1RS arm of Secale cereanum ‘Kriszta’ confers resistance to stripe rust, improved yield components and high arabinoxylan content in wheat

Wheat-rye T1BL.1RS translocation is widespread worldwide as the genes on 1RS arm have positive effect on stress resistance, grain yield and adaptation ability of wheat. Nowadays, the T1BL.1RS wheat cultivars have become susceptible to rust diseases because of the monophyletic (‘Petkus’) origin of 1RS. Here we report and discuss the production and detailed investigation of a new T1BL.1RS translocation line carrying 1RS with widened genetic base originating from Secale cereanum. Line ‘179’ exhibited improved spike morphology traits, resistance against stripe rust and leaf rust, as well as higher tillering capacity, fertility and dietary fiber (arabynoxylan) content than the parental wheat genotype. Comparative analyses based on molecular cytogenetic methods and molecular (SSR and DArTseq) makers indicate that the 1RS arm of line ‘179’ is a recombinant of S. cereale and S. strictum homologues, and approximately 16% of its loci were different from that of ‘Petkus’ origin. 162 (69.5%) 1RS-specific markers were associated with genes, including 10 markers with putative disease resistance functions and LRR domains found on the subtelomeric or pericentromeric regions of 1RS. Line ‘179’ will facilitate the map-based cloning of the resistance genes, and it can contribute to healthy eating and a more cost-efficient wheat production.

Interspecific hybridization is one of the most promising way to improve the genetic diversity of bread wheat 1 . The most widely known wheat-alien introgression has been the spontaneous translocation T1BL.1RS that is traced back to the cross of wheat 'Criewener 104' × rye 'Petkus Roggen' made in Germany between 1920 and 1930 2 . Its effect on morphology and baking quality of bread wheat has been investigated in numerous studies. The positive effect of this translocation on grain yield of hexaploid wheat 3,4 and durum wheat 5,6 is commonly accepted, though in some cases no such effect was shown 7,8 . Association between the presence of 1RS and increase in root biomass, water use efficiency as well as harvest index was also found in glasshouse and field experiments 9,10 .
1RS arm also contains genes (Pm8, Lr26, Sr31 and Yr9) providing resistance against powdery mildew (Erysiphe graminis f. sp. tritici), leaf rust (Puccinia triticina), stem rust (Puccinia graminis f. sp. tritici), and stripe or yellow rust (Puccinia striiformis f. sp. tritici), respectively 11,12 , a reason that led to the widespread utilization of the T1BL.1RS translocation in wheat improvement. However, the resistance genes Lr26, Yr9 and Pm8 are no longer effective against new virulent biotypes of the pathogens in Europe 13 . Virulence to the Sr31 resistance gene has also been reported from Uganda 14 , and from Kenya, Ethiopia, Sudan, and Iran 15 . Spread of the Sr31-virulent pathotype in countries where wheats carrying the T1BL.1RS chromosome are cultivated can cause serious problems for agriculture as the genetic vulnerability of T1BL.1RS cultivars is the consequence of the lack of allelic variation on the 'Petkus'-derived 1RS arm 4 . The importance of widening the genetic base of cultivated wheat by introducing new rye genetic resources into resistance breeding programmes has grown 16,17 .  Table S1).
Lines '179' , 'C5' and 'D5' were classified as T1BL.1RS translocations, as the hybridization pattern of probe pSc119.2 was typical for the rye chromosome arm 1RS (strong signal on the satellite and another one proximal from the secondary constriction) and for the long arm of 1B (Fig. 2a) in all the three genotypes.
As '179' was the first and from the breeder's point of view a promising (uniform) genotype we selected, in the subsequent parts of our study we focus on the presentation of this line. Genotypes 'C' and 'D' were involved only in studies on allelic differences between the 'Petkus'-and 'Kriszta'-originated 1RS chromosome arms.
