Introgression of Two Quantitative Trait Loci for Stripe Rust Resistance into Three Chinese Wheat Cultivars

: Wheat stripe rust, caused by Puccinia striiformis f. sp. tritici (Pst) , is one of the most devastating diseases in wheat. Due to the large-scale and widely-distributed planting pattern of wheat, the directional selection pressure of the pathogen is very strong. Therefore, it is urgent to pyramid more stripe rust resistance genes in wheat cultivars to enhance resistance durability and ensure wheat production safety. In this study, two quantitative trait loci (QTL) for adult plant resistance (APR) to stripe rust, QYr.nafu-2BL and QYr.nafu-3BS , were validated and introgressed from wheat line P9897 into three Chinese elite wheat cultivars, Chuanmai 42, Xiangmai 25, and Zhengmai 9023, through marker validation. The three Chinese elite varieties were used as the female parent to cross with wheat line P9897, and they were selfed to the F 6 generation. A total of 114 lines were then selected based on ﬁeld agronomic traits and stripe rust resistance. Four markers ( Xcfd73, Xgwm120, Xbarc87 and Xbarc133 ) linked with the QTL’s regions were employed to screen the 114 F 6 lines. Subsequently, 27 lines combining two target QTL from P9897 were selected. The combination of agronomic traits and disease resistance results showed that 13 of these selected lines had favorable application prospects. The promising lines selected in this study could enrich the genetic resources of wheat stripe rust resistance genes, as well as provide material support and a theoretical basis for the prevention and control of wheat stripe rust in China.


Introduction
Wheat is a cereal crop widely cultivated all over the world. It is planted in over 200 million hectares per year with a production of 751.1 million tons, and human consumption of wheat is increasing every year [1]. By 2050, the global population will exceed 9.5 billion, according to the World Bank, and global wheat production needs to increase by 60% to meet food demand [2,3]. Wheat stripe rust is caused by Puccinia striiformis f. sp. tritici (Pst). The fungi can spread long distances on air currents, and its occurrence and damage can be characterized as long-term, fulminant, epidemic, or variable. The disease is distributed throughout global wheat-producing areas. In China, wheat stripe rust occurs QTL or more. The agronomic traits and disease resistance will be evaluated to obtain which lines have high application value.

Plant Material
The donor parent was the wheat line P9897, which carries two adult plant stripe rust-resistant QTL (QYr.nafu-2BL and QYr.nafu-3BS) and has maintained a good APR to stripe rust over many years of testing [34]. Its stability and resistance QTL make it an excellent resistant germplasm resource. The Chinese elite varieties Chuanmai 42 (from Sichuan Province and approved by the National Crop Variety Approval Committee (NCVAC) in 2004, and known to carry the Yr26 gene), Xiangmai 25 (from Hubei Province and approved by the NCVAC in 2008), and Zhengmai 9023 (from Henan Province and approved by the NCVAC in 2003) were the recipient parents. These three varieties have had high yields and strong disease resistance in the past, but with the emergence of new races of stripe rust, especially the emergence and prevalence of CYR34, their resistance to stripe rust has gradually been lost. However, due to their exceptional agronomic traits, they can continue to be used as recipient parents [35][36][37].

Cross Combination and Offspring Screening
P9897 was crossed with Chuanmai 42, Xiangmai 25, and Zhengmai 9023, respectively, with P9897 as the male parent and the three elite varieties as the female parents. The F 1 seeds were obtained and planted in the field. The F 2 , F 3 , F 4 , and F 5 seeds of each cross combination were planted over 30 rows, with approximately 80 seeds in a 2 m row with 30 cm between rows. We performed a bulk harvest before the F 5 generation for each cross to retain all possible genotypes, and after several generations of accumulation, we selected the F 5 generation plants with resistance to stripe rust, moderate plant height, and a high number of tiller and spikelet numbers. A total of 114 F 6 lines were obtained from the F 5 generation of three cross combinations in 2018. Through artificial screening and selecting in this way, the agronomic traits of the F 6 lines we obtained became stable ( Figure 1). combine two QTL or more. The agronomic traits and disease resistance will be evaluated to obtain which lines have high application value.

