Introduction

Wheat stripe rust (yellow rust), an airborne disease caused by Puccinia striiformis f. sp. tritici (Pst), is one of the most damaging diseases of wheat (Triticum aestivum L.) worldwide (Chen 2005; Wellings 2011). In China, the disease occurs annually in most wheat growing areas, especially in northwest, southwest, and the Huang-Huai wheat production region, which is considered as the largest stripe rust epidemiologic region in the world (Wan et al. 2007). Although the use of fungicides can prevent the disease, the application of fungicides adds a significant extra cost to farmers. The chemicals may cause adverse effects to the environment and may not be fully effective if missing the best application time. Growing resistant cultivars is the most effective, environmentally sound, economic, and consistently used method of controlling stripe rust. To improve stripe rust resistance, two types of resistance are generally used in wheat breeding programs. One type is race specific, also known as all-stage resistance, and the other is non-race specific or adult plant resistance. All-stage resistance expresses from seedling through all developmental stages of wheat, facilitates breeding selection, and is widely employed in wheat cultivars. However, race-specific resistance can be easily overcome by new virulent races, due to the high variability of the stripe rust pathogen, resulting in rust epidemics (Li and Zeng 2002; Line 2002). Adult plant resistance is durable but may not be adequate when rust infection is too severe. Both resistance types need to be combined to attain the better disease control (Chen 2005). With the variation of Pst virulence, the resistance genes in the majority of breeding programs in China have varied at least eight times since the first nation-wide stripe rust epidemics in the 1950s (Hou et al. 2013; Li and Zeng 2002; Wan et al. 2003, 2007). In the past decade, resistance in wheat cultivars Hybrid 46, Suwon 11, and ‘Yr24’ were overcome by the predominant Pst races, and only a few resistance genes are effective against the current Chinese Pst population (Cheng et al. 2006; Han et al. 2010, 2012; Wan et al. 2003; Wang et al. 2007). The need is great to identify new, effective resistance genes for breeding programs in stripe rust epidemic areas, such as Sichuan, Yunnan, Shaanxi, and Gansu, China (Wan et al. 2004; Cheng et al. 2006; Yang et al. 2003).

Landraces constitute a valuable genetic resource, possessing various desirable agronomic traits, including high-level resistance to wheat stripe rust. In the 1950s, Prof. Zhao used a Chinese landrace Maza crossed with an elite cultivar Biyu to develop a set of wheat cultivars with good resistance to stripe rust in China (Li and Zeng 2002; Wan et al. 2007). However, the landrace has some poor agronomic traits, requiring an extended period for selective introgression of desirable traits through the breeding process. In the past several decades, molecular techniques have provided useful tools for modern wheat breeding. Identification of molecular markers associated with resistance genes then using these markers in breeding process will facilitate successful selection. Molecular markers have been developed for many stripe rust resistance genes, such as Yr5, Yr10, Yr15, Yr26, Yr45, Yr53, Yr64, and Yr65 (Yan et al. 2003; Wang et al. 2002; Li et al. 2011; Wang et al. 2008; Xu et al. 2013; Yaniv et al. 2015; Cheng et al. 2014). Among the molecular markers, simple sequence repeat (SSR) marker is most widely utilized in wheat disease resistance gene mapping and genetic improvement, due to its high polymorphism, repeatability, and low cost (Li et al. 2009; Ren et al. 2010; Wang et al. 2009; Zhang et al. 2010; Cheng et al. 2014).

Lankao 5, a wheat landrace collected in Henan province of China, showed a high-level resistance to stripe rust in our field trials over many years. In this study, we analyzed the resistance inheritance of Lankao 5 to the prevalent races in the Chinese Pst population and constructed a genetic map of the stripe rust resistance gene in Lankao 5 with SSR markers using F1 plants and F2:3 populations from cross of Lankao 5 and Chinese Spring. These findings will be important for understanding the genetic basis of rust resistance in Lankao 5 and for using marker-assisted selection to transfer this resistance gene to other wheat cultivars in improving control of wheat stripe rust.

