QTL mapping of adult plant resistance to stripe rust and leaf rust in a Fuyu 3/Zhengzhou 5389 wheat population

Tkele Welu Gebrewhi, Peipei Zhng, Yue Zhou, Xiocui Yn, Xinchun Xi,Zhonghu He,e, Dqun Liu,, Zifeng Li,

aDepartment of Plant Pathology, College of Plant Protection, Hebei Agricultural University, Biological Control Center for Plant Diseases and Plant Pests of Hebei, Baoding 071001, Hebei, China

bCollege of Agriculture, Aksum University-Shire Campus, Tigray, Ethiopia

cBaoding University, Baoding 071001, Hebei, China

dInstitute of Crop Sciences, National Wheat Improvement Center, Chinese Academy of Agricultural Sciences, Beijing 100081, China

eInternational Maize and Wheat Improvement Center (CIMMYT) China Office, c/o CAAS, Beijing 100081, China

ABSTRACT Stripe or yellow rust (YR) and leaf rust (LR) cause large losses in wheat production worldwide. Resistant cultivars curtail the levels of losses. The present study aimed to identify quantitative trait loci(QTL)for YR and LR resistance in 147 F2:6 recombinant inbred lines (RIL) derived from the cross Fuyu 3/Zhengzhou 5389. The RIL population and parents were genotyped with the Wheat55K single nucleotide polymorphism (SNP) array and simple sequence repeat (SSR) markers. All materials were also phenotyped for YR severity at Mianyang in Sichuan province and Baoding in Hebei province in the 2015/2016,2016/2017,and 2017/2018 cropping seasons, and LR severity at Zhoukou in Henan province and at Baoding in 2017/2018.Eleven QTL for YR resistance and five for LR resistance were detected using inclusive composite interval mapping (IciMapping). Four of these QTL on chromosomes 1BL,2BS,3AL, and 5AL conferred resistance to both YR and LR. The QTL on 1BL was Lr46/Yr29, and that on 7BL might be Lr68. The QTL on chromosome 2BS was detected at a similar position to previously detected loci. QYr.hebau-3AL/QLr.hebau-3AL, QYr.hebau-5AL/QLr.hebau-5AL,QYr.hebau-7DL,QYr.hebau-4BS,QYr.hebau-6DL,and QYr.hebau-2AS are likely to be new. An SSR marker for QYr.hebau-7DL was developed and validated in a diverse wheat panel from China, suggesting effectiveness in different genetic backgrounds. These QTL with closely linked SNP and SSR markers could be useful for marker-assisted selection in wheat breeding programs targeting durable resistance to both diseases.

1. Introduction

Wheat stripe or yellow rust (YR) and leaf rust (LR), caused by Puccinia striiformis f. sp. tritici (Pst) and P. triticina (Pt), respectively, are important diseases of common wheat (Triticum aestivum) worldwide. Epidemics of these diseases lead to devastating losses under favorable conditions. Wheat is vulnerable to these diseases because of race changes and aerial dispersal over long distances[1].

Stripe rust can cause 100% yield loses under extreme conditions following initial infection at an early growth stage[2]. Epidemics in China begin mostly in autumn-sown winter wheat areas in the southwest and northwest, and in springsown spring wheat regions in the northwest and in southwestern provinces of Sichuan, Shanxi and Gansu, which represent important over-summering areas for survival of the pathogen [3,4].

Leaf rust can cause up to 40% yield losses under favorable situations[5,6].Severe LR epidemics were recorded in China in 1969, 1973, 1975, 1979, and 2012, causing substantial losses[7,8]. In 2012 and 2015 yield losses were recorded in some regions of Gansu,Sichuan,Shanxi,Henan,and Anhui[5,9].LR has become increasingly important in major wheat production regions of Northern China, possibly because of climate change.