Plant Material
The donor parent was the wheat line P9897, which carries two adult plant stripe rust-resistant QTL (QYr.nafu-2BL and QYr.nafu-3BS) and has maintained a good APR to stripe rust over many years of testing [34]. Its stability and resistance QTL make it an excellent resistant germplasm resource. The Chinese elite varieties Chuanmai 42 (from Sichuan Province and approved by the National Crop Variety Approval Committee (NCVAC) in 2004, and known to carry the Yr26 gene), Xiangmai 25 (from Hubei Province and approved by the NCVAC in 2008), and Zhengmai 9023 (from Henan Province and approved by the NCVAC in 2003) were the recipient parents. These three varieties have had high yields and strong disease resistance in the past, but with the emergence of new races of stripe rust, especially the emergence and prevalence of CYR34, their resistance to stripe rust has gradually been lost. However, due to their exceptional agronomic traits, they can continue to be used as recipient parents [35][36][37].

Cross Combination and Offspring Screening
P9897 was crossed with Chuanmai 42, Xiangmai 25, and Zhengmai 9023, respectively, with P9897 as the male parent and the three elite varieties as the female parents. The F1 seeds were obtained and planted in the field. The F2, F3, F4, and F5 seeds of each cross combination were planted over 30 rows, with approximately 80 seeds in a 2 m row with 30 cm between rows. We performed a bulk harvest before the F5 generation for each cross to retain all possible genotypes, and after several generations of accumulation, we selected the F5 generation plants with resistance to stripe rust, moderate plant height, and a high number of tiller and spikelet numbers. A total of 114 F6 lines were obtained from the F5 generation of three cross combinations in 2018. Through artificial screening and selecting in this way, the agronomic traits of the F6 lines we obtained became stable ( Figure 1). Figure 1. The scheme of development, validation, and selection of the wheat lines approach for the introgression of stripe rust-resistance QTL in three elite cultivars (Chuanmai 42, Xiangmai 25, and Zhengmai 9023) using the same donor (P9897). Chuanmai 42 is red, Xiangmai 25 is yellow, Zhengmai 9023 is blue, and P9897 is green; the color of their cross offspring is the same as their respective female parents. Figure 1. The scheme of development, validation, and selection of the wheat lines approach for the introgression of stripe rust-resistance QTL in three elite cultivars (Chuanmai 42, Xiangmai 25, and Zhengmai 9023) using the same donor (P9897). Chuanmai 42 is red, Xiangmai 25 is yellow, Zhengmai 9023 is blue, and P9897 is green; the color of their cross offspring is the same as their respective female parents.

Genotyping Wheat Lines by Molecular Markers
The parents and 114 F 6 lines of the three cross-combinations were genotyped using molecular markers linked to the two QTL (QYr.nafu-2BL and QYr.nafu-3BS) in P9897 to selected lines with a single QTL or two QTL ( Table 1). The SSR markers Xcfd73 and Xgwm120, which are closely linked to QYr.nafu-2BL, and Xbarc87 and Xbarc133, which are closely linked to QYr.nafu-3BS, were used to detect the F 6 lines with QYr.nafu-2BL, QYr.nafu-3BS, or both QTL. Two molecular markers, WE-173 and Xbarc181, which are linked to Yr26 [39], were used to detect the lines containing Yr26 in the Chuanmai 42/P9897 cross-combination (Table 1).

DNA Extraction and PCR
In 2019, the fresh leaves of the 114 F 6 lines and four parents were collected in a Mianyang experimental field. The DNA were extracted by a modified cetyltrimethyl ammonium bromide (CTAB) method [44]. The polymerase chain reaction (PCR) program followed conditions described in previous research [45,46]. PCR was performed in 10µL reaction mixtures containing 2µL (100ng/µL) template DNA, 1µL 10×PCR buffer (containing mg 2+ ), 0.8µL 2.5 mM of each dNTP, 1µL (2µM) of each primer solution, 0.2µL Taq DNA polymerase solution (2.5 unit/µL), and 4µL sterilized dd H 2 O. The PCR amplification profile consisted of a denaturation (4 min at 94 • C) followed by 35 cycles (94 • C for 30 s, 55 • C for 20 s, and 72 • C for 30 s) and an extension for 8 min at 72 • C. The PCR amplification products of the linked markers were separated using 6% polyacrylamide gel electrophoresis (PAGE) (Beijing Solarbio Science & Technology Co., Ltd. Beijing, China) [47].

