Physical Mapping of Stem Rust Resistance Gene Sr52 from Dasypyrum villosum Based on ph1b-Induced Homoeologous Recombination

Wheat stem rust caused by Puccinia graminis f. sp. tritici (Pgt) had been a devastating foliar disease worldwide during the 20th century. With the emergence of Ug99 races, which are virulent to most stem rust resistance genes deployed in wheat varieties and advanced lines, stem rust has once again become a disease threatening global wheat production. Sr52, derived from Dasypyrum villosum and mapped to the long arm of 6V#3, is one of the few effective genes against Ug99 races. In this study, the wheat–D. villosum Robertsonian translocation T6AS·6V#3L, the only stock carrying Sr52 released to experimental and breeding programs so far, was crossed with a CS ph1b mutant to induce recombinants with shortened 6V#3L chromosome segments locating Sr52. Six independent homozygous recombinants with different segment sizes and breakpoints were developed and characterized using in situ hybridization and molecular markers analyses. Stem rust resistance evaluation showed that only three terminal recombinants (1381, 1380, and 1392) containing 8%, 22%, and 30% of the distal segment of 6V#3L, respectively, were resistant to stem rust. Thus, the gene Sr52 was mapped into 6V#3L bin FL 0.92–1.00. In addition, three molecular markers in the Sr52-located interval of 6V#3L were confirmed to be diagnostic markers for selection of Sr52 introgressed into common wheat. The newly developed small segment translocation lines with Sr52 and the identified molecular markers closely linked to Sr52 will be valuable for wheat disease breeding.


Introduction
Common wheat (Triticum aestivum L., 2n = 6x = 42) is one of the most important crops serving as a daily staple for human consumption worldwide. Stem rust caused by the fungus Puccinia graminis f. sp. tritici (Pgt) used to be one of the most destructive fungal diseases of common wheat [1,2]. During the twentieth century, several severe stem rust epidemics occurred in all major wheat growing areas, causing significant yield losses ranging from 10 to 59% [2][3][4]. In the past several decades, epidemics of stem rust were successfully controlled mainly due to widespread employment of resistance genes in wheat varieties [5]. However, a new Pgt race, Ug99 and its 13 variants known as the Ug99 race group, has emerged since 1999 [6,7]. These are virulent to the overwhelming majority of stem rust resistance genes that have been deployed in wheat breeding programs worldwide, including Sr24 [8], Sr31 [6], Sr36 [9], and Sr38 [10]. Therefore, currently only a few effective resistance genes against the Ug99 race group are present in common wheat varieties and advanced lines [11]. Thus, there is an urgent need to identify novel resistance genes from wild relatives of common wheat to broaden stem rust resistance resources in wheat breeding programs [12].
Deleterious linkage drag, caused by the presence of additional genes on the alien chromosome segments introgressed into wheat, may lead to unfavorable agronomic and end-use quality traits limiting the direct utilization of alien genes in wheat improvement programs. The most effective method to reduce linkage drag includes developing wheat-alien recombinants with small alien segments by inducing recombination between the alien chromatin and its homoeologous region of wheat chromosome [22]. However, the Ph genes which control homologous chromosome pairing in common wheat, such as the Ph1 gene on chromosome 5BL, can prevent homoeologous chromosomes from pairing and recombining [23,24]. Lack of or suppression of Ph genes in common wheat can promote meiotic homoeologous recombination between wheat and its wild relatives [25]. A deletion mutant of Ph1 (ph1b) has been reported to be most effective to induce wheat-alien homoeologous recombination [16,26,27]. Some stem rust resistance genes, such as Sr32 and Sr39 from Aegilops speltoides [28,29], Sr43 from Thinopyrum ponticum [30], and Sr53 from Ae. geniculata [31], have been successfully introduced into common wheat from wild relatives based on ph1b-induced homoeologous recombination.
The objectives of this study were to develop wheat-D. villosum 6V#3L translocation lines with shortened 6V#3L segments carrying Sr52 by using ph1b-induced homoeologous recombination, to physically map Sr52 onto the shortened 6V#3L segment and also to select diagnostic polymerase chain reaction (PCR)-based molecular markers for Sr52 detection. Our work provides three novel Sr52 germplasms with small 6V#3L segments and three Sr52 diagnostic molecular markers for wheat-breeding programs, and lays a foundation for further cloning of the stem rust resistance gene Sr52.

