Fine Mapping of Wheat Stripe Rust Resistance Gene Yr26 Based on Collinearity of Wheat with Brachypodium distachyon and Rice

The Yr26 gene, conferring resistance to all currently important races of Puccinia striiformis f. sp. tritici (Pst) in China, was previously mapped to wheat chromosome deletion bin C-1BL-6-0.32 with low-density markers. In this study, collinearity of wheat to Brachypodium distachyon and rice was used to develop markers to saturate the chromosomal region containing the Yr26 locus, and a total of 2,341 F2 plants and 551 F2∶3 progenies derived from Avocet S×92R137 were used to develop a fine map of Yr26. Wheat expressed sequence tags (ESTs) located in deletion bin C-1BL-6-0.32 were used to develop sequence tagged site (STS) markers. The EST-STS markers flanking Yr26 were used to identify collinear regions of the rice and B. distachyon genomes. Wheat ESTs with significant similarities in the two collinear regions were selected to develop conserved markers for fine mapping of Yr26. Thirty-one markers were mapped to the Yr26 region, and six of them cosegregated with the resistance gene. Marker orders were highly conserved between rice and B. distachyon, but some rearrangements were observed between rice and wheat. Two flanking markers (CON-4 and CON-12) further narrowed the genomic region containing Yr26 to a 1.92 Mb region in B. distachyon chromosome 3 and a 1.17 Mb region in rice chromosome 10, and two putative resistance gene analogs were identified in the collinear region of B. distachyon. The markers developed in this study provide a potential target site for further map-based cloning of Yr26 and should be useful in marker assisted selection for pyramiding the gene with other resistance genes.


Introduction
Wheat (Triticum aestivum L.) is an important crop and a primary food source for humans. Stripe rust, caused by the fungal pathogen Puccinia striiformis Westend. f. sp. tritici Erikss. (Pst), is an important disease of wheat in China and many other countries. To date, 53 stripe rust resistance genes (Yr1-Yr53) and numerous temporarily designated genes have been reported in wheat (http://wheat.pw. usda.gov/cgi-bin/graingenes). Most of these genes have been mapped on chromosomes and/or specific chromosomal regions, and many of them have been used in wheat breeding programs worldwide. However, with the spread of Pst race CYR32, a large number of known resistance genes are no longer effective in China [1].
Despite considerable progress in the identification and mapping of stripe rust resistance genes, only two adult plant resistance (APR) genes, Yr18 [2] and Yr36 [3], have been cloned. Yr26 has been widely used in wheat breeding programs in China for developing stripe rust resistant cultivars [4,5], varieties with Yr26 are grown on more than 3.4 million hectares in China. As the gene is still effective against the current Pst populations, cloning Yr26 is important for understanding the molecular mechanisms of resistance. The Yr26 gene, which is present in the common wheat line 92R137, was derived from Chinese T. turgidum landrace c80-1 [6]. The gene was previously mapped near the centromere region, putatively on the short arm of wheat chromosome1 B with SSR markers Xgwm11, Xgwm18 and Xgwm413 [6]. A recent study located Yr26 to the deletion bin C-1BL-6-0.32 with molecular markers WE173 and Xbarc181 [7]. The genetic distances between Yr26 and the two closest flanking markers were 1.4 and 4.3 cM, respectively. Although several markers have been mapped to the Yr26 region, the number of the markers is still limited, and more are needed for more efficient marker-assisted selection, fine mapping and map-based cloning of Yr26.
A perception is that fine mapping and map-based cloning in hexaploid wheat (T. aestivum, 2n = 6x = 42, genomic formula AABBDD) faces enormous challenges because of the huge genome size (17 Gb), polyploidy and highly repetitive sequences (.80%) within the genome. This problem can be solved, at least partially, by leveraging the physically mapped wheat ESTs [8,9] and conserved syntenic relationship between wheat and model grass species [10,11,12,13]. Collinearity of chromosome regions between wheat and model species, such as rice and B. distachyon, is well characterized [9,14]. The available whole genomic sequences of rice and B. distachyon provide useful information for developing molecular markers, identifying candidate genes for traits of interest, predicting biological functions of genes and cloning genes. Such comparative genomic approaches have been used in map-based cloning of many wheat genes, including Yr18 [2] and Yr36 [3] for stripe rust resistance, and vernalization response genes Vrn1, Vrn2 and Vrn3 [15,16,17]. A particular challenge to mapbased cloning of Yr26 is its proximity to the centromere and that small recombination distances in such regions may correspond to huge physical distances at the DNA level.
Towards fine mapping and map-based cloning of Yr26, the objective of this study was to saturate the chromosome region containing Yr26 through comparative genomics analysis using genomic sequences of rice and B. distachyon and available wheat ESTs. Markers closely linked to Yr26 should be useful for markerassisted selection and contribute towards map-based cloning of this gene.