Disease Resistance and Agronomic Trait Evaluation in the Field
The F 5 plants, F 6 lines, and parents were evaluated for adult plant stripe rust response in Mianyang in the Sichuan province (31 • 33 N, 104 • 55 E, altitude 485 m) during the 2017-2018 and 2018-2019 cropping seasons. The F 5 plants of each cross combination were planted across 30 rows, with approximately 80 seeds in a 2 m row with 30 cm between rows. The F 6 lines were arranged in randomized complete blocks with three replications. Each line, with approximately 80 seeds, was planted in a 2 m row with 30 cm between rows. The parents and the susceptible control Mingxian 169 (M169) were planted after every 20 rows throughout the field. To increase field inoculum, one column of Mingxian169 was also planted perpendicularly and adjacent to the test rows. Mianyang is natural over-wintering region for stripe rust in China, and nurseries regularly become infected without artificial inoculation. The main prevalent races in Sichuan Province currently include CYR32, CYR33, and CYR34. In recent years, CYR34 has gradually become the predominant race [9]. Infection type (IT) data were collected based on the 0-9 scale of Line and Qayoum [48]. Disease severity (DS) scores were based on the modified Cobb scale [49]. Both IT and DS were recorded two times for parents, all F 5 plants, and F 6 lines when stripe rust severity on M169 reached between approximately 50% and 90% around 1 to 20 April in Mianyang. IT numbers of 0, 1-3, 4-6, and 7-9 were considered immune, resistant, intermediate, and susceptible, respectively. DS was recorded as the percentage of the diseased leaf area [43,50]. For the F 5 plants, we selected resistant plants with full spikes, plant height less than 1 m, and a tiller number of 4 or greater. For the F 6 lines, we selected 10 plants from the middle row of each repeating plot to evaluate agronomic traits. After the milk stage, we measured the average plant height of the selected plants with a ruler. The plant height of each plant was measured from the ground to the top of the spike. After maturity, the spikelets on each main spike, including the sterile spikelets, and the effective tillers of each plant were counted, and the average was taken. After harvest, we randomly selected 200 seeds from each wheat line and recorded their weight. The procedure was repeated three times for each line, and the average weight was taken as the 1000 grain weight.

Results
A total of 114 F 6 lines were selected due to their resistance to stripe rust, moderate plant height, and high tiller and spikelet numbers. Of these, 19 lines were selected from the cross of Chuanmai 42/P9897, 62 lines were selected from the cross of Xiangmai 25/P9897, and 33 lines were selected from the cross of Zhengmai 9023/P9897.

Quantitative Trait Locus Detection
Corresponding to the QTL detection results ( Figure S1), the 114 F 6 lines from three cross-combinations were divided into four groups: carrying QYr.nafu-2BL, carrying QYr.nafu-3BS, carrying both QTL, and carrying no target QTL ( Table 2). The results of QTL detection showed that among the F 6 lines of cross-combination Chuanmai 42/P9897, four lines carried QYr.nafu-2BL, three lines carried QYr.nafu-3BS, seven lines carried both QTL, and five lines did not carry a QTL; the mean IT and mean DS for the lines were 1.5, 1.7, 1.6, 2.8 and 4.3, 3, 3, and 40, respectively ( Figure 2). In addition, two markers (WE-173 and Xbarc181) linked to Yr26 were used to detect the F 6 lines of Chuanmai 42/P9897, identifying that 63.2% (12 in total) of the lines contained Yr26 in the Chuanmai 42/P9897 cross-combination. Among them, a total of six lines combined Yr26 with the two target QTL from P9897. The F 6 lines of cross-combination Xiangmai 25/P9897 consisted of 25 lines carrying QYr.nafu-2BL, 9 lines carrying QYr.nafu-3BS, and 15 lines carrying both QTL, while 13 did not carry a QTL. The mean IT and mean DS were 2.1, 1.8, 1.6, 3.6 and 8, 6.7, 5.5, and 34.3, respectively ( Figure 2). In cross-combination Zhengmai 9023/P9897, consisted of eight lines carrying QYr.nafu-2BL, 10 lines carrying QYr.nafu-3BS, and five lines carrying both QTL, while 10 did not carry a QTL. The mean IT and mean DS were 2.3, 1.7, 1.6, 3.4 and 8, 6.3, 4.8, and 23.3, respectively ( Figure 2). Among the F 6 lines of the three cross-combinations, the total number of lines containing two QTL was 27. Apart from this, at least one QTL was detected in a total of 86 F 6 lines, and in 28 F 6 lines a QTL was not detected. As an example, Figure S1 shows that marker Xgwm120 linked to QYr.nafu-2BL (a), marker Xbarc87 linked to QYr.nafu-3BS (b), and marker WE-173 linked to Yr26 was present or absent in P9897, Chuanmai 42, Xiangmai 25, and Zhengmai 9023, and segregated in the F 6 lines.  The phenotypes of all of the lines in which a QTL was detected were resistant, as the ITs of Chuanmai 42, Xiangmai 25, and Zhengmai 9023 were 7, 6, and 7 (Figure 3), respectively. These results demonstrate that introgression of the QTL from P9897 into the three leading varieties significantly enhanced the disease resistance of its offspring (Figure 2).  The phenotypes of all of the lines in which a QTL was detected were resistant, as the ITs of Chuanmai 42, Xiangmai 25, and Zhengmai 9023 were 7, 6, and 7 (Figure 3), respectively. These results demonstrate that introgression of the QTL from P9897 into the three leading varieties significantly enhanced the disease resistance of its offspring (Figure 2). The phenotypes of all of the lines in which a QTL was detected were resistant, as the ITs of Chuanmai 42, Xiangmai 25, and Zhengmai 9023 were 7, 6, and 7 (Figure 3), respectively. These results demonstrate that introgression of the QTL from P9897 into the three leading varieties significantly enhanced the disease resistance of its offspring (Figure 2).

