Fine mapping of a stripe rust resistance gene YrZM175 in bread wheat

A stripe rust resistance gene YrZM175 in Chinese wheat cultivar Zhongmai 175 was mapped to a genomic interval of 636.4 kb on chromosome arm 2AL, and a candidate gene was predicted. Stripe rust, caused by Puccinia striiformis f. sp. tritici (PST), is a worldwide wheat disease that causes large losses in production. Fine mapping and cloning of resistance genes are important for accurate marker-assisted breeding. Here, we report the fine mapping and candidate gene analysis of stripe rust resistance gene YrZM175 in a Chinese wheat cultivar Zhongmai 175. Fifteen F1, 7,325 F2 plants and 117 F2:3 lines derived from cross Avocet S/Zhongmai 175 were inoculated with PST race CYR32 at the seedling stage in a greenhouse, and F2:3 lines were also evaluated for stripe rust reaction in the field using mixed PST races. Bulked segregant RNA-seq (BSR-seq) analyses revealed 13 SNPs in the region 762.50–768.52 Mb on chromosome arm 2AL. By genome mining, we identified SNPs and InDels between the parents and contrasting bulks and mapped YrZM175 to a 0.72-cM, 636.4-kb interval spanned by YrZM175-InD1 and YrZM175-InD2 (763,452,916–764,089,317 bp) including two putative disease resistance genes based on IWGSC RefSeq v1.0. Collinearity analysis indicated similar target genomic intervals in Chinese Spring, Aegilops tauschii (2D: 647.7–650.5 Mb), Triticum urartu (2A: 750.7–752.3 Mb), Triticum dicoccoides (2A: 771.0–774.5 Mb), Triticum turgidum (2B: 784.7–788.2 Mb), and Triticum aestivum cv. Aikang 58 (2A: 776.3–778.9 Mb) and Jagger (2A: 789.3–791.7 Mb). Through collinearity analysis, sequence alignments of resistant and susceptible parents and gene expression level analysis, we predicted TRITD2Bv1G264480 from Triticum turgidum to be a candidate gene for map-based cloning of YrZM175. A gene-specific marker for TRITD2Bv1G264480 co-segregated with the resistance gene. Molecular marker analysis and stripe rust response data revealed that YrZM175 was different from genes Yr1, Yr17, Yr32, and YrJ22 located on chromosome 2A. Fine mapping of YrZM175 lays a solid foundation for functional gene analysis and marker-assisted selection for improved stripe rust resistance in wheat.

1 3 analysis and marker-assisted selection for improved stripe rust resistance in wheat.

Introduction
Wheat (Triticum aestivum L.) is a staple food crop worldwide, providing substantial amounts of nutrients for humans (Shewry 2009;Shewry and Hey 2015). The estimated global wheat production in 2021 was 770 million tonnes (FAO, http:// www. fao. org/ world foods ituat ion/ en/). Wheat stripe rust (or yellow rust), caused by Puccinia striiformis f. sp. tritici (PST), is a widely distributed disease that causes millions of tonnes of yield losses every year (Chen and Kang 2017). Stripe rust occurred in 1.78 million ha of wheat in China during 2011-2016 and has since increased to about 5 million ha annually (Huang et al. 2020). Resistant cultivars are the most environment-friendly approach to control stripe rust (Xia et al. 2007). Although many resistance genes have been identified, relatively few of them, such as Yr5, Yr15, and Yr18, continue to confer resistance in China. Therefore, it is necessary to mine new stripe rust resistance genes and to deploy them in wheat cultivars, preferably in multiple gene combinations.
Wheat cultivar Zhongmai 175 is widely grown in the North Winter Wheat Zone of China; it has high yield and good level of resistance to stripe rust. It also shows moderate to high resistance in Sichuan and Gansu provinces, the hotspots for stripe rust epidemics and year-round survival in China. One of the parents of Zhongmai 175 was the French line VPM1 which had high resistance to powdery mildew, stem rust, leaf rust and stripe rust, and was widely used as a disease resistant germplasm by breeders. Zhongmai 175 carries Yr17 from VPM1 and a putatively new stripe rust resistance gene YrZM175 that was located in chromosome 2A . The aims of this study were to fine-map YrZM175, predict a candidate gene for further analysis, and to develop gene-specific markers for research and breeding.

Plant materials
A total of 7,325 F 2 plants derived from the cross of Avocet S/Zhongmai 175 were used for fine mapping of YrZM175. One hundred and seventeen random F 2:3 lines were evaluated for stripe rust reaction in the greenhouse and field to confirm the results from the F 2 population.

Evaluation of stripe rust reaction
Fifteen F 1 , 7,325 F 2 plants and 117 F 2:3 lines of Avocet S/ Zhongmai 175 and the parents were inoculated at the seedling stage in controlled greenhouse conditions with PST race CYR32, kindly provided by Dr. Gangming Zhan (Northwest A & F University, Yangling, Shaanxi). CYR32 is avirulent to Zhongmai 175 and virulent to Avocet S. Ten to 15 seeds were planted in a 9 × 9 × 9 cm plastic pot, along with three seeds of susceptible cultivar Mingxian 169 as control. Seedlings were inoculated with CYR32 at 1.5-leaf stage (7-8 days old) by brushing fresh urediniospores from fully infected Mingxian 169 leaves. Inoculated plants were kept in plastic boxes with 100% humidity at 10 °C in darkness for 24 h, then transferred to 15 ± 2 °C conditions with 16 h of light (20,000 lx) daily. Infection types (ITs) were scored 14-18 days after inoculation based on a 0-4 scale (0-2 as resistant, 3-4 as susceptible) (Bariana and McIntosh 1993).
Fifteen F 1 plants and 117 F 2:3 lines along with both parents with 30-40 seedlings each were also evaluated for stripe rust reaction at Pidu Experimental Station, Sichuan Academy of Agricultural Sciences in Chengdu, in the 2017-2018 and 2018-2019 cropping seasons. F 2:3 lines were sown in two replications. Avocet S was used as a susceptible control and spreader and inoculated with mixed PST races (CYR32, CYR33, and CYR34). ITs and disease severity (DS) were recorded twice at 4-to 5-day intervals when the DS of Avocet S reached 90-100%. DS was scored as the percentage of infected leaf area. ITs were scored similarly to those at seedling stage mentioned above.

DNA and RNA extraction
Leaf samples of plants were used to extract genomic DNA using the CTAB method (Saghai-Maroof et al. 1984). For F 2:3 lines, equal amounts of leaf tissue from 8-10 plants 1 3 of each line were mixed for DNA extraction. Because of large genome differences between Zhongmai 175 and Avocet S and the absence of some genes in the target region in Avocet S, we later chose Mingxian 169 as a susceptible control for RNA isolation. Fresh leaf tissues of Zhongmai 175 and Mingxian 169 inoculated with CYR32 were collected before inoculation (0 h) and at every two hours in 12 h post-inoculation (hpi), and at 10 a.m daily in 1 to 14 days post-inoculation (dpi) with three biological replications for RNA extraction using the Plant RNA Kit (Omega Biotek, Inc. Norcross, Georgia USA).