Genetic Analysis of Stripe Rust Response
Seedlings of 92R137 were resistant (IT 0;) and those of AvS were susceptible (IT 4) in a seedling test with race CYR32. The F 2 population segregated in 1,747 resistant and 594 susceptible, fitting a 3:1 ratio (x

Development of Yr26-linked EST-STS Markers from Wheat ESTs
Six EST-STS markers (WE173, WE171, WE177, WE201, WE202 and WE210) linked with Yr26 in an F 2 population of 92R1376Yangmai 5 [7] were tested for polymorphism in cross AvS692R137. Only  Figure 1). The codominant markers STS-CD77 and WE173 were detected using both agarose gel and polyacrylamide gel electrophoresis. All ten EST-STS markers (Table 1) were used to genotype the entire F 2 population of 2,341 plants.

Development of Conserved Markers through Comparative Genomics of Wheat with B. distachyon and Rice
To develop more markers for Yr26, 169 wheat ESTs in deletion bin C-1BL-6-0.32 were used to identify their similar genomic sequences in B. distachyon and rice; 126 had significant similarities To accurately characterize the collinearity between the Yr26 region and the genomic regions of B. distachyon and rice, ten sequences corresponding to the mapped wheat ESTs were used as queries to perform a BLAST search against the rice and B. distachyon genome sequences. Orthologs of four wheat ESTs, BQ165938, CD453471, BQ160383 and BE443531, were detected on B. distachyon chromosome 3, and the first three were detected on rice chromosome 10. The other six wheat ESTs, BQ160738, BE493918, BQ169964, CD490549, BE497109 and BF474347, either had significant similarities to sequences on other chromo-somes of B. distachyon and rice, or the scores and E values were not in accordance with the search parameters (Table S1)   collinear regions was displayed using the Artemis Comparison Tool [30]. As shown in Figure 3, most gene orders were conserved, but there were some rearrangements. Genes located in the collinear regions of B. distachyon and rice were used as queries to search for orthologous wheat ESTs in the wheat EST database (http://wheat.pw.usda.gov/GG2/blast.shtml) and the identified wheat ESTs were used to design primers. A total of 358 conserved primers were designed using Conserved Primers 2.0 [26] and used to determine polymorphisms between the parents and bulks. Twenty one conserved markers were found to be polymorphic (Table 1). Among the 21 conserved polymorphic markers (Table 1) most, such as CON-3, CON-6, CON-8 and CON-11, were codominant (Figure. S1). All 21 conserved markers were used to genotype the 43 recombinants between STS-CD28 and STS-BQ33.

High Resolution Map for Yr26 and Collinearity Relationships of Wheat EST Markers with Orthologs in B. distachyon and Rice
A high resolution map for Yr26 in deletion bin C-1BL-6-0.32 ( Figure 4A) was constructed with 31 markers, including the 10 EST-STS markers developed directly from wheat ESTs and 21 conserved markers developed through synteny analysis with B. distachyon and rice ( Figure 4B, C, D; Table 1). The ten EST-STS markers were closely linked to Yr26 with genetic distances ranging from 0.43 to 2.14 cM ( Table 2). The conserved markers, which further greatly saturated the linkage map ( Figure 4B), were found to be closely linked with the Yr26 locus and fell within a genetic interval of 1.16 cM (0.39 and 0.77 cM on two sides of the gene), and six of them, CON-6, CON-7, CON-8, CON-9, CON-10 and CON-11, cosegregated with Yr26. Two conserved markers, CON-4 and CON-12, flanked the Yr26 locus at genetic distances of 0.08 and 0.17 cM ( Figure 4B).
Comparative genomic analysis revealed that 23 and 17 wheat ESTs had similarities on B. distachyon chromosome 3 and rice chromosome 10, respectively (Figure 4C, D; Table 3), again revealing high levels of collinearity of the Yr26 region with B. distachyon chromosome 3 and rice chromosome 10 ( Figure 4B, C, D). The orders of these markers were highly conserved between wheat and B. distachyon, but there was a rearrangement between wheat and rice. The rearrangement was observed between markers CON-5 (CD936328) and CON-4 (CJ883804) (Figure 4B, D). The two most closely linked markers CON-4 (CJ883804) and CON-12 (BJ280972) narrowed the genomic region carrying Yr26 to a 1.92 Mb (Bradi3g28410 -Bradi3g29600) on B. distachyon chromosome 3 and 1.17 Mb (Os10g0470700 -Os10g0489800) on rice chromosome 10. There are 135 and 68 genes in the narrowed collinear regions of B. distachyon and rice, respectively. No typical NBS-LRR resistance gene analog was found in the collinear regions of rice (Os10g0470700 -Os10g0489800) and B. distachyon (Bradi3g28410 -Bradi3g29600). However, Bradi3g28590 was annotated as ''leucine-rich repeat (LRR) protein kinase'', and Bradi3g29120 was annotated as ''protein kinase''. The relationships between the putative LRR and protein kinase genes and Yr26 need to be examined in more detail.

