The wheat powdery mildew resistance gene *Pm4* also confers resistance to wheat blast

Wheat blast, caused by Magnaporthe oryzae pathotype Triticum (MoT), has rapidly spread from South America to Asia and Africa, posing a severe threat to wheat production in regions such as India and China. Only a few race-specific resistances to MoT are known (for example Rmg7, Rmg8, and the Aegilops ventricosa-derived 2NS translocation), with 2NS being the only robust field resistance in international trials. However, virulent MoT isolates can overcome 2NS and some resistances are temperature- or background-sensitive. The effector AVR-Rmg8 is widespread in the pandemic MoT lineage (B7) found in Bangladesh and Zambia, and both Rmg7 and Rmg8 recognize this effector. No MoT-effective resistance gene had been cloned to date. This study aimed to identify and clone the wheat gene that recognizes AVR-Rmg8, define its allelic variation and tissue/temperature specificity, and assess its effectiveness against different AVR-Rmg8 alleles and against field-relevant isolates, thereby providing deployable resistance for breeding.

Prior studies mapped Rmg7 (2A) and Rmg8 (2B) and showed both recognize AVR-Rmg8, with Rmg8 retaining activity at higher temperatures while Rmg7 does not. Genome-wide association studies across international nurseries found 2NS to be the only potent and robust field resistance, yet some isolates are virulent on 2NS cultivars and 2NS efficacy varies by background. Wheat blast isolates from Bangladesh and Zambia belong to a clonal lineage (B7) that uniformly carries AVR-Rmg8, making the cognate gene a prime breeding target. Previous work identified host-specificity resistance genes Rwt3 (an NLR) and Rwt4 (a wheat tandem kinase) acting against non-MoT pathotypes. In mildew, Pm4 was cloned as a race-specific resistance featuring alternative splice variants encoding a serine/threonine kinase–MCTP fusion protein; both isoforms are required for resistance to Bgt. Multiple Pm4 alleles with distinct specificities have been described, some introgressed from wild relatives. Collectively, the literature underscores the scarcity of effective MoT resistances, the importance of AVR-Rmg8 recognition, and the precedent for kinase-fusion R genes conferring resistance to multiple pathogens.

- Pathogen isolates and phenotyping: Two MoT isolates were used to phenotype AVR-Rmg8-dependent resistance: NO6047 carrying a non-recognized AVR-Rmg8 allele (ell) and its transformant NO6047+AVR8 expressing the recognized el allele, plus field isolate Py 15.1.018 (AVR-Rmg8 el) virulent on 2NS cultivars Jagger and CBFusarium ENT014. A diversity panel of 320 wheat lines (including 300 Watkins landraces) and several cultivars with genome assemblies were screened in detached seedling leaf and spike assays at controlled temperatures (22 °C or 26 °C), scoring disease on a 0–6 scale with multiple biological replicates. - AgRenSeq: NLR-enriched k-mer-based association genetics was performed using SY-Mattis as reference, analyzing phenotypes against NO6047+AVR8 and Py 15.1.018 to identify genomic associations. - Haplotype analysis: The 5.3 Mb interval associated on 2AL was interrogated using a 50 kb-window similarity heat map across 827 Watkins accessions and 218 modern cultivars to define shared haplotypes (regions 1 and 2) and further refine the candidate interval. Short-read alignments (bowtie2), variant filtering (samtools), and visualization (IGV) were used to detect candidate SNPs. - Candidate gene and expression: Gene models and RNA-seq (three-leaf stage) within the refined interval were examined to prioritize expressed genes. Sequence polymorphisms were compared between resistant and susceptible lines. Alternative splicing and protein identity were assessed by BLAST. - Functional validation: Germplasm resources for Pm4 were leveraged, including Federation NILs (Fed-Pm4a, Fed-Pm4b), eight EMS loss-of-function mutants in Fed-Pm4b targeting V1/V2-specific exons, and Bobwhite S26 transgenics overexpressing Pm4b_V1 and Pm4b_V2 (Nr#3 and Nr#52). Isogenic MoT transformants differing only in AVR-Rmg8 (Br48Δel vs Br48Δel+el; also transformants with el, ell, ell') were used to test allele-specific responses in leaf and spike tissues at 22 °C and 26 °C. - Allele prevalence and typing: PCR/KASP assays were designed to detect Pm4 (exon 7-based marker distinguishing functional Pm4 from a homoeologue). Prevalence was assessed in the Watkins landraces and the Gediflux European variety collection. Allelic series were assigned by sequence comparison; two novel alleles (A50E and W446*) were designated Pm4i and Pm4j. - Expression analysis: qRT–PCR quantified Pm4b_V1 and Pm4b_V2 expression in spikes of transgenic lines and compared V1/V2 expression between Pm4b and Pm4f carriers; statistical analyses (GLM, t-tests) evaluated differences.

