Characterization of synthetic wheat line Largo for resistance to stem rust

Abstract Resistance breeding is an effective approach against wheat stem rust caused by Puccinia graminis f. sp. tritici (Pgt). The synthetic hexaploid wheat line Largo (pedigree: durum wheat “Langdon” × Aegilops tauschii PI 268210) was found to have resistance to a broad spectrum of Pgt races including the Ug99 race group. To identify the stem rust resistance (Sr) genes, we genotyped a population of 188 recombinant inbred lines developed from a cross between the susceptible wheat line ND495 and Largo using the wheat Infinium 90 K SNP iSelect array and evaluated the population for seedling resistance to the Pgt races TTKSK, TRTTF, and TTTTF in the greenhouse conditions. Based on genetic linkage analysis using the marker and rust data, we identified six quantitative trait loci (QTL) with effectiveness against different races. Three QTL on chromosome arms 6AL, 2BL, and 2BS corresponded to Sr genes Sr13c, Sr9e, and a likely new gene from Langdon, respectively. Two other QTL from PI 268210 on 2DS and 1DS were associated with a potentially new allele of Sr46 and a likely new Sr gene, respectively. In addition, Sr7a was identified as the underlying gene for the 4AL QTL from ND495. Knowledge of the Sr genes in Largo will help to design breeding experiments aimed to develop new stem rust-resistant wheat varieties. Largo and its derived lines are particularly useful for introducing two Ug99-effective genes Sr13c and Sr46 into modern bread wheat varieties. The 90 K SNP-based high-density map will be useful for identifying the other important genes in Largo.


Introduction
Wheat (Triticum aestivum L., 2n ¼ 6x ¼ 42, AABBDD) stem rust, caused by Puccinia graminis Pers.:Pers. f. sp. tritici Eriks & E. Henn (Pgt), is a major threat to wheat production worldwide. Since the Pgt race TTKSK (also known as Ug99) was first identified in Uganda in 1999, a total of 13 variants within the Ug99 lineage, commonly known as to the Ug99 race group, have been detected across 13 countries in East Africa and the Middle East over the last two decades (Pretorius et al. 2000;Singh et al. 2015;Patpour et al. 2016;Terefe et al. 2019). Li et al. (2019) recently proposed that high genetic diversity of Ug99 races largely resulted from somatic hybridization and nuclear exchange between dikaryons, which likely is a driving force for the emergence of new pathotypes in asexual fungal populations. In addition, there are a few non-Ug99 lineage Pgt races such as TRTTF, TTTTF, JRCQC, and TKTTF, known to carry virulence against frequently deployed stem rust resistance (Sr) genes such as Sr9e, Sr25, Sr36, SrTmp, and Sr1RS Amigo (Jin 2005;Jin and Singh 2006;Olivera et al. 2012Olivera et al. , 2015Olivera Firpo et al. 2017;Patpour et al. 2017). Olivera et al. (2019) also found different virulent combinations among the Pgt races collected from Georgia. Together these diverse Pgt races pose a serious threat to global food security. Development of resistant wheat varieties is an effective approach to counter these threats. To achieve this goal, the wheat research community continuously searches for new resistance genes.
The wheat primary gene pool has been considered the best resource of resistance (R) genes due to minimal deleterious effects caused by linkage drag. The hexaploid wheat D-genome progenitor Aegilops tauschii Coss. (2n ¼ 2x ¼ 14, DD) is known to be a great resource of R genes for various diseases and insect pests (Ogbonnaya et al. 2013;Arora et al. 2019). To utilize Ae. tauschii accessions for the development of the resistant wheat lines/cultivars against biotic stresses, Dr. Leonard R. Joppa (USDA-ARS, retired) developed over 40 synthetic hexaploid wheat (SHW) lines by crossing durum wheat [T. turgidium L. subsp. durum (Desf.) Husn., 2n ¼ 4x ¼ 28, AABB)] "Langdon" with different Ae. tauschii accessions (Xu et al. 2010). Among these Langdon-derived SHW germplasm, one line was released and named Largo (CI 17895), which carries the Gb3 gene for greenbug (Schizaphis graminum, Rondani) resistance derived from Ae. tauschii accession PI 268210 (Joppa and Williams, 1982). Since its release, Largo and its derivatives have been the primary source of greenbug resistance in the winter wheat germplasm and varieties in Texas (Lazar et al. 1996(Lazar et al. , 1997Rudd et al. 2014). Largo was also identified to carry resistance to wheat curl mite (Aceria tosichella Keifer) (Dhakal et al. 2018) and several fungal diseases, including Septoria tritici blotch (Adhikari et al. 2015), Fusarium head blight (Szabo-Hever et al. 2018), and stem rust (Friesen et al. 2008).