Materials and methods

Plant and pathogen materials

Lankao 5 is a wheat landrace collected from Henan province of China, which is resistant to most races of stripe rust in China. Chinese Spring (CS) is a wheat germplasm susceptible to all Pst races at the seedling stage. Lankao 5 was crossed with CS and a set of 21 monosomic CS lines (3D monosomic line was absent in further analyses due to F1 failed to produce F2 seeds). Chinese Spring monosomic lines and F1 plants were confirmed cytologically using the Feulgen staining method (Xu et al. 2013). For each of the 21 crosses, the F1 plants were selfed to develop F2 populations; then, the F2 plants were grown to maturity and harvested for F2:3 seeds. The seeds of parents, F1, F2, and F2:3 generations, were propagated in a controlled greenhouse.

Six predominant Chinese Pst races (Chen et al. 2009; Li et al. 2016), Sun11-4, Sun11-11, CYR29, CYR30, CYR31, and CYR32, were used to test Lankao 5, CS, and F1 plants and F2 populations of CS monosomic lines at the seedling stage in greenhouse.

Seedling tests

Each F2 population from crosses between Lankao 5 and the 20 CS monosomic lines were tested in parallel with race Sun11-11 to analyze the chromosome location of the resistance gene in Lankao 5. F2:3 generations from Lankao5 × CS 1D monosomic which were selected to construct a genetic linkage map were tested with race Sun11-11.

Seedling tests were conducted under controlled greenhouse conditions. About 15 seeds from each parent and F1 generations, about 200 seeds from each F2 population, and 20 seeds from each of the F3 lines were used for tests. Fifteen to 20 seeds were planted per 10 cm diameter × 12 cm deep pot. Plants were grown at day/night temperatures of 25/15 °C and day/night lights of 16/8 h cycle.

Pathogen inoculations were performed when the first wheat leaves were fully expanded. The inoculum was made of spore/talc mixture at a 1/20 ratio and gently dusted on leaves. Inoculated plants were kept in a dew chamber at 10 °C for about 24 h, then moved to a growth chamber under a daily cycle of 16 h of light at 18 °C and 8 h of darkness at 10 °C. Plant infection type (IT) were scored based on the scale of 0, 0;, 0;+, 1, 1+, 2, 2+, 3−, 3, 3+, and 4, where 0 means no visible infection, 0; means necrotic/chlorotic flecks without sporulation, 0;+ means necrotic/chlorotic stripes without sporulation, 1 means necrotic/chlorotic stripes with scattered small sporulation, 1+ means necrotic/chlorotic stripes with many small sporulation, 2 means a few moderate sporulation with necrotic/chlorotic flecks, 2+ means many moderate sporulation with necrotic/chlorotic stripes, 3− means abundant sporulation with necrotic stripes, 3 means abundant sporulation with chlorotic stripes, 3+ means abundant sporulation with chlorosis, and 4 means abundant sporulation without chlorosis (Bariana and McIntosh 1993; Li et al. 2006a). Plants with IT of 0 to 2+ were considered resistant, while those with ITs of 3− to 4 were treated as susceptible (Liu 1988; Li et al. 2006a). The susceptible wheat cv. Chinese Spring served as a universal test control.

DNA extraction and bulk segregant analysis

The parents and 141 F2 plants from Lankao 5 × Chinese Spring 1D monosomic were selected for SSR analysis. Genomic DNA was extracted from leaves of parents and F2 plants using the CTAB method (Riede and Anderson 1996). Based on F2:3 phenotypic data, resistant and susceptible bulks were made with equal amounts of DNA by ten homozygous resistant and ten homozygous susceptible F2 plants, respectively.

Microsatellite marker analysis

Because the resistance gene in Lankao 5 to Pst race Sun11-11 was cytogenetically located on chromosome 7B by CS monosomic lines, 44 SSR markers specific to wheat chromosome 7B were used to identify polymorphism and to confirm the chromosomal region of the resistance gene in Lankao 5. SSR primer sequences were obtained from Grain Genes (http://wheat.pw.usda.gov), then synthesized by Sangon Biotech, Inc. (Shanghai, China). The PCR method for each SSR marker followed previous reports (Röder et al. 1998; Somers et al. 2004).

