Bulked segregant RNA sequencing and linkage mapping identify a Pm4 allele conferring stable powdery mildew resistance in wheat breeding line YZ16

Abstract

Background

Wheat powdery mildew, caused by Blumeria graminis f. sp. tritici (Bgt), remains a major constraint on wheat production. Resistance genes with diagnostic markers and deployed without obvious agronomic disadvantages are essential for durable breeding. Here, we genetically and molecularly characterized powdery mildew resistance in the common wheat line YZ16 through inheritance analysis, bulked segregant RNA sequencing (BSR-Seq), linkage mapping and allele validation.

Results

YZ16 showed high seedling resistance to Bgt isolate E09 and resistance to nine of ten additional Bgt isolates, indicating a relatively broad resistance spectrum. Under field inoculations, YZ16 also displayed stable adult-plant resistance. Genetic analysis of an F₂ population and F_2:3 families indicated a single dominant resistance gene controlling resistance, temporarily designated PmYZ16. BSR-Seq localized the resistance locus to a distal interval on chromosome 2 A (766.9–784.6 Mb), and linkage mapping revealed complete co-segregation between resistance and JS717×JS718, a functional marker of the cloned Pm4 gene. Sequencing comparison showed that the encoded protein is identical to Pm4d/Pm4e, indicating that YZ16 carries a Pm4 allele. Expression profiling further indicated dynamic regulation of two splice variants (PmYZ16_V1 and PmYZ16_V2) after Bgt inoculation. Across two seasons, YZ16 showed generally comparable performance for key yield-related traits.

Conclusions

Powdery mildew resistance in YZ16 is conferred by the Pm4 allele on chromosome arm 2AL and can be efficiently tracked using a functional marker. Together with resistance at both seedling and adult stages generally competitive agronomic performance under the tested conditions, YZ16 represents a valuable donor for marker-assisted selection and resistance gene pyramiding in wheat breeding.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-08903-1.

Background

Wheat (Triticum aestivum L.) is a cornerstone of global food systems, and sustaining wheat productivity is central to food security. Genetic improvement, supported by strategic use of diverse germplasm, has repeatedly reduced losses from major diseases and helped limit reliance on fungicides [1, 2]. Nevertheless, wheat production remains persistently constrained by fungal diseases, the incidence and severity of which are shaped by agronomic intensification and shifting environments [3].

Among major foliar diseases of wheat, powdery mildew caused by Blumeria graminis f. sp. tritici (Bgt) is widespread and can substantially reduce yield and grain quality [4]. Host resistance is widely regarded as the most economical and environmentally sustainable long-term strategy for disease control, and modern breeding increasingly relies on combining phenotypic screening with molecular tools to accelerate resistance deployment [5]. To date, numerous powdery mildew resistance genes have been reported across the wheat genome, and meta-analyses further highlight genomic “hotspots” where resistance loci cluster and where candidate defense genes, such as nucleotide-binding site leucine-rich repeat (NBS-LRR) and receptor-like kinases, are enriched [4, 6].

Two major constraints continue to limit the durable use of powdery mildew resistance. First, many race-specific genes are rapidly overcome when deployed singly over large areas, motivating multilocus strategies and careful deployment schemes [7–9]. Second, alien introgressions may carry linkage drag and reduce breeding efficiency [10–13]. Therefore, resistance alleles already embedded in adapted wheat backgrounds are often more readily deployable. In contrast, resistance alleles already embedded in adapted common-wheat backgrounds are often more readily deployable [14]. Mechanistically, wheat resistance is not limited to canonical NLR receptors; diverse protein architectures and complex regulation, including alternative splicing, can underpin effective immune responses and may offer additional options for resistance stacking [6, 15].

Recent advances in high-throughput sequencing technologies, together with the availability of high-quality reference genomes for common wheat and its progenitor species, have greatly accelerated resistance gene discovery [16–18]. In particular, RNA sequencing has improved the resolution of genetic dissection by enabling the joint analysis of sequence polymorphisms and expression variation, thereby supporting more efficient locus anchoring and candidate prioritization [19–22]. In the present study, we investigated YZ16, a common wheat breeding line adapted to the Huang–Huai wheat region of China that exhibits stable resistance to powdery mildew. We aimed to (i) determine the inheritance of resistance, (ii) localize the resistance locus using bulked segregant RNA sequencing (BSR-Seq) and linkage mapping, (iii) clarify its relationship to known resistance loci on chromosome arm 2AL, and (iv) evaluate expression signatures and agronomic performance to assess breeding utility. Together, these analyses provide a marker-enabled resource for resistance deployment and pyramiding in wheat breeding.

Results

Resistance evaluation of YZ16 at seedling and adult plant stages

At the seedling stage, the wheat line YZ16 exhibited a high level of resistance to powdery mildew isolate E09, with an infection type (IT) of 0, whereas the susceptible control Mingxian 169 displayed profuse sporulation with ITs of 3–4 (Fig. 1). Resistance tests using ten additional Bgt isolates revealed that YZ16 was resistant to the majority of tested isolates, indicating a relatively broad resistance spectrum (Table 1).

Fig. 1.

Seedling reactions of YZ16 and Mingxian 169 (MX169) to Blumeria graminis f. sp. tritici (Bgt) isolate E09

Table 1.

Response of YZ16 and Mingxian 169 to 10 Blumeria graminis f. sp. tritici (Bgt) isolates

Bgt isolate	YZ16	Mingxian 169	Bgt isolate	YZ16	Mingxian 169
F02	0	4	F28	3	4
F06	0	4	E20	2	4
F07	0	4	E21	1	4
F16	0	4	E23-1	0	4
F19	0	4	E32	0	4

 Responses were recorded as infection types at the seedling stage using a 0–4 scale, in which 0–2 were considered resistant and 3–4 susceptible

Adult plant resistance was further evaluated under field conditions. After inoculation at the jointing stage with a mixture of Bgt isolates together with naturally occurring populations collected from northern China during the 2022–2024 cropping seasons, YZ16 consistently exhibited strong resistance at the adult stage. Disease scores ranged from 0 to 2 at both heading and milk stages, whereas Mingxian 169 was highly susceptible, with disease scores ranging from 7 to 9. These results indicate that YZ16 confers stable resistance to powdery mildew across developmental stages.

Inheritance of powdery mildew resistance in YZ16

To determine the genetic basis of resistance, an F₂ population derived from a cross between YZ16 and Mingxian 169 was evaluated at the seedling stage using Bgt isolate E09. Among the 298 F₂ individuals, 215 plants were resistant and 83 were susceptible, fitting the expected 3:1 segregation ratio (χ² = 0.15, P > 0.05), which indicated that resistance in YZ16 is controlled by a single dominant gene.

Segregation in 291 F_2:3 families further supported this conclusion: 75 families were homozygous resistant, 136 were segregating, and 80 were homozygous susceptible, fitting the expected 1:2:1 ratio (χ² = 1.62, P > 0.05). The resistance locus was temporarily designated PmYZ16.

BSR-Seq analysis anchors the resistance locus to a physical interval on chromosome 2 A

To rapidly localize the resistance locus, BSR-Seq was conducted using two extreme bulks, each consisting of 40 homozygous resistant or susceptible F₂ individuals. High-quality transcriptome sequencing generated 22.47 Gb of clean data for the resistant bulk and 20.40 Gb for the susceptible bulk, with Q30 values exceeding 96% for both bulks. The average sequencing depth reached 65.93× and 48.78×, and the clean reads showed high mapping efficiencies to the wheat reference genome. Using Euