Wheat stripe rust, also known as yellow rust, caused by the air-borne fungus Puccinia striiformis f. sp. tritici (Pst), is considered the primary biotic threat to wheat production globally (Todorovska et al., 2009). In 2017, 88% of the world’s wheat production was susceptible to stripe rust, causing yield losses estimated to be 1 billion USD (Brandt et al., 2021), and a countrywide epidemic affected about 5.5 million ha of wheat in China (Zeng et al., 2022). It shows regional characteristics and occurs frequently in the southwestern and northwestern of China, especially in the southeastern Gansu, a stripe rust hotspot in northwestern of China (Bai et al., 2012). The management of stripe rust includes the deployment of resistant wheat and the use of fungicide, and breeding resistant wheat cultivars is the most effective and economic strategy for controlling stripe rust (Wen et al., 2008).

Genetic resistance to stripe rust in wheat can be broadly divided into two phenotypically, mechanistically, and genetically distinct categories: (a) seedling resistance (all-stage resistance, ASR), which is detected at the seedling stage but is also expressed at all developmental stages. It usually shows a major effect and is race specific (Roberts and Caldwell, 1970; Lin and Chen, 2007); (b) adult plant resistance (APR), which is expressed at the adult plant stages and appears to be durable. Some APR genes are less characterized and are difficult to select phenotypically (Ren R. et al., 2012).

To date, most identified stripe rust resistance genes conferring “seedling resistance” encode classic nucleotide-binding site leucine-rich repeat (NBS-LRR) R proteins that recognize effectors and trigger a defense response to resist disease. While it is easy to overcome the recognition of a single classical R gene via a single genetic variation in an avirulence gene (Steuernagel et al., 2016; Schwessinger, 2017). The major epidemics causing severe economic losses have occurred in various regions of the world including United States, China, and Australia (Nsabiyera et al., 2018) due to virulence on seedling resistance genes including Yr2, Yr6, Yr9, Yr17, and Yr27 (Wellings and McIntosh, 1990; Nsabiyera et al., 2018). Only a few seedling stripe rust resistance genes (i.e., Yr5 and Yr15) are still effective to some Pst races in China, which drives the demand for persistent disease-resistant cultivars.

Adult plant resistance is generally known as more durable than ASR, since a single genetic variation appears insufficient to overcome this type of resistance in the asexual stage of Pst (Schwessinger, 2017). In general, APR delays infection and production of spore, leading to slow rusting phenotypes instead of complete immunity. This type of resistance gene usually encodes allele-specific protein variants that are molecularly unrelated to NBS-LRR proteins. In the case of these genes, Yr18 and Yr46 encode two types of transporters (Krattinger et al., 2009; Moore et al., 2015). In addition, Yr36, a chloroplast-localized kinase WKS1 regulating, reactive oxygen species production (Ellis et al., 2014; Gou et al., 2017). Since the durability of R genes is usually defined by the global genetic diversity of the pathogen, the combination of APR genes with ASR genes is the best way for a wheat breeder to develop the durable stripe rust resistant wheat variety at all plant stages. To date, more than 83 Yr genes or alleles (Yr1–Yr83) have been formally named in wheat (Li et al., 2020; Draz et al., 2021). Most of them are race-specific ASR genes, and only a few are APR genes. Lr34/Yr18/Pm38/Sr57 (Dyck, 1991), Lr46/Yr29/Pm39/Sr58 (Singh et al., 1998), and Lr67/Yr46/Pm46/Sr55 (Herrera-Foessel et al., 2011) confer pleiotropic APR to stripe rust, leaf rust, powdery mildew, and stem rust. QTL analysis has been widely used to dissect complex traits through identifying the genomic location and effects of genes contributing to quantitative variation (Young, 1996). Over the past 20 years, more than 320 stripe rust resistance QTLs have been reported using different molecular markers, including diversity array technology (DArT), single sequence repeats (SSRs), and single-nucleotide polymorphisms (SNPs) (Lan et al., 2017); the genetic locations of these QTLs are continually being refined through fine mapping studies (Lan et al., 2017; Draz et al., 2021).