Each 15 μL PCR mixture consisted of 1.5 μL 10× buffer, 2.1 μL DNA template (25 ng/μL), 0.3 μL dNTPs (2.5 mM each), 1.2 μL MgCl2 (25 mM), 1.5 μL each of the forward and reverse primers (10 μM each), 0.15 μL Taq DNA polymerase (5 U/μL) and 6.75 μL ddH2O. All applied PCR reagents were purchased from Sangon Biotech, Inc. PCR amplification was performed using a MJ Research PTC-200 thermal cycler by the following protocol: initial denaturation at 94 °C for 5 min, followed by 35 cycles of 30 s each under denaturation at 94 °C, 1 min of annealing at 50–68 °C (depending on primers), and 45 s at 72 °C for extension, then, a finally extension at 72 °C for 10 min. After amplification, 6 μL of formamide loading buffer [98% formamide, 10 mM EDTA (pH 8.0), 0.5% (w/v) xylene cyanol, and 0.5% (w/v) bromophenol blue] was added to the PCR product. About 5–7 μL mixture of the PCR product and loading buffer for each sample was loaded for electrophoresis on 8% polyacrylamide gel, then visualized by applying silver stain (Bassam et al. 1991; Chen et al. 1998).

Data analysis

Segregation of the markers and resistance was both tested for fitness to expected ratios using the chi-squared (χ 2) tests on Excel data analysis program (Microsoft Office 2010). Linkage analysis and map construction of SSR markers and resistance locus were performed with Mapmaker 3.0 (Lander et al. 1987). The recombination rate was converted to map distance (in centimorgan, cM) according to the Kosambi mapping function (Kosambi 1944). The linkage map was drawn using Mapdraw V2.1 (Liu and Meng 2003).

Results

Resistance evaluation and inheritance analysis of Lankao 5 resistance to stripe rust

Infection types (ITs) of two races (CYR32 and Sun11-4) on seedlings of Lankao 5 were resistant with no apparent symptoms (IT 0), while those of four Pst races (CYR29, CYR30, CYR31, and Sun11-11) exhibited necrotic flecks without sporulation (IT 0;).The results indicated that Lankao 5 was highly resistant to the test Pst races. In contrast, ITs of six Pst races on Chinese Spring displayed full sporulation and absent in any necrosis (IT 4), indicating Chinese Spring was highly susceptible (Table 1).

Table 1 Reaction types of wheat cultivars Lankao 5 and Chinese Spring to different races of the wheat stripe rust fungus, Puccinia striiformis f. sp. tritici (Pst)
Full size table

The resistance inheritance patterns to wheat stripe rust were controlled by various genes in Lankao 5. In tests on Pst races Sun11-11 and CYR30, F1 plants were resistant, while F2 populations were segregating as at a ratio 3R:1S when tested on Sun11-11and at a ratio 13R:3S when tested on CYR30 (Table 2). The preceding tests indicated that resistance in Lankao 5 to Pst race Sun11-11 was controlled by a dominant gene (temporarily designated YrLk), while resistance to Pst race CYR30 was controlled by one dominant gene and one recessive gene, independently.

Table 2 Genetic analysis of stripe rust resistance to six races of Puccinia striiformis f.sp. tritici in Lankao 5 (P1) × Chinese Spring (P2) F1 and F2 populations
Full size table

In tests on Pst races CYR31, CYR32, and Sun11-4, all F1 plants exhibited resistance, while F2 populations segregated at a ratio 15R:1S (Table 2), demonstrating that resistance in Lankao 5 to these three races all was controlled by two independently inherited dominant genes.

In terms of resistance to CYR29, all F1 plants were susceptible, while F2 populations segregated at a ratio of 7R:9S (Table 2), indicating that resistance to CYR29 in Lankao 5 was controlled by two recessive genes of independent action.