Resistant cultivars are the most economically and environmentally friendly strategy for controlling both rusts, but the problem is lack of durability of a particular source of resistance when it is widely deployed. Seedling or all-stage resistance and adult plant resistance (APR) are major categories used to describe the responses of cereal plants to rusts[2,10] and some other diseases. All-stage resistance is detectable at the seedling stage and continues to provide protection throughout the entire growth cycle. However, this kind of resistance tends to be more frequently overcome by virulent races of the respective pathogens [11]. APR is expressed during the post-seedling growth stages and can be race specific or race non-specific.Race non-specific resistance tends to be quantitatively inherited and is more likely to be durable. It frequently confers partial or slow-rusting resistance[1,12]and may confer an inadequate level of protection to avoid at least some losses.To date,79 LR resistance and 80 YR resistance genes have been identified and catalogued in wheat [13]. In addition, there are reports of ＞249 LR [14] and 327 YR [15]quantitative trait loci(QTL).

The search for rust resistance genes is an ongoing process due to the ever-changing pathogen populations. Researchers must also be aware of the responses of contemporary cultivars and breeding materials to the prevailing pathogen population in order for application of effective selection strategies.The current consensus of opinion is that resistance should be based on combinations of genes that are best achieved by marker-assisted breeding schemes.Thus,closely linked DNA markers need to be identified and used not only for breeding but also for basic research aimed at understanding the nature of different types of resistance and isolation of the underlying resistance genes. Over the past 20 years,simple sequence repeat (SSR) markers became the preferred marker type for linkage mapping due to the advantages of codominance,accuracy,high repeatability,high polymorphism,chromosome specificity, and ease of handling [16]. With development of comprehensive genetic maps QTL mapping became a widely used method to dissect the genetic variation underlying quantitative variation, including disease resistance [1]. QTL mapping is a valuable preliminary way to detect chromosomal regions associated with resistance;these regions are delimited by linked markers. The challenge then was to find further markers to reduce the size of the QTL region and to increase the precision of analysis and therefore use of the resistance gene(s). Single nucleotide polymorphisms (SNP) with their almost unlimited numbers provided the next opportunity.

High-density SNP gene-chip technologies provide a superior approach for QTL mapping due to less errors in evaluation, higher accuracy and particularly, higher density than SSR markers[17].The Wheat55K and Wheat660K arrays were both designed by Chinese Academy of Agricultural Sciences researchers and commercialized by Affymetrix. The Wheat660K is highly effective for constructing high-density wheat genetic maps. However, the Wheat55K (Affymetrix Axiom Wheat55)is more important for QTL discovery because all 53,063 tags were specifically selected from the 660 K array.The tags are evenly distributed across all wheat chromosomes such that each chromosome has about 2600 SNPs with an average genetic distance of 0.1 cM and an average physical distance of ＜300 kb[18].

Fuyu 3 (Yumai 18/80(6)-3-3.10) is a spring wheat selected from Yumai 57 at the Agricultural Science and Technology Experimental Station in Pingyu County, Henan province. It is susceptible to some Pt races at the seedling stage but is highly resistant to both rusts at the adult stage [19]; however, the genetic basis of its resistance remained unknown.Zhengzhou 5389 is highly susceptible to both diseases at all growth stages.The objective of the present study was to map QTL for APR to YR and LR using a recombinant inbred line (RIL) population derived from the cross Fuyu 3/Zhengzhou 5389.

2. Materials and methods

2.1.Plant materials and pathogens

One hundred and forty-seven F2:6RIL were used to map QTL for resistance to YR and LR. Zhengzhou 5389 and Mingxian 169, both highly susceptible to LR and YR, were used as susceptible checks in disease nurseries. Isolates of Pst races CYR32 and CYR33 and/or Pt races THJL, THJC,and PHGP were used in the field trials.

2.2.Field trials

All 147 RILs and parents were grown at Mianyang(31.48°N and 104.68°E, Sichuan province) and Baoding (38.85°N and 115.47°E) in the 2015/2016 and 2017/2018 cropping seasons(hereafter referred as YR2016My, YR2017My and YR2018My,and YR2017Bd and YR2018Bd) for evaluation of YR severities.Similarly, LR trials were conducted in Zhoukou (33.80°N and 114.53°E, Henan province) and Baoding in 2017/2018 (hereafter,LR2018Zk and LR2018Bd).Inoculations were carried out at the beginning of January at Mianyang, around April 1 at Zhoukou and April 10 at Baoding. The field trials were conducted as randomized complete blocks n with two replications. Plots consisted of single 1.5 m rows spaced 0.25 m apart, and approximately 60 seeds were planted in each row.Highly susceptible line Zhengzhou 5389 for LR trials or Mingxian 169 for YR trials were planted after every ten rows as checks and to aid the spread of urediniospores. Spreader rows of Zhengzhou 5389 for LR and Mingxian 169 for YR were planted perpendicular and adjacent to the test rows. YR and LR epidemics were initiated by spraying the spreader rows at the tillering stage with aqueous suspensions of equal amounts of urediniospores of each race to which a few drops of Tween 20 (0.03%) were added. Disease severities as percentages of leaf area covered with uredinia were scored twice in each environment according to the modified Cobb scale[20].Field data for YR from Baoding in 2016 was excluded from the final analysis due to inadequate disease development. Final disease severity (FDS) data were used for QTL analysis. Phenotypic correlation coefficients between FDS in different environments were calculated by SAS software(SAS Institute,Cary,NC).

2.3. Molecular genotyping

Genomic DNA was extracted from non-infected seedling leaves of the RILs and parents by the CTAB method [21].DNA concentrations were determined using a Thermo Scientific NanoDrop 2000.All RILs and parents were genotyped with the Affymetrix Wheat55K iSelect SNP Array at CapitalBio Technology Corporation, Beijing (http://www.capitalbio.com).Monomorphic and SNP loci with ＞10% missing values, vague SNP calling, and minor allele frequencies ＜5% were excluded from further analysis [22]. Twenty-seven SSR markers were also used to genotype the entire population. PCR for SSR markers was performed following Helguera et al. [23].

Redundant SNP markers displaying identical segregation patterns were deleted using the “BIN” tool in inclusive composite interval mapping (IciMapping) V4.1 [24], and a single marker was selected to denote each bin on the basis of the minimum amount of missing data(arbitrary choice when all data were equal). The filtered SNP markers were used to construct linkage maps with the “MAP” tool in IciMapping V4.1. The algorithm of group order was “nnTwoOpt” and the Kosambi function was used to calculate map distances from recombination frequencies [25]. The criterion and window size for Rippling were “SARF” and “5”, respectively. Linkage maps were graphically visualized with MapChart 2.3[26].

2.4.Construction of linkage maps and QTL analysis

QTL mapping was performed using the inclusive composite interval mapping(ICIM)method in IciMapping software[24].A logarithm of odd (LOD) threshold of 3.0 was set to declare significant QTL through 1000 permutations at P ＜0.01. Stepwise regression was used to detect the percentages of phenotypic variance explained (PVE) by individual QTL and additive effects at the LOD peaks.QTL identified in individual environments with overlapping 20 cM intervals were considered identical [27]. The identified flanking sequences of all SNP and SSR probes were subjected to BLAST against the Chinese Spring wheat reference sequence (IWGSC RefSeq v1.0,https://urgi.versailles.inra.fr/blast_iwgsc/blast.php)to hit their physical positions.

2.5. Designing SSR primer and QTL validation using diverse wheat panels

The genomic sequence between flanking SNP markers AX-110938697 and AX-111591751 for QYr.hebau-7DL was aligned from the Chinese Spring wheat genome sequence (IWGSC,http://wheat-urgi.versailles.inra.fr/) through a BLAST search to design SSR primer pairs.The SSR primers were designed by Batch Primer 3 software (http://probes.pw.usda.gov/cgi-bin/batchprimer3/batchprimer3.cgi). Primers were premeditated according to:(a)primer length from 18 to 28 bp with 23 as the optimum;(b)PCR product size from 150 to 300 bp;(c)annealing temperature from 55°C to 65°C with an optimum of 60°C;and(d) GC contents from 45% to 55%, with 50% as optimum. The primer pairs were synthesized by Sangon(Shanghai).

