An exome capture panel of the Triticeae D genome facilitates defining the introgression landscape of <i>Aegilops tauschii</i> ‐wheat derivatives

To diagnose introgression from ancestor species Aegilops tauschii into common wheat, the ATPanExomeV1 genotyping panel was developed based on the genomes of Ae. tauschii and modern wheat. The panel showed significant potential for recognizing introgression segments, structural variations, and discovering new genes or haplotypes in the derivatives of Ae. tauschii. As a limited number of Aegilops tauschii accessions were involved in the formation of common wheat, the D-subgenome showed the narrowest genetic diversity compared to the A and B subgenomes (Zhou et al., 2021; Cavalet-Giorsa et al., 2024). Practically, introgression of favorable genes from Ae. tauschii into common wheat can enhance genetic diversity and is a typical strategy to overcome this genetic bottleneck (Cox et al., 2017; Zhou et al., 2021; Li et al., 2024). Despite the generation of numerous introgression lines by both synthetic hexaploid wheat (SHW) and synthetic octaploid wheat (SOW) approaches, challenges related to the high-throughput evaluation of these lines have limited the extensive application of introgression breeding. Although whole-genome sequence-based approaches could facilitate the identification of introgression lines (Zhou et al., 2021), resequencing of large populations remains costly for the complex hexaploid wheat genome. Furthermore, current array-based genotyping platforms were primarily developed using hexaploid wheat species, and the limited genetic diversity in the D subgenome of hexaploid wheat has constrained their effectiveness in analyzing the derivatives of Ae. tauschii. Specifically, Ae. tauschii was classified into two main phylogenetic lineages (L1 and L2) and one tiny lineage (L3), and substantial variations present in Ae. tauschii were not detected in wheat D genome (Zhou et al., 2021; Li et al., 2022; Cavalet-Giorsa et al., 2024). Comparatively, genotyping by target sequencing (GBTS) is a cost-effective technology based on targeted sequence capture and high-throughput sequencing. Therefore, a standardized critical methodology is essential to accurately detect the untapped Ae. tauschii genomic resources in wheat improvement using GBTS. To achieve whole-genome coverage of both common wheat and Ae. tauschii, we selected five assemblies from sublineages of Ae. tauschii (T093, AY61, XJ02, AY17, and AL8/78) and the D genome of common wheat cultivar AK58 for exon capture probe design, targeting both high- and low-confidence genes (Figure S1A) (Luo et al., 2017; Zhao et al., 2017; Zhou et al., 2021; Jia et al., 2023). The six stocks contained a total of 49,465 homoeologous gene sets, of which 17,478 were shared across all six genomes (Figure S1B). The probe set were designed to target protein-coding regions after six rounds of alignment until specific CDS from each genome were added and a final template set was constructed (Figure 1A). In this process, genome Ⅰ (T093) served as the starting template, with an initial coding sequence size of 48.51853 Mb. Genomes Ⅱ–Ⅶ (AL8/78 L, AK58, AL8/78 J, AY17, AY61, and XJ02) contributed additional CDS sizes of 14.8747, 4.0985, 3.781, 1.4324, 5.1588, and 0.7605 Mb, respectively (Table S1). An adaptable probe set containing 102 Mb of CDS and adapters was ultimately generated and designated as ATPanExomeV1. A high on-target rate was examined on five Ae. tauschii accessions and wheat cultivar AK58 (Table S2), and capture efficiency and coverage of the panel exceeded 96% (Figure S2; Tables S3, S4). Design and applications of the ATPanExomeV1 Panel (A) Workflow for developing the exome capture panel of the Triticeae D genome: CDS sequences specific to each genome were obtained using Blast/Blat/Minimap2 to generate a pan-genome exome. (B) Development process of the Ae. tauschii-common wheat introgression lines. (C) A profile to recognize introgressed fragments based on calculated SNP ratios between introgression line and parent line. (D) Introgression segments derived from both Ae. tauschii accessions with and without reference genomes were precisely anchored using the ATPanExomeV1 Panel. (E) Two significant loci for plant height were identified using