Analysis of morphological traits. The 1BL.1RS translocation significantly modified the morphology of the wheat parent. Compared to Mv9kr1, line '179' possesses awned and longer spikes with higher number of spikelets (Fig. 2b, Table 2, Supplementary Fig. S2). The improved spike architecture of '179' , due to the higher number of tillers, resulted in an approximately 50% increase in the number of seeds per main spike and an increase of about 80% in the number of seeds per plant. Artificial disease resistance tests. As the rye chromosome arm 1RS harbours resistance genes against leaf rust (Puccinia triticina) and powdery mildew (Blumeria graminis f. sp. tritici), artificial resistance tests were performed to assess infection response of '179' to these pathogens (Table 1). Infections were carried out using pathotypes with known virulence/avirulence formula (Supplementary Table S2 Quality measurements. Knowing that rye has higher dietary fiber content than wheat, we also investigated the effect of 1RS arm on the arabinoxylan (AX, a major cell wall polysaccharide) content and the protein composition of the wholemeal in '179' together with the wheat and rye parents (Table 1). Line '179' exhibited significantly higher (increased by 40.8%) total arabinoxylan (TOT-AX) content than the parental wheat Mv9kr1 reaching the level of the parental rye variety 'Kriszta' ( Table 3). The increase in the level of water extractable arabinoxylans (WE-AX) was even higher (66.8%) in '179' relative to the wheat parent.
An increase (20.3%) in the total protein content was also measured in the line '179' (Table 3). The glutenin/ gliadin ratio decreased, while the ratio of the high molecular weight (HMW) and low molecular weight (LMW) glutenin subunits (HMW/LMW) increased significantly in this genotype compared to Mv9kr1. A significant decrease in the unextractable polymeric protein (UPP) content was also found. The molecular weights of ω-secalin monomers are very close (48-55 kilodalton) 46 , consequently their separation was only possible by using A-PAGE based on differences in protein charge density at low pH ( Supplementary  Fig. S3). Bands at identical positions in Mv9kr1 and line '179' proved the Mv9kr1 (maternal) origin of our translocation line. Components of the ω-secalin block encoded by the alleles Sec-1 (Gli-R1) in 'Kriszta' , 'Petkus' , 'Mv Magdaléna' , and line '179' were identical, except that an additional (sixth) faint band was also expressed in the '179' .

Comparative cytogentic analysis of 'Petkus'-and 'Kriszta'-derived 1RS arms. As line '179'
showed altered resistance traits compared to 'Mv Magdaléna' , a study was carried out in order to reveal putative differences between the 'Petkus'-and 'Kriszta'-derived 1RS arms at the cytomolecular level. After analysing ten preparations per genotype ('Mv Magdaléna' , '179' , 'Kriszta' , and its parental genotypes 'Várda' and S. strictum ssp. anatolicum 'R797'), and ten cells per preparation, we consistently found that 1RS arm of '179' showed a double subterminal pSc119.2 signal on the satellite, while that of 'Mv Magdaléna' had a single band at the same position ( Fig. 4). Similar double subterminal pSc119.2 pattern was observed in some individuals of the parental S. cereanum genotype 'Kriszta' that has 1R chromosomes with different pSc119.2 hybridization patterns in homo-or heterozygous form ( Supplementary Fig. S4). Among the 'Kriszta' plants studied, we detected 1R chromosomes whose FISH pattern was typical for S. strictum ('R797'), but their morphology (size and visibility of the satellite) corresponded to S. cereale ('Várda'), while other 1R chromosomes corresponded morphologically to 'R797' , but they lacked any hybridization signals at distal position on the satellite (Fig. 4). Based on its pSc119.2 hybridization pattern, we concluded that the 1RS arm of '179' derived from a recombination between that of S. cereale (strong interstitial signal proximal to the secondary constriction) and S. strictum ssp. anatolicum (double sub-terminal pSc119.2 signal). DArTseq analysis and functional annotation. We wanted to investigate allelic composition differences between the 1RS chromosome arms of the T1BL.1RS wheat genotypes '179' and 'Mv Magdaléna' , as well as between the 1RS arm fixed in the stripe rust resistant lines '179' , 'C5' and 'D5' , and the sensitive 1R addition line DA1R in much more detail. For this purpose, a rye DArTseq genotyping platform has been applied which provides much higher marker coverage for 1RS.    www.nature.com/scientificreports www.nature.com/scientificreports/ After quality filtration by high call rate and reproducibility of the 258,090 Silico-DArT markers obtained in the present study, 175,558 (68%) were scored as '0' (absence of genomic representation) for 'Mv9kr1' , and as '1' (6401 markers) or '0' (169,157 markers) for the Mv9kr1-S. cereanum DA1R line susceptible to stripe rust. Then we selected 5312 putative 1RS-specific Silico-DArT markers using the Mv9kr1-'Kriszta' T1BL.1RS genotypes ('179' , 'C5' and 'D5') and the DA1R line according to criteria summarized in Supplementary Data S1. We obtained a total of 71,177 SNP markers, as well. After quality filtration and removing markers having alleles in the wheat genotype 'Mv9kr1' , 6027 markers remained of which 1755 were selected as putative 1RS-specific SNP markers.