PCR followed the protocol developed by Helguera et al.[23].Amplification of genomic DNA was carried out in 10 μL reaction volumes containing 5 μL 2× Taq PCR Master Mix(Tiangen Biochemical Incorporation,Beijing),3 μL ddH2O,1 μL(4 mol μL?1)of primer,and 1 μL(4 ng μL?1)of template DNA.PCR products for SSR primers were separated in 6% polyacrylamide denaturing gels that were silver-stained for band detection.

t-Tests were conducted to compare presence (+QTL) and absence(?QTL)of QYr.hebau-7BL and QYr.hebau-7DL effects on stripe rust FDS data from 2016/2017 and 2017/2018 using a diverse panel of 25 genotypes from China(Table S1).

Table 2-Pearson correlation coefficients (r) for two-way comparisons of stripe rust and leaf rust responses in all environments.

Table 3-QTL for final disease severity for stripe rust and leaf rust identified by IciMapping of data from the Fuyu 3/Zhengzhou 5389 RIL population.

3. Results

3.1. Phenotyping of stripe rust and leaf rust responses

The FDS for YR and LR of susceptible controls Mingxian 169 and Zhengzhou 5389 ranged from 80% to 100% across environments. The mean FDS of Fuyu 3 and Zhengzhou 5389 for YR were 5.2% and 88%, respectively, across five environments and for LR were 5.0% and 97.5%, respectively, in two environments. The average FDS of the RILs across environments ranged from 39.9% to 59.9% for YR and 50.1% to 51.9% for LR (Table 1). The minimum and maximum FDS of the RIL population across all environments ranged from 1% to 100%,respectively (Table 1). The frequency distributions for YR and LR FDS in each environment were continuous but skewed towards resistance(Fig.S1).

FDS scores for YR across environments YR2016My,YR2017My, YR2018My, and YR2018Bd were significantly correlated with coefficients of 0.50 to 0.72 (P ＜0.001) (Table 2).However,the coefficients of correlation for YR2017Bd were from 0.39 to 0.49(P ＜0.05)(Table 2),probably due to marginal environmental conditions for YR development. The temperature at Baoding is comparatively high in mid-May (30 °C)which is not optimal for YR development. The FDS scores for LR in two environments were also significantly correlated with a coefficient of 0.49 (P ＜ 0.001). The coefficients of correlation between YR and LR disease scores were 0.39 to 0.59 across trials, indicating the possibility of genes with pleiotropic effects.

3.2. Linkage map construction

After removing SNPs that were monomorphic, with ＞10% missing rate, and with distorted segregation, 6185 polymorphic markers,comprising 2005 for the A genome,2170 for the B genome and 2010 for the D genome,were used to construct linkage maps of 1767.4,1796.6,and 1654.9 cM for the A,B,and D genomes,respectively.The marker densities were 0.88,0.83,and 0.82 cM/marker, respectively (Table S2).

3.3.QTL mapping of rust resistance

3.3.1.Potential pleiotropic rust resistance QTL

Four potential pleiotropic rust resistance QTL from Fuyu 3 were identified (Table 3, Fig. 1). QYr.hebau-1BL/QLr.hebau-1BL was detected in three environments,explaining 8.7%-21.6% of the phenotypic variance; QYr.hebau-2BS/QLr.hebau-2BS accounted for 10.5%-12.6% and 13.5%-14.5% of phenotypic variances for YR and LR, respectively, across four environments; QYr.hebau-3AL/QLr.hebau-3AL with relatively lower effects on disease response explained 4.9%-8.9% of the phenotypic variances in YR2016My and LR2018Bd, respectively, and QYr.hebau-5AL/QLr.hebau-5AL explained 5.1%-5.8% of the phenotypic variances in two environments.

