Genetic analysis of Aegilops tauschii-derived seedling resistance to leaf rust in synthetic hexaploid wheat

Seedling resistance to leaf rust available in the synthetic hexaploid wheat line Syn137 was characterised by means of cytogenetic and linkage mapping. Monosomic analysis located a single dominant gene for leaf rust resistance on chromosome 5D. Molecular mapping not only confirmed this location but also positioned the gene to the distal part of the long arm of chromosome 5D. A test of allelism showed that the gene, tentatively named LrSyn137, is independent but closely linked to Lr1. It appears that Syn137 is occasionally heterogeneous for Lr1 since the analysis of the Lr1-specific marker RGA567-5 in the genetic mapping population indicated the presence of Lr1. Syn137 represents another source of genetic variation that can be useful for the diversification of leaf rust resistance in wheat cultivars.


Introduction
Leaf rust, caused by the fungus Puccinia triticina (Pt), is a foliar wheat disease of global significance. The most effective, economical, and environmentally sound means of controlling this disease is the deployment of resistant wheat cultivars. To date, formally designated leaf rust resistance genes have been catalogued at 76 loci (Lr1-Lr79;McIntosh et al. 2013McIntosh et al. , 2017Qureshi et al. 2018). The release of cultivars with resistance based upon single major resistance genes leads to the emergence of pathotypes with matching virulence. Hence, the identification of new sources of resistance to leaf rust becomes an ongoing process to maintain resistance diversity in released cultivars. Various strategies for prolonging resistance in commercial cultivars have been proposed. These include the deployment of different combinations of major and/or adult plant resistance genes within single cultivars, referred to as gene pyramiding or gene stacking, or between different plants within the wheat crop such as in agronomically similar cultivar mixtures or genetically related multiline varieties (Burdon et al. 2014).
Synthetic hexaploid wheat lines (2n = 6x = 42, AABBDD sub-genomes) produced as chromosomally doubled hybrids (via colchicine treatment) between Triticum turgidum (2n = 4x = 28, AABB sub-genomes) and Aegilops tauschii (2n = 2x = 14, DD genome) are important genetic resources enabling the direct exploitation of genetic variation present in both the AABB sub-genome progenitors and close relatives and the D genome progenitor of cultivated wheat. Of the currently named leaf rust resistance genes in wheat, five were derived from Ae. tauschii. The cloned leaf rust resistance gene Lr21 (Huang et al. 2003) was first made available in a synthetic line RL5406 (Rowland and Kerber 1974;McIntosh et al. 1995) before being backcrossed in cultivar Thatcher and other genotypes. Introgression of genes Lr22a (Rowland and Kerber 1974) and Lr32 (Kerber 1987) was achieved in a similar way, whereas Lr39 (Raupp et al. 2001) and Lr42 (Cox et al. 1993) were transferred by direct hybridisation with common wheat and embryo rescue from the F 1 hybrids (Gill and Raupp 1987). Previously designated genes Lr40 and Lr41 were shown to be Lr21 and Lr39, respectively, whereas wheat stock WGRC16 reported to have Lr43 carried gene combination Lr21 and Lr39 (Gill et al. 2008). Therefore, these gene designations were deleted from the Catalogue of Gene Symbols for Wheat (McIntosh et al. 2013).
In the present study, we used both traditional and molecular genetic approaches to investigate the genetic basis of resistance to leaf rust in the synthetic hexaploid wheat line Syn137.

Materials and methods
Plant materials and genetic analysis Syn137 (68.111/RGB-U//Ward/3/Ae. tauschii (WX629)), a leaf rust resistant entry in the CIMMYT 1st AB×D Elite Synthetics Programme, was crossed with each of the 21 Chinese Spring (CS) monosomic lines originally developed by, and obtained from, E.R. Sears, University of Missouri, USA. Cytologically confirmed monosomic F 1 plants were grown in the greenhouse to obtain F 2 seeds. The location of genes by monosomic analysis depends on the identification of an abnormal genetic ratio in one cross (the 'critical' cross) in which the resistance gene is located on the non-pairing monosomic chromosome, compared with normal disomic inheritance of the resistance gene in the 20 'non-critical' crosses. To confirm F 2 segregation of the critical cross, F 2:3 lines (34 plants each) of 20 resistant F 2 plants were assessed for response to leaf rust. A total of 93 F 2:3 lines originating from the disomic cross between CS and Syn137 were used to analyse linkage between molecular markers and the resistance gene. Chi-squared tests for goodness of fit were used to test for deviation of observed data from theoretically expected segregation ratios. Chi-squared values were corrected for continuity (http://vassarstats.net/csfit.html). A test of allelism between the gene in Syn137 and Lr1 in the Thatcher derivative RL 6003 involved 54 F 2:3 lines (24 to 30 seedlings each) and deployed rust isolates S12 (avirulent to both resistance genes) and Pt60 (virulent to Lr1 and avirulent to the gene in Syn137). A Chi-squared test of independence using a genetic ratio of 11 (homozygous resistant + segregating 15:1): 2 (segregating 3:1, resistant to both isolates): 2 (segregating 3:1 to one isolate, homozygous susceptible to the other): 1 (homozygous susceptible) was applied to assess linkage of the two resistance genes. A set of 13 single-gene lines with known leaf rust resistance genes was used to compare leaf rust responses to Syn137.