In order to confirm that the selected Silico-and SNP-DArT markers are specific for rye chromosome 1R, we used trimmed marker sequences for BLASTn against the rye Lo7 genomic sequences. Out of the 5312 Silico-DArT markers, 679 were aligned to the rye Lo7 WGS 1R contigs, while out of the 1755 SNP markers 168 gave hits against 1R contigs of rye (Supplementary Data S2). These (679 + 168 = ) 847 markers were used in the subsequent analysis to select markers specific for the short arm of 1R, and to order them using the recent version of the rye virtual gene order map (Rye Genome Zipper v.2) 50 . As Bauer et al. (2017) 50 mapped the 1R centromere to the position of 60.72 cMorgan (cM), we checked the region ranging from 0 cM to 60.72 cM (considered as 1RS) for the presence of marker-specific Lo7 WGS contigs. Out of the 847 1R-specific markers mentioned above, 233 (27.5%) were located on the short arm of 1R. Out of the 233 markers, 129 (111 Silico-DArT + 18 SNP; 55.36%) were specific for the stripe rust resistant genotypes '179' , 'D5' , and 'C5' (Fig. 5a), and 104 markers (44.64%) were specific for the susceptible DA1R line. This pronounced difference in the allelic composition suggests that the 1RS arm fixed in the stripe rust resistant genotypes is different from that of DA1R.
We also compared the allelic composition of the 1RS arm of the line '179' with the 'Petkus'-derived 1RS of 'Mv Magdaléna' . Out of the above-mentioned 233 1RS-specific markers 225 gave unambiguous allelic results in both wheat genotypes. Thirthy-seven (16.4%) of the 225 markers showed different alleles in '179' relative to 'Mv Magdaléna' (Fig. 5b).   www.nature.com/scientificreports www.nature.com/scientificreports/ Theoretically, a significant amount of the DArTseq markers is associated with coding sequences or other parts of the "genic" region (promoters, introns, 3′ UTR sequences), which is the consequence of a methyl filtration step (complexity reduction) during the process of DArTseq marker production. After homology search of the 233 1RS-specific markers against the cDNA sequences of rye or the related cereal species, we obtained significant hits  Fig. 6.

Discussion
Line '179' presented in this study provides new allelic variations for the rye chromosome arm 1RS introgressed into wheat as it carries a pair of T1BL.1RS chromosomes in which the 1RS arm derives from the perennial rye cultivar 'Kriszta' , a hybrid of S. cereale and S. strictum ssp. anatolicum.
Effects on morphological and quality traits. The novel genetic variation on the 1RS arm of line '179' caused positive changes in tillering capacity, lentgth of spikes, number of spikelets, and fertility that led to a 72-87% increase in the number of seeds per plant, confirming the findings of previous studies 34 .
