Discovery of the New Leaf Rust Resistance Gene Lr82 in Wheat: Molecular Mapping and Marker Development

Breeding for leaf rust resistance has been successful worldwide and is underpinned by the discovery and characterisation of genetically diverse sources of resistance. An English scientist, Arthur Watkins, collected pre-Green Revolution wheat genotypes from 33 locations worldwide in the early part of the 20th Century and this collection is now referred to as the ‘Watkins Collection’. A common wheat genotype, Aus27352 from Yugoslavia, showed resistance to currently predominating Australian pathotypes of the wheat leaf rust pathogen. We crossed Aus27352 with a leaf rust susceptible wheat selection Avocet S and a recombinant inbred line (RIL) F6 population of 200 lines was generated. Initial screening at F3 generation showed monogenic segregation for seedling response to leaf rust in Aus27352. These results were confirmed by screening the Aus27352/Avocet S RIL population. The underlying locus was temporarily named LrAW2. Bulked segregant analysis using the 90K Infinium SNP array located LrAW2 in the long arm of chromosome 2B. Tests with molecular markers linked to two leaf rust resistance genes, Lr50 and Lr58, previously located in chromosome 2B, indicated the uniqueness of LrAW2 and it was formally designated Lr82. Kompetitive allele-specific polymerase chain reaction assays were developed for Lr82-linked SNPs. KASP_22131 mapped 0.8 cM proximal to Lr82 and KASP_11333 was placed 1.2 cM distal to this locus. KASP_22131 showed 91% polymorphism among a set of 89 Australian wheat cultivars. We recommend the use of KASP_22131 for marker assisted pyramiding of Lr82 in breeding programs following polymorphism check on parents.


Introduction
Leaf rust, caused by Puccinia triticina Eriks. (Pt), is a major disease of wheat (Triticum aestivum L.) in many parts of the world. A recent global survey of the impact of pests and pathogens in wheat rated leaf rust as the most damaging globally [1]. Deployment of host resistance has been and continues to be an effective measure to control rust diseases, including leaf rust [2,3]. Sources of resistance that condition near-complete protection against avirulent pathogen isolates throughout the entire life of the plant are referred to as all stage resistance (ASR), whereas those that provide low to moderate levels of resistance at the post-seedling stages are classified as adult plant resistance (APR). A combination of more than two APR genes is needed to condition commercially acceptable level of resistance [4,5]. Combinations of ASR and APR genes are desirable to achieve durable control of rust pathogens. In the 1980s, two groups of wheat cultivars became popular in north-eastern Australia that were derivatives of cultivars Hartog (Pavon S) and Cook. Hartog carries the resistance gene combination Lr1 and Lr13 and in addition the APR gene Lr46. In contrast, Cook carries ASR gene Lr3a and APR gene Lr34. Leaf rust resistance gene Lr24 was backcrossed into both backgrounds. Lr24 and Sr24 are located on an alien segment and inherit together [6]. Wheat cultivars possessing Lr24 covered approximately 28% and 20% of 1999 and 2000 crop season receivals by the Australian Wheat Board in eastern Australia, respectively [7]. Similarly, backcross derivatives of cultivars Cook and Hartog with leaf rust resistance gene Lr37 covered 33.2% and 21.2% receivals in Queensland and New South Wales, respectively, during the 1999 crop season [7].
Eighty-one leaf rust resistance loci have been formally named [14]. Most of these genes belong to the ASR category and follow the gene-for-gene hypothesis [15,16]. Virulence shifts in Pt populations have reduced the effectiveness of several leaf rust resistance genes [17,18], and in some cases has allowed the recycling of defeated genes due to their resistance to newly evolved/exotic pathotypes and the concurrent decline of older pre-existing pathotypes. For example, Lr23 was ineffective to dominant pre-1984 Pt pathotypes in Australia, and the 1984 exotic introduction 104-(2),3,(6),(7),11 and its derivatives carry partial virulence for this gene (https://www.sydney.edu.au/content/dam/corporate/documents/sydneyinstitute-of-agriculture/research/plant-breeding-and-production/cereal_rust_report_2012 _vol_10_3.pdf (accessed on 20 May 2022)). Virulence for Lr23 has been rare or absent in Australia since 2004 (R.F. Park personal observation). These pathotypic changes stress the need for continuous discovery and characterisation of APR and ASR loci from diverse germplasm (landraces and wheat wild relatives) collections for sustained wheat production [19].
A set of 838 pre-Green Revolution common wheat landrace genotypes, referred to as the 'Watkins Collection' [20], was available in Australia at the Australian Winter Cereal Collection, Tamworth, New South Wales (now Australian Grains Genebank, Horsham, VIC, Australia). These genotypes were screened for leaf rust response at the adult plant growth stages in artificially inoculated leaf rust field nurseries and at seedling growth stages under greenhouse conditions. Entry Aus27352, originally collected from Yugoslavia, showed resistance to the currently predominant Pt pathotypes including 104-1,3,4,6,7,8,10,12+Lr37. This investigation covers mode of inheritance, molecular mapping and identification of markers closely linked with ASR to leaf rust in Aus27352.

