual levels of PTI were proposed to provide basal resistance against adapted pathogens [8,9], which would place PTI within nonerace-specific resistance. Triticeae PRRs remain largely unidentified, although some candidates have been pinpointed [10]. Most race-specific R genes identified encode nucleotide-binding domain leucine-rich repeat (NLR) immune receptors that recognize pathogen strainespecific effectors. The resulting effectortriggered immunity efficiently stops pathogen spread.
Hundreds of R genes have been genetically described in the major cereal crops [11] and remain as an essential pillar for disease resistance breeding. Recent gold standard genomic resources [12e15] alongside innovative gene cloning strategies have greatly facilitated R gene cloning [16], providing tangible opportunities to broaden disease resistance diversity in crops. Here, we focus on recent progress in the molecular identification of race-specific resistance genes in wheat, barley, rye, and wild relatives. These genes include the canonical R genes encoding NLR proteins, but also an increasing number of novel immune receptors with unique domain architectures, possibly revealing novel aspects of hoste pathogen interactions. Importantly, the term resistance gene is not associated with a specific molecular characteristic, but rather based on a phenotypic and genetic characterization (Box 1).

NLR-based race-specific resistance in
Triticeae is highly diversified and polymorphic A growing number of wheat, barley, and rye resistance genes have been molecularly isolated [16,17], including genes from wild relatives that are functional in wheat (Table 1). They are mostly active against fungal diseases but can also confer resistance to aphids [18]. Some rye genes have been cloned from rye chromosomal translocations introgressed into wheat [19e21]. This has also revealed that members of the same NLR gene cluster have evolved into stem rust resistance genes in wheat (Sr33) and rye (Sr50) or powdery mildew resistance genes in barley and wheat (Mla allelic series), raising interesting evolutionary implications on conserved effector function and recognition [22]. For most of the isolated genes, forward genetic screens revealed only one complementation group. This suggests that additional genes are not necessary for race-specific resistance. Alternatively, these genes might be redundant or essential, thereby escaping detection by mutagenesis. Based on the results from the molecular identification of 40 R genes, we estimate that around 85% of wheat racee specific resistance genes encode NLRs. However, this might be an overestimation as molecular identification has focused on genes that are used in breeding, excluding genes in older landraces or nondomesticated wheat, as well as genes with narrower specificity potentially less useful for breeding. To study the molecular interactions of NLRs with pathogen molecules, with interaction specificity sometimes depending on a single amino acid [23], several avirulence genes have been recently isolated (Table 1). However, the identification of avirulence genes lags behind resistance genes.
Recent genomic analysis has revealed the complete sets of NLR coding genes in wheat. In the reference genome of cultivar Chinese Spring, 3400 full-length NLR loci were detected, with 1560 of them expressed and with intact open reading frames [24]. The pan-'NLRome' of wheat, that is the complete diversity of NLR coding genes in the gene pool, was estimated by comparing 10 high-quality genomes from diverse wheat elite varieties. Among the ten genomes studied, only 31e34% of the NLR signatures were found across all genomes, whereas the number of unique NLR signatures ranged from 22 to 192. Furthermore, it was estimated that 10 wheat genomes reveal about 90% of all NLR genes present in the wheat gene pool, which is estimated to consist of 5905 to 7780 unique NLR genes [25]. Thus, the theoretical maximal number of NLR-based, race-specific resistance genes (excluding allelic variants) in wheat is around 7,000, of which less than 10% would have been genetically described until now. However, the wheat pan-'NLRome' could be even larger. The lines selected in the 10þ Wheat Genomes Project cover well the global genetic diversity of elite wheat cultivars, but they possibly do not represent the complete diversity present in landraces and in the diploid and tetraploid genomes of wheat.
Some NLR genes have unique functional aspects only partially understood. For example, both expression as well as resistance phenotype increased with temperature for the Sr21 gene [26]. Moreover, Sr21 expression and resistance were lower in wheat genotypes with a D genome. D genomeebased suppression was also observed for stem rust resistance in the cultivar 'Canthatch'. However, it is not known if the suppressed stem rust resistance gene encodes an NLR. The suppressor on the D genome of 'Canthatch' was identified as a subunit of the mediator complex which is conserved in eukaryotes and regulates gene expression [27]. Suppression of the rye NLR Pm8 by an ortholog in wheat was found to occur at the protein level, a mechanism which might be responsible for the frequent suppression of genes introgressed into bread wheat from diploid or tetraploid wheat relatives [28]. The genetic definition of a resistance gene Resistance genes have been identified by phenotypic and genetic studies for more than hundred years, long before gene cloning was possible. They are widely used in crop breeding programs and remain as important sources of disease resistance. In wheat, several hundred genes are known for resistance to the three rust diseases leaf, stem, and stripe rust as well as powdery mildew. The term 'resistance gene' has been used to describe the polymorphic genetic component that makes the crucial difference between resistant and susceptible genotypes within a species. At the molecular level, the product of a resistance gene frequently recognizes (directly or indirectly) the presence of the pathogen, but this must not necessarily be the case. For example, a nonpolymorphic gene acting upstream of the resistance gene might detect a pathogen molecule, and a downstream component required for signal transduction and defense response induction could be the component whose presence or absence would determine if a genotype is resistant. This downstream component would be then called the resistance gene. It would be polymorphic in the gene pool of the crop species, whereas the upstream component sensing the pathogen, e.g. an NLR protein, would be nonpolymorphic. Therefore, not all R genes necessarily encode typical immune receptors such as NLRs or receptor-like proteins, and race specificity does not depend on a specific protein structure such as NLR. It is mostly unknown how the non-NLR proteins described in this review provide racespecific resistance, i.e. if they are immune receptors, are downstream components in a signaling cascade, or have other molecular functions. It is expected that the study of their functions will reveal novel biological aspects of plant disease resistance. Table 1 List of cloned race-specific resistance genes against fungal pathogens isolated in wheat, barley, rye, and wild relatives and their corresponding cloned Avr genes. h Resistance genes for which molecular identification was carried out based on forward genetic screens, and only one complementation group was found. Barley Rph1 is the exception with additional complementation groups found.