Initial Screening of Wheat-D. villosum 6V#3L Recombinants with Proximal and Distal Markers
A total of 250 BC 1 F 2 plants derived from BC 1 F 1 individuals with homozygous ph1b plus monosomic 6AS·6V#3L and 6A were screened by using 6V#3L-specific molecular markers BE422631/HaeIII, located in proximal bin C-6AL4-0.55, and 6L11/MboI, located in distal bin 6AL8-0.90-1.00. Six putative recombinants were selected based on disassociation of the two molecular markers, of which three plants (1386, 1382, and 1385) were positive for the proximal marker BE422631/HaeIII but negative for the distal marker 6L11/MboI, and the remaining three plants (1381, 1380, and 1392) had only distal 6V#3L marker 6L11/MboI. Genomic in situ hybridization (GISH) analyses of these six plants confirmed that the first three plants, lacking distal marker 6L11/MboI, were interstitial recombinants, whereas the last three plants, missing proximal marker BE422631/HaeIII, were terminal recombinants ( Figure 1).

Analyses of Segment Sizes and Breakpoints of Chromosome 6V#3L
Homozygous progenies derived from six recombinant plants were used to further identify the segment sizes and breakpoints of the translocated chromosomes by eight 6V#3L-specific molecular markers described in Table 1, GISH, and nondenaturing fluorescence in situ hybridization (ND-FISH). The results of molecular markers and cytogenetic analyses revealed that all six recombinants were different in both segment size and breakpoints of chromosome 6V#3L (Table 2).

Analyses of Segment Sizes and Breakpoints of Chromosome 6V#3L
Homozygous progenies derived from six recombinant plants were used to further identify the segment sizes and breakpoints of the translocated chromosomes by eight 6V#3L-specific molecular markers described in Table 1, GISH, and nondenaturing fluorescence in situ hybridization (ND-FISH). The results of molecular markers and cytogenetic analyses revealed that all six recombinants were different in both segment size and breakpoints of chromosome 6V#3L (Table 2). The combined GISH and ND-FISH patterns revealed that recombinants 1386, 1382, and 1385 were all interstitial translocation Ti6AS·6V#3L-6AL, and their breakpoints were at FL 0.70, FL 0.78, and FL 0.92, being equal to a segment size of 70%, 78%, and 92% of the long arm length of chromosome 6V#3, respectively ( Figure 2). Molecular marker analyses of these three interstitial translocation lines using eight primer sets (Table 1)

Physical Mapping of the Gene Sr52
All six homozygous translocation lines were evaluated for stem rust resistance, together with their recipient parent CS as a susceptible control, and CS-D. villosum 6V#3 disomic addition line TA7682 and CS-D. villosum T6AS·6V#3L translocation line TA5617 as resistant controls. The results showed that all three terminal translocation lines (1381, 1380, and 1392) were resistant to stem rust (ITs; to 1) (Figure 4), whereas all three interstitial translocation lines (1386, 1382, and 1385) were susceptible (ITs 3 to 4) (Figure 4). Since chromosome 6V#3L bin FL 0.92-1.00 was the only bin shared by all the resistant translocation lines, and moreover was absent from the sensitive translocation lines, the stem rust resistance gene Sr52 from D. villosum must be physically mapped to the interval of FL 0.92-1.00 of chromosome 6V#3L ( Figure 5). Three 6V#3L-specific molecular markers (CINAU1532, 6L-4, and 6L11/MboI) present in all terminal translocation lines are also located in the same bin carrying Sr52 (Figure 3). GISH patterns showed that recombinants 1381, 1380, and 1392 belonged to terminal translocation with breakpoints at FL 0.92, 0.78, and 0.70, which indicated that 6V#3L segment size was 8%, 22%, and 30% of the 6V#3L length, respectively ( Figure 2). FISH analyses displayed that the translocated chromosomes were T6AS·6AL-6V#3L in recombinants 1381 and 1392 ( Figure 2D,F), whereas it was T6DS·6DL-6V#3L in line 1380 ( Figure 2E), being composed of full wheat chromosome arm 6DS, centromere, and proximal segment of chromosome 6DL conjoined to the distal segment of 6V#3L. In addition, recombination occurred between wheat chromosome 6A and 6D, producing a pair of T6AS.6AL-6DL translocated chromosomes in line 1380. Molecular marker analyses of these three terminal translocation lines revealed that six and five distal 6V#3L-specific markers were present in 1392 and 1380, respectively, whereas only three distal 6V#3L-specific markers were present in 1381 ( Figure 3). Thus, the order of the segment sizes of the translocated chromosome 6V#3L in these three terminal translocation lines was 1381 < 1380 < 1392.