Discussion
Despite the increasing numbers of stripe rust resistance genes identified and deployed in wheat breeding programs, only two have been cloned and characterized [2,3]. In the present study we established a high resolution map of Yr26 using a comparative genomics approach to provide a sound basis for further progress in map-based cloning of this gene.
There is collinearity among wheat chromosome 1B, rice chromosome 10 and B. distachyon chromosome 3 [9,14,31]. In the present study, most of the 169 wheat ESTs in the deletion bin C-1BL-6-0.32 were found to have significant similarities with genes on B. distachyon chromosome 3 and rice chromosome 10, confirming a close syntenic relationship and indicating that the genomic sequences of B. distachyon and rice should be useful for comparative analysis wheat genes. Rice was the first selected grass species for genome sequencing [32,33] and B. distachyon is considered as the best model for wheat at present [34,35]. In the present study, we found a higher number of orthologs between wheat and B. distachyon than between wheat and rice. This is consistent with the relationships among the three species as reported in the above studies.
Nevertheless, many exceptions to collinearity were observed in the comparisons of wheat, rice and B. distachyon due to rearrangements involving gene transposition, duplication, deletion and inversion [13,36,37]. Such anomalies in collinearity complicated the use of model species for genetics. These model grass genomes may not always provide sequence information to assist in identification of candidate gene. In the present study, gene deletions were observed when comparing the orthologous regions of rice and B. distachyon, in the collinear regions of rice (Os10g0462900 -Os10g0524500) and B. distachyon (Bradi3g28070 -Bradi3g31630), 46 rice genes had no orthologs in the corresponding region of the B. distachyon genome, and 121 genes PLOS ONE | www.plosone.org predicted in B. distachyon had no orthologs in the corresponding region of the rice genome. Within the narrowed collinear regions between markers CON-4 and CON-12, only two genes, Bra-di3g28590 and Bradi3g29120 annotated as LRR and protein kinases, were present in the 1.92Mb region (Bradi3g28410 -Bradi3g29600) of B. distachyon, but these were absent in the collinear region of rice (Os10g0470700 -Os10g0489800).
Even if a target gene has no orthlogs in rice and B. distachyon, the flanking genes in those species are sufficiently conserved to provide useful information for developing conserved markers to saturate the target gene region in wheat. With 25 wheat genes found to have orthologs in B. distachyon and rice and six cosegregated markers for Yr26, the present study demonstrates a comparative genomics approach using the B. distachyon and rice sequences is effective for identifying markers for genes in wheat.
High resolution physical maps of wheat chromosomes showed that most disease resistance genes are arranged in clusters and are present mainly in the distal parts of the chromosomes [38]. Resistance genes cloned by map-based cloning, such as leaf rust resistance genes Lr21 [39] and stripe rust resistance gene Yr36 [3], are all distally located. In contrast, the target gene Yr26 in this study maps to deletion bin C-1BL-6-0.32, a region that is near the centromere of chromosome 1B. Because recombination is limited around the centromere regions, with a consequent inflation in physical/genetic distances [40], map-based cloning of Yr26 will be extremely difficult. However, we believe the difficulty can be overcome by integrating comparative genomics with BAC based chromosome walking torward the gene. The cosegregating and closely linked markers identified in the present study will be useful for screening the BAC library and identify BAC clones containing Yr26. Based on the high level of effectiveness and some evidence of race specificity, we hypothesize that Yr26 could be a NBS-LRR type gene. The resistance gene candidates of the NBS-LRR type can be tested for resistance functions using gene silencing, mutation and transformation.
Cloning of the Yr26 gene may contribute to understanding the mechanism of resistance at the molecular level and to a better understanding of this gene and its possible alleles. New markers developed in this study are diagnostic for Yr26 and should facilitate rapid detection of Yr26 (and putative alleles) in wheat cultivars and breeding lines, and therefore, can be used for pyramiding Yr26