The durum wheat variety Langdon was developed using a modified backcross procedure to transfer stem rust resistance from Khapli emmer (T. turgidum subsp. dicoccum Schrank) during the stem rust outbreak of the 1950s in the Northern Great Plains (Heyne 1959). Previous studies indicated that Langdon carries at least four Sr genes (Salazar and Joppa 1981). However, besides Sr13c (Zhang et al. 2017;Gill et al. 2021), other Sr genes in Langdon have not been unambiguously identified and confirmed. Because Langdon is one of the founders of modern durum germplasm and varieties in the US, identification of the Sr genes it harbors will enhance our understanding of the Sr genes present in modern durum wheat germplasm. Similarly, Ae. tauschii accession PI 268210 was previously identified to be resistant to all Pgt races tested, including TTKSK (Friesen et al. 2008;Zhang 2013). However, the Sr gene(s) in PI 268210 has also not been identified.
In addition to its high value in bread wheat breeding, Largo should be a useful parental line of a permanent mapping population that can be used for identification, mapping, and marker development for the agronomically important genes derived from durum Langdon and Ae. tauschii PI 268210. We conducted this study intending to identify the genes controlling stem rust resistance by developing, genotyping, and phenotyping a recombinant inbred line (RIL) population from a cross between Largo and the bread wheat line ND495.

Plant material and stem rust screening
A population of 226 RILs developed from a cross between a hard spring wheat line ND495 and SHW line Largo was used for genotypic and phenotypic analysis. Largo (CI 17895) was developed from a cross between durum wheat Langdon and Ae. tauschii accession PI 268210 (Joppa and Williams 1982). ND495 was developed at North Dakota State University (Fargo, ND, USA) and has a pedigree of Justin*2/3/ND 259/Conley//ND 112 (Anderson et al. 1999). The RILs along with parental lines ND495, Largo, PI 268210, and Langdon were phenotyped for seedling resistance in two biological replications (5 plants/replication) with Pgt races TTKSK (04KEN156/04), TRTTF (06YEM34-1), and TTTTF (01MN84A-1-2). The virulence/avirulence details of the three races are listed in Table 1.
The stem rust screening experiment was performed under controlled greenhouse conditions at the USDA-ARS Cereal Disease Laboratory, St. Paul, MN using the procedure described by Hundie et al. (2019). Briefly, the primary leaves of the seedling plants at 7 to 9 days after planting were inoculated with the Pgt urediniospores. After inoculation, the plants were moved into a greenhouse maintained at 18 6 2 C with a 16 hours photoperiod. The plants were scored for infection type (IT) at 14 days post inoculation based on the Stakman et al. (1962) 0-4 scale followed by the additional symbols ( þ and -) for the pustule size (Roelfs and Martens 1988). To identify the regions harboring quantitative trait loci (QTL) associated with resistance to the three Pgt races, the IT scores of each RIL for individual races were converted to the linearized IT (LIT) scores in a 0-9 scale as described by Zhang et al. (2014), where a score of 0 to 5 was considered as resistant and 6 to 9 considered as susceptible. To determine the repeatability of stem rust test for the RIL population, we conducted correlation analysis using LIT scores between two reps for each race. The t-tests (least significant difference) were also conducted to detect the RIL lines that significantly differ from the parents. The statistical analysis was conducted by using the PROC GLM procedure in SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). The mean of the linearized IT (LIT) scores of two replications were used for the development of histograms and QTL analysis.

Genotypic analysis
Out of 226 RILs used for the phenotypic analysis, 188 RILs were randomly selected for the genotypic analysis to avoid the bias in the marker data set. DNA extraction of 188 RILs along with parental lines ND495, Largo, PI 268210, and Langdon was done according to the procedure described in Faris et al. (2000). For genotyping, the wheat Infinium 90 K SNP iSelect array (Wang et al. 2014) was used and whole-genome linkage maps were developed by using the MapDisto 1.8.2.1 software package (Lorieux 2012) with a logarithm of odds (LOD) cut-off value of 3.0, and mapping distances were measured using the Kosambi mapping function (Kosambi 1943). The order of steps used for the linkage map development was followed as described in Sharma et al. (2019a). Briefly, linkage groups were first identified and then followed by fixing the marker order within each group by using the command "order sequence." Next, "check inversions," "ripple order," and "drop locus" commands were used to generate robust linkage maps. For purposes of generating figures, linkage maps with few non-redundant loci were developed by using the software Mapchart 2.32 (Voorrips 2002).