Chromosomal location of the resistance gene

To determine the chromosomal location of resistance gene YrLk in Lankao 5 to Pst race Sun11-11, monosomic analysis was applied. Because F1 plants of Lankao 5 × CS 3D monosomic line failed to produce F2 seeds, none further analysis was conducted on their F2 populations. The other 20 crosses of Lankao 5 and CS monosomic lines were examined with race Sun11-11. Each F2 population from crosses between Lankao 5 and 19 CS monosomic lines was segregating at a ratio of 3 resistant:1 susceptible, except the Lankao 5 × CS 7B monosomic line (Table 3). The corresponding Lankao 5 × CS 1D monosomic line (chose by which showed lowest χ 2 value on monosomic analysis (Table 3)) F3 populations was segregating at a ratio of 1R:2Seg:1S. Plant observations fit the expected ratio, complying with the chi-squared test. For example, Table 3 shows the test results for the F2 Lankao 5 × CS 1D monosomic line population (χ 2 = 0.36 < χ 2 0.05 = 5.99). The preceding results further supported the conclusion that one dominant gene in Lankao 5 conferred resistance to Sun11-11. Segregation of F2 plants from crosses of Lankao 5 × CS 7B monosomic line significantly deviated from the 3R:1S ratio (χ 2 = 43.64 > χ 2 0.05 = 3.84), indicating that chromosome 7B carries the resistance gene YrLk.

Table 3 Expected ratios, observed resistant (R) and susceptible (S) F2 plants from crosses of resistant cultivar Lankao 5 with Chinese Spring (CS) monosomic lines, inoculated with the Chinese wheat stripe rust Puccinia striiformis f.sp. Tritici race Sun11-11, and probabilities of chi-squared test for fit to theoretical ratios
Full size table

Molecular mapping of the stripe rust resistance gene

Forty-four SSR markers covering chromosome 7B were screened on Lankao 5, resistant bulk, the susceptible parent Chinese Spring, and susceptible bulk to identify those associated with the YrLk locus. Five markers, Xwmc476, Xbarc267, Xwmc396, Xgwm131, and Xwmc517, generated polymorphism bands between the resistant and susceptible bulks, as well as their parents. All of the five markers are located on the long arm of chromosome 7B, based on the wheat consensus SSR map (Somers et al. 2004).

These five SSR markers were further used to genotype the 141 F2:3 plants. The results showed that each of the five markers was co-dominant and well fitted a segregating ratio of 1 (homozygous resistant parental allele):2 (heterozygous allele):1 (susceptible parental allele), indicating each marker is a single locus and reliable for use to construct a linkage map (Table 4). Linkage analysis showed that resistance gene YrLk was linked with the five SSR loci; therefore, YrLk was also located on the long arm of chromosome 7B. A linkage map was composed of YrLk locus and five SSR loci, spanning 21.6 cM. The marker order obtained for the five SSRs was compatible with that determined by Somers et al. (2004). The two closest-flanking loci of YrLk were Xwmc396 and Xbarc267, with genetic distances of 3.3 and 4.4 cM, respectively (Fig. 1). For example, Fig. 2 displays the segregation of marker alleles in the F2:3 populations of the two closest-flanking markers. Xwmc396 and Xbarc267 are located near the centromere, based on wheat SSR maps (Somers et al. 2004), delineating the position of YrLk as proximal to the centromere of chromosome 7BL.

Table 4 Observed molecular marker numbers and alleles of Lankao 5 (A), Chinese Spring (B) and their heterozygous (H) progeny in F2 population from Lankao 5 × Chinese Spring, and probabilities of chi-squared test of fit to theoretical ratios
Full size table
Fig. 1
figure 1

Linkage map of stripe rust resistance gene YrLk

Full size image
Fig. 2
figure 2

PCR profiles for the F2 segregated population and parents, amplified by the YrLk flanking markers Xbarc267 (a) and Xwmc396 (b) in polyacrylamide gel. M DNA size ladder, Pr resistant parent Lankao 5, Ps susceptible parent Chinese Spring, Br resistant DNA bulk, Bs susceptible DNA bulk, R resistant plant, S susceptible plant, H heterozygous resistant plants