BSR-seq assay
Based on phenotypic evaluations at 14 dpi, 50 resistant (IT = 0) and 50 susceptible plants (IT = 4) were separately mixed to form resistant (R) and susceptible (S) pools for RNA extraction and sequencing. The pooled leaf samples and parents were sent to Novogene (Tianjin, China) for RNA-seq. After quality testing, single RNA libraries were constructed for each sample and sequenced in an Illumina HiSeq4000 platform, generating 150 bp paired-end reads. BWA (Burrows-Wheeler Aligner) (Li et al. 2009) was used to align the clean reads of each sample against the reference genome of Chinese Spring, Ae. tauschii ssp. strangulata accession AL8/78 , T. urartu accession G1812 (Ling et al. 2018), T. turgidum ssp. dicoccoides accession Zavitan (Avni et al. 2017), T. turgidum ssp. durum cv. Svevo (Maccaferri et al. 2019), and T. aestivum cv. Aikang 58 and Jagger (Walkowiak et al. 2020) (https:// www. ncbi. nlm. nih. gov/ assem bly/? term= triti ceae), respectively, after a series of quality control (QC) procedures. SNPs/InDels calling between the two pools was performed using the Unified Genotyper function in GATK (McKenna et al. 2010). The homozygous SNPs/InDels were extracted to calculate the SNP/InDel index. The sliding window method was used to present the SNP/InDel index of polymorphic sites in whole genome. The differences in SNP/InDel indices between the two pools were calculated as the delta SNP/ InDel index to determine the target region of the resistance gene. SNPs/InDels with delta SNP/InDel index of 1 or -1 are considered highly reliable.

Marker development through BSR-seq and genome mining
Highly reliable SNPs from BSR-seq analyses were converted into cleaved amplified polymorphic sequence (CAPS) markers following Thiel et al. (2004) based on their 700-1,000 bp flanking sequences identified in the Chinese Spring reference genome RefSeq v1.0 (IWGSC 2018) and the six other genome sequences mentioned above. Chromosome-specific primers for each SNP were designed with DNAMAN 8.0 software. Due to the limited number of SNPs from BSRseq, more variations between the parents were identified from gene sequences in the target region in reference to the RefSeq v1.0 (IWGSC 2018). According to gene annotations from RefSeq v1.0 (IWGSC 2018) and the six other genome sequences, gene-specific primers were designed from putative disease resistance genes in the target scaffolds to isolate gene sequences in the parents. PCR sequencing results for the parents were aligned to identify the polymorphic SNPs/ InDels using Geneious Prime 2019.0.3 software. InDels between the parents were converted to sequence-tagged site (STS) markers, whereas CAPS markers were developed from SNPs.

Genetic map construction
Chi-squared (χ 2 ) tests were conducted to determine the goodness-of-fit of the observed and expected ratios of segregation in the F 2 and F 2:3 populations using SAS 9.2 (SAS Institute, Cary, NC, USA). All susceptible F 2 plants were tested by molecular markers in the target region for linkage analysis with the resistance gene. Genetic distances between polymorphic SNP, STS and SSR markers linked to the stripe rust resistance gene were calculated using the Kosambi mapping function (Kosambi 1943) computed by Mapmaker 3.0 with an LOD threshold of 3.0 (Lincoln et al. 1993). The linkage map was graphically drawn with MapDraw V2.1 (Liu and Meng 2003).

Collinearity analysis of homologous sequences
Collinear alignment of genes and homoeologous intervals was performed using TGT website (http:// wheat. cau. edu. cn/ TGT/) by blasting markers flanking YrZM175 to obtain physical positions of homologous sequences in related species in Triticum multi-omics center databases on WheatOmics 1.0 (http:// wheat omics. sdau. edu. cn/) and to confirm corresponding homoeologous intervals. Gene annotations within homoeologous intervals were searched and downloaded in EnsemblPlants (http:// plants. ensem bl. org/ index. html), and related disease resistance genes were checked.

Genome walking to get sequences of candidate genes
Genome walking is an effective technique to obtain unknown sequences adjacent to known sequences. To find the physical position of the start point with different sequences in the target region between Zhongmai 175 and Chinese Spring to perform genome walking, we designed chromosome-specific primers at 763.83 Mb upstream of TraesCS2A02G562800 in Chinese Spring RefSeq v1.0 to perform PCR with Zhongmai 175 and Chinese Spring as templates. If sequences of PCR products were common between Zhongmai 175 and Chinese Spring we continued to design primers to amplify downstream sequences from 763.83 Mb until sequences of PCR products showed polymorphisms between Zhongmai 175 and Chinese Spring. Then the position of the polymorphic sequence was set as the start point for genome walking using the Genome Walking Kit (TaKaRa, Dalian). Walking primer designing and PCR amplifications were performed following the manufacturer's protocol. Purified PCR products were ligated with pEASY-T5 Blunt-Zero cloning vectors at 25 °C for 2 h and transformed into Trans1-T1 competent cells by the heat shock (TransGen Biotech Co., Ltd., Beijing). After coating the plate and incubating at 37 °C for 24 h, 6-10 positive clones from each PCR product were randomly selected and sequenced by Shanghai Sangon Biotech Co., Ltd (http:// www. sangon. com/).