Various approaches can be used to identify genetic loci and genes in wheat. High-density SNP markers using gene-chip technology provide a superior approach for QTL mapping due to their high-throughput, efficiency, allele specificity, and high-resolution capacity (Yu et al., 2011; Draz et al., 2021). The Axiom wheat 55K SNP array with 53,063 SNP probes was selected from the wheat 660K SNP array, which is considered to be more appropriate for wheat genetic research (Ren et al., 2018; Huang et al., 2019). Based on the advantages of lower costs, higher accuracy, and its medium density, this SNP array has been widely used in QTL identification in wheat (Huang et al., 2019; Zhang et al., 2019). Also, whole-genome exome sequencing (WES) has been successfully applied for the identification of genetic loci and isolating genes in wheat (Henry et al., 2014; Mo et al., 2018). For instance, candidate natural variants were identified using a 110 Mb exome capture assay at the Yr6 locus responsible for stripe rust resistance (Gardiner et al., 2016). A BSA-based exome capture sequencing pipeline for rapid gene cloning has been developed by Chengdu Tcuni Technology, which is an alternative method to effectively sequence coding regions with low cost in wheat (Dong et al., 2020). Therefore, SNP chip and WES approaches can be the effective strategies to find new genetic loci and candidate genes associated with important traits in wheat.

The Italian common winter wheat cv. Pascal (Lan Yin 1), introduced to Gansu Province of China in 1995, showed a high level of APR to stripe rust in the field but was susceptible to CYR32 and CYR33 at the seedling stage. However, the seedling gene locations and inheritance of APR to stripe rust in Pascal were not clear. Thus, the objectives of this study were to (1) map seedling stripe rust resistance gene in the Huixianhong/Pascal F₅ RIL population; (2) identify the APR to stripe rust in the same population; (3) determine the interaction effect between the identified resistance loci on stripe rust in the adult plant stage; and (4) develop molecular markers of new and stable resistance QTL for the wheat breeder to breed durable stripe rust-resistant wheat variety.

“Pascal” was conferred by two recessive genes effective against two Pst races CYR 31 and CYR 32, respectively, at the seedling stage (Cao et al., 2006). In this study, we found both parents, Huixianhong and Pascal, susceptible to three Pst races of CYR32, CYR33, and CYR34 at the seedling stage (IT varied from 7 to 8). Most of the stripe rust resistance genes conferring seedling resistance can easily lose their effect since selection pressure favors the virulence in the pathogen population, which could explain that Pascal does not provide seedling resistance to stripe rust in this study. The constant evolution of pathogens and thus the breakdown of seedling resistance make our effort to continuously identify the sources of adult plant resistance increasingly important. We mapped three adult plant resistance QTLs for stripe rust in the Huixianhong/Pascal RIL population, and Pascal showed stable resistant over the last 50 years. There were five stripe rust resistance genes on chromosome 3D, viz. Yr45 (Li et al., 2011), Yr49 (Ellis et al., 2014), Yr66 (Bariana et al., 2022), Yr71 (Bariana et al., 2016), and Yr73 (Dracatos et al., 2016). Yr45, Yr71, and Yr73 were mapped on the 3DL, and the other two genes were mapped on 3DS. Yr45 was discovered from a common spring wheat “PI181314,” flanked by RGAP markers wgp118, wgp115, and two SSR markers Xwmc656 and Xbarc6 which were mapped 11.7 and 12.6 cM proximal to Yr45. We tried to test the two parents using markers wgp115, Xwmc656, and Xbarc6 (Li et al., 2011). However, none of them are polymorphic, and the physical interval of Yr45 was from 476.9 to 514.9 Mb on chromosome 3D, whereas QYr.gaas-3DL was located around 354.6 Mb on the same chromosome. Besides, Yr45 is an all-stage resistant gene, whereas QYr.gaas-3DL was an APR locus in wheat, and therefore, Yr45 could be different from QYr.gaas-3DL. Yr71 is an adult plant resistance gene mapped on 3DL in the Australian cultivar Sunco, and the physical position of its flanking marker gwm114b was at around 603.4 Mb (Bariana et al., 2016). KASP marker Yr71_16434, 1.8 cM distal to Yr71, was at 608 Mb on chromosome 3D. There was no polymorphic for Xgwm114b and Yr71_16434A in the “Huixianhong/Pascal” RILs population when we tried to detect the population, whereas QYr.gaas-3DL was located near the centromeric region of chromosome 3D. Another gene Yr73, a complementary gene on 3DL in the Australian cv. Avocet R, conferred seedling resistance to stripe rust. These results indicate that QYr.gaas-3DL is likely different from genes Yr71 and Yr73.