NLR resistance genes with integrated domains (IDs)
In addition to the domains found in typical NLR proteins, some wheat and barley NLRs contain IDs that may be involved in receptor activation or downstream signaling [29]. Some of these chimeric genes are ancient, and they originated before the speciation of grass lineages. For example, rice contains NLRs with zinc-finger BED domains [30,31]. In wheat, two active resistance genes encoding ID-NLRs have recently been cloned. First, the wheat stripe rust resistance genes Yr5, Yr7,a n dYrSP encode proteins with an N-terminal noncanonical zincfinger BED domain [32]. The BED domain replaces the coiled coil domain present in canonical NLR proteins, and it is followed by the NB-ARC and the LRR domains. Mutant analysis shows that this BED domain is critical for resistance and displays a high degree of sequence conservation among the BEDeNLR proteins encoded by the Yr5, Yr7,andYrSP genes, implying the BED domain plays a major role in protein function. Moreover, each gene has a distinct recognition specificity, attributable to the numerous polymorphisms in the C-terminal LRR domain. Therefore, it is assumed that race-specificity in BEDeNLR proteins is controlled similarly to canonical NLR proteins ( Figure 1).
The wheat YrU1 gene encodes an ID-NLR with ID domains at both its C and N termini: an N-terminal ankyrin repeat and a C-terminal WRKY domain [33]. This type of ID-NLR protein is only found in wheatrelated species, and it self-associates in vivo and in planta through the CC and ANK domains. The WRKY domain is a putative transcriptional domain that might be involved in recognition of a stripe rust effector to activate immune response similarly to the Arabidopsis resistance protein complex RPS4/RRS1 [34]( Figure 1).
Tandem kinase proteins (TKPs) can confer both race-specific as well as non-racespecific resistance  Figure 1). Here, it is believed that malfunction of some of Pto response network genes would result in such resistance variation.
TKP proteins contain a kinase domain with serine/threonine specificity with strong homology to Pto and PRRs ( Figure 2). Pto functions together with the NLR protein Prf to confer resistance against bacterial pathogens in tomatoes [47]. It could be that the diverse, genetic backgroundedependent resistance responses by Pto result from the presence/absence of as yet unknown genetic components also involved in TKP-mediated resistance. In Rpg1-mediated resistance, E3 ubiquitin ligase SCF (Skp1-cullin 1-F-box) complex components seem to be involved in resistance function [48]. It is likely that additional genetic components modulate TKP-mediated resistance. Finally, it has been hypothesized that the pseudokinase domain serves as decoy for the effectors, and after interaction, the pseudokinase activates the kinase domain to phosphorylate downstream components resulting in resistance [36]( Figure 2). Evidently, the elucidation of TKP-mediated signaling will require additional work to establish the molecular mechanism underlying this resistance.

Novel types of race-specific resistance genes
The development of nonbiased gene isolation strategies in cereals such as MutChromSeq [59] has resulted in  Lr14a also shows similarity to the human transient receptor potential ankyrin channels that are Ca 2þ -permeable nonselective cation channels (Figure 1). The Pm4 gene encodes a chimeric protein of a serineethreonine kinase and multiple C2-domains and transmembrane regions [61]. Functional analysis of Pm4 revealed that two protein variants resulting from constitutive alternative splicing are needed for resistance and that the two encoded protein variants interact biochemically forming an endoplasmic reticulumeassociated complex ( Figure 1). Pm4 shows homology to Arabidopsis proteins located in plasmodesmata, suggesting the unidentified AvrPm4 effector could be recognized at the plasmodesmata. Both Lr14a and Pm4 will undoubtedly reveal novel molecular mechanisms for achieving race-specific plant immunity.