3.3.2.QTL mapping of APR to stripe and leaf rust

Seven additional QTL for YR and one QTL for LR were detected in the population and designated QYr.hebau-1AL, QYr.hebau-2AS, QYr.hebau-4AL, QYr.hebau-4BS, QYr.hebau-6DL, QYr.hebau-7BL,and QYr.hebau-7DL and QLr.hebau-5DL,respectively(Table 3, Fig. 1). Most resistance alleles were contributed by Fuyu 3,whereas that on 4AL was from Zhengzhou 5389. No QTL was previously reported in Zhengzhou 5389.

QTL QYr.hebau-7BL and QYr.hebau-7DL were stably in all seasons in the Mianyang and Baoding trials,explaining 2.9%-21.9% and 16.4%-28.4% of the phenotypic variance, respectively (Table 3, Fig. 1). QYr.hebau-7BL and QYr.hebau-7DL reduced stripe rust by 16.9%-22.5% and 14.5%-16.4% in 2017 and 2018 cropping seasons, respectively (Table S1). QTL QYr.hebau-4BS and QYr.hebau-6DL, were identified in three environments and accounted for 9.6%-16.2% and 12.5%-22.6% phenotypic variance, respectively. QYr.hebau-2AS was detected in two environments, explaining 4.5%-9.3% of the phenotypic variance. QYr.hebau-1AL and QYr.hebau-4AL were each identified in a single environment with minor effects explaining 2.8% and 3.5% of phenotypic variance,respectively.

Table 4-Comparison of the physical positions of QTL identified in the present study with those reported previously.

QLr.hebau-5DL for LR resistance was identified in LR2018Zk with a LOD value of 3.7 and explained 9.2% of the phenotypic variance.

3.3.3.SSR marker development and validation in Chinese wheat panels

A pair of SSR primers temporarily named HEBAU-7DL (forward primer: 5′-GTTGTTTTGTCAAACTCTTGC-3′; reverse primer: 5′-GCAACTCACAGTCAAAATACC-3′) was designed for detection of QYr.hebau-7DL. The primer length, annealing temperature (°C), GC (%) content and PCR product size were 21,55,45,and 150,respectively.HEBAU-7DL was used to test a panel of 25 diverse Chinese wheat accessions,including Fuyu 3 (Table S3) to validate QYr.hebau-7DL and determine its diagnostic value for marker-assisted selection (MAS). Seventeen accessions carried the same QYr.hebau-7DL allele as Fuyu 3.The wheat panels contained QYr.hebau-7DL allele were also found with low level of FDS to stripe rust in 2017 and 2018 cropping seasons (Table S1). Therefore, this marker has a great value in MAS for YR screening of large set of materials and/or segregating materials. Wheat panels contained both QYr.hebau-7DL and QYr.hebau-7BL alleles showed better resistance than those panels with either QYr.hebau-7DL or QYr.hebau-7BL allele during both cropping seasons.

4. Discussion

The present research using a RIL population was a continuation of that of Qi et al.[19]who conducted a study of F2:3lines from the same cross using SSR markers.The objective was to repeat the analysis with increased precision.

Qi et al. [19] reported three SSR-based LR-resistance QTL(QLr.hbu-1BL.2, QLr.hbu-2BS.2, and QLr.hbu-7BL) using F2:3lines of Fuyu 3/Zhengzhou 5389, and the resistance alleles were from Fuyu 3.Using F2:6RILs from the same cross we identified 11 and 5 APR genes for YR and LR, respectively. Fuyu 3 contributed resistance alleles at pleiotropic loci QYr.hebau-1BL/QLr.hebau-1BL and QYr.hebau-2BS/QLr.hebau-2BS, that involved the same SSR loci as for QLr.hbu-1BL.2 and QLr.hbu-2BS.2.The QTL for LR resistance on chromosome 7BL was not confirmed, but QYr.hebau-7BL was located at the same position as QLr.hbu-7BL. The present results should be more reliable than those of Qi et al. [19] because of near homozygosity (F2:6RILs) and inclusion of SNPs spanning the whole wheat genome in addition to SSRs for genotyping.

