Fine mapping of powdery mildew resistance gene MlWE74 derived from wild emmer wheat (Triticum turgidum ssp. dicoccoides) in an NBS-LRR gene cluster

Powdery mildew resistance gene MlWE74, originated from wild emmer wheat accession G-748-M, was mapped in an NBS-LRR gene cluster of chromosome 2BS. Wheat powdery mildew, caused by Blumeria graminis f. sp. tritici (Bgt), is a globally devastating disease. Wild emmer wheat (Triticum turgidum var. dicoccoides) is a valuable genetic resource for improving disease resistance in common wheat. A powdery mildew resistance gene was transferred to hexaploid wheat line WE74 from wild emmer accession G-748-M. Genetic analysis revealed that the powdery mildew resistance in WE74 is controlled by a single dominant gene, herein temporarily designated MlWE74. Bulked segregant analysis (BSA) and molecular mapping delimited MlWE74 to the terminal region of chromosome 2BS flanking by markers WGGBD412 and WGGBH346 within a genetic interval of 0.25 cM and corresponding to 799.9 kb genomic region in the Zavitan reference sequence. Sequence annotation revealed two phosphoglycerate mutase-like genes, an alpha/beta-hydrolases gene, and five NBS-LRR disease resistance genes that could serve as candidates for map-based cloning of MlWE74. The geographical location analysis indicated that MlWE74 is mainly distributed in Rosh Pinna and Amirim regions, in the northern part of Israel, where environmental conditions are favorable to the occurrence of powdery mildew. Moreover, the co-segregated marker WGGBD425 is helpful in marker-assisted transfer of MlWE74 into elite cultivars.


Introduction
1 3 temperate climates. Breeding for resistance is the most economical and effective strategy to control powdery mildew. Up to now, more than a hundred powdery mildew resistance genes/alleles have been documented, and some of them have played important roles in stabilizing wheat yield, such as Pm21 (He et al. 2018;Xing et al. 2018), Yr18/Lr34/Pm38/Sr57 (Krattinger et al. 2009), and Pm46/Yr46/Lr67/Sr55 (Moore et al. 2015). However, the emergence of new virulent pathotypes of Bgt reduces the resistance conferred by resistance (R) genes (Singh et al. 2016). Recent studies indicated that Pm2, Pm3a,Pm3b,Pm3f,Pm4a,Pm6,Pm8, and Pm17 have been overcome in part or all of the USA, while Pm1a, Pm3a, and Pm8 were defeated in Australia, China, and Egypt (Parks et al. 2008;Cowger et al. 2018). Therefore, in response to the newly evolved Bgt virulent isolates, it is necessary to continuous search for new powdery mildew resistance genes.
Among the powdery mildew resistance genes that were currently reported, more than half of them are derived from diploid or tetraploid wild relatives of wheat including Aegilops squarrosa, Ae. speltoides, Ae. longissima, Ae. ovata, Dasypyrum villosum, T. urartu, T. turgidum var. dicoccoides, T. turgidum var. dicoccum, T. turgidum var. durum, T. timopheevii, T. monococcum, Thinopyrum intermedium, and cereal rye (Secale cereale L.) (http:// wheat. pw. usda. gov/). The wild relatives of wheat are important sources for discovering wheat powdery mildew resistance genes. Transfer of alien genes is still an effective strategy for increasing the genetic diversity of powdery mildew resistance in wheat breeding.
Wild emmer wheat accession G-748-M is resistant to Bgt isolate E09. The powdery mildew resistance gene from this accession was transferred to hexaploid wheat by crossing and backcrossing with susceptible common wheat cultivars, resulting in common wheat line WE74 (YD1817/G-748-M//7*ND015). Line WE74 conferred highly resistance to Bgt isolate E09 at the seedling stage in the greenhouse and the adult plant stage in fields. The objectives of this study are to fine map the powdery mildew resistance gene in WE74, with an ultimate goal of cloning the powdery mildew resistance gene and providing breeders with markers that can be used in marker-assisted breeding.