Development of a Mapping Population
Aus27352 was crossed with the wheat selection Avocet S (AvS) and an F 2:6 recombinant inbred line (RIL) population was developed using the single head descent method. Briefly, Genes 2022, 13, 964 3 of 9 a single head was harvested from each plant from the F 2 generation onwards and two seeds from each family were planted in the F 3 , F 4 and F 5 generations and a single head was harvested from each family. The whole plant was harvested at the F 6 generation to generate 200 F 2:6 RILs.

Molecular Mapping
Genomic DNA was extracted from parents and the entire RIL population using a modified CTAB method [22]. DNA was quantified with the Nanodrop 1000 (Thermofisher Technologies, Inc., Waltham, MA, USA) and quality of DNA was tested by agarose gel electrophoresis. Equal amounts of DNA were pooled from 40 homozygous resistant and 40 homozygous susceptible RILs to prepare resistant and susceptible DNA bulks. Both DNA bulks, parental lines and an artificial F 1 (DNA from 40 random RILS were mixed in equal quantity) were subjected to genotyping using 90K Infinium SNP array at AgriBio, La Trobe University, Bundoora, VIC, Australia to conduct bulked segregant analysis (BSA). Linked SNPs were converted to Kompetitive Allele Specific PCR (KASP) markers using the software Polymarker (http://www.polymarker.info (accessed on 20 May 2022)). These KASP markers were tested on the entire RIL population following a procedure described by Nsabiyera et al. [23].

Statistical Analysis
Chi-squared (χ 2 ) analyses were performed to check the goodness-of-fit of the observed leaf rust response and marker loci segregation to the expected genetic ratios. Recombination fractions were computed using MapDisto [24]. A genetic linkage map was drawn using MapChart software version 2.3 [25] to show the graphical representation of locus order.

Molecular Mapping of LrAW2
BSA with the 90 K Infinium SNP wheat array was performed. Thirty-nine SNPs from the long arm of chromosome 2B differentiated resistant and susceptible bulks. These SNPs spanned the 761,279,407 to 794,886,621 bp region of the Chinese Spring physical map (IWGSC RefSeq_V2.0). KASP marker assays were designed for these SNPs and were tested on parental lines. Fourteen SNPs that produced clear clusters (Table 3) were genotyped on the entire RIL population. LrAW2 was flanked by KASP_22131 (0.8 cM) on the proximal side (towards centromere) and KASP_11333 (1.2 cM) on the distal side in the long arm of chromosome 2B (Figure 1).

Testing of Flanking Markers for Polymorphism on a Wheat Panel
Markers KASP_22131 and KASP_11333 were assayed on a set of 89 Australian wheat cultivars to assess their roles in marker assisted selection of LrAW2 in wheat breeding programs. KASP_22131 and KASP_11333 amplified G:G and A:A alleles, respectively, in the resistant parental stock Aus27352 and the alternate alleles in the susceptible parent AvS (Table 4). KASP_22131 was polymorphic in 81 cultivars and produced the AvS allele (A:A), whereas KASP_11333 was polymorphic in 64 cultivars with the AvS allele (G:G). The amplification of LrAW2-linked allele of the closer marker KASP_22131 occurred in nine cultivars (Correll, Espada, LRPB Kittyhawk, Orion, Chief CL Plus, Gladius, Impose CL Plus, LRPB Arrow and Wedin). Cultivar Correll may carry LrAW2 or another gene with similar pathogenic specificity based on leaf rust response data against several pathotypes. Kittyhawk lacked this gene and the presence of other effective leaf rust resistance genes in the remaining cultivars did not allow postulation of this locus. Taking into consideration the polymorphism, we recommend that KASP_22131 can be used for pyramiding of LrAW2 with other marker-tagged ASR and APR genes for leaf rust resistance.