Future research directions and open questions
In contrast to the well-studied interactions of NLRs with effectors, the molecular analysis of noneNLR-based racespecific resistance is at an early stage. It will be important to isolate the corresponding pathogen avirulence genes, which will also allow to identify host targets and to understand their relationship with resistance genes. AvrStb6 is the only known avirulence gene corresponding to a non-NLR protein [51,52]. It has the typical characteristics of a short, secreted protein with no homologies to known proteins, very similar to avirulence proteins recognized by NLRs. Avr gene identification might rapidly advance for stem rust and powdery mildew resistance genes where there has been progress in the identification of several avirulence genes recognized by NLR immune receptors (Ta b l e 1 )[62e66]. Such work is essential to determine if non-NLR race-specific resistance genes, particularly the group encoding kinase domain proteins, function independently of NLR action or if they are guardees of nonpolymorphic and possibly redundant NLR proteins, similar to the PtoePrf interaction in tomatoes where the Pto kinase is the target of the pathogen effector [47] (Figure 1).
It will also be important to study the resistance phenotypes of single non-NLR genes in defined, susceptible genetic backgrounds, either by backcrossing or by the development of transgenic lines. Furthermore, the relevance of the genetic background for gene function must be studied in detail: for example, the protein encoded by Lr14a confers a unique resistance phenotype, which depends on several modifier genes [67]. The further characterization of modifier genes will give insight into molecular mechanisms of gene function. Finally, all the novel types of resistance proteins must be explored for use in agriculture and for possible improvement by mutational changes. The future isolation of a large number of the genetically described resistance genes in cereals and their corresponding Avr genes will establish the whole interactome of ReAvr proteins and provide the basis for the development of effective and durable strategies to combat cereal diseases.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Multiple wheat genomes reveal global variation in modern breeding. Nature 2020, https://doi.org/10.1038/s41586-020-2961-x. Ten chromosome pseudomolecule genomes of elite wheat varieties are presented in this seminal paper of wheat genomics that will facilitate breeding. It reports a detailed multi-genome-derived NLR protein repertoire that will assist in the cloning of resistance genes. Stem rust resistance in wheat is suppressed by a subunit of the mediator complex. Nat Commun 2020, https://doi.org/ 10.1038/s41467-020-14937-2. This paper describes the genetic basis of resistance suppression by identifying SuSr-D1, a suppressor of wheat stem rust resistance. Authors present how wheat sub-genomes impact on regulatory processes at the transcriptional level, which could help to transfer resistance genes present in close relatives of wheat, frequently suppressed in hexaploid backgrounds. BED-domain-containing immune receptors confer diverse resistance spectra to yellow rust. Nat Plants 2018, 4:662-668. This paper reports on the identification of two major yellow rust resistance genes encoding proteins containing a zinc-finger BED domain followed by canonical NB-ARC and LRR domains, demonstrating the involvement of ID-NLRs in race-specific resistance in wheat.

*
. Wang H, Zou S, Li Y, Lin F, Tang D: An ankyrin-repeat and WRKY-domain-containing immune receptor confers stripe rust resistance in wheat. Nat Commun 2020, https://doi.org/ 10.1038/s41467-020-15139-6. This paper reports on the identification of the yellow rust resistance gene YrU1 that encodes an NLR protein with ID domains at both protein termini: an N-terminal ankyrin-repeat and a C-terminal WRKY domain, which could be involved in effector recognition.   A membrane-bound ankyrin repeat protein confers racespecific leaf rust disease resistance in wheat. Nat Commun 2021, 12:956. This paper reports on the cloning of a race-specific resistance gene to leaf rust encoding a membrane-bound protein with multiple ankyrin domains. Lr14a is structurally similar to non-selective cation channels.The cloning of Lr14a is one of the examples how gene cloning is supported by high-quality genome assemblies. Together with [56] it describes a novel protein family involved in race-specific resistance in wheat, which forms the basis to exploit this type of proteins in resistance breeding. Wheat Pm4 resistance to powdery mildew is controlled by alternative splice variants encoding chimeric proteins. Nat Plants 2021, 7:327-341. This paper describes the cloning of the wheat powdery mildew Pm4 gene encoding a kinase-MCTP protein, a chimeric protein resulting from an MCTP and serine-threonine kinase. The gene undergoes constitutive alternative splicing to generate two isoforms, both required for resistance and contributing equally to resistance. Both isoforms create an ER-associated complex revealing a novel and unique molecular and subcellular basis for race-specific resistance in a major crop. Together with [55] describes a novel protein family involved in race-specific resistance in wheat, which forms the basis to exploit this type of proteins in resistance breeding. , this paper describes the first wheat stem rust Avr gene, which is directly recognised by the corresponding NLR immune receptor. This finding provides tools for molecular surveillance and early detection of virulent races that can assist in pathogen-informed breeding strategies.