GWAS. Red dots represent variants in candidate regions. (F) Pairwise linkage disequilibrium (LD) model, with color gradients representing LD strength (R² value), red for high LD and yellow for low LD. (G) The locus on chromosome 2D was mapped to a region of 534 kb, containing a known gene Rht8. Using Ae. tauschii T093-common wheat AK58 introgression lines, a variation-based approach was established to track the introgressed fragments (Figure 1B, C). A total of 1,170,009 SNPs were identified between parent line T093 and AK58, and served as a common denominator (Figure S3A–C). For each introgression line, shared SNPs originating from T093 served as a numerator and a calculated ratio was used to represent the introgression proportion (Figure 1C). Taking line AT9059 as an example, 10-gene-based windows and 5-gene-based intervals were used to delineate introgressed fragments (Figure S4), and the segments were found to be increasingly fragmented and reduced in size from the BC1F1 generation to the BC4F1 generation (Figure S5). To precisely determine the introgression boundaries, we dissected example line AT10020 with a 21.76 Mb sized homozygous segment on chromosome 2D (Figure S6A), and anchored the start position to a 5-gene interval (TraesAK58CH2D01G049100–TraesAK58CH2D01G049500) and the end position between TraesAK58CH2D01G099600 and TraesAK58CH2D01G099800 (Figure S6B, C). Furthermore, we examined five randomly selected lines derived from Ae. tauschii accessions used for panel construction, and lines from four other accessions (AY60, AY73, SX09, and SX06) without genome assemblies. A series of homozygous and heterozygous introgression segments were precisely anchored as well (Figure 1D; Table S5), demonstrating that the panel could be widely and efficiently utilized for genotyping introgression derived from most accessions of Ae. tauschii species. To evaluate probe capture efficiency and validate karyotype variations for different polyploid wheat with D subgenomes, we genotyped representative lines of synthetic tetraploid wheat (HG27458, AADtDt), synthetic autoploid tetraploid (HG27090, D1D1D2D2), synthetic hexaploid wheat (HG27480, AABBDtDt), and synthetic octaploid wheat (HG27506, AABBDDDtDt) using the panel. The stability of chromosomes and recombination events was diagnosed by fitting copy number variations (CNV) and karyotype analysis. The presence of a trisomy on chromosome 2D was confirmed in HG27458, while HG27090 was trisomic on 2D1 and 4D2, monosomic on 2D2 and 4D1, and carried homoeologous recombination events at the distal region of most chromosomes (Figure S7A, B). No structural or numerical changes in chromosomes were anchored in HG27480, while HG27506 was monosomic on 4B, 1Dt, and 6Dt, and trisomic on 1D, along with some homoeologous recombination events (Figure S7C, D). This demonstrated that ATPanExomeV1 Panel-based CNV could effectively evaluate chromosomal integrity and homoeologous recombination events for Ae. tauschii-derived polyploidies. The ATPanExomeV1 Panel was used to genotype an Ae. tauschii T093-common wheat AK58 introgression population in order to assess its utility in the genetic dissection of important traits. A total of 233,823 SNPs and 25,618 indels were generated in the D subgenome. Taking plant height (PH) as an example, two significant loci were identified using a genome-wide association study (GWAS) on chromosome 2D (23,588,718–25,032,468 bp) and 4D (16,138,170–31,264,074 bp), respectively (Figure 1E, F). An introgression line AT20779 (with a single 11.89 Mb introgressed segment of chromosome 2D from Ae. tauschii T093) carrying the locus on 2D showed a significantly higher PH, kernel length (KL), and spike length (SL) compared to AK58 (Figure S8A–C). A secondary population was thus constructed by backcrossing AT20779 with AK58, and the region was fine-mapped to 534 kb (23,715,853–24,250,253 bp) by KASP-32 and KASP-17, containing a known gene Rht8 (TraesAK58CH2D01 G066900) regulating PH (Figure 1G; Table S6), and T093 showed a distinct haplotype compared with common wheat (Figure S9). For the locus on 4D, an interval was further anchored to the 17,013,180–20,077,182 bp region based on the introgression map (Figure S10), and Rht-D1 (TraesAK58CH4D01G046600) was speculated to be the