Evaluation of Disease Resistance and Agronomic Traits
The combined results of two surveys of the lines showed different degrees of disease resistance. P9897 showed a high level of resistance (IT = 2; DS = 5%) ( Figure 3); however, Chuanmai 42 (IT = 7; DS = 60%), Xiangmai 25 (IT = 6; DS = 50%), and Zhengmai 9023 (IT = 7; DS = 50%) were shown to be susceptible ( Figure 4A). Based on our many years of disease resistance evaluation in the field, these three recipient parents are still partially resistant to stripe rust (IT 5-8; DS 50%-80%). Li and Xu reported that the ITs of Chuanmai 42, Xiangmai 25, and Zhengmai 9023 were 0, 2, and 2, respectively at seeding stage [51,52]. Therefore, we speculate that these three varieties contain unknown resistance genes. In the F 6 lines of Chuanmai 42/P9897 (19 total), 18 lines were resistant, 1 line was of intermediate resistance, and no line was immune (Table 3, Figure 4B). In the F 6 lines of Xiangmai 25/P9897 (62 total), the number of immune, resistant, and intermediate lines were 3, 55, and 4, respectively (Table 3, Figure 4C). In addition, in the F 6 lines of Zhenagmai 9023/P9897 (33 total), the number of immune, resistant, and intermediate lines were 1, 31, and 1, respectively (Table 3, Figure 4D). There were no susceptible lines in the three cross-combinations, except for Xiangmai 25/P9897-30 (IT = 6, DS = 70%), in which the DS of all the F 6 lines were less than 50%.   The three recipient parents, as the elite cultivars in China, have very good agronomic traits, which are used as standard references. Lines with a plant height between 80 and 100 cm (considering factors such as lodging resistance and easy harvesting), a spikelet number of 17 or greater, a tiller number of 4 or greater, and 1000 grain weight larger than 40 g were selected. At last, 13 lines were selected according to the above criteria (Table 4).
In summary, in the 114 F6 lines of the three cross combinations, we utilized the MAS method to select 27 lines which combined two QTL from P9897, including some lines combining Yr26 or some unknown resistance QTL. All the selected lines were resistant to stripe rust, according to the results of the disease resistance evaluation. Then, combining the evaluation of agronomic traits, we narrowed our selection to 13 (Table 4). These lines possessed high resistance to stripe rust, moderate plant height, and high tiller and spikelet numbers, and their 1000 grain weights were similar or surpassed the parent lines. These final selected lines have promising application prospects. Table 4. The results of the evaluation of resistance to stripe rust and agronomic traits and QTL detection of the susceptible control, the parents, and the selected lines. The three recipient parents, as the elite cultivars in China, have very good agronomic traits, which are used as standard references. Lines with a plant height between 80 and 100 cm (considering factors such as lodging resistance and easy harvesting), a spikelet number of 17 or greater, a tiller number of 4 or greater, and 1000 grain weight larger than 40 g were selected. At last, 13 lines were selected according to the above criteria (Table 4).
In summary, in the 114 F 6 lines of the three cross combinations, we utilized the MAS method to select 27 lines which combined two QTL from P9897, including some lines combining Yr26 or some unknown resistance QTL. All the selected lines were resistant to stripe rust, according to the results of the disease resistance evaluation. Then, combining the evaluation of agronomic traits, we narrowed our selection to 13 (Table 4). These lines possessed high resistance to stripe rust, moderate plant height, and high tiller and spikelet numbers, and their 1000 grain weights were similar or surpassed the parent lines. These final selected lines have promising application prospects.