Wheat Genotypes and Evaluation of Stripe Rust Reactions
A F 2 population of 2,341 plants and 551 F 3 line progenies with 30-40 plants in each, derived from a cross between susceptible genotype Avocet S (AvS) and resistant line 92R137 (Yr26), were used for genetic analysis and fine mapping of Yr26. For 92 F 2 plants identified as recombinants between markers WE201 and STS-BQ6 flanking Yr26 [7], 30-40 plants in each of their F 2:3 families were tested with Pst race CYR32 to confirm the phenotypes of the corresponding F 2 plants. A total of 41 wheat genotypes were used to validate the molecular markers identified to be linked to the Yr26 locus, including 13 Yr near-isogenic lines (NILs) of Avocet S (AvS), 13 Chinese wheat cultivars with Yr26, 13 wheat genotypes with known Yr genes and 2 genotypes (92R137 and AvS) as positive and negative controls for the Yr26 allele (Table 4).
A predominant Chinese Pst race CYR32, which was avirulent on the AvSYr26 NIL and virulent on AvS, was used to test the F 2 and F 2:3 populations and their parents. Seedlings grown in the greenhouse under controlled conditions were inoculated with fresh urediniospores when second leaves were fully expanded. Inoculated plants were incubated at 9uC and 100% relative humidity for 24 h and then transferred into a greenhouse with 14 h light (22,000 lx) at 17uC and 10 h of darkness at 12uC. Infection types (IT) were scored on a 0-4 scale [18] 15 days after inoculation when stripe rust symptoms were fully developed on the susceptible parent.

DNA Extraction and Bulked Segregant Analysis
Genomic DNA was extracted from F 2 seedlings of cross 92R1376AvS and the wheat genotypes described above using the sodium lauroylsarcosine protocol [19,20]. Based on stripe rust response phenotypes, 10 resistant and 10 susceptible F 2 plants with the same infection types as the resistant (IT 0;) and susceptible (IT 4) parents were selected to establish the resistant (BR) and susceptible (BS) bulks for bulked segregant analysis [21].

Development of EST-STS Markers
Because Yr26 was previously assigned to wheat chromosome deletion bin C-1BL-6-0.32 with six EST-STS markers (WE201, WE202, WE210, WE171, WE173 and WE177) [7], these markers were used to test for polymorphisms between the present parents and bulks. In addition to those markers, 163 new pairs of EST-STS primers were designed from wheat ESTs mapped in the deletion bin (http://www.wheat.pw.usda.gov/index.shtml) using Primer Premier 5 software, and used in the bulked segregant analysis.

Comparative Genomic Analysis and Conserved Marker Development
To develop more markers for Yr26, a comparative genomics approach was used. First, all of the 169 wheat ESTs assigned to deletion bin C-1BL-6-0.32 were used in BLASTn searching to identify collinear regions in the genomes of B. distachyon (http:// www.brachypodium.org/) and rice (http://rice.plantbiology.msu. edu/cgi-bin/gbrowse/rice/). Homologous sequences of B. distachyon and rice were selected using an expected value of 1E 210 and identity 80% as cutoff points. Then ten mapped ESTs sequences were selected to identify collinear regions between the Yr26 region, B. distachyon and rice based on the BLASTn results. The genes of B. distachyon and rice located in the collinear regions were used as queries to search the wheat EST database (http:// wheat.pw.usda.gov/GG2/blast.shtml) using BLASTn. A total of 358 wheat ESTs were identified and used to design conserved markers [22,23,24,25] using Conserved Primers 2.0 software [25].

PCR Amplification and Electrophoresis
PCR was performed in a S1000 Thermal Cycler (BIO-RAD) for each DNA sample in a volume of 15 ml containing 1.0 U Taq DNA polymerase, 1.5 ml of 106 buffer (50 mmol KCl, 10 mmol Tris-HCl, pH 8.3), 2.0 mmol MgCl 2 , 200 mmol of each dNTP, 0.6 mmol of each primer and 50-100 ng of template DNA. The PCR conditions were: denaturation at 94uC for 4 min, followed by 35 cycles of 94uC for 1 min, 55uC for 1 min, 72uC for 1 min and a final extension for 10 min at 72uC. PCR products were separated in 6% denaturing polyacrylamide gels, 8% non-denaturing polyacrylamide gels or 1.5% agarose gels, depending upon the marker, visualized using silver staining [26] for polyacrylamide gels or ethidium bromide for agarose gels and photographed. Table 4. Presence (+) and absence (2) of 11 molecular markers that can distinguish Yr26 from other Yr genes in wheat genotypes.

Statistical Analysis and Genetic Linkage Map
Chi-squared analysis (x 2 ) was used to test agreement of expected and obtained segregation ratios. The genetic distances between markers and the Yr26 locus were calculated with software JOINMAP version 4.0 [27] using the Kosambi mapping function [28] and a LOD score of 3.0 as a threshold. The genetic linkage map was drawn with the software Mapdraw V2.1 [29]. Figure S1 Examples of PCR products amplified with four conserved markers. CON-1 (a), CON-4 (b), CON-6 (c) and CON-7 (d); RP, 92R137; RB, resistant bulk; SP, AVS; SB, susceptible bulk; R, resistant plants; S, susceptible plants; Arrow indicated the polymorphic amplification products. (TIF)