candidate gene. T093 presented a unique haplotype (Figure S11), and introgression line AT20152-1 carrying this haplotype displayed a significantly higher PH compared to AK58 (Figure S12A–C). This panel thus demonstrated significant potential for discovering new genes or untapped Ae. tauschii haplotypes in the modern wheat background. In conclusion, a GBTS-based genotyping panel ATPanExomeV1 was developed and showed high efficiency to determine introgressed segments, structural variations, and recombination events in the derivatives of Ae. tauschii, providing an effective strategy for other wild relatives of polyploid crops as well. We hope for the widespread application of the panel in the future to enable the effective utilization of introgression stocks in wheat breeding programs and to facilitate advanced functional genomics studies. This work was supported by the National Key Research and Development Program of China (2023YFF1002200, 2022YFF1001602), the National Natural Science Foundation of China (32441052, 32230079), the Natural Science Foundation of Henan Province (242300421101), the Key Research and Development Program of Henan Province (231111110200), the Program for Innovative Research Team (in Science and Technology) in University of Henan Province (26IRTSTHN002), Key Research Projects of Higher Education Institutions in Henan Province (24A180003), and the Key Research Project of the Shennong Laboratory (SN02-2024-02). The authors declare no conflicts of interest. C.-P.S., Y.Z., and H.L. designed the experiments, supervised the study, and revised the manuscript. H.L. J.J.X., C.L., and Y.F.L. performed most of the research and drafted the manuscript. L.L.Z., F.N., Q.Q.D., Y.Z., and Z.L. performed some experiments and developed the strains. J.C.D., X.X.S., and X.Y.L. performed plant trait measurements. H.L. and Z.X.C. developed the exome panel. S.L.B., G.H.G., and K.W. reviewed the manuscript. All authors read and approved the final version of the paper. Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.70106/suppinfo Figure S1. The ATPanExomeV1 panel was designed by obtaining target segments from five reference genomes of Ae. tauschii accessions from each sublineage and common wheat AK58. Figure S2. The quality of ATPanExomeV1 was assessed by coding region coverage in five Ae. tauschii accessions from each sublineage, common wheat AK58 and derived SOW. Figure S3. Distribution of differential variation sites between Ae. tauschii T093 and common wheat AK58. Figure S4. Statistics of exogenous fragments of Ae. tauschii T093-common wheat AK58 introgression line AT9059 based on different window sizes. Figure S5. Detection of introgression segments in a single family of Ae. tauschii T093-common wheat AK58 introgression lines from generations BC1F1 to BC4F1 based on ATPanExomeV1. Figure S6. Determination of recombination breakpoints of Ae. tauschii T093-common wheat AK58 introgression line AT10020. Figure S7. Chromosome variation detection based on ATPanExomeV1 and FISH. Figure S8. Agronomic performance of AT20779 carrying a single introgression segment of chromosome 2D from Ae. tauschii T093. Figure S9. Compared with common wheat CS and AK58, the Rht8 gene of T093 (AetT093_2Dv1G106300) showed a distinct haplotype, carrying a six-nucleotide insertion at positions 1906–1907, and a G-to-A transition at codon 966. Figure S10. Linkage analysis mapped the plant height locus on chromosome 4D to a region of 17,013,180–20,077,182 bp. Figure S11. Haplotype analysis of Rht-D1 among common wheat and Ae. tauschii. Figure S12. Agronomic performance of AT20152-1 carrying a single introgression segment of chromosome 4D from Ae. tauschii T093. Table S1. Template genome information for designing ATPanExomeV1. Table S2. Capturing data information of five Ae. tauschii accessions from each sublineage and common wheat AK58 using ATPanExomeV1. Table S3. Coverage of captured data in five Ae. tauschii accessions. Table S4. Coverage ratios of five Ae. tauschii accessions under different depth conditions. Table S5. Introgressed segment information of introgression lines derived from Ae. tauschii accessions with and without reference genomes. Table S6. Primer sequence of KASP markers. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.