Quantitative real-time PCR
First-strand cDNA was synthesized using the PrimeScript RT Reagent Kit with gDNA Eraser (TaKaRa, Kyoto, Japan). The cDNA products of Zhongmai 175 and Mingxian 169 were used for quantitative real-time PCR (qPCR) assays in a BioRad CFX system with the iTaq Universal SYBR Green Supermix (BioRad). Gene-specific primers for candidate genes and the ACTIN gene as internal control were designed using DNAMAN 8.0 software. Each sample had three biological replicates. Relative gene expression was calculated using the 2 -∆∆Ct equation (Schmittgen and Livak 2008).

Fine mapping of YrZM175
BSR-seq analysis revealed 13 highly reliable SNPs in the genomic region 762.50-768.52 Mb on chromosome arm 2AL based on the Chinese Spring RefSeq v1.0 (IWGSC 2018) (Figs. 1A and S2). We initially screened 35 simple sequence repeat (SSR) markers on chromosome 2A and identified nine markers polymorphic between the parents; among them, Xwmc658, Xgwm382, and Xgwm311 on 2AL were linked to YrZM175 with genetic distances of 6.51 cM, 5.12 cM, and 5.12 cM, respectively (Figs. 1B and S3). The physical positions of Xwmc658 and Xgwm382 (Xgwm311) are at 771.16 Mb and 772.96 Mb, respectively, in agreement with the results of BSR-seq. The 13 polymorphic SNPs were mainly concentrated in two physical positions, i.e., four SNPs at 762.50-762.64 Mb and six around 768.52 Mb on 2AL (Table S1). We developed CAPS markers YrZM175-CAPS1 and YrZM175-CAPS2 (Table S2), based on a SNP at 762.50 Mb and another at 768.52 Mb, with genetic distances of 0.81 cM and 2.57 cM from YrZM175, respectively.
Many polymorphic SNPs/InDels between parents were identified by sequencing genes in the target region and were converted to CAPS or STS markers (Table S2). YrZM175 was then mapped in a genetic interval of 1.41 cM between YrZM175-CAPS1 and YrZM175-InD1, spanning a 1.58 Mb (762.50-764.08 Mb) genomic region containing 35 highconfidence genes, including six putative disease resistance genes (Table S3) based on IWGSC RefSeq v1.0. After designing gene-specific PCR primers (Table S4) and sequencing six putative resistance genes in Zhongmai 175, Avocet S and Mingxian 169, TraesCS2A02G560600 showed . There were nine high-confidence genes in this genomic region based on IWGSC RefSeq v1.0, including putative disease resistance genes TraesC-S2A02G562800 and TraesCS2A02G563200 (Table S5).

Analysis of genes in the 636.4 kb interval of Chinese Spring
The NBS-LRR gene TraesCS2A02G563200 showed no difference between Zhongmai 175, Avocet S and Mingxian 169, indicating that it was not a candidate. No PCR product was obtained following amplification of kinase gene TraesCS2A02G562800 in all three genotypes, exhibiting that this gene also was not a candidate. The genome sequence was quite different or absent in the three cultivars compared with Chinese Spring (Fig. 2B) Sequencing following amplification of a 3,000-bp promoter sequence upstream of the ATG start codon by genomic-specific primers (Table S7) referring to Svevo revealed three SNPs, single 1-bp and 11-bp insertions, and single 9-bp, 11-bp and 23-bp deletions in Avocet S compared with Zhongmai 175 (Fig. S4). Annotation of the promoter sequence of Zhongmai 175 on database PLACE (https:// www. dna. affrc. go. jp/ PLACE/? action= newpl ace) indicated that the SNPs/InDels could change several important cis-acting regulators such as ABA responsive element (ABRE), carbohydrate metabolite signal responsive element 1 (CMSRE-1) and salicylic acid regulatory element GT-1 motif. However, tests of all susceptible F 2 plants with a gene-specific marker developed from a polymorphic site between Zhongmai 175 and Avocet S in the promoter, and the marker located 0.21 cM from the resistance gene failed to show an association. We concluded that TRITD2Bv1G265970 was not the candidate of YrZM175.