Physical Mapping of the Gene Sr52
All six homozygous translocation lines were evaluated for stem rust resistance, together with their recipient parent CS as a susceptible control, and CS-D. villosum 6V#3 disomic addition line TA7682 and CS-D. villosum T6AS·6V#3L translocation line TA5617 as resistant controls. The results showed that all three terminal translocation lines (1381, 1380, and 1392) were resistant to stem rust (ITs; to 1) (Figure 4), whereas all three interstitial translocation lines (1386, 1382, and 1385) were susceptible (ITs 3 to 4) (Figure 4). Since chromosome 6V#3L bin FL 0.92-1.00 was the only bin shared by all the resistant translocation lines, and moreover was absent from the sensitive translocation lines, the stem rust resistance gene Sr52 from D. villosum must be physically mapped to the interval of FL 0.92-1.00 of chromosome 6V#3L ( Figure 5). Three 6V#3L-specific molecular markers (CINAU1532, 6L-4, and 6L11/MboI) present in all terminal translocation lines are also located in the same bin carrying Sr52 (Figure 3).

Validation of New Molecular Markers Linked to Sr52 on the Shortened D. villosum 6V#3L Chromosome Segments
The three molecular markers CINAU1532, 6L-4, and 6L11/MboI located in the same 6V#3L segment carrying Sr52 were further validated with 14 common wheat varieties and advanced breeding lines ( Figure 6). The results showed that primer sets for all three markers generated polymorphic bands only in the materials carrying Sr52 including CS-D. villosum 6V#3 disomic addition line TA7682, CS-D. villosum T6AS·6V#3L translocation line TA5617, and the terminal small segment recombinants 1381, 1380, and 1392, but not in the susceptible CS or in the other 14 wheat cultivars and advanced breeding lines. Therefore, molecular markers CINAU1532, 6L-4, and 6L11/MboI can serve as diagnostic markers to perform marker-assisted selection of Sr52 in future

Validation of New Molecular Markers Linked to Sr52 on the Shortened D. villosum 6V#3L Chromosome Segments
The three molecular markers CINAU1532, 6L-4, and 6L11/MboI located in the same 6V#3L segment carrying Sr52 were further validated with 14 common wheat varieties and advanced breeding lines ( Figure 6). The results showed that primer sets for all three markers generated polymorphic bands only in the materials carrying Sr52 including CS-D. villosum 6V#3 disomic addition line TA7682, CS-D.
villosum T6AS·6V#3L translocation line TA5617, and the terminal small segment recombinants 1381, 1380, and 1392, but not in the susceptible CS or in the other 14 wheat cultivars and advanced breeding lines. Therefore, molecular markers CINAU1532, 6L-4, and 6L11/MboI can serve as diagnostic markers to perform marker-assisted selection of Sr52 in future wheat resistance breeding programs.