Leaf rust reaction tests
Disease testing was carried out on primary leaves of 10-day old host seedlings according to the method of Felsenstein et al. (1998). The 3-cm-leaf segments were cultured in clear polystyrene boxes on 6 g/l agar and 35 mg/l benzimidazole. The experiments used Pt isolates derived from single spores, most of which were collected in Europe. Pt isolate Race 9 was originally provided by P.L. Dyck, Winnipeg, Canada. Inoculum was produced on the susceptible wheat cultivar Kanzler, collected and dispersed above the exposed leaves in a settling tower at densities of 400-500 spores/cm 2 . Plates with inoculated leaf segments were wrapped in paper towel wetted with distilled water, and then enclosed in black plastic for 24 h. The leaf segment boxes were maintained under continuous light in a growth chamber at 17°C and at 60-80% relative humidity. Disease response was measured 10 days after inoculation and followed the 0-4 infection type (IT) scoring system, in which IT '0' indicated no visible symptoms. IT ';' indicated hypersensitive flecks; IT '1' indicated small uredinia with necrosis; IT '2' indicated small to medium-sized uredinia with green islands and surrounded by necrosis or chlorosis; IT '3' indicated medium to large-sized uredinia with chlorosis; IT '4' indicated large uredinia without chlorosis and IT 'X' indicated heterogeneous ITs, similarly distributed over a given leaf. Plus and minus signs were used to indicate higher and lower response than average for a given IT. Infection types '3' or higher were regarded as compatible (high IT), whereas ITs of '2' or lower were regarded as incompatible (low IT).

Molecular mapping
Bulked segregant analysis (Michelmore et al. 1991) was used to identify microsatellite marker loci from wheat chromosome 5D with linkage to the gene in Syn137. Resistant and susceptible bulks consisted of DNA from eight homozygous resistant and eight homozygous susceptible F 2:3 lines of the CS × Syn137 mapping population. Fifteen individuals of each F 2:3 line were collected for DNA extraction (Huang et al. 2000b). Analysis of microsatellite markers from chromosome 5D was carried out as described in Huang et al. (2000a). Primer information and PCR conditions for the Lr1-specific marker RGA567-5 were taken from Cloutier et al. (2007). Following segregation analysis of leaf rust response, 5 microsatellite markers, and RGA567-5 in the mapping population, a partial linkage map was computed with the program JoinMap 5.0. Map distances were calculated using the Haldane function. Charts of genetic linkage maps were drawn with the computer program MapChart 2.1 (Voorrips 2002).

Leaf rust response
The IT response pattern of Syn137 to 9 Pt isolates was different to those obtained for the 13 reference genotypes (Table 1). The resistance gene in Syn137 was characterised by low ITs to all Pt cultures ranging from '0;' to ';2='.

Monosomic analysis
F 2 populations from the monosomic F 1 hybrids were tested with Pt isolate S12. Segregation in all crosses, except that involving chromosome 5D, corresponded to that expected for 3 resistant: 1 susceptible, indicating a single dominant gene for resistance (Table 2). Segregation in the 5D cross deviated significantly from 3:1 (χ 2 3:1 = 21.49, P < 0.0001, df = 1), with only 1 of 76 seedlings scored as susceptible. In this critical cross, it was expected that the disomics (RR) and monosomics (R−) were resistant, whereas the nullisomics (-) were susceptible. Hence, it is assumed that the susceptible plant was a nullisomic indicating that the resistance gene was located on chromosome 5D. In addition, 20 F 3 lines derived from randomly selected resistant F 2 plants of the critical cross were progeny tested. It has been well documented that the accuracy of individual F 2 plant classification can be established on the basis of progeny testing. Where a single chromosome conferring resistance is involved, F 2 progenies of non-critical crosses segregate 1 resistant: 2 segregating: 1 susceptible, while the progeny of a critical cross should have a reduced number of the latter (McIntosh 1987). In the present result, none of the F 2 progeny showed a 1:2:1 segregation confirming that the resistance gene is located on chromosome 5D.