4.1. QYr.hebau-1BL/QLr.hebau-1BL

Several QTL have been reported on chromosome 1B [5],including the pleiotropic Lr46/Yr29 locus [28]. QLr.caas-1BL[29], QLr.pser-1BL [30], QLr.hbu-1BL.2 [19], and QLr.csiro-1BL [31]are among QTL identified on chromosome 1BL. Based on the positions of flanking markers QYr.hebau-1BL/QLr.hebau-1BL should be Lr46/Yr29 (Table 4). Furthermore, Fuyu 3 was also positive for the Lr46/Yr29 marker csLv46g22 [28]. The resistance genes reported on chromosome 1BL are widely distributed in the CIMMYT wheat germplasm [32] and have persisted effective for ＞60 years [33]. Chapingo, one of the parents of Fuyu 3 was introduced from Mexico [19] and probably contained QYr.hebau-1BL/QLr.hebau-1BL.

4.2.QYr.hebau-2BS/QLr.hebau-2BS

Several QTL for YR have been identified on chromosome 2BS.QYrid.ui-2B.1 in IDO444[34], QYr.sgi-2B.1 in Kariega[35], QYrlo.wpg-2BS in Louise [36], QYr2B in Opata 85 [37], QYr.inra-2BS in Renan[38],QYrlu.cau-2BS1 and QYrlu.cau-2BS2 in Luke[39]and QYr.cim-2BS in Francolin [40], mapped at 73.6, 100.8, 172.7,47.6, 1.2, 20.6, 100.8, and 166.5 Mb physical positions,respectively (Table 4). QYr.hebau-2BS/QLr.hebau-2BS(123.7-133.9 Mb)appeared to represent a different position.

Genes Lr48 [41], Lr35 [42], and QLr.hbu-2BS.2 [19] located at 91.8, 268.5, and 133.5 Mb, respectively (Table 4) were mapped on chromosome 2BS.It is likely that QYr.hebau-2BS/QLr.hebau-2BS is different from Lr48 and Lr35. Given that QLr.hbu-2BS.2[19] and QYr.hebau-2BS/QLr.hebau-2BS were identified in the same cross they should be the same.

4.3.QYr.hebau-3AL/QLr.hebau-3AL

QLr.fcu-3AL in TA4152-60 [43] was mapped at 709.4 Mb on chromosome 3AL (Table 4). Crossa et al. [44] identified a QTL associated with YR, LR and grain yield at 15.4 Mb on chromosome 3AL (Table 4). QYr.hebau-3AL/QLr.hebau-3AL for YR and LR resistance was located at 717.6-719.2 Mb near to QLr.fcu-3AL[43].

4.4.QYr.hebau-5AL/QLr.hebau-5AL

Five QTL for YR, QYr.caas-5AL.2 [29], QYr.caas-5AL [45], QYr.ucw-5AL [46], QYrtb.pau-5A [47], and QYr.cimmyt-5AL [48] were mapped on chromosome 5AL at 687.7, 678.2, 687.7-698.1,558.3-562.8, and 671.3 Mb, respectively (IWGSC RefSeq v1.0,https://urgi.versailles.inra.fr/blast_iwgsc/blast.php) (Table 4).In addition, one QTL QLr.cimmyt-5AL on chromosome 5AL contributed by Avocet was flanked by wPT-0373 and wPT-0837[48] closely linked to Vrn-A1 at 587.0 Mb. QYr.hebau-5AL/QLr.hebau-5AL at 436.1-441.1 Mb was likely represented a different potentially new gene.