Full size image

Discussion

In addition to YrLk, seven officially named Yr genes (Yr2, Yr6, Yr39, Yr52, Yr59, Yr63, and Yr67 reported by McIntosh et al. 1998; El-Bedewy and Robbelen 1982; Lin and Chen 2007; Ren et al. 2012; Zhou et al. 2014; McIntosh et al. 2014), five temporarily named Yr genes (YrZh84, YrC591, YrShan515, yrMY37, and YrSu-1, reported by Li et al. 2009; Li et al. 2006b; Zhang et al. 2011; Ren et al. 2015; Yan et al. 2009), and 11 quantitative-trait loci (QTL) (Maccaferri et al. 2015) have been reported on chromosome 7B. Yr6 and Yr63 are located on the short arm of chromosome 7B (El-Bedewy and Robbelen 1982; McIntosh et al. 2014), a different chromosomal arm from YrLk. The Yr6 near-isogenic line AvSYr6NIL and Yr6 carrier cultivars Heines Peko and Heines Kolben (Calonnec et al. 1997) were susceptible to Pst race Sun11-11 in our greenhouse test, while YrLk is resistant. YrLk should be different from Yr6.

Yr39, Yr52, and Yr59 are high-temperature adult-plant (HTAP) resistance genes, which are susceptible to Pst races at the seedling stage (Lin and Chen 2007; Ren et al. 2012; Zhou et al. 2014). Yr52, Yr59, Yr67, YrZh84, and YrC591 are located on the telomere of the long arm of chromosome 7B (Li et al. 2006b; McIntosh et al. 2014; Ren et al. 2012; Zhou et al. 2014; Xu et al. 2014), while YrLk is proximal to the centromere of 7B. Based on the resistance type and chromosome location, YrLk may be different from these four genes.

In 1962, Yr2 was first reported as located on chromosome 7B in Heines VII (McIntosh et al. 1998) and was later mapped on 7BL, with a genetic distance to Xwmc364 of 5.6 cM (Lin et al. 2005). Xwmc364 does not exhibit polymorphism between Lankao 5 and Chinese Spring, and cannot be mapped in our population. YrSu-1 was reported on chromosome 7BL with the genetic distance from Xwmc396 to YrSu-1 of 21.1 cM (Yan et al. 2009). The distance from YrLk to Xwmc396 is 3.3 cM. When Heines VII was tested with Sun11-11 at the seedling stage in greenhouse condition, it was susceptible (IT 3–4), and the carrier cultivar of YrSu-1, Suwon11, was susceptible to Sun11-11 (Chen et al. 2009). The rust response reaction and relative chromosomal position of three genes suggested that Yr2 and YrSu-1 are different from YrLk.

YrShan515 was identified in Chinese landrace Shan 515, conferring resistance to Pst race Sun11-4 (Zhang et al. 2011). The genetic distances of flanking markers Xbarc267 and Xwmc653 to YrShan515 are 3.0 and 3.4 cM, respectively (Zhang et al. 2011), while the genetic distance from Xbarc267 to YrLk is 4.4 cM. It appears that YrShan515 and YrLk are genetically proximal. However, resistance by Shan 515 to race Sun11-11 is reported as controlled by two recessive genes (Zhang 2010), different from results for YrLk in this study. Furthermore, Shan 515 is susceptible to race CYR32 while Lankao 5 is resistant to CYR32. These results indicate high likelihood that YrLk and YrShan515 are different genes.

yrMY37 was identified in Chinese wheat cultivar Mianmai 37, conferring resistance to Pst race v26 (Ren et al. 2015). The genetic distances of flanking markers Xgwm297 and Xbarc267 to yrMY37 are 0.78 and 0.38 cM, respectively (Ren et al. 2015), while the genetic distance from Xbarc267 to YrLk is 4.4 cM. But yrMY37 was reported as a recessive gene, while YrLk is a dominant gene. The reaction type of Mianmai 37 (yrMY37) to race CYR32 is 1, and to CYR31 is 2 (Ren et al. 2015), while reaction type of Lankao 5 to race CYR32 is 0, and to CYR31 is 0:. These results indicate that YrLk and yrMY37 are different genes.

The above cultivars content of Yr genes originate in very different places. Yr2, Yr6, Yr63, and Yr67 are derived from European cultivars. Yr39 was identified from the US cultivar Alpowa (Lin and Chen 2007). The carrier cultivars for Yr52, Yr59, and YrC591 originate in Pakistan, Iraq, and India, respectively (Ren et al. 2012; Zhou et al. 2014; Li et al. 2009). YrZhou84 was identified from wheat cultivar, Zhou8425B. The stripe rust resistance in Zhou8425B originated in an Italian cultivar Milan (Li et al. 2006b). YrSu-1 is derived from a Korean cultivar Suwon 11. Shan 515, YrShan515 gene carrier, is a wheat landrace from Shaanxi, China (Zhang 2010). yrMY37 is derived from Chinese wheat cultivar Mianmai 37.

Overall, YrLk shows great differences in various aspects from other Yr genes on chromosome 7B, including chromosomal position, origin of gene carrier, and resistance against Pst races. It is very likely that YrLk is a novel, effective stripe rust resistance gene. Figure 3 shows the relative locations of YrLk (D) and Yr2 (B), Yr39 (C), Yr52 (I), Yr59 (J), Yr67 (G), YrC591 (K), YrZh84 (F), YrSu-1 (A), YrShan515 (E), and yrMY37 (L) on 7BL compared with common markers on the chromosome 7BL linkage maps.

Fig. 3
figure 3

Linkage maps for stripe rust resistant genes on the long arm of chromosome 7B and comparisons with the consensus map. The dashed lines connect common markers on different maps. Genetic map showing the relative genetic locations of YrLk (D) with other Yr genes in this study. Yr2 in Lin et al. (2005) (B), Yr39 in Lin and Chen (2007) (C), Yr52 in Ren et al. (2012) (I), Yr59 in Zhou et al. (2014) (J), Yr67 in McIntosh et al. (2014) (G), YrC591 in Xu et al. (2014) (K), YrZh84 in Li et al. (2006b) (F), YrSu-1 in Yan et al. (2009) (A), YrShan515 in Zhang et al. (2011) (E), and yrYM37 in Ren et al. (2015) on chromosome 7BL, and the shared markers in each linkage map and the corresponding chromosomal bin locations (H)

Full size image

Landraces are important genetic sources and may harbor valuable stripe rust resistance (Sthapit et al. 2014; Kertho et al. 2015). Wang et al. (2012) identified over 200 stripe rust resistance landraces from 28 countries and developed 70 germplasms in the background of the elite spring wheat cultivar ‘Avocet’, and these germplasms have various resistance types to the Pst populations in the USA. Most of these germplasm lines have high resistance to current Chinese Pst races (Zhou et al. 2015). The resistance genes Yr45, Yr52, Yr53, Yr59, and Yr62 in these landraces were mapped with molecular markers (Li et al. 2011; Xu et al. 2013; Ren et al. 2012; Zhou et al. 2014; Lu et al. 2014). Some Chinese wheat landraces play important roles in the stripe rust epidemic areas where modern cultivars are unable to adapt to the local ecosystem, such as Pingyan 50, Caoxuan 5, and Baidatou (Lan et al. 2010; Zhou et al. 2015; Ma et al. 2015). However, most landraces have poor agronomic traits or low yield potential and are thus not suited to modern cultivation conditions. It is necessary to develop techniques to transfer desirable traits, such as stripe rust resistance, from these landraces to new cultivars, while excluding negative traits. Undoubtedly, the closely linked molecular markers of resistance genes constitute an efficient tool for selection during wheat breeding. The genomic region of YrLk is expressed as a gene hotspot, which allows gene exchange to occur. As a result, combining different Yr genes into a new germplasm may prove to be relatively easy.

In this study, results confirm that Lankao 5 is resistant to most of the currently prevalent Chinese Pst races at the seedling stage. At least four genes in Lankao 5 conferring resistance to stripe rust were genetically determined, one dominant gene (YrLk) resistant to race Sun11-11, one dominant gene and one recessive gene to CYR30, two dominant genes resistant to races CYR31 and CYR32 (one of which maybe YrLk), and two recessive genes resistant to CYR29. Lankao 5 is an excellent stripe rust resistance wheat landrace, which is one of the best candidates for disease resistance; furthermore, the closely flanking markers, Xwmc396 and Xbarc267, of YrLk allow YrLk for high efficient use in marker-assisted selection for stripe rust resistance in breeding programs.