Discussion
Wild relatives of common wheat possess a valuable genetic pool of beneficial traits that could be used for wheat improvement. Development of compensating wheat-alien translocation lines with small alien segments has been considered an effective method to transfer the desirable genes from wild relatives to wheat cultivars in crop improvement programs [30,36]. The gene Sr52 from D. villosum is one of the few resistance genes effective against Ug99 races and was previously mapped to the long arm of 6V#3 [15]. In the present study, we attempted to reduce Sr52 linkage drag using ph1b-induced homoeologous recombination and selecting for the Sr52 recombinant with the least amount of alien chromatin. Of the three translocation lines (1381, 1380, and 1392) carrying Sr52, 6V#3L segments were shortened by approximately 92%, 78%, and 70%, respectively, compared with the original stock TA5617. These recombinants carrying Sr52 with substantial reduction of 6V#3L chromatin should be valuable germplasms for wheat stem rust resistance breeding as well as for eventual isolation of Sr52.
Integrating cytogenetic and molecular marker analyses play an important role in physical mapping of desirable agronomic genes from wild relatives of common wheat. For example, the powdery mildew resistance gene Pm57 was cytogenetically mapped to 2S s #1L bin FL 0.75-0.87 [37]. A resistant gene(s) against powdery mildew from Agropyron cristatum was physically mapped to 2PL bin 0.66-0.86 [38]. The wheat yellow mosaic virus resistance gene Wss1 from D. villosum was located in the bin FL0.78-1.00 of 4VS [16]. The gene(s) conferring blue-grained character from Th. ponticum was located in bin 4AgL-6 of FL 0.75-0.89 [39]. In the present study, six wheat-D. villosum 6V#3L translocation lines possessing different segment sizes of D. villosum chromatin were identified. By combining cytogenetic and molecular marker analyses and evaluation of sensitivity to stem rust, the gene Sr52 was further mapped into the chromosome bin of FL 0.92-1.00, the very distal region of D. villosum 6V#3L.
Development and utilization of molecular markers closely linked to targeted genes was an effective method for selecting desirable traits in the early generation of the breeding program [21]. In this study, three molecular markers (CINAU1532, 6L-4, and 6L11/MboI) tightly linked to Sr52 were validated to be present only in lines carrying Sr52 but not in all 14 tested wheat varieties and advanced lines. These three markers will greatly facilitate the utilization of the recombinants with

Discussion
Wild relatives of common wheat possess a valuable genetic pool of beneficial traits that could be used for wheat improvement. Development of compensating wheat-alien translocation lines with small alien segments has been considered an effective method to transfer the desirable genes from wild relatives to wheat cultivars in crop improvement programs [30,36]. The gene Sr52 from D. villosum is one of the few resistance genes effective against Ug99 races and was previously mapped to the long arm of 6V#3 [15]. In the present study, we attempted to reduce Sr52 linkage drag using ph1b-induced homoeologous recombination and selecting for the Sr52 recombinant with the least amount of alien chromatin. Of the three translocation lines (1381, 1380, and 1392) carrying Sr52, 6V#3L segments were shortened by approximately 92%, 78%, and 70%, respectively, compared with the original stock TA5617. These recombinants carrying Sr52 with substantial reduction of 6V#3L chromatin should be valuable germplasms for wheat stem rust resistance breeding as well as for eventual isolation of Sr52.
Integrating cytogenetic and molecular marker analyses play an important role in physical mapping of desirable agronomic genes from wild relatives of common wheat. For example, the powdery mildew resistance gene Pm57 was cytogenetically mapped to 2S s #1L bin FL 0.75-0.87 [37]. A resistant gene(s) against powdery mildew from Agropyron cristatum was physically mapped to 2PL bin 0.66-0.86 [38]. The wheat yellow mosaic virus resistance gene Wss1 from D. villosum was located in the bin FL0.78-1.00 of 4VS [16]. The gene(s) conferring blue-grained character from Th. ponticum was located in bin 4AgL-6 of FL 0.75-0.89 [39]. In the present study, six wheat-D. villosum 6V#3L translocation lines possessing different segment sizes of D. villosum chromatin were identified. By combining cytogenetic and molecular marker analyses and evaluation of sensitivity to stem rust, the gene Sr52 was further mapped into the chromosome bin of FL 0.92-1.00, the very distal region of D. villosum 6V#3L.
Development and utilization of molecular markers closely linked to targeted genes was an effective method for selecting desirable traits in the early generation of the breeding program [21]. In this study, three molecular markers (CINAU1532, 6L-4, and 6L11/MboI) tightly linked to Sr52 were validated to be present only in lines carrying Sr52 but not in all 14 tested wheat varieties and advanced lines. These three markers will greatly facilitate the utilization of the recombinants with the shortened D. villosum 6V#3L segment carrying Sr52. Additionally, it could assist tracking Sr52 when pyramiding Sr52 with other useful disease resistance genes or superior agronomic traits-related loci in the future.
In conclusion, the gene Sr52 was further mapped to D. villosum chromosome 6V#3L distal bin FL 0.92-1.00 by integrated analyses of molecular markers, GISH, FISH, and stem rust sensitivity evaluation of six newly developed homozygous wheat-D. villosum translocation lines in this study. The three translocation lines carrying Sr52 with shortened D. villosum 6V#3L chromatin will provide potentially useful germplasms to breeders for wheat breeding programs aimed at eliminating susceptibility to stem rust. In addition, we identified three molecular markers closely linked to Sr52, which will be helpful for marker-assisted introgression of Sr52 into wheat varieties and breeding lines.