Plant materials
Wild emmer wheat accession G-748-M was kindly provided by Dr. ZK Gerechter-Amitai, Agricultural Research Organization, The Volcani Centre, Israel. G-748-M was resistant, with infection type (IT) 0, to Bgt isolate E09, a local prevailing Bgt isolate virulent to Pm1, Pm3 and Pm8 but avirulent to Pm2, Pm4, Pm5e, Pm6, Pm21, Pm24, Pm36, and Pm41 (Cowger et al. 2018;Xie et al. 2020;He et al. 2018;Xing et al. 2018;Lu et al. 2020;Blanco et al. 2008;Li et al. 2020). Yanda 1817 and ND015 were used as susceptible parental lines for crossing and backcrossing to transfer resistance gene from G-748-M to common wheat, resulting in the powdery mildew resistant line WE74 (YD1817/G-748-M//7*ND015). WE74 was highly resistant to Bgt isolate E09 both at the seedling and adult growth stages. WE74 was crossed to the susceptible wheat Xuezao (XZ), developing 165 F 2 plants and their F 2:3 families for genetic analyses and genetic mapping. A large F 2 population including 2107 plants from a cross between WE74 and XZ was used to construct a high-density linkage map. A collection of 461 wild emmer wheat accessions from different geographical collections were used to test the distribution of the powdery mildew resistance gene identified in WE74.

Powdery mildew evaluations
The parental lines WE74, XZ and the corresponding F 1 , F 2 and F 2:3 materials and the recombinant families from the mapping populations were evaluated for response to powdery mildew at two-leaf stage. The inoculated plants were grown under a daily cycle of 16 h of light and 8 h of darkness at 22 ± 2 °C in a greenhouse. The resistant and susceptible parents were planted in the middle of each tray as the resistant and the susceptible controls, respectively. Seedlings with unfolded first leaves were inoculated with Bgt E09 by dusting of conidiospores. Infection types (ITs) were evaluated after 15 d on a scale of 0-4, in which, 0, 0; 1, 2, 3, and 4 represented immune, necrotic fecks, high resistance, moderate resistance, moderate susceptibility and high susceptibility, respectively. Phenotypes were classified into two groups, resistant (R, IT 0-2) and susceptible (S, IT 3-4) (Liu et al. 1999). WE74, IW170 and Pm26-40, carrying MlWE74, MlIW170 and Pm26, respectively, were also challenged by 11 Bgt isolates collected from different regions of China.

Genomic DNA isolation and marker analysis
Genomic DNA was extracted from parental lines, F 2 plants, F 2:3 families and wild emmer wheat accessions following the CTAB method (Devi et al. 2013). For bulked segregant analysis, separate DNA bulks were assembled using equal amounts of DNA from ten homozygous resistant and ten homozygous susceptible F 2 plants, respectively. Wheat microsatellite markers (Xgwm, Xwmc, Xbarc, Xcfa, Xcfd and Xcau series) mapped on A and B genome chromosomes (Graingenes, http:// wheat. pw. usda. gov/) were chosen for marker analyses. Polymorphic markers indicative of linkage with the powdery mildew resistance gene were further used to genotype the entire F 2:3 mapping population to determine genetic linkage between the gene and the markers. Based on genomic locations of the identified SSR markers in the Zavitan reference sequence , SSR, STS and InDel markers in the target region of the powdery mildew resistance gene were developed for linkage analysis.

Data analysis
Genetic analysis was performed to examine the expected segregation ratios in the F 2 and F 2:3 from WE74 × XZ populations using a Chi-squared (χ2) test. MAPMAKER 3.0 (Lander et al. 1987) was used to construct a linkage map, with a LOD score of 3.0 as the threshold. The genetic map was drawn with the software Mapdraw V2.1 (Liu and Meng 2003).

Geographical distribution analysis
A total of 461 accessions of wild emmer wheat collected from natural populations, representing a wide range of ecogeographic distribution of wild emmer wheat in Israel, Lebanon, Syria, Turkey, and its vicinity were used to profile the distribution of the powdery mildew resistance gene. The co-segregating markers were used to detect the collection of wild emmer wheat.

Inheritance of powdery mildew resistance in WE74
Genetic analysis was carried out to investigate the inheritance mode of powdery mildew resistance in WE74. WE74 was resistant (IT 0), XZ was susceptible (IT 4), and the F 1 plants were resistant (IT 0;), indicating dominance of the resistance (Fig. 1). We initially phenotyped 165 F 2 plants, of which 120 were resistant and 45 were susceptible. The powdery mildew resistance was shown to segregate as a single  (Table 1). To verify this result, the corresponding F 2:3 progenies segregated as 38 homozygous resistant: 82 segregating: 45 homozygous susceptible families, fitting to the ratio 1:2:1. Therefore, these results indicate that the powdery mildew resistance in WE74 was controlled by a single dominant gene, temporarily designated MlWE74.