Discussion
In this study, we combined conventional breeding methods with molecular marker validation and the introgression of two QTL from P9897 into three domestic leading varieties: Chuanmai 42, Xiangmai 25, and Zhengmai 9023. After the QTL were detected, we selected 27 lines from the three crosses that showed good stripe rust resistance, and of those, 13 that also possessed excellent agronomic traits. These results illustrate the application value of these germplasm resources.
In the results, some lines that contained only one QTL also showed a high level of resistance, but considering the more durable resistance and wider resistance spectrum, the lines that combined two or some unknown QTL from Chuanmai 42, Xiangmai 25, and Zhengmai 9023 have a higher application value in the long term. According to previous research [53], Chuanmai 42 contains the Yr26 resistance gene, which confers seedling resistance but has been overcome by the CYR34 PST race, which has recently increased in prevalence [54]. Therefore, combining the ASR gene (Yr26) with the APR genes (QYr.nafu-2BL and QYr.nafu-3BS) may form a complementary effect that can widen the resistance spectrum [55] and enhance resistance. Studies focused on Xiangmai 25 and Zhengmai 9023 have not identified genes for stripe rust resistance; however, they are presumed to contain unknown resistance genes based on field resistance evaluation ( Figure 4A). Combining these unknown resistance genes with the two QTL from P9897 can provide a long-lasting and high-level resistance to stripe rust [2]. For example, Xiangmai 25/P9897-34 has a higher resistance (IT = 0, DS = 0%) than the donor parents P9897 (IT = 2, DS = 10%), so we speculate that this line combines the unknown resistance genes from Xiangmai 25 with the two QTL from P9897.
In this study, some lines did not detect resistance QTL of P9897, but their field phenotype still showed resistance or immunity. The first likely reason for this is because the molecular markers applied to MAS should be co-segregated or closely linked to the target trait (1 cM or less) [56], while the molecular markers used in this study had a genetic distance greater than 1 cM. This means the probability of recombination between the marker and the gene increased, so the detection rate was low. Secondly, the population used in this study differed from the population used in the P9897 resistance QTL mapping, so different recombinations occurred between the genes and the molecular markers. As a result, the amount of lines containing QTL that were detected by molecular markers was less than the actual number that were present. Therefore, these resistant offspring lines without detectable target QTL still have further application value.
According to the results of 1000 grain weight analysis of the lines of the three cross combinations, the lines with higher level resistance (IT = 0~2) had a lower mean 1000 grain weight than the lines with lower level resistance (IT = 3~6). These results support the theory that an active immune response results in yield penalties for crops fighting pathogens [57]. Therefore, in the breeding process, it is necessary to consider the disease resistance of the line, and also to screen other agronomic traits in order to select varieties suitable for wide application.
At present, in the study of using marker validation to screen target genes in wheat breeding in order to eliminate undesirable donor traits, many backcrossing and selection steps are required, making the introgression of resistance genes more complicated [58]. However, since the target genes to be introgressed in this study were two QTL located on different chromosomes, we performed a bulk harvest before the F 5 generation, retaining all possible genotypes and reducing the loss of ideal genes that were still highly heterozygous [59]. After several generations of accumulation, we used artificial screening to select the F 5 generation plants according to phenotypes, such as disease resistance and agronomic traits. By the F 6 generation, the traits had stabilized and combined with molecular marker detection, so the selected target lines have promising application prospects. Although F 6 was not tested in multiple environments in this study, the agronomic traits and resistance to stripe rust of these cross combinations became stable after the preliminary evaluation and selection of F 5 . Then, through the screening and evaluation of F 6 , the agronomic traits and resistance of the recombinant inbred lines (RILs) were finally screened out to be stable in inheritance and expression, so the results of this experiment were reliable. With the emergence of new races of stripe rust, especially the emergence and prevalence of CYR34, these three elite varieties have gradually lost their resistance to stripe rust, which impacted yield. In this study, two resistance QTL were introgressed into wheat cultivars to enhance their resistance without compromising their excellent agronomic traits, so that they can continue to be used commercially.

Conclusions
This study combines conventional breeding with marker validation. Through the evaluation of the field agronomic traits and molecular detection results of resistant QTL, 27 wheat lines with two or more QTL and showing high resistance to wheat stripe rust were successfully selected.
Among the 27 lines selected, a total of 13 lines also possessed several excellent agronomic traits, such as moderate plant height, spikelet number, number of tillers, and 1000 grain weights that were similar to or better than the three leading wheat varieties. This will enrich the genetic resources of wheat stripe rust resistance in China and provide material support and a theoretical basis for the prevention and control of wheat stripe rust in China.