Replacement of 1BS with 1RS in wheat leads to quality deficiencies. Changes in the protein composition, particularly the reduction in the concentration of low-molecular-weight (LMW) glutenin and gliadin proteins, cause weaker dough strength and stickiness 51 . Although Kumlay et al. (2003) 52 found that the presence of the secalin genes had greater negative impact than the loss of Glu-B3 and Gli-B1 loci on 1BS, contribution to the effects mentioned above of Sec-1 alleles is not entirely clear 53 . Analysing highly informative quality parameters in doubled haploid T1BL.1RS translocation lines it was concluded that the lower quality caused by the translocation chromosome can be counterbalanced by beneficial alleles at Glu-A1, Glu-B1 and Glu-D1, particularly in lines with high protein content 54 . In this respect, line '179' may be a promising breeding material having elevated total protein content and a unique A-PAGE profile of ω-secalins. Amino acid sequence analysis revealed that ω-secalin genes from rye (S. cereale), hexa-and octoploid triticale, and T1BL.1RS translocation lines are homologous and highly conserved 55   www.nature.com/scientificreports www.nature.com/scientificreports/ were associated with significantly increased amounts of soluble AX, while the additions of 1R, 3R, 4R and 7R resulted in significantly lower levels of soluble DF and AX. Contrary to this, Cyran et al. (1996) 57 reported that chromosomes 4R had an impact on high AX expression in wheat comparable to that of rye AX content. Our results suggest that the role of chromosome arm 1RS in increasing both TOT-AX and WE-AX content in wheat is also significant, presumably due to the presence of S. strictum chromatin On the basis of our findings we can conclude that the line '179' carrying a new T1BL.1RS translocation is an important genetic resource for breeding programmes aimed at producing wheat cultivars with improved grain yield and increased level of dietary fibers.
Effects on disease resistance. Stripe rust, earlier considered infectious in regions of lower temperature, nowadays causes severe epidemics in warmer wheat growing areas (including Central Europe) as well, owing to the emergence of new races with expanded virulence profiles and increased aggressiveness especially at higher temperature 58 . Identification and incorporation of diverse sources of resistance (particularly of the durable type) against this pathogen is an urgent and challenging task for resistance breeding. So far several successful attempts exploiting the genetic variation of wild and cultivated relatives have been reported to improve disease resistance of bread wheat (for review see 1 ). Regarding rye genepool as resistance source for wheat improvement, Wang et al. (2009) 59 transferred 1RS arm from the rye cultivar German White into the chinese wheat cultivar Xiaoyan 6, and www.nature.com/scientificreports www.nature.com/scientificreports/ produced a T2BL.1RS wheat-rye chromosome translocation line resistant to stripe rust and powdery mildew. Using Chinese wheat and rye cultivars, various T1BL.1RS translocation lines resistant to stripe rust 60 or both stripe rust and powdery mildew [61][62][63] were also developed. In case of wild relatives of rye, only two papers reporting on the transfer of stripe rust resistance from S. africanum have been published 64,65 , where the resistance genes were mapped on the chromosome arms 1RS 64 and 2RL 65 .
In the present study, the novel T1BL.1RS translocation line '179' was resistant not only to stripe rust, but also to leaf rust, which is consistent with the fact that resistance genes Lr26 and Yr9 (together with Sr31), form a closely linked gene cluster mapped on the satellite of 1RS arm in S. cereale 12,66 . It is well known that LRR domains are important parts of the plant immune receptors 67 . Consistently with this, we found that DArTseq markers associated with genes carrying the LRR domain, and polymorphic between the stripe rust resistant '179' and the sensitive DA1R were located on the subtelomeric 1.21-16.71 cM interval of Lo7 1RS arm. For the time being we are uncertain whether the genes providing resistance against stripe rust and leaf rust in the genotype '179' are variants of Yr9 and Lr26 or they are new resistance genes. Genetic mapping of resistance against stripe rust and leaf rust using an M9kr1-'Kriszta' T1BL.1RS line '179' (resistant) x M9kr1-'Kriszta' DA1R (susceptible) biparental mapping population will help to identify the position of the resistance genes in the line '179' . Later, production of knockout mutants in '179' , sequencing of T1BL.1RS chromosomes flow-sorted from the resistant genotype '179' and from its susceptible mutants (MutChromSeq), and combination of the sequences with the mapping data will open the way for the positional cloning of the resistance genes introgressed into the line '179' 68 .