Collinearity analysis of the target genomic interval
Based on Triticum genome databases (Ae. tauschii, T. urartu, T. dicoccoides, T. turgidum and 10 + Genomes) at TGT website, we performed collinearity analysis of the closest marker YrZM175-InD2 and the targeted genomic region according to the genome sequence of Chinese Spring RefSeq v1.0 (IWGSC 2018). The 636.4-kb mapping interval showed a good collinearity with 10 + Genome varieties, whereas

Analysis of genes in the collinear interval of other wheat varieties and related species
Based on BSR-seq analysis using genome sequence information from five species (Ae. tauschii, T. urartu, T. dicoccoides, T. turgidum, and T. aestivum cv. Aikang 58 and Jagger), there was an obvious single peak on homoeologous group 2 chromosomes in all species (Fig. S7). Based on two polymorphic SNPs between R and S pools from BSR-seq analysis in the collinear interval on chromosome 2AL referring T. dicoccoides and T. turgidum genome sequences, respectively, molecular markers were developed and tested on susceptible F 2 plants from cross Avocet S/ Zhongmai 175, and the linkage distances of two SNPs with YrZM175 were 0.12 cM. The results confirmed sequence similarity between Chinese Spring, T. dicoccoides and T. turgidum. Based on collinearity and homology analysis of genes in the collinear interval at website TGT, there were ten putative disease resistance genes in the region upstream of marker YrZM175-InD2 on chromosomes 2A or 2D based on four Triticum reference genomes (T. turgidum, T. dicoccoides, T. urartu, and Ae. tauschii), whereas these genes were absent in Chinese Spring (Fig. 3). Amplifying and sequencing all ten genes (TRITD2Av1G295090, TRIT-D2Av1G295160, TRITD2Av1G295250, TRIDC2AG082340, T R I D C 2 A G 0 8 2 3 6 0 , T R I D C 2 A G 0 8 2 4 3 0 , TuG1812G0200006348, TuG1812G0200006350, AET2Gv21292200, AET2Gv21292500) in Zhongmai 175 and Avocet S indicated that these genes were also not present in Zhongmai 175 and Avocet S. Therefore, we inferred that Zhongmai 175 does not contain these gene sequences on chromosome arm 2AL.
A putative disease-resistance-like gene isolated in the target interval of chromosome arm 2AL in Zhongmai 175 by genome walking was almost the same (with two SNPs in the intron) as a kinase gene TRITD2Bv1G265970 on chromosome arm 2BL in T. turgidum. This indicated that chromosome arm 2AL in Zhongmai 175 might carry genes originating from chromosome arm 2BL in T. turgidum. Analysis of 11 putative disease resistance genes (Table S8) in the collinear interval of 2BL in T. turgidum in Zhongmai 175 and Mingxian 169 detected differential expression of TRITD2Bv1G264470 and TRITD2Bv1G264480. The expression levels of TRITD2Bv1G264470 were significantly higher at 5-14 dpi than at 0 dpi in both cultivars, with higher expression in Mingxian 169 (Fig. 4A). The expression of TRITD2Bv1G264480 was first observed in Zhongmai 175 at 2 hpi, then significantly increased at 1-14 dpi, with constantly higher expression level than in Mingxian 169 (Fig. 4B). TRITD2Bv1G264470 and TRITD2Bv1G264480 are NBS-LRR genes. Amplifying parental lines with gene-specific primers (Table S9) indicated no sequence difference at CDS domain of TRITD-2Bv1G264470 between Zhongmai 175 and Mingxian 169 or Avocet S. The ORF sequence of TRITD2Bv1G264480 was 23,559 bp with a 5,661-bp CDS domain in T. turgidum. The homoeologous CDS sequences of TRITD-2Bv1G264480 were amplified in the target interval on chromosome arm 2AL in Zhongmai 175, Mingxian 169 and Avocet S (Fig. S8). There were 30 SNPs at CDS domain between Zhongmai 175 and Mingxian 169, resulting in 19 amino acid variations. The deletion of 3,890 bp in Mingxian 169 led to a frameshift mutation (Fig. S9). Based on one SNP at CDS domain between Zhongmai 175 and Avocet S, we developed a molecular marker CAPS6 (Table S2) and tested all susceptible F 2 plants. CAPS6 co-segregated with the resistance gene. Therefore, we considered that TRITD2Bv1G264480 was a candidate of YrZM175.