Genetic mapping
Assuming single-gene segregation, the F 3 population of cross CS × Syn137 displayed distorted segregation when tested with isolates S12 and Race 9 (26 homozygous resistant: 61 heterozygous: 6 susceptible; χ 2 1:2:1 = 17.65, P = 0.0001, df = 2). Based on the results of monosomic analysis, microsatellite markers evenly distributed across chromosome 5D were used for molecular analysis. The five microsatellite marker loci Xbarc177, Xgwm269, Xgwm272, Xgwm565, and Xgwm654 from the long arm of chromosome 5D were polymorphic in bulked segregant analysis. Linkage analysis of phenotypic and molecular data in 93 F 2:3 lines from CS × Syn137 refined location of the resistance gene distal to Xgwm272 (Fig. 1). As the dominant leaf rust resistance gene Lr1 is also known to be located on chromosome 5DL, marker RGA567-5 functional for Lr1 was assayed on the parental lines. Syn137 showed amplification of the RGA567-5 marker fragment, whereas CS was null. Segregation analysis across the population showed that the Lr1-specific marker mapped 5.4 cM proximal to the studied resistance locus (Fig. 1). The new leaf rust resistance gene was temporarily designated LrSyn137. As observed for the studied phenotype, all marker loci deviated significantly from Mendelian expectations (Table 3). While all loci showed a deficiency of CS-the female parent-homozygotes, LrSyn137 showed an excess of heterozygotes, and the molecular marker loci were skewed towards homozygous Syn137 genotypes.

Discussion
Cytogenetic and linkage mapping located a dominant leaf rust resistance gene in Syn137 on chromosome 5D. In the course of determining the identity of the resistance gene on chromosome 5DL, a functional marker for Lr1 was given priority to be assayed on the mapping population. The Lr1-specific marker RGA567-5 was found to map proximal to LrSyn137 indicating distinctiveness of LrSyn137 from Lr1. In addition, we could confirm the close linkage of RGA567-5, and thus Lr1, proximal to microsatellite marker locus Xgwm272 (Ling et al. 2003). Despite a limited number of progeny, a genetic test of allelism between Lr1 and LrSyn137 further supported that the two genes are linked to each other. The study provides a good basis for the development of single nucleotide polymorphism-based kompetitive allele specific PCR markers for marker-assisted selection of LrSyn137. The experiments conducted, however, suggested that Syn137 is heterogeneous for resistance gene Lr1. However, heterogeneity for Lr1 seems to be rare as only the line that was used for establishing the mapping population seemed to have carried Lr1; besides the successful allelic cross, all non-critical monosomic crosses-Pt isolate S12 was avirulent to both LrSyn137 and Lr1showed single-gene segregation. The mapping population showed segregation distortion of all loci on chromosome 5DL. Therefore, it appears that compared to lines possessing only LrSyn137, the line carrying both LrSyn137 and Lr1, additionally carries genes on chromosome 5DL generating a distortion in normal segregation in favour of themselves. Similar to our observations, Faris et al. (1998) and Li et al. (2015) reported distorter loci in Ae. tauschii and common wheat, respectively, located in the same genomic region on chromosome 5DL.  Three formally designated genes were located on chromosome 5D: Lr1, shown to be available in many wheat cultivars (McIntosh et al. 1995) and Ae. tauschii accessions (e.g., Ling et al. 2004), Lr57 from Ae. geniculata (Kuraparthy et al. 2007), and Lr70 from common wheat (Hiebert et al. 2014), of which the latter two were assigned to the short arm of chromosome 5D. Qi et al. (2015) described leaf rust resistance gene LrLB88 on chromosome 5DL that co-segregated with Lr1 but showed a reaction pattern to 13 Chinese Pt pathotypes that was clearly distinct to Lr1. Whether LrSyn137 and LrLB88 are independent genes or LrLB88 is an allele or closely linked to Lr1 needs to be determined in follow-up research. However, satisfactory evidence is presented, similar to powdery mildew (Miranda et al. 2006(Miranda et al. , 2007, that diverse leaf rust resistance genes are located in the terminal region of chromosome 5DL. Synthetic hexaploid wheats were mainly used for the transfer of genes controlling resistance to biotic stress because of their mostly simple inheritance and ease of detection, but they have also emerged as a valuable resource for enhancing tolerance to abiotic stresses, nutritional value, and grain quality attributes (van Ginkel and Ogbonnaya 2007;Li et al. 2018). However, new avenues must be taken for increasing allele diversity and recombination in structured populations to exploit the wealth of information available in synthetic wheats. Therefore, population types such as the multiparent advanced generation intercross (Cavanagh et al. 2008) or multiple synthetic derivatives (Gorafi et al. 2018) will support introgression breeding and accelerate gene discovery.
Acknowledgements Open Access funding provided by Projekt DEAL.
Author contributions VM and SLKH conceived the research. SLKH and FJZ performed cytogenetic mapping and allelism test. VM and MS performed marker analysis and molecular mapping. VM and SLKH drafted the manuscript and MS and FJZ commented on it.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflicts of interest.
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.