The explanation why the M9kr1-'Kriszta' T1BL.1RS line '179' is resistant to stripe rust and leaf rust while the M9kr1-'Kriszta' DA1R line is susceptible may be that diverese 1R chromosomes exist in the open-pollinated population of parental rye cultivar Kriszta, and the 1RS arm fixed in the stripe rust resistant genotypes is different from that fixed in the Mv9kr1-'Kriszta' DA1R line. This idea was supported by the fact that more than half (55.36%) of the 233 1RS loci identified by the DArTseq technology were specific for the resistant genotype, and 44.64% were specific for the susceptible 1R addition line. This supposition was further supported by the cytomolecular results that different 1R chromosomes with altered pSc119.2 FISH patterns were detected in 'Kriszta' rye, which agreed well with a previous study, where four out of five'Kriszta' plants investigated by the 1RS-specific SSR marker RMS13 showed polymorphic amplicon patterns 16 . Based on this finding, it is assumable that during meiosis in the descendants of the S. cereale × S. strictum ssp. anatolicum hybrid several crossing-over events occurred resulting in different 1RS chromosome arms that carry chromatins from the two species in various proportions. Our hypothesis on the recombination was further confirmed by molecular marker analyses. Due to the high transferability of SSR markers between S. cereale and S. strictum 45  www.nature.com/scientificreports www.nature.com/scientificreports/ we obtained a similar level of polymorphism (16.4%) between the 1RS loci in '179' and 'Mv Magdaléna' when DArTseq markers were used in higher coverage for genotyping (37 out of the 225 were polymorphic).
Knowing that the Pm8 locus (in S. cereale) is mapped only 1.7 cM distal to the Lr26/Sr31/Yr9 gene cluster 69 , and the artificially inoculated 'Kriszta' was resistant to powdery mildew, the susceptibility of line '179' was unexpected. The most probable explanation of this observation may be the phenomenon of genetic suppression of resistance genes, which often manifests itself when these genes are transferred from related species with lower ploidy level to hexaploid bread wheat [70][71][72] or synthetic wheat 73,74 . Expression of the 1RS-derived Pm8 is suppressed by translated gene products from the Pm3 locus, a wheat ortholog of Pm8, located on wheat chromosome arm 1AS 75 . Resistance suppression or non-suppression of Pm8 depends on the wheat germplasm in which the 1RS chromosome arm is introgressed 13,76,77 . Based on the findings of the reports mentioned above, we have speculated about the following possibilities: (i) suppression activity of Pm3 on chromosome arm 1AS in wheat genotype Mv9kr1 (in contrast with many Chinese wheat cultivars) is expressed, (ii) the gene on 1RS in line '179' must be Pm8 carrying allelic variation for resistance against powdery mildew, (iii) if line '179' carries a gene of S. stricum origin different from Pm8, its expression is also influenced by the gene products of Pm3.
Final conclusion. The S. cereale x S. strictum recombinant 1RS originating from Secale cereanum cv. Kriszta confers novel genetic diversity in the T1BL.1RS translocation chromosome of line '179' reported here. The new, morphologically uniform wheat genotype possessing altered spike morphology, increased fertility, high protein and arabynoxylan content, and resistance to stripe rust and leaf rust is a promising gene source for wheat breeding. It has been involved in a breeding program aiming transfer of the new T1BL.1RS translocation into modern winter wheat cultivars, therby it can contribute to healthy eating and a more cost-efficient wheat production. Line '179' will also facilitate the map-based cloning of the resistance genes located on the recombinant 1RS chromosome arm.

Methods
For reasons of clarity, a short overview of the aims as well as methods and genotypes used in the present study is given in Table 1.  Supplementary Fig. S1. The wheat-S. cereanum T1BL.1RS translocations cytogenetically identified in the BC 2 F 8 generation have been multiplied in the field since then.
Preparation of metaphase chromosome spreads. Chromosome preparations from roots of germinating seeds of 'Várda' , 'R797' , 'Kriszta' , lines '179' , 'C5' , 'D5' , and'Mv Magdaléna' followed the method described by Endo and Gill (1984) 80 with minor modifications. Roots kept in Carnoy's solution I (absolute ethanol:glacial acetic acid 3:1 v/v) at 37 °C for seven days were stained with 1% (w/v) acetocarmine for at least 2 h, fixed again and stored at −20 °C until use. Root tips were squashed in 45% acetic acid. After removing the coverslips by freezing in liquid nitrogen, the preparations were dehydrated in ethanol series, air-dried overnight and stored at −20 °C.
