Objective 1. Development of a high-quality RenSeq platform for the soybean research community. It has been well known that the Williams 82 soybean reference genome includes only a fraction of genes within the natural soybean population. This is particularly true for the NBS-LRR genes. To ensure the SoyRenSeq platform to be developed in this project has desirable power in capturing NBS-LRR gene clusters in any soybean varieties, we will first extract all NBS-LRR gene clusters from the 26 representative wild and cultivated soybean genomes sequenced by long-read sequencing approach. Then we will identify the redundant and NBS-LRR gene unique cluster sequences. Finally, we will use the complete set of non-redundant NBS-LRR gene cluster sequences to design a set of biotinylated RNA baits through coordination with Arbor Bioscience.
Objective 2. Sequencing and assembly of NBS-LRR gene clusters in major soybean lines carrying resistance to prevalent soybean pathogens in the Midwest region. We will use Soy_myBait1.0 to enrich NBS-LRR gene cluster sequences from soybean lines each carrying resistance to a specific pathogen as well as soybean lines routinely used as susceptible parents for mapping population development. We anticipate most of the disease resistances targeted in this project are controlled by NBS-LRR genes and can be captured by SoyRenSeq, but some, particularly the resistances showing quantitative variation, are likely controlled by non-NBS-LRR genes. Nevertheless, the data generated from all the resistant lines as well as the susceptible lines would be valuable for understanding the origin and dynamic variation of NBS-LRR genes and for pinpointing candidate R genes in NBS-LRR gene clusters underlying race-specific resistances. The NBS-LRR gene cluster sequences will be selected for fragments larger than 3 kilobase pairs, and the resulting DNA samples will be multiplexed and sequenced using PacBio Sequel SMRT sequencing platform. The sequences generated by SMRT sequencing will be assembled, annotated, and then anchored to chromosomal regions.
Objective 3. Analysis of R gene expression and responses to various soybean pathogens. Plant R genes are generally responsive to pathogen infection and can be detected by profiling of gene expression – the process by which the information encoded in a gene is used to make RNA molecules (called transcripts) that code for proteins. We will detect genome-wide gene expression changes in responses to each specific pathogen using two RNA sequencing methods. We will use SoyRenBaits1.0 to enrich full-length NBS-LRR transcripts and then conduct Iso-Seq. To capture R genes that do not belong to the NBS-LRR gene family, we will conduct short-read RNA-seq. Finally, the expression of strong candidate R genes will be further measured by real-time quantitative PCR, a low-cost technique used to detect COVID-19.
Objective 4. Evaluation of resistance to various pathogens and mapping of major R genes and QTL. Our investigators in the six states each will tackle a single or multiple soybean diseases targeted in respective research programs. Briefly, PI Ma and co-PI Cai in Indiana will identify candidate genes for RpsUN1 and RpsUN2, and candidate genes for resistance against Fusarium graminearum, Pythium ultimum and Pythium irregulare, as well as a novel Rbs gene for brown stem resistance, Rcs3 for frogeye leaf spot resistance, and Rbs3 for brown stem rot resistance. Co-PI Bhattacharyya in Iowa will identify candidate genes for Rps6, Rps12, Rps13, and two additional novel Rps genes. Co-PIs Wang and Lin in Michigan will primarily focus on identification of candidates for Rpsan1 and a novel Rcs gene from PI 532464 for frogeye leaf spot resistance. Co-PIs Lorenz in Minnesota and Miranda in North Dakota will work on identification of QTL for white mold resistance and brown stem rot resistance. For resistance to a particular disease with chromosomal locations defined in NBS-LRR gene clusters, we may be able to directly pinpoint candidate genes by the SoyRenSeq approach. For resistance to a particular disease with a mapping population available, we will conduct fine mapping. For resistance to a particular disease without a mapping population or previous knowledge about chromosomal location, we will construct a mapping population. Due to the short duration of the project, we may only be able to validate 1-2 R genes, if pinpointed in the early stage of the project, through genetic transformation, although PI Ma’s lab has full capability for soybean transformation.
Objective 5. Development of candidate R-gene-based molecular markers for precision breeding. Once candidate R genes for specific pathogens are identified, we will design R- gene-based molecular markers and then validate their effectiveness by using the segregating mapping population. Such markers will be used for selection of the R genes in the breeding programs led by the soybean breeders in our team.