QTL analysis
To detect genomic regions associated with stem rust resistance, a QTL analysis was conducted using QGENE (4.3.10) software (Joehanes and Nelson 2008) and the single-trait multiple interval mapping (MIM) method (Kao et al. 1999). Based on the MIM statistical model (Kao et al. 1999), we assumed that there are m QTL (Q 1 , Q 2 , . . . Q m ) for controlling resistance to stem rust in the RIL population. The resistance phenotype value Y for a RIL, i, can be related to the m putative QTL by the model (Kao et al. 1999).
where m is the mean, x ij is coded as 1 =2 (Q j Q j ) or -1 =2 (q j q j ) for the genotype of Q j , a j is the main effect of Q j , and w jk is the epistatic effect between Q j and Q k , d jk is the indicator for epistasis between Q j and Q k , and e i is the error that is assumed to follow N(0, r 2 ). The LOD value 3.0 was set as the cut-off for the QTL detection. After identification of the gene-associated regions, simple sequence repeat (SSR) markers from marker sets BARC (Song et al. 2005), CFD (Guyomarc'h et al. 2002;Sourdille et al. 2003;Somers et al. 2004), WMC (Somers et al. 2004), and GWM (Rö der et al. 1998) were further used to map the specific chromosomes. Four (cfd15, cfd61, cfd72, and wmc429) and three (barc18, gwm388, and wmc154) SSR markers were mapped on chromosomes 1 D and 2B, respectively. For the Sr46 gene region, 10 previously known SSRs (barc124, barc95, cfd36, cfd43, gwm102, gwm210, gwm261, gwm455, wmc112, and wmc25) were mapped on chromosome arm 2DS. In addition, three SSRs (Xrwgs46, Xrwgs47, and Xrwgs49) developed based on reference genome sequences were also mapped ( Table 2). The primers of these markers were designed using the Primer-BLAST suite (http://www.ncbi.nlm.nih.gov/tools/primer-blast/, last accessed June 15, 2021 ) based on sequences within the Sr46 region of chromosome arm 2DS in Ae tauschii (AL 8/78) and hexaploid wheat (Chinese Spring) (IWGSC 2018). The SSR genotyping assays were performed using 6% non-denatured poly-acrylamide gels as described in Saini et al. (2018).

Data availability
The plant materials are available upon request. All data necessary for confirming the conclusions of the article are present within the article, figures, tables, and supplementary files. Supplemental material is provided at figshare: https://doi.org/10. 25387/g3.14450454. Supplementary File S1 contains IT and LIT scores of all lines. Supplementary Files S2-S4 present the results of LSD tests for mean LIT scores of RILs and their parental lines tested with Pgt race TTKSK, TRTTF, and TTTTF, respectively. Supplementary File S5 contains whole-genome linkage maps and Supplementary File S6 contains genotypic data for 188 RILs genotyped with wheat Infinium 90 K SNP iSelect array and SSR markers.

Results
Stem rust screening showed that Largo exhibited low infection types to Pgt races TTKSK (IT 2), TRTTF (ITs ;2 À and ;12 À in replicates 1 and 2, respectively), and TTTTF (IT 22 À ) ( Table 3). ND495 was susceptible to TTKSK (IT 3 þ ), moderately susceptible to TRTTF (ITs 31 þ and 1 þ 3 À in replicates 1 and 2, respectively), and resistant to TTTTF (IT ;3 and 0;13 in replicates 1 and 2, respectively). The Pearson correlation coefficients between the two replications for TTKSK, TRTTF, and TTTTF were 0.90, 0.83, and 0.95, respectively, which were highly significant (P < 0.0001), indicating high repeatability of the two replicates in the tests with each race. Therefore, the mean linearized infection type (LIT) scores from both reps for each race were used in the subsequent analysis.