- Genetic mapping: AgRenSeq identified a strong association on wheat chromosome arm 2AL (788.8–794.1 Mb), shared for both NO6047+AVR8 and Py 15.1.018 phenotypes, defining a 5.3 Mb interval. Haplotype analysis pinpointed a 1 Mb "region 1" (788.55–789.55 Mb) and narrowed resistance to the proximal 400 kb (788.55–788.95 Mb). - Candidate gene: Among ten genes in the 400 kb interval, five were expressed. Only TraesSYM2A03G00828360 harbored polymorphisms segregating with susceptibility: W446* (stop) in two susceptible lines and A50E in two others. The gene is alternatively spliced into 560 aa (V1) and 747 aa (V2) isoforms and is essentially identical to the known mildew resistance gene Pm4. - Functional validation: Federation NILs carrying Pm4a or Pm4b were resistant in seedling leaves to Br48Δel+el but susceptible in spikes at 22 °C. All eight EMS mutants in Pm4b (exon 6 V1-specific and exon 7 V2-specific) lost resistance, demonstrating both isoforms are required for MoT resistance. Bobwhite S26 lines overexpressing Pm4b_V1+V2 were resistant in leaves; the higher-copy line (Nr#52) was also resistant in spikes and showed significantly higher spike expression of Pm4b_V1 vs Nr#3 (P<0.001), indicating expression threshold/copy number can enable spike resistance. - Spectrum against AVR-Rmg8 alleles: Most Pm4 alleles (Pm4a, Pm4b, Pm4d, Pm4f, Pm4h, Pm4i) conferred seedling resistance to AVR-Rmg8 el, ell, and ell' transformants. Pm4g and the truncated Pm4j were ineffective. Pm4i showed weaker response to ell' and was susceptible to complex isolates (Py 15.1.018 and NO6047+AVR8), suggesting suppression by additional effectors. - Tissue and temperature effects: At seedlings, carriers of Pm4b, Pm4d, and Pm4f were resistant at 22 °C and 26 °C. In spikes at 22 °C, Pm4d and Pm4f conferred resistance, whereas Pm4b did not. At 26 °C, Pm4d conferred moderate spike resistance while Pm4b and Pm4f were susceptible; note all Pm4d carriers also had 2NS, potentially contributing to spike resistance. - Efficacy against Bangladeshi isolate: In detached spike assays at 22 °C, Pm4f carriers showed strong resistance to Bangladesh isolate BTJP4-1 (AVR-Rmg8 el); Pm4d carriers showed moderate resistance (likely influenced by 2NS); Pm4b carriers were susceptible. - Allele prevalence: Pm4 was rare in Watkins landraces (3.4%; 28/827) but more frequent in Gediflux European cultivars (15.5%; 67/432). Among Pm4-positive modern cultivars, Pm4b was most common (71%), while Pm4f predominated in Watkins (21/28) but was absent from modern lines. Pm4d was only found in modern varieties together with the 2NS segment, consistent with co-introduction via VPM1. - Novel alleles: Two newly identified Pm4 variants were named: Pm4i (A50E) and Pm4j (W446* truncation). - Breeding relevance: Pm4f provides non-2NS spike resistance effective against the Bangladeshi lineage carrying AVR-Rmg8 el, offering a critical complementary source to 2NS.

The study answers the central question of which wheat gene recognizes AVR-Rmg8 by showing that Pm4, a serine/threonine kinase–MCTP fusion previously known for mildew resistance, confers recognition and resistance to MoT carrying AVR-Rmg8. This provides the first cloned MoT resistance gene and reveals cross-pathogen specificity of a non-canonical R gene. Allelic and expression-dependent differences explain tissue specificity: while Pm4a/Pm4b act at the seedling stage, certain alleles (notably Pm4f) can protect spikes, and elevated expression (or copy number) can extend efficacy to spike tissues. Most Pm4 alleles recognize multiple AVR-Rmg8 variants, suggesting resilience to some effector polymorphism; however, alleles like Pm4i can be suppressed by other effectors, highlighting the complexity of host–pathogen interactions. The predominance of Pm4f in landraces and its absence in elite European germplasm underscore the value of landrace diversity for identifying deployable resistances. Given that 2NS dominates current field resistance yet can be overcome, Pm4—especially Pm4f—offers a key complementary resistance for regions impacted by the B7 lineage (Bangladesh and Zambia). The findings also motivate exploration of additional mildew or rust resistances that may confer blast resistance, and emphasize maintaining Rwt4 in breeding due to PWT4-mediated suppression of Rmg8 in rwt4 backgrounds.

Pm4 is identified and validated as the wheat gene recognizing the MoT effector AVR-Rmg8, representing the first cloned resistance effective against wheat blast. Allelic diversity within Pm4 shapes tissue- and temperature-specific effectiveness; notably, Pm4f confers robust spike resistance against the Bangladeshi lineage, providing a crucial non-2NS source for breeding. Overexpression can enable spike resistance for Pm4b, indicating that expression thresholds are important. These results open avenues to pyramid Pm4 (preferably Pm4f) with 2NS and maintain Rwt4 to enhance durability, and to search in wild tetraploids for new Pm4 alleles with broader and more stable efficacy across AVR-Rmg8 variants and environments. Future work should test performance under hotter field conditions typical of Bangladesh and Zambia, dissect mechanisms of cross-pathogen recognition by kinase-fusion R genes, and evaluate additional mildew/rust resistances for blast control.

- Tissue and temperature dependence: Several Pm4 alleles (for example Pm4b) confer only seedling resistance and/or lose spike efficacy at elevated temperatures (26 °C), raising concerns for hot environments. - Pathogen variability and suppression: Although many Pm4 alleles recognize multiple AVR-Rmg8 variants, certain alleles (for example Pm4i) can be suppressed by other MoT effectors, and effectiveness depends on the presence of AVR-Rmg8 in field populations. - Confounding by 2NS: Spike resistance observed in Pm4d carriers may be confounded by the concurrent presence of 2NS, complicating attribution of effects to Pm4d alone. - Expression threshold: Spike resistance in Pm4b required higher expression/copy number, suggesting potential expression-level constraints in typical genetic backgrounds. - Environmental relevance: The maintenance of Pm4-mediated spike resistance at temperatures exceeding 26 °C remains to be established for target regions such as Bangladesh and Zambia.