Identification of candidate genes for YrZM175
The targeted region of YrZM175 according to the Chinese Spring genome contained a cluster of putative resistance genes. After gene cloning and sequence analysis of six putative disease resistance genes in this interval, we performed transgenic assays of TraesCS2A02G560600 using two transgenic vectors; one was the CDS domain driven by Ubi promoter and another was the 6,251-bp ORF (including coding and intron region) with a 2,770-bp presumed native promoter and 1,500-bp terminator (3'-UTR). Unfortunately, all the positive T 1 plants were susceptible to race CYR32 in seedling stage tests, indicating that TraesCS2A02G560600 was not a candidate of YrZM175. We amplified five and two genes upstream and downstream, respectively, of TraesC-S2A02G562800 according to the Chinese Spring genome and again failed to obtain PCR products, indicating great sequence differences of this genomic region between Zhongmai 175 and the Chinese Spring reference genome. With the release of reference genomes such as T. turgidum ssp. durum cv. Svevo (Maccaferri et al. 2019), T. aestivum cv. Zang1817 (Guo et al. 2020), 10 + Genome (Walkowiak et al. 2020), and T. aestivum cv. Fielder (Sato et al. 2021) we were able to extend our investigation. We performed genome walking to obtain the unknown gene sequences in the 636.4-kb interval on chromosome arm 2AL of Zhongmai 175. Firstly, we cloned gene sequences at 763.83 Mb of chromosome arm 2AL in Zhongmai 175 referring to Chinese Spring genome to confirm that it was definitely from the genome sequence of chromosome arm 2AL. Then, we cloned the downstream gene sequences from 763.83 Mb step by step with genomespecific primers (Table S10) and determined the positions of amplified gene sequences based on Chinese Spring reference genome. Finally, after two steps of genome walking, we obtained a gene sequence that was almost the same as a genome sequence (TRITD2Bv1G265970) on chromosome arm 2BL of durum wheat Svevo. However, the linkage distance between TRITD2Bv1G265970 and the resistance gene was 0.21 cM, indicating that TRITD2Bv1G265970 was not the candidate of YrZM175.
We amplified and sequenced ten putative disease resistance genes in the collinear chromosome arm 2AL interval in five Triticum genomes in Zhongmai 175 and obtained no candidate gene sequences. Finally, we assessed TRIT-D2Bv1G264480 on chromosome arm 2BL in T. turgidum as a candidate gene based on a co-segregating marker, sequence alignment of resistant and susceptible parents and gene expression patterns. The marker CAPS6 for TRITD-2Bv1G264480 was co-segregating with YrZM175 in the F 2 population with 7,325 plants (Fig. S10). CAPS6 was a dominant marker that could not distinguish homozygous susceptible plants and heterozygous plants. We sequenced the CAPS6 PCR products of 10 recombinant susceptible F 2 plants that were identified by the marker YrZM175-InD2 with a genetic distance of 0.12 cM to YrZM175, and the result indicated that 10 plants contained the homozygous susceptibility allele at TRITD2Bv1G264480 locus (Fig.  S11). Our research indicated that Zhongmai 175 has complicated genetic origins that caused difficulties in mapping the target genes according to single reference genome. In addition to the Chinese Spring genome sequence, it is also very necessary to refer other available genome sequences of wheat and related species for fine mapping and cloning of disease resistance genes in bread wheat.