Testing of Flanking Markers for Polymorphism on a Wheat Panel
Markers KASP_22131 and KASP_11333 were assayed on a set of 89 Australian wheat cultivars to assess their roles in marker assisted selection of LrAW2 in wheat breeding programs. KASP_22131 and KASP_11333 amplified G:G and A:A alleles, respectively, in the resistant parental stock Aus27352 and the alternate alleles in the susceptible parent AvS (Table 4). KASP_22131 was polymorphic in 81 cultivars and produced the AvS allele (A:A), whereas KASP_11333 was polymorphic in 64 cultivars with the AvS allele (G:G). The amplification of LrAW2-linked allele of the closer marker KASP_22131 occurred in nine cultivars (Correll, Espada, LRPB Kittyhawk, Orion, Chief CL Plus, Gladius, Impose CL Plus, LRPB Arrow and Wedin). Cultivar Correll may carry LrAW2 or another gene with similar pathogenic specificity based on leaf rust response data against several pathotypes. Kittyhawk lacked this gene and the presence of other effective leaf rust resistance genes in the remaining cultivars did not allow postulation of this locus. Taking into consideration the polymorphism, we recommend that KASP_22131 can be used for pyramiding of LrAW2 with other marker-tagged ASR and APR genes for leaf rust resistance.

