Stripe rust, caused by Puccinia striiformis f. sp. tritici (Pst), significantly reduced wheat production worldwide. Identification of stripe rust resistance genes is the foundation of improving wheat resistance breeding and revealing its genetic mechanism.

A multi-omics approach combined with genome-wide association study (GWAS) was used for dissecting adult plant stripe rust resistance for wheat advanced breeding lines collected from International Maize and Wheat Improvement Center (CIMMYT) and International Centre for Agricultural Research in the Dry Areas (ICARDA) bread-wheat breeding programs. In the present study, a diversity panel of 411 wheat lines from CIMMYT and ICARDA was used for genome-wide association study and a major locus on chromosome arm 2BL was identified. In order to verify the stability of the locus, the resistant line Z501 with the resistance allele of the locus was crossed by the susceptible line Jinmai 79, and the locus tentatively named YrZ501 was successfully confirmed using linkage mapping based on F_2:3 genetic population of Jinmai 79×Z501. Then we performed candidate gene analysis based on gene annotation, comparative genome, transcriptome and gene-based association analysis.

Combining GWAS and linkage mapping results, the YrZ501-2BL was located in the physical interval of 0.26 Mb (575.706-576.587 Mb) on chromosome 2B. According to the annotation information of Chinese Spring reference genome IWGSC v1.1, there were six high confidence genes of 12 genes in this region. Using online website, the target interval in the Chinese spring reference genome was compared with other published different ploidy wheat genomes. The six high-confidence genes within this interval can basically be found homologous in other wheat lines, and the genes arranged in the same order, indicating that the interval may not have large fragment insertions, deletions and inversions. The above results showed that we can perform candidate gene prediction analysis based on the reference genome information. After analysis of their transcriptomic data between the resistant parent Z501 and susceptible parent Jimai 79, only three genes, TraesCS2B02G406400, TraesCS2B02G406500 and TraesCS2B02G406600 showed variable expression levels and were induced by stripe rust infection. Further, they encode GATA transcription factor, SH3 domain-containing protein 2 and zinc finger protein, respectively. Gene-based association analysis revealed that there was a significant SNP (G1369A) in TraesCS2B02G406500 that was associated with stripe rust responses. Although this SNP (G1369A) did not cause amino acid coding changes (both TCG and TCA encode serine), it may be associated with alternative splicing. Moreover, it showed significant differences of the stripe rust responses between the different haplotypes (G1369A). Further analysis revealed two other variants G1377A and G1431A, that caused amino acid changes, i. e. valine (GTT) to isoleucine (ATT) and valine (GTG) to methionine (ATG), respectively. However, the two SNPs were rare variants as they accounting for only 0.87% of the 455 re-sequencing wheat accessions and they were not tested for significance. In summary, TraesCS2B02G406500 was preliminarily considered as an important candidate gene of YrZ501-2BL. In addition, the corresponding AQP markers were developed based on the SNPs among the YrZ501 candidate regions, which can be used to marker-assisted selection in molecular breeding application of wheat stripe rust resistance.

A candidate causal gene TraesCS2B02G406500 associated with stripe rust resistance was successfully identified on wheat chromosome 2B using an integrated method of multi-omics and association analysis, which laid a solid foundation for further gene cloning and functional verification.