Candidate gene TRITD2Bv1G264480
The putative resistance gene TRITD2Bv1G264480 encodes an NBS-LRR protein that belongs to two overlapping homologous superfamilies: P-loop containing nucleoside triphosphate hydrolase, and apoptotic protease-activating factors (helical domain) (http:// www. ebi. ac. uk/ inter pro/ entry/ Inter Pro/ IPR00 2182/). The predicted protein of TRIT-D2Bv1G264480 is closest to TRIDC2BG090350 on chromosome 2B from T. dicoccoides (Fig. S12). The paralogous genes on chromosome 2A in genomes of seven bread wheat varieties clustered together phylogenetically (Fig. S12), with 89% and 66% sequence similarities in the CDS domain to TRITD2Bv1G264480 on chromosome 2B and the paralogous gene TRITD2Av1G294930 on chromosome 2A in T. turgidum, respectively, indicating much closer phylogenetic relationship with TRITD2Bv1G264480 from chromosome 2B in T. turgidum, whereas the orthologous genes on chromosome 2B in T. aestivum cv. CDC Landmark, cv. CDC Stanley, cv. Fielder and cv. Julius showed 67.4% similarity with TRIT-D2Bv1G264480. In addition, paralogous genes on chromosomes 2A in T. Urartu and 2D in Ae. tauschii and orthologous genes on chromosome 2B in T. spelta also had quite distant phylogenetic relationships with TRITD2Bv1G264480 (Fig. S12). Therefore, we inferred TRITD2Bv1G264480 was transferred from chromosome 2B to 2A during evolution from an old crossover event of bread wheat. The CDS domain of candidate gene in Zhongmai 175 showed 97.8% and 96.8% similarities with TRITD2Bv1G264480 (T. turgidum) and TRIDC2BG090350 (T. dicoccoides), respectively, whereas the sequence similarities with homologues genes on chromosome 2A in other bread wheat varieties (Zang1817, Jagger, Fielder, Mace, CDC Landmark and Norin 61) were relatively low (all below 88.3%), indicating multiple allelic variations or haplotypes of the candidate gene present in bread wheat during evolution.