4.5.QYr.hebau-7BL

Several APR genes/QTL have been mapped on chromosome 7BL; Yr52 in PI 183527 [49], Yr59 in PI 178759 [50], Yr79 in PI 182103 [51], YrZH84 in Zhou 8425B [52], Yr39 in Alpowa [53],Yr67 in C591 [54], and Lr68 in Parula [55] were identified at 732.4, 723.9, 750.6, 723.9-750.6, 604.7, 699.9-718.4, and 734.2 Mb, respectively (IWGSC RefSeq v1.0, https://urgi.versailles.inra.fr/blast_iwgsc/blast.php, Table 4). It is likely that QYr.hebau-7BL (712.3-723.2 Mb) is different from Yr52,Yr79, and Yr39, but could be one or other of Yr59, YrZH84, or Yr67. QYr.caas-7BL in SHA3/CBRD [29], QLr.csiro-7BL [31] and QLr.hbu-7BL [19] were identified at 557.0, 411.3, and 723.8 Mb,respectively(Table 4).Of these,QLr.hbu-7BL was considered to be Lr68(734.2 Mb)(Table 4).QYr.hebau-7BL was likely located at a similar position to QLr.hbu-7BL. Although there is no report to indicate that Lr68 is associated with resistance to YR, but the results of both the present study and Qi et al. [19]suggested a pleiotropic locus. The relationships among Yr59,YrZH84,Yr67,Lr68,and QYr.hebau-7BL require further analysis.

4.6. QYr.hebau-7DL

Yr33 conferring APR in Batavia was located on chromosome 7DL flanked by Xgwm437 and Xgwm11 [56] at 450.5 Mb (Table 4). QYr.hebau-7DL, identified in all YR environments, was flanked by AX-110938697 and AX-111591751(Table 3,Fig.1)at 591.0 and 597.1 Mb, respectively. QYr.hebau-7DL appears to be a novel QTL. SSR marker HEBAU-7DL was developed and validated in a diverse wheat panel (Table S3), could be useful in MAS to screen germplasm and identify the sources of YR resistance.

4.7. QYr.hebau-4BS and QYr.hebau-6DL

The stable QYr.hebau-4BS appears be new because no YR resistance gene/QTL was previously located on chromosome 4BS.QYr.ufs-6D with minor effect was located on chromosome 6DL at about 79.9 Mb (Table 4) by Agenbag et al. [57]. QYr.hebau-6DL at 464.6-472.0 Mb is clearly different.

4.8. QYr.hebau-2AS

Hao et al. [58] identified QYr.uga-2AS at about 28.2 Mb on chromosome 2AS (Table 4). Yr17 from Ae. ventricosa is on chromosome 2AS [59]. Moreover, tests with VENTRIUP/LN2 marker[59]indicated that Yr17 locus was not present in Fuyu 3 or Zhengzhou 5389 (data not shown). QYr.hebau-2AS at 36.85-36.87 Mb was quite close to QYr.uga-2AS.

5. Conclusions

Eleven QTL for YR resistance and five QTL for LR resistance were detected in this study. QYr.hebau-7BL and QYr.hebau-7DL were major QTL identified in all stripe rust environments. SSR marker for QYr.hebau-7DL, was developed and validated in a diverse panel of Chinese wheat genotypes. Average effects of these QTL for resistance to stripe rust severities in the wheat panels from China were significant in 2017 and 2018 cropping seasons. Potentially new QTL identified in this study must be validated in other populations and environments. The closely linked SNP and SSR markers can be used in MAS and pyramiding of APR genes in wheat breeding programs to improve YR and LR resistance.

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2019.09.013.

Acknowledgments

We thank Prof. R.A. McIntosh, University of Sydney, for the critical review of an early draft of this manuscript. This study was supported by the National Natural Science Foundation of China (31361140367, 31571662, and 31601299).