Synthetic wheat reveals broad and isolate-specific blotch resistance loci

A multi-isolate genome-wide association study in synthetic hexaploid wheat has identified a chromosome 3D region associated with reduced Septoria tritici blotch symptoms against four of five tested fungal isolates, alongside several loci detected only in response to individual pathogens. The experiment also showed that suppression of the fungus’s reproductive structures, called pycnidia, can be partly separated from the necrosis visible on infected leaves, pointing to genetically distinct components of disease resistance in wheat.

Septoria tritici blotch, caused by the fungus Zymoseptoria tritici, is a major constraint on wheat production. The pathogen’s large, genetically diverse populations can overcome host resistance and develop resistance to fungicides. Synthetic hexaploid wheat, produced by combining tetraploid durum wheat with the wild D-genome progenitor Aegilops tauschii, offers a potential reservoir of new resistance alleles. The study used a three-family nested association mapping population developed by crossing the UK winter wheat variety Robigus with synthetic donors SHW.035, SHW.054 and SHW.075.

For the association analysis, researchers evaluated 171 nested association mapping lines with both high-quality genotype and phenotype data. Seedlings were inoculated with five British isolates of Z. tritici collected from wheat between 2015 and 2017: HT-18, HT-44, HT-53, HT-74 and HT-96. Disease was assessed repeatedly from 10 to 29 days after inoculation. The area under the disease progress curve was calculated separately for leaf necrosis and pycnidia coverage, allowing the study to distinguish damage to host tissue from the production of structures that carry fungal spores.

The synthetic donors strongly restricted pycnidia development, and several descendant lines showed little or no pycnidia across the isolate panel. By contrast, visible necrosis varied more among the genetic backgrounds. Broad-sense heritability was moderate to high for both traits: estimates ranged from 0.76 to 0.88 for necrosis and from 0.65 to 0.79 for pycnidia. Correlations among isolate-specific necrosis responses were also stronger than correlations between necrosis and pycnidia. Across isolates, Spearman correlations ranged from 0.57 to 0.80 for necrosis but only from 0.18 to 0.46 between the two disease components, supporting the conclusion that visible damage and pathogen reproduction have partly independent genetic control.

The strongest and most reproducible association was designated qSTB-3D.1. It was detected for necrosis after inoculation with HT-18, HT-53, HT-74 and HT-96, but not HT-44, and the favourable allele was contributed by SHW.035. In each responsive isolate, the peak occurred at the same SNP, AX-95,090,074, located at 609.21 megabases on chromosome 3D. Because linkage disequilibrium declined to r² of 0.2 only after about 50 megabases, the associated signal defined a broad 23.7-megabase candidate interval from 587.09 to 610.79 megabases. Its repetition across four pathogens indicates broad-spectrum activity, but its absence for HT-44 shows that the effect is not universal.

The qSTB-3D.1 interval contains 425 genes and multiple plausible components of plant immunity. About 19 megabytes from the strongest peak is a cluster of seven cysteine-rich receptor-like kinase genes that includes the cloned resistance gene Stb16q. Closer to the association peak is the wall-associated kinase gene TraesCS3D02G533100 at 608.92 megabases. Other wall-associated kinases, receptor-like kinases, nucleotide-binding leucine-rich repeat proteins and receptor-like proteins also occur within the interval. The available mapping resolution cannot distinguish Stb16q from the nearby kinase or another linked determinant, so the study identifies qSTB-3D.1 as a resistance-associated interval rather than assigning it to a single causal gene.

Additional associations depended on the particular isolate. For necrosis, qSTB-6D.1 and qSTB-3A.1 met the study’s permutation-based genome-wide significance threshold, while qSTB-1B.1 and qSTB-2D.1 met the false-discovery-rate threshold but not the permutation threshold. The HT-18-specific regions lay on chromosomes 1B and 6D, the HT-53-specific region lay on chromosome 3A, and the HT-74-specific region lay on chromosome 2D. The chromosome 3A interval contained the cloned wall-associated kinase gene Stb6, while the chromosome 2D interval included a cysteine-rich receptor-like kinase and two other kinase genes. None of these candidate genes was established as causal.

Only one genome-wide association was detected for pycnidia coverage: qSTB-1B.2, found in response to HT-96 on chromosome 1B and associated with an allele contributed by Robigus. No obvious immune receptor-like gene occurred in the corresponding interval. The scarcity of associations did not indicate that pycnidia suppression lacked genetic control. Instead, the trait showed moderate-to-high heritability but was strongly suppressed across much of the synthetic-derived material, leaving limited phenotypic variation for association mapping. Discrete visual scoring in 20-percentage-point increments, high frequencies of favourable alleles and limited within-family segregation may also have reduced the analysis’s power to detect smaller effects.

A separate segregation analysis supported the distinction between the two disease components. Researchers crossed the resistant SHW.035-derived line SHW.BC.144.12.1.4.1 with susceptible Robigus and assessed 256 BC2F2 seedlings after inoculation with HT-74. Among 256 plants, 191 were classified as resistant for pycnidia and 65 as susceptible, a pattern consistent with a single dominant factor segregating in a 3:1 ratio (chi-squared = 0.021, p = 0.885). Necrosis did not fit either a 3:1 or 1:3 single-gene expectation, indicating more complex inheritance. This result applies to the specific biparental cross and does not establish that the same factor governs pycnidia suppression throughout the wider population.

The findings identify promising breeding resources but remain subject to several limitations. The association population included only 171 lines, phenotyping was unbalanced across 11 batches and was not formally randomised, and visual disease scoring limited sensitivity to small effects. All assays were conducted on seedlings under controlled conditions, whereas adult-plant resistance may involve additional loci and environmental interactions. The broad linkage-disequilibrium interval around qSTB-3D.1 also requires fine mapping, sequencing and functional tests, while the isolate-specific intervals need replication in independent populations and field trials. Even so, the study shows that multi-isolate phenotyping can separate resistance effective against diverse pathogen genotypes from effects confined to particular isolates, and that selecting for reduced pycnidia production may provide a useful target distinct from reducing visible leaf damage.

Subject of Research: Genetic control of Septoria tritici blotch resistance in synthetic hexaploid wheat

Article Title: Multi-isolate GWAS identifies broad-spectrum and isolate-specific resistance-associated loci for Septoria tritici blotch in synthetic hexaploid wheat

Article References: Blower, A., Horsnell, R., Howell, P. M., Wright, T. I. C., Ray, R., Rawsthorne, S., Leigh, F. J., & Kanyuka, K. (2026). Multi-isolate GWAS identifies broad-spectrum and isolate-specific resistance-associated loci for Septoria tritici blotch in synthetic hexaploid wheat. Theoretical and Applied Genetics, 139(11), Article 309. https://doi.org/10.1007/s00122-026-05423-2

Image Credits: AI Generated

DOI: 10.1007/s00122-026-05423-2

Keywords: synthetic hexaploid wheat, Septoria tritici blotch, Zymoseptoria tritici, genome-wide association study, disease resistance, plant immunity, chromosome 3D, pycnidia suppression, fungal virulence, wheat breeding, Multi-isolate, GWAS