The mean LIT scores of the RIL population for race TTKSK ranged from 4.5 to 9.0, with Largo and ND495 scoring 5.0 and 9.0, respectively ( Figure 1A and Supplementary Files S1 and S2). There was no significant transgressive segregation detected even though one line (NL025) had slightly increased levels of resistance over Largo (Supplementary File S2). For TRTTF, the RILs had mean LIT scores that ranged from 0.0 to 9.0 with Largo and ND495 scoring 1.0 and 5.0, respectively ( Figure 1B). Twelve lines showed increased levels of resistance (mean LITs 0.0 and 0.5) over Largo, but the increases were not significant (P 0.05). However, 22 RILs (mean LITs 7.0-9.0) were significantly more susceptible than ND495 (Supplementary File S3), indicating the presence of transgressive segregation in the population. For reactions to TTTTF, the RIL population had mean LIT scores also ranging from 0.0 to 9.0 even though both parents were in the resistant range, with ND495 being more resistant than Largo ( Figure 1C and Supplementary Files S1 and S4). In the population, 35 lines (mean LITs 0.0-1.0) had significantly lower mean LIT scores than ND495 (2.5), whereas 44 lines (6.5-9.0) had significantly higher mean LIT scores than Largo (5.0), indicating a strong transgressive segregation for resistance to TTTTF in the population.
Linkage maps were developed for the entire genome with 8203 (90 K SNP þ SSR) markers representing 1739 loci across the 21 chromosomes and map density ranging from 0.9 cM/locus for chromosome 1B to 4.4 cM/locus for chromosome 4D (Table 4 and  Supplementary Files S5 and S6). Two QTL regions associated with TTKSK resistance were identified on chromosome arms 2DS and 6AL designated as QSr.rwg-2D and QSr.rwg-6A, respectively (Table  5 and Figure 2). The QSr.rwg-2D QTL was positioned at 4 cM, flanked by Xrwgs46 and IWB43851 with a LOD value of 52.4 and explained 62.1% of the phenotypic variation (R 2 Â 100). This region of chromosome arm 2DS is known to carry Sr46. The second TTKSK-associated QTL, QSr.rwg-6A, was positioned at 98 cM and flanked by IWA441 and IWB51469. It had a LOD value of 23.5 and explained 21.3% of phenotypic variation. The gene Sr13 is known to lie within this genomic region. Both QSr.rwg-2D and QSr.rwg-6A had positive additive values of 1.2 and 0.7, respectively, indicating that TTKSK resistance was derived from Largo. For TRTTF, four QTL were identified on chromosome arms 2BS, 2BL, 2DS, and 6AL, designated as QSr.rwg-2B.1, QSr.rwg-2B.2, QSr.rwg-2D, and QSr.rwg-6A, respectively (Table 5 and Figure 2). QSr.rwg-2B.1 (42.0 cM) was flanked by IWB7072 and IWB2380 with a LOD value of 22.9 and it explained 33.3% of the phenotypic variation. The second TRTTF-specific QTL, QSr.rwg-2B.2 (74.0 cM), was identified on chromosome arm 2BL and was flanked by IWB71742 and IWB73196. This QTL had a LOD value of 4.0 and explained 16.2% of phenotypic variation. The third TRTTFassociated QTL, QSr.rwg-2D (LOD ¼ 3.5), was located on chromosome arm 2DS and was similar to the TTKSK QTL located in the Sr46 region, however, its effect for TRTTF was less compared with TTKSK with an explained 3.6% of phenotypic variation. Likewise, the fourth QTL, QSr.rwg-6A, also coincided with the TTKSK and TRTTF QTL and explained 18.5% of the phenotypic variation for TRTTF resistance and had a LOD value of 13.8. The positive additive values for all the QTL regions suggest that resistance was derived from Largo (Table 5).
A total of four QTL were identified for resistance to Pgt race TTTTF on chromosome arms 1DS, 2BS, 4AL, and 6AL and were designated as QSr.rwg-1D, QSr.rwg-2B.1, QSr.rwg-4A, and QSr.rwg-6A, respectively (Table 5 and Figure 2). Among these four QTL, only the QSr.rwg-4A associated resistance was derived from ND495, whereas all others were derived from Largo. The QSr.rwg-1D (LOD ¼ 7.8) was located at 16.0 cM and flanked by IWB22674 and IWB31245, explaining 16.8% of phenotypic variation. QSr.rwg-2B.1 was located at 38 cM and flanked by IWA413 and IWA2571. This QTL explained 4.6% of phenotypic variation and was adjacent to the TRTTF QTL located at 42 cM on chromosome arm 2BS. QSr.rwg-4A (LOD ¼ 6.6), which explained 16.1% of the phenotypic variation, was identified on chromosome arm 4AL at 114 cM and flanked by IWB9431 and IWB5461 located in the region known to be associated with Sr7. The QSr.rwg-6A region was common among the three Pgt races tested in this study, and for TTTTF it has maximum LOD at position 100 cM (distorted from TTKSK and TRTTF QTL peak). It explained 17.0% of the phenotypic variation and had a LOD value of 8.7.