Fluorescence and genomic in situ hybridization. Total rye (Secale cereale L.) genomic DNA was labeled with digoxigenin-11-dUTP (Roche Diagnostics, Mannheim, Germany) by nick translation and used as a probe for genomic in situ hybridization (GISH). Rye and wheat chromosomes were identified with the rye subtelomeric heterochromatic sequence pSc119.2 31 and FISH probe combination of pSc119.2, Afa-family 81 and the 45 S rDNA clone pTa71 82 , respectively. PCR-amplified pSc119.2 83 and Afa-family were labeled with biotin-16-dUTP (Roche) and digoxigenin-11-dUTP, respectively, while pTa71 was co-labeled with these nucleotides. GISH and fluorescence in situ hybridization (FISH) were carried out simultaneously. The hybridization mixture per slide (total volume = 30 µL) contained 30 ng labeled rye genomic DNA, 50 ng pSc119.2 repetitive DNA probe, 50% v/v formamide, 2 × SSC (0.15 mol/L NaCl plus 0.015 mol/L sodium citrate), 10% w/v dextran sulphate, 1.4 µg salmon sperm DNA and 0.1% w/v sodium dodecyl sulphate. The mixture did not contain wheat blocking DNA. Chromosome preparations were denatured at 75 °C for 6 min and hybridized overnight at 42 °C. Signals of the digoxigenin-and biotin-labeled probes were detected using anti-digoxigenin-Rhodamin Fab fragments (Roche) and streptavidin-FITC (Roche), respectively. Slides were counterstained with 1 µg/mL DAPI (4' ,6-diamidino-2-phenylindole, Amersham, Germany). Images were acquired through a Zeiss Axioskop 2 fluorescence microscope equipped with filter sets appropriate for DAPI (filter set 1), FITC (filter set 10), Rhodamin (filter set 15) and for the simultaneous detection of FITC and Rhodamine (double filter set 24) with a Spot CCD camera (Diagnostic Instruments, Sterling Heights, MI, USA) and processed with Image-Pro Plus software (Media Cybernetics, Silver Spring, Md, USA). www.nature.com/scientificreports www.nature.com/scientificreports/ Observations were carried out on ten preparations (i.e. ten individuals) of each genotype and on ten cells per preparation.
Field observations. As stripe rust isolates for artificial inoculation are not available in Hungary, assessment and comparision of infection responses of the lines '179' , 'C5' and 'D5' , the T1BL.1RS cultivar Mv Magdaléna, the disomic addition line DA1R carrying a whole 1R chromosome as well as the parental genotype 'Kriszta' was only possible under field conditions. In order to ensure the natural infection, the genotypes mentioned above and the highly susceptible parental Mv9kr1 (as possible infection source) were grown in neighbouring plots (2 m 2 each) of the high-input pre-breeding (PB) nursery of the Agricultural Institute, Centre for Agricultural Research (Martonvásár, Hungary; geographic coordinates: 47°19'58"N 18°47'08"E) in three consecutive growing seasons from 2014 to 2016. Infection severity (0-100%) was scored according to the modified Cobb scale 84 based on the percentage of infected leaf area. Ratings were recorded weekly for a month from the first appearance of the disease on the susceptible Mv9kr1.
Morphological comparision of line '179' with the parental wheat genotype Mv9kr1 was carried out in the PB nursery in 2015, and in both the PB nursery and a low-input location (LI; Tükrös nursery, Martonvásár, Hungary; 47°18'40"N 18°46'56"E) in 2019. Fifty seeds of each genotype were sown in 1 m rows with 10 seeds per row and a row distance of 15 cm. Characteristics were taken from ten randomly selected plants of each genotype. Plant height and tillering (spikes per plant) were measured in the field, the length of the main spike, number of spikelets per main spike, number of seeds per main spike, number of seeds per plant and thousand kernel weight (TKW) were recorded after harvest.