Chromosomal location of MlWE74
A total of 352 wheat SSR markers (Xgwm, Xwmc, Xbarc, Xcfa, Xcfd and Xcau) mapped to A and B genomes were screened for polymorphism between the resistant and susceptible F 2 DNA bulks for bulked segregant analysis. Polymorphic SSR markers were selected to genotype the 165 F 2 plants from WE74 × XZ cross DNA samples. Nine polymorphic SSR markers, Xgwm210, Xgwm614, Xbarc297, Xcau357, Xwmc243, Xgwm257, Xbarc7, Xbarc55, and Xwmc477, were detected linked to MlWE74 (Table S1), and a genetic linkage map was constructed. All the nine SSR markers were located on chromosome 2B (Fig. 2a). Gene MlWE74 was localized to a 1.8 cM genetic interval between flanking markers Xcau357 and Xwmc243 at genetic distances of 1.2 cM (distal) and 0.6 cM (proximal), respectively (Fig. 2b). These results verified that the powdery mildew resistance gene MlWE74 was located on chromosome arm 2BS.

Fine mapping of MlWE74
In order to construct a high-resolution genetic linkage map for the powdery mildew resistance gene MlWE74, a large segregating population including 2107 F 2 plants derived from the cross between WE74 and XZ was developed. The two flanking molecular markers Xcau357 and Xwmc243 were used to genotype the entire F 2 population, resulting in 76 F 2 recombinants between them. The F 2:3 families of these recombinants were tested for powdery mildew resistance and then used for genotyping by the newly developed polymorphic markers. From 102 SSR and InDel primer pairs designed according to the WEW Zavitan (v2.0) and Chinese Spring reference genomes, nine polymorphic markers were developed and integrated into the genetic linkage map of MlWE74 after genotyping the 165 F 2:3 families of the low-resolution mapping population and all the 76 recombinants identified from the fine genetic mapping population (Tables S1 and S2). Finally, MlWE74 was narrowed down into a 0.25 cM genetic interval franked by markers WGGBD412 and WGGBH346 (Fig. 2c and d). Molecular marker WGGBD425 co-segregated with MlWE74 in the WE74 and XZ population (Table S1; Fig. 2c and d; Fig.  S1a).

Gene annotation of the MlWE74 genomic region
The flanking markers WGGBD412 and WGGBH346 were used to identify the corresponding genomic region of the WEW Zavitan (v2.0) reference genome sequence. Finally, MlWE74 was delimited to an approximate 799.9 kb physical interval. Eight protein-coding genes in this physical interval were predicted (Fig. 2e). These predicted genes included 2 phosphoglycerate mutase-like proteins (TRIDC2BG005010 and TRIDC2BG005040), 5 NBS-LRR disease resistance proteins (TRIDC2BG005090, TRIDC2BG005100, TRID-C2BG005110, TRIDC0UG006500 and TRIDC2BG005230) and an alpha/beta hydrolases protein (TRIDC2BG005120). It has an obvious NBS-LRR gene cluster in the mapping region of MlWE74.

Micro-collinearity analysis of the MlWE74 locus among tetraploid and hexaploid wheat genomes
The MlWE74 co-segregated marker WGGBD425 and two flanking markers WGGBD412 and WGGBH346 were used to search against the durum wheat cv. Svevo reference genome and hexaploid wheat genomes, including Chinese Spring, Fielder, Julius_MAGIC3, Landmark, Jagger, ArinalFor, Mattis, Longreach Lancer, Mace, Norin61 and spelt wheat PI 190962, to define the corresponding physical intervals of locus MlWE74 in these genomes. The collinear relationship of the protein-coding genes in these intervals revealed highly micro-collinearity among tetraploid and hexaploid wheat genomes (Table S3). Six genes, TRIDC2BG005010, TRIDC2BG005040, TRIDC2BG005090, TRID-C2BG005100, TRIDC2BG005110, TRIDC2BG005110, and TRIDC2BG005230, in Zavitan genome were syntenic among different genomes, whereas TRIDC2BG005120 and TRIDC0UG006500 were only present in Zavitan ( Fig. 3; Table S4).

Geographical distribution analysis of MlWE74
The distribution of MlWE74 locus was investigated by screening the geographically diverse accessions of wild emmer wheat using the co-segregating marker WGGBD425. The positive amplicon was found in 15 WEW accessions that were mainly present in the region of Rosh Pinna and Amirim in Israel belong to the northern WEW population of Israel ( Fig. 4; Table S5). The frequency of MlWE74 is only 3.3% in 461 wild emmer wheat accessions tested. Powdery mildew tests confirmed that all these WEW accessions carrying MlWE74 co-segregated marker were resistant to Bgt E09 (Table S6).