For the three Pgt races used in this study, the QTL QSr.rwg-2B.2 located on chromosome arm 2BL was only effective against TRTTF (Figure 2). There have been several Sr genes reported on chromosome arm 2BL (Sr9, Sr16, Sr28, and Sr883-2B) (McIntosh 1995;Hiebert et al. 2010;Sharma et al. 2019b). For Sr9, seven alleles have been identified: Sr9a, Sr9b, Sr9d, Sr9e, Sr9f, Sr9g, and Sr9h (Green et al. 1960;Knott 1966;McIntosh and Luig 1973;Loegering 1975;Rouse et al. 2014). Among all these reported genes and their alleles, Sr9e is known to be present in many durum wheat varieties including Langdon (Luig 1983;Singh et al. 1992), and it has a minor effect against TRTTF (Saini et al. 2018). Sr16 is not effective against TRTTF , and the Sr28 gene is known to confer resistance against TTKSK, but QSr.rwg-2B.2 did not condition resistance to this race. Based on the consensus map location of SNPs, the Sr883-2B gene reported by Sharma et al. (2019b) is located some distance from the QSr.rwg-2B.2 (Wang et al. 2014). Sr9h is effective against Pgt race TTKSK , but QSr.rwg-2B.2 resistance was not associated with TTKSK and TTTTF. Because Largo carries the TRTTF-effective gene Sr9e from Langdon and the other genes known to reside on 2BL can essentially be ruled out, it is most certain that the Sr gene underlying QSr.rwg-2B.2 is Sr9e.
The TTKSK-and TRTTF-specific QTL QSr.rwg-2D was located near the distal end of the chromosome arm 2DS, which is a region known to harbor Sr32 (Mago et al. 2013) and Sr46 (Yu et al. 2015;Arora et al. 2019). Both genes are effective against TTKSK and TRTTF (Olivera et al. 2012). The Sr32 gene was originally derived from Ae. speltoides Tausch (Friebe et al. 1996) and should not be the gene underlying the QTL QSr.rwg-2D because this gene had not been introduced into any of the parental lines (i.e., Langdon, ND495 and Largo). QSr.rwg-2D was located proximal to gwm210 and distal to cfd36 (Figure 2 and Supplementary File S5), which corresponds to the Sr46 location based on the map developed in the F 2 population derived from the Ae. tauschii cross CIae 25 Â AL8/78 (Yu et al. 2015), suggesting that Sr46 is likely the gene underlying QSr.rwg-2D. Sr46 is effective against TTKSK, TRTTF, and TTTTF (Yu et al. 2015), however, in the current analysis the QSr.rwg-2D was not associated with the TTTTF resistance. Based on this phenotypic difference, we speculate that the Sr gene underlying QSr.rwg-2D derived from Largo may be a different allele of Sr46. However, Sr46 was mapped using the diploid Ae. tauschii F 2 population, whereas the QSr.rwg-2D was identified in the  hexaploid RIL population. Because genomic interaction in allopolyploid wheat often causes the reduction or suppression of resistance of some Sr genes (Hiebert et al. 2020), it is also possible that the different reactions of Sr46 and QSr.rwg-2D to TTTTF were caused by different ploidy levels. Therefore, further study is needed to determine the identity of the gene for QSr.rwg-2D. Among all the QTL identified in this study, only QSr.rwg-4A positioned on chromosome arm 4AL was derived from ND495 ( Figure 2). This QTL conditioned resistance against Pgt race TTTTF and was located in the physical region known to be associated with the Sr7 locus and a TTKSK-effective gene SrND643 (Basnet et al. 2015;Saini et al. 2018). Because QSr.rwg-4A is not resistant to TTKSK, SrND643 is apparently not the candidate gene for QSr.rwg-4A. To date, two alleles, Sr7a and Sr7b, have been reported at the Sr7 locus (McIntosh et al. 1995). Sr7b is not effective against TTTTF, whereas Sr7a is effective against TTTTF and it is nearly fixed in the wheat breeding germplasm in the Northern Great Plains (  2018). As QSr.rwg-4A was located to the Sr7 region and has resistance to TTTTF, most likely Sr7a is the underlying gene for this region.