Quality measurements. Crude protein content was assessed from whole grain flour of 40 mg seeds per genotype by the Kjeldahl method 85 using Kjeltec 1035 Analyzer. Glutenin, gliadin, albumin + globulin, and unextractable polymeric protein (UPP% = insoluble glutenin/soluble + insoluble glutenin) content were determined by size-exclusion high-performance liquid chromatography (SE-HPLC) 86 , relative amounts of the HMW glutenin subunits were determined by reversed-phase high-performance liquid chromatography (RP-HPLC) 87 , while total and water extractable pentosans (arabinoxylans -AX) were determined by colorimetry using modified methods described earlier in details 88 .
Electrophoresis of prolamin storage proteins. Prolamin storage protein (gliadin and/or secalin) composition of wheat genotypes Chinese Spring (as reference), Mv9kr1, 'Mv Magdaléna' and line '179' , and rye cultivars 'Kriszta' and 'Petkus' was determined by acid-polyacrylamid gel electrophoresis (A-PAGE, pH = 3.1) using the method of Jackson et al. (1996) 89 . Prolamins were extracted from single seeds with 70% ethanol at 65 °C shaken for 30 minutes. Samples were run on 12,5% polyacrylamide gel with an acrylamide/bisacrylamide ratio of 1:32. Gel polymerization was initiated by hydrogen peroxide. Gels were run on 30 mA for 10 minutes than on 70 mA for 3 hours using reverse poles. Gels were dyed with Brilliant Blue G solution (Sigma).
Artificial disease resistance tests. Resistance Supplementary Table S2 online. Before brush-inoculation with water-based leaf rust urediospore suspension, the wax layer was washed off from the leaf surface. To ensure high humidity conditions, the inoculated plants were covered with polythene bags for 48 hours. Infection types were recorded 14 days after inoculation using the 0 to 4 Stakman-scale 90 .
In the case of powdery mildew, the isolates (LH07-14, LH14-14) were formerly propagated on a susceptible wheat cultivar (Carsten V). Ten plants of each genotype were grown in a 50 × 40 cm wooden isolator box with 80-90% relative air humidity, and 16 h daily natural and illuminated light conditions at 18 °C. Heavily sporulating colonies on 'Carsten V' plants were shaken into the test box. Infection levels were scored on a 0 to 4 scale (0-2: resistant, 3-4: susceptible) 91 after ten days. The generated results were listed in tables for Silico-and SNP-DArT markers. Markers with call rate > 95% [available from the BioStudies database (https://www.ebi.ac.uk/biostudies/studies/S-BSST280; Accession number: S-BSST280)] were selected for subsequent analysis. Silico-DArT markers were scored as binary data (1 or 0) reflecting the presence or absence of a marker in the genomic representation of each sample 92 . We used 1-row Mapping Format for the SNP markers scored as '−' , '0' , '1' and '2' representing absence of a marker, reference allele (Rye_v2; provided by Diversity Arrays Technology Pty Ltd.) only, alternative allele, and both reference and alternative alleles, respectively. Our objective was to identify markers in order to differentiate between wheat chromosomes and chromosome 1R of S. cereanum, and investigate polymorphism between stripe rust resistant and stripe rust susceptible wheat-S. cereanum lines. In order to separate markers specific for S. cereanum, Silico-DArT markers, as well as SNP alleles present in the wheat (Mv9kr1) genome were discarded. Genotyping data of the Mv9kr1-S. cereanum T1BL.1RS ('179' , 'C5' , 'D5') and 1R addition (DA1R) lines were then analysed to select putative 1 R specific markers, as well as markers polymorphic between stripe rust resistant ('179' , 'C5' , 'D5') and stripe rust susceptible (DA1R) genotypes using criteria summarized in Supplementary Data S1 online.
Translated cDNA sequences were used for functional annotations. The Gene Ontology information was extracted from the Universal Protein Resource (ftp://ftp.uniprot.org; UniProt release 2019_04) database, the resulting protein collections were subsequently scanned with the Hidden Markov Model (HMM)-based HMMER 3.0 software package (http://eddylab.org/software/hmmer/) 93

Data availability
The datasets analysed during the current study are available in the BioStudies repository [https://www.ebi. ac.uk/biostudies/studies/S-BSST280] with the accession code S-BSST280, and also included in supplementary information files (Supplementary Data S1-S3) of this article.