3
The relationship between YrZM175 and other Yr genes on chromosome 2A Zhongmai 175 has been widely grown in the North Winter Wheat Zone of China since 2008. It was derived from the cross between Jing 411 and BPM27 (Fig. S13). BPM27 is a disease-resistant line developed by Prof. Zuomin Yang at China Agricultural University in the 1980s; one of its parents was VPM1, a French line with 2NS translocation derived from crosses of Ae. ventricosa, T. turgidum L. var. carthlicum (T. persicum) and common wheat cv. Marne (Fig.  S13). Bariana and McIntosh (1994) reported that the stripe rust resistance of VPM1 was conferred by Yr17 derived from Ae. Ventricosa. Lu et al. (2016) mapped YrZM175 on chromosome arm 2AS using 344 F 2 plants and 147 F 2:3 lines derived from cross Lunxuan 987/Zhongmai 175 tested with CYR29. The present study used many more F 2 plants from a cross of Avocet S and Zhongmai 175 but employed race CYR32. The discrepancy of gene locations can be attributed to different mapping populations, molecular markers and PST races used in the two studies. Unfortunately, the PST race CYR29 used previously  was no longer available to test the present population, thus we are not very sure whether the resistance genes are the same or not. The race CYR32 is virulent to VPM1 (Yr17) ( Table S11). Screening of the 1,793 susceptible F 2 plants in the Avocet S/Zhongmai 175 population with the marker URIC/LN2 (Helguera et al. 2003) as proxy for Yr17 indicated linkage of 55.2 cM. YrZM175 is located at the distal end (763,452,916-764,089,317 bp) of chromosome arm 2AL, whereas Yr17 is located at the distal (25.0-38.0 cM linkage interval) of chromosome arm 2AS (Cruz et al. 2016) (Fig. S14). Lu et al. (2016) performed multi-pathotype test on VPM1 (Yr17) and Zhongmai 175 (Yr17 and YrZM175), the two lines showed different reactions to seven PST races at the seedling stage (Table S11). A test with the marker URIC/LN2 in 92 susceptible F 2 plants indicated that the linkage distance between Yr17 and YrZM175 was 39.0 cM , being similar to the present study.
To date, three Yr genes have been mapped on chromosome arm 2AL, i.e., Yr1, Yr32 and YrJ22 (Lupton and Macer 1962;Eriksen et al. 2004;Chen et al. 2016). Yr1 is linked to Xgwm382 (Fig. 5B), with a genetic distance of 5.6 cM (Zheng et al. 2017;Bansal et al. 2009 Chen et al. 2016) (Table S11), we believe YrZM175 and Yr1 are different. Based on the linkage maps and physical positions of linked markers, the physical position of YrZM175 is close to Yr1, possibly indicating allelism. To clarify their relationship, tests of allelism and gene cloning should be carried out. Xgwm382 is also closely linked to YrJ22 with a genetic distance of 2.5 cM ( Dr. Can Chen), a 5 Mb physical distance from YrZM175. As Jimai 22 and Zhongmai 175 have different pedigrees and different arrays of reactions to PST races (Table S11), we assume that YrZM175 and YrJ22 are different genes or alleles. Eriksen et al. (2004) found that the marker Xcdo678 linked with Yr32 in a genetic distance of 35 cM, while Xcdo678 is co-segregating with Pm4 (Ma et al. 1994). Pm4 and Yr1 are closely linked to each other with a genetic distance of 2.0 cM (McIntosh and Arts 1996). As a result, the genetic distance between Yr32 and Yr1 is about 33 cM (Fig. 5A) (Yang et al. 2019), indicating that Yr32 is far away from YrZM175 and Yr1. Based on linkage maps and molecular marker analyses, we conclude that YrZM175 is different from Yr1, Yr17, Yr32, and YrJ22.

Complexity of the distal genome region of chromosome 2AL
Structural genomic variation such as presence-absence variations (PAVs) and copy number variations (CNVs) has been recognized to have the potential to generate phenotypic variation in maize ) and barley (Muñoz-Amatriaín et al. 2013). Megabase-scale PAVs of Tripsacum origin were confirmed to be under selection during maize domestication and adaptation (Huang et al. 2021). Rimbert et al. (2018) also found high frequencies of PAVs in the distal regions of wheat chromosomes. Akhunov et al. (2003a, b) found that duplicated loci were most frequently located in the distal regions of chromosomes, and their distribution was positively correlated with recombination rate. The higher recombination rate indicates that these regions are fast-evolving in adapting to biotic and abiotic stresses. Besides, there are higher polymorphisms in the distal regions of chromosomes with the decreased levels of synteny between homoeologous chromosomes as distance from centromeres increase (Akhunov et al. 2003a, b). Our experimental results on chromosome arm 2AL in Zhongmai 175 were in agreement with these related reports mentioned above; this is reflected by the result that the candidate gene TRITD2Bv1G264480 for YrZM175 likely originated from chromosome arm 2BL of T. turgidum. The distal region of chromosome has great structural genomic variations and therefore is easier to evolve new alleles that trigger phenotypic variation.
In conclusion, YrZM175 is a stripe rust resistance gene that confers moderate to high resistance to stripe rust in the field. We fine mapped YrZM175 and identified a candidate gene based on a high-resolution linkage map, collinearity analysis and gene expression analysis. Transgenic assays are ongoing to validate the functions of candidate of YrZM175. The identification and fine mapping of YrZM175 provide options for deployment in combination with other effective resistance genes.