The QTL QSr.rwg-6A derived from Langdon is located on chromosome arm 6AL, which carries three known TTKSK-effective Sr genes, Sr13, Sr26, and Sr52. Among these known genes, Sr26 and Sr52 were originally transferred into wheat from wild species Thinopyrum ponticum (Podp.) Barkw. & D.R. Dewey [Agropyron elongatum (Host) Beauv.] (Knott 1961;Dundas et al. 2007) and Dasypyrum villosum (L.) Candargy (Qi et al. 2011), respectively. Because Sr26 and Sr52 have not been transferred into durum wheat Langdon, they can be ruled out as the causal gene for QSr.rwg-6A. This QTL was known to be physically associated with Sr13 and effective against all three Pgt races used in the current analysis (McIntosh 1995;Simons et al. 2011;Periyannan et al. 2014b;Zhang et al. 2017;Gill et al. 2021). Gill et al. (2021) identified Sr13 as the causal gene for the stem rust resistance in an accession PI 387696 of T. turgidum subsp. carthlicum (Neyski) Á . Lö ve & D. Lö ve. By comparing the QSr.rwg-6A region to the Sr13 region in the study by Gill et al. (2021), we found that six SNP markers (IWB61092, IWB50538, IWB7048, IWB28546, IWB37898, and IWB34398) in the two regions were in common in both 90 K SNPbased high-density maps. Two of the markers, IWB37898 and IWB34398, that are tightly linked to Sr13 in the study by Gill et al. (2021) are also located in the QSr.rwg-6A region. Zhang et al. (2017) identified Sr13 as a coiled-coil nucleotide-binding leucinerich repeat (NLR) gene. They identified three resistant (R1-R3) and 10 susceptible (S1-S10) haplotypes of this gene based on the reactions to TTKSK and designated R1/R3 and R2 as Sr13a and Sr13b, respectively, based on their resistant and susceptible reactions to JRCQC. Gill et al. (2021) re-designated the R1 and R3 haplotypes as Sr13a and Sr13c, respectively, based on their susceptible and resistant reactions to TCMJC. Among different diploid, tetraploid, and hexaploid wheat accessions that have been characterized for these haplotypes, Langdon was categorized as having the R3 haplotype of Sr13 (Zhang et al. 2017;Gill et al. 2021). Because Langdon is present in the Largo background, Sr13c is the gene underlying QSr.rwg-6A.
In summary, we mapped three known Sr genes Sr9e (QSr.rwg-2B.2), Sr13c (QSr.rwg-6A), and Sr7a (QSr.rwg-4A) in the ND495 Â Largo RIL population. In addition, there were three other genomic regions associated with stem rust resistance. Of these three Sr regions, QSr.rwg-1D (likely a new gene, temporarily designated as SrLargo1D) and QSr.rwg-2D (possibly a new allele of Sr46) were derived from the Ae. tauschii parent of Largo. The QSr.rwg-2B.1 derived from Langdon is located in a region with no known Sr genes. Therefore, QSr.rwg-2B.1 is probably associated with a new Sr gene (temporarily designated as SrLangdon2B) against Pgt races TRTTF and TTTTF. As no evaluation was previously performed to identify the Sr gene(s) in ND495, identification of Sr7a (QSr.rwg-4A) in this study suggests that ND495 carries Sr gene(s) with minor effects. The identification of these Sr genes in Largo will guide the future efforts to stack multiple resistant genes. Among the Ug99-effective Sr genes, both Sr13c and Sr46 had resistance to a broad spectrum of Pgt races. However, they are among a few genes from the primary gene pool that have not been utilized or deployed in modern bread wheat germplasm. Several NIL lines such as NL143, NL159, and NL193 with resistance to the three Pgt races were found to carry all the six genes, they may serve as the donors for simultaneously introducing Sr13c, Sr46, and four other genes into adapted bread wheat germplasm and varieties. Because Largo has resistance to other fungal diseases, the 90 K SNP marker data set and the high-density linkage map developed in this study will be useful for identifying and mapping the genes controlling other agronomically important traits derived from durum wheat (Langdon), bread wheat (e.g., ND495), and Ae. tauschii (PI 268210).