Plant Materials
The plant materials used in this study included common wheat CS TA3808, CS ph1b mutant stock TA3809, which lacked the Ph1 gene and thereby elevated homoeologous recombination, CS-D. villosum 6V#3 disomic addition line TA7682, CS-D. villosum T6AL·6V#3S translocation line TA5618, and T6AS·6V#3L translocation line TA5617. The long arm of chromosome 6V#3 in lines TA7682 and TA5617 carried Sr52 which conferred resistance to stem rust of wheat (Table 3). All materials were kindly provided by the Wheat Genetics Resource Center (WGRC) at Kansas State University and maintained at the experimental station of Henan Agricultural University, China.

Molecular Marker Analyses
Genomic DNA (gDNA) was extracted from fresh leaves at the two-leaf stage following the method of   [38]. The concentration and purity of DNA were measured with the NanoPhotometer P360 (Implen GmbH, München, Germany).
PCR reactions were conducted in a 15 µL volume containing 2.0 µL template gDNA (100 ng/µl), 1.0 µL of each primer (5.0 µmol/L), 7.5 µl Taq MasterMix (CW Bio Inc., Beijing, China), and 3.5 µl ddH 2 O. The PCR program was conducted at 94 • C for 5 min, followed by 35 cycles of melting at 94 • C for 30 s, annealing at 55-66 • C (depending on the Tm of the primers) for 30 s, and extension at 72 • C for 1 min, with a final extension at 72 • C for 10 min. The PCR products were digested with restriction enzymes. Five microliters of a restriction enzyme mixture containing 2.85 µL of ddH 2 O, 2.0 µL of CutSmart buffer, and 0.15 µL of an enzyme stock solution were added to 15 µL of PCR products and incubated for 3.5 h at 37 • C. The PCR or restricted PCR products were separated on a 2.0% agarose gel, stained with ethidium bromide, and visualized by Tanon 2500 Gel Imaging System (Tanon Science & Technology Co., Ltd., Shanghai, China).

Cytogenetic Analyses
Chromosome preparations of root tip cells at the mitotic metaphase were obtained as described by Huang et al. (2018) [42]. The cytological observations were performed using a BX51 Olympus phase contrast microscope (Olympus Corporation, Tokyo, Japan).
The technique of GISH was performed according to Liu et al. (2017) [37]. Total gDNA of D. villosum labeled with fluorescein-12-dUTP was used as probe, and gDNA of common wheat CS was used as blocking for GISH. Fluorescent images were captured with an AxioCam MRc5 CCD camera using a Zeiss Axio Scope A1 fluorescence microscope (Carl Zeiss AG, Oberkochen, Germany). Images were processed using Photoshop CS 3.0 (Adobe Inc., San Jose, CA, USA).

Stem Rust Evaluation
Evaluation of stem rust resistance was performed at the seedling stage using the major dominant race 21C3CTTTM in China as inoculants following the procedure of   [44]. Fourteen days after inoculation, infection types (ITs) were assessed using a 0-4 scale as described by Roelfs et al. (1988) [45], with 0 as immune, as necrotic flecks, 1 as small necrotic pustules, 2 as small to medium-sized chlorotic pustules with green islands, 3 as medium-sized chlorotic pustules, and 4 as large pustules without chlorosis. Plants with ITs 0-2 were classified to be resistant while plants with ITs 3-4 were scored as susceptible.

Conflicts of Interest:
The authors declare no conflict of interest.