The QTL QYr/Lr.cim-3DC for stripe rust near the centromeric region of chromosome 3D was identified in common wheat line UC1110 (Lan et al., 2017). It provided both of the leaf rust and stripe rust resistance in the population. QYr.gaas-3DL was also located near the centromeric region of chromosome 3D, but we did not detect the effect on leaf rust. Qyrsicau-3DL, a high-confidence APR locus was detected in a genome-wide association study (GWAS) in Sichuan wheat (Ye et al., 2019). The SNP marker AX-109329567 closely linked with Qyrsicau-3DL was located at 595 Mb on chromosome 3D, which was about 240 Mb away from QYr.gaas-3DL. Huang et al. (2021) identified an APR QTL (QYrsn.nwafu-3DL) on 3DL explaining 5.8–12.2% of the phenotypic variation from Shannong 33 (SN33). The physical position of its flanking marker AX-109582945 was around 407.4 Mb on chromosome 3D, which is around 52 Mb away from QYr.gaas-3DL. Furthermore, QYrsn.nwafu-3DL was present in 7.4% of a panel of 420 current Chinese wheat cultivars, which was much lower than the presence of QYr.gaas-3DL in a collection of 153 Chinese germplasm resources in our studies. Therefore, QYrsn.nwafu-3DL could be different from the APR resistance QYr.gaas-3DL.

The stripe rust resistance genes and QTL located on chromosome 5AS in the previous studies are extensive. A minor effect QTL (QPst.jic-5A) for field stripe rust resistance on chromosome 5AS was identified in a stripe rust susceptible winter wheat cv. Brigadier (Jagger et al., 2011). QPst.jic-5A was located between marker Xwmc752 and Xgwm786, placing it near the centromere on the short arm of chromosome 5A. Since QYr.gaas-5AS was mapped at the interval of 59.1–62.2 Mb and far away from the centromere of chromosome 5A, QPst.jic-5A could be different from QYr.gaas-5AS. A major QTL, QYr.cau-5AS, exhibited 24.1–35.3% of PVE in winter wheat AQ24788-83 at the tillering growth stage (Quan et al., 2013). This QTL was flanked by markers Xcfa2250 and Xwmc705, which were located around 245 Mb, suggesting this QTL was different from QYr.gaas-5AS.

QYr.cim-5AS was a minor APR QTL derived from Avocet and mapped near the SSR marker Xcfa2104 and Xbarc186 (Ren et al., 2017). The physical position of those SSR markers was around 22–46 Mb, placing near QYr.gaas-5AS in the present studies. However, it only explained 4.0–5.1% of YR variation, but QYr.gaas-5AS had PVE values ranging from 11.0 to 17.3%. It is possible that QYr.gaas-5AS is different from this QTL. However, this needs to be confirmed by allelism test in the future. Qyrsicau-5AS, a minor APR QTL with a PVE of 7.1–8.1%, were detected by the GWAS in Sichuan wheat (Ye et al., 2019). The SNP marker AX-111623511 linked with Qyrsicau-5AS was located at 101 Mb on chromosome 5A, which was about 40 Mb from QYr.gaas-5AS, indicating QYrsicau-5AS could be different from the APR resistance QYr.gaas-5AS. Therefore, QYr.gaas-5AS is possibly a new QTL for APR to stripe rust.