Discussion
Leaf rust is prevalent in almost all wheat growing areas globally [2] and causes widespread and at times severe damage to wheat crops [1]. Most of the race specific genes for leaf rust resistance have been overcome by the evolution in local Pt populations and/or by exotic introductions in Australia [8,11,12]. These events however shaped gene-based control measures for leaf rust. The Australian wheat industry relied on the Lr1 and Lr13 combination, Lr24 was introduced in the 1980s following the development of white-seeded recombinants [6]. It was then supplemented by Lr37 in the 1990s [26]. The value of the APR genes Lr34 and Lr46 became widely evident in the 21st century. These observations led to the realisation that durable leaf rust control could be achieved by combinations of ASR and APR genes [3,27,28].
This study identified a new leaf rust resistance locus LrAW2 in the long arm of chromosome 2B and it is effective against Pt pathotypes 104-1,3,4,6,7,8,10,12 +Lr37 and 76-3,5,7,9,10,12,13 +Lr37, which along with several derivative pathotypes currently prevail in Australia. The markers flanking LrAW2, KASP_21133 and KASP_11333, are located in the 788 and 790 Mb regions of the physical map of Chinese Spring, respectively (IWGSC_RefSeq_v2.0). There are two known leaf rust resistance genes located in the long arm of chromosome 2B, Lr50 [29] and Lr58 [30]. Lr50 was introgressed from T. timopheevi and Lr58 from Aegilops triuncialis. Lr50 (gwm382) and Lr58 (ncw-Lr58-1) linked markers were tested on the parental lines and 22 RILs from the population. Marker gwm382 is a dominant marker and produces 139 bp and none of the test lines produced 139 bp amplicon. The Lr58-linked marker ncw-Lr58-1 is a co-dominant marker and produced about 400 bp amplicon in lines carrying Lr58 and 250 bp in non-carriers. All test lines produced 250 bp products. Both Lr50 and Lr58 follow dominant inheritance, whereas LrAW2 has recessive inheritance. Based on these results LrAW2 is unlikely to be either of these genes. Hence a permanent gene symbol Lr82 was allocated to LrAW2.
Annotated genes located between the markers flanking LrAW2 were extracted from the IWGSC genome assembly of wheat cv. Chinese Spring v1.0 using the tool Pretzel (https: //plantinformatics.io (accessed on 20 May 2022). Functional annotations for these genes were obtained from the IWGSC RefSeq data repository at INRA (https://urgi.versailles. inra.fr/download/iwgsc/IWGSC_RefSeq_Annotations/v1.0 (accessed on 20 May 2022). Ninety annotated genes (36 high-confidence and 54 low-confidence) were identified between the two KASP markers that flanked Lr82. Of these, two genes (TraesCS2B01G608500 and TraesCS2B01G608800) are predicted to encode TIR-NBS-LRR disease resistance proteins, based on the IWGSC RefSeq gene annotation. These could be candidate genes for Lr82 (Figure 2).
The 'Watkins Collection' has been a rich source of new rust resistance genes that are yet to be deployed in agriculture. Previously leaf rust resistance gene Lr52 was formally named by Canadian workers in a landrace from Iran [31]. Bansal et al. [32] showed close association of Lr52 with a new stripe rust resistance locus Yr47. The Yr47/Lr52 combination is currently being used in Australian and Indian wheat breeding programs (H.S. Bariana personal communication with breeders). Several stripe rust resistance genes have been discovered from this collection [19].
The concept of triple rust resistance is often not addressed holistically, with new cultivars lacking adequate resistance to one or the other of the three rust pathogens [19]. The long arm of chromosome 2B carries several rust resistance genes, two of which that are intriguing from the exotic introduction point of view and can protect Australia against the Puccinia graminis f. sp. tritici pathotype Ug99 and its derivatives are Sr28 [33] and Sr9h [34]. Although both genes are ineffective to extant pathotypes of P. graminis f. sp. tritici in Australia, they would assume importance if one or more of the Ug99 group of pathotypes were to be detected in Australia. Sr9e could be more useful against predominating Australian pathotypes of P. graminis f. sp. tritici. In addition, stem rust resistance genes Sr36 and Sr39 could be useful candidates for pyramiding [35]. Similarly, stripe rust resistance genes Yr5a, Yr5b, Yr43, Yr44, Yr53 and Yr72 are effective against a majority of P. striiformis f. sp.
tritici pathotypes and are located in the long arm of chromosome 2B [36][37][38][39]. Development of recombinants carrying combinations of Lr82 with leaf rust, stem rust and stripe rust resistance genes can lead to achievement of durable triple rust resistance.
Genes 2022, 13, x FOR PEER REVIEW 7 of 10 genes were obtained from the IWGSC RefSeq data repository at INRA (https://urgi.versailles.inra.fr/download/iwgsc/IWGSC_RefSeq_Annotations/v1.0 (accessed on 20 May 2022). Ninety annotated genes (36 high-confidence and 54 low-confidence) were identified between the two KASP markers that flanked Lr82. Of these, two genes (TraesCS2B01G608500 and TraesCS2B01G608800) are predicted to encode TIR-NBS-LRR disease resistance proteins, based on the IWGSC RefSeq gene annotation. These could be candidate genes for Lr82 ( Figure 2). The 'Watkins Collection' has been a rich source of new rust resistance genes that are yet to be deployed in agriculture. Previously leaf rust resistance gene Lr52 was formally named by Canadian workers in a landrace from Iran [31]. Bansal et al. [32] showed close association of Lr52 with a new stripe rust resistance locus Yr47. The Yr47/Lr52 combination is currently being used in Australian and Indian wheat breeding programs (H.S. Bariana personal communication with breeders). Several stripe rust resistance genes have been discovered from this collection [19].
The concept of triple rust resistance is often not addressed holistically, with new cultivars lacking adequate resistance to one or the other of the three rust pathogens [19]. The long arm of chromosome 2B carries several rust resistance genes, two of which that are intriguing from the exotic introduction point of view and can protect Australia against the Puccinia graminis f. sp. tritici pathotype Ug99 and its derivatives are Sr28 [33] and Sr9h [34]. Although both genes are ineffective to extant pathotypes of P. graminis f. sp. tritici in Australia, they would assume importance if one or more of the Ug99 group of pathotypes were to be detected in Australia. Sr9e could be more useful against predominating Australian pathotypes of P. graminis f. sp. tritici. In addition, stem rust resistance genes Sr36 and Sr39 could be useful candidates for pyramiding [35]. Similarly, stripe rust resistance genes Yr5a, Yr5b, Yr43, Yr44, Yr53 and Yr72 are effective against a majority of P. striiformis Markers have been developed for many ASR genes conditioning resistance to leaf rust; Lr23 [40], Lr24 [41], Lr42 [42], Lr52 [21], Lr53 [43], Lr57 [44], Lr76 [44], Lr80 [14] and APR genes Lr34 [45], Lr48 [23], Lr49 [46], Lr67 [47] and Lr68 [48]. These ASR and APR genes can be pyramided in different combinations to combat evolution in Pt populations. In particular, the combination of Lr82 with Lr23 and Lr24 and at least one APR gene would contribute towards long-lasting control of this disease in Australia.

Conclusions
This study identified and characterised a new leaf rust resistance gene, Lr82, in the common wheat Yugoslavian landrace Aus27352. Lr82 was mapped 0.8 cM distal to KASP_22131 in the long arm of chromosome 2B. KASP_22131 amplified the allele alternate to that linked with Lr82 in 91% of 89 Australian wheat cultivars. These results support the implementation of KASP_22131 in marker assisted pyramiding of Lr82 with other marker tagged rust resistance genes to achieve durable triple rust control in new wheat cultivars.