Comparison of MlWE74 with known powdery mildew resistance genes on 2BS
A dominant powdery mildew resistance gene MlWE74 derived from wild emmer wheat was identified and finely mapped to a 799.9 kb genomic region on chromosome 2BS according to the reference genome of WEW cv.
Zavitan (v2.0) (Fig. 2). Up to now, eight powdery mildew resistance genes in wheat are reported on chromosome arm 2BS, including three permanently designated genes Pm26 (Rong et al. 2000),  (Fig. 5a and c). The smallest physical intervals of the other six powdery mildew resistance genes, Pm42, Pm68, Ml5323, PmL962, pmWE99  . Ml5323 derived from Triticum turgidum ssp. dicoccum has different origins with MlWE74. Moreover, allelism test showed that resistance genes Pm26 and Ml5323 are not allelic (Piarulli et al. 2012). The recessive gene pmWE99, originated from wheat-Thinopyrum intermedium, was mapped between chromosome 2BS distal terminus to marker Xgwm148. MlWE74 is a dominant powdery mildew resistance gene derived from wild emmer wheat. Hence, MlWE74 is different from those known genes on chromosome 2BS according to their difference in physical locations, effects and origins. Powdery mildew resistance gene MlIW170 derived from wild emmer wheat was located on the physical interval Chr2B_Zavitan v2.0: 26.41-27.25 Mb in the reference genome (Fig. 5h). The physical interval of MlIW170 is different from the region of MlWE74, Chr2B_Zavitan v2.0: 25.48-26.28 Mb (Fig. 5h). The Pm26-derived from wild emmer wheat was mapped to 2BS, co-segregating with RFLP marker Xwg516 (Rong et al. 2000). The STS marker Xcau516 was developed from Xwg516 co-segregated with MlIW170 in an F 2 population. These results indicate that Pm26 and MlIW170 are likely to be identical, or allelic (Liang et al. 2015). The reactions of the three genes to 11 Bgt isolates revealed different infection types (Table S7).
The MlWE74 co-segregated marker WGGBD425 amplified null allele in Pm26-40 and IW170, indicating the presence and absence variation (PAV) in the genomic intervals harboring the three genes (Fig. S1b). Pm26 was described as a recessive gene (Rong et al. 2000). MlIW170 was reported as an incompletely dominant gene (Liu et al. 2012). However, genetic analysis of these two genes was performed in hexaploid and tetraploid levels, respectively. The polyploid level and genetic background difference may influence inheritance pattern of the two genes. In this study, MlWE74 was found as a dominant gene in common wheat genetic background. From the current available data, it is difficult to figure out if Pm26, MlIW170, and MlWE74 are the different alleles in the same locus or different loci in the same genomic region. Beside of gene cloning, an allelism test is necessary to clarify their relationship in future.

Genomic structure and micro-collinearity analysis of the MlWE74 locus
The genomic region of MlWE74 (799.9 kb) according to the reference genome of WEW cv. Zavitan (v2.0) corresponds to the physical intervals from 430.1 kb to 491.5 kb in the genomes of Svevo, Chinese Spring, Fielder, Julius_MAGIC3, Landmark, Jagger, ArinalFor, Mattis and PI 190962, and from 1663.9 kb to 1669.8 kb in the genomes of Longreach Lancer, Mace and Norin61   Fig. 3). Since the protein-coding genes are relatively conserved in those genomes, the physical region difference from 400 to 800 kb and from 800 kb to 1600 kb may mainly resulted from transposons and retrotransposons amplifications in those genomes, together with some NBS-LRR gene duplications.

Geographical distribution of MlWE74 and potential value in wheat breeding
Wild emmer wheat is a valuable source in improving both durum and common wheat due to its direct ancestry and rich genetic diversity (Nevo et al. 2014). It is mainly distributed in Israel, Syria, Jordan, Lebanon, south-east Turkey, northern Iraq, and western Iran (Nevo et al. 2014). We used 461 accessions of wild emmer wheat, representing a wide range of ecogeographic distribution in Near-Eastern Fertile Crescent natural populations, to profile the distribution of MlWE74. A low frequency of MlWE74 was detected in the region of Rosh Pinna and Amirim in Israel that were assigned to narrow area in Mount Hermon region, which is sometimes considered to be the southernmost extension with higher and cooler elevation (up to 1600-1900 m) and favorable climatic conditions for disease development (Li et al. 2020b). The pathogens environment of high selective pressure facilitates the co-evolution of disease-resistant genes. The absence of MlWE74 in the WEW natural populations from southeastern Turkey, where wheat is believed to be domesticated, suggests that MlWE74 may not participate in the process of gene transfer into cultivated wheat during wheat domestication and polyploidization. Therefore, MlWE74 derived from WEW would be a valuable resource for disease resistance. The co-segregating molecular marker WGGBD425 could be used for markerassisted selection in wheat breeding program.