Previous studies have detected several QTLs for stripe rust on chromosome 1AL in wheat. A minor APR QTL QYr.sgi-1A was located on chromosome 1A in the population of Kariega/Avocet S, contributing 6–12% to the phenotypic variance, and flanked by markers s15m19D and s23m18E, but was inconsistently detected across environments (Ramburan et al., 2004). Unfortunately, we failed to obtain the physical location of this QTL. Whereas QYr.gaas-1AL had PVE values ranging from 11.3 to 23.1%, it was stable across all environments in our studies, indicating that QTL QYr.sgi-1A was unlikely to be the same locus as QYr.gaas-1AL. Bariana et al. (2010) identified QYr.sun-1A controlling APR to stripe rust in Kukri/Janz-derived doubled haploid (DH) population, which explained 6–7% phenotypic variation for stripe rust. This QTL was flanked by marker Xgwm164 and its physical position was around 280.6 Mb. But QYr.gaas-1AL was located in the interval of 505.3–507.9 Mb, which indicated that QYr.gaas-1AL was different from QYr.sun-1A.

Ren Y. et al. (2012) identified a QTL (QYr.caas-1AL) contributed APR to stripe rust in German spring wheat cultivar Naxos. It was located in marker interval XwPt-2406–Xwmc59 at 575.4 Mb, which was roughly 70 Mb away from QYr.caas-1AL. Besides, QYr.caas-1AL only explained 8.2% of the phenotypic variance in one environment, which showed lower effect and stability than QYr.gaas-1AL. QYrid.ui-1A, a minor high-temperature adult plant (HTAP) resistance QTL from the hard red winter wheat germplasm IDO444, were significant only for IT in single environment (Chen et al., 2012). QYr.gaas-1AL had high effect of stripe rust resistance and were stable across all environments. Gebrewahid et al. (2020) identified an APR QTL on chromosome 1A in spring wheat Fuyu 3 designated QYr.hebau-1AL, which was located on the interval of AX-109403007 to AX-110502416 around 286.6–289.3 Mb. This QTL should be different from QYr.gaas-1AL based on a physical distance of about 220 Mb between them. Grover et al. (2022) identified an ASR locus Yraci in the European winter wheat cultivar “Acienda” on the distal end of wheat chromosome 1A. Besides, two SNPs AX-95162217 and AX-94540853 linked with Yraci were mapped in the bin at 54.04 cM on chromosome 1A, whereas QYr.gaas-1AL covered a physical interval of 505.3–506.8 Mb on chromosome 1A was far away from Yraci. Therefore, Yraci and QYr.gaas-1AL were not the same QTL for strip rust. Further research is needed to determine the novelty of QYr.gaas-1AL from Pascal.

Combining BSA and the exome sequence strategy could accelerate gene mapping, especially in wheat with a large and complex genome. BSE-Seq is helpful for the construction of a linkage map across the whole genome, and it could be easily used to identify the linked interval regardless of the multiple gene copies and obtain most of the variations existing in the coding regions of genes (Dong et al., 2020). Using bulked segregant analysis and the exome sequence strategy, Mo et al. (2018) identified a clear peak region on chromosome 4BS associated with increased plant height. Harrington et al. (2019) identified a locus controlling an environmentally dependent chlorosis phenotype in the Durum wheat cv. Kronos. Martinez et al. (2020) finely mapped a novel TaMKK3 allele conferring the wheat ERA8 ABA-hypersensitive germination phenotype in a wheat backcross population. In our studies, BSE-Seq helped us to confirm the reality of QTL and select candidate genes in QTL QYr.gaas-1AL region. TraesCS1A02G313700, TraesCS1A02G313800, and TraesCS1A02G314900 had more than 10 SNPs on their exon regions. Of these genes, TraesCS1A02G314900 encodes metacaspase, which has been reported to modulate autophagy to confine cell death to the target cells during Arabidopsis vascular xylem differentiation (Escamez et al., 2016), which suggests that this gene may be a strong candidate related to disease resistance. Further research is needed to determine the relationship between these genes and stripe rust resistance.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

BB initiated the project, designed the experiment, and contributed to phenotype data. CL designed the experiment and finalized the manuscript. ZL assisted in the data analysis and contributed to drafting the manuscript. HW and XD performed the sample preparation and DNA extraction. LW and JD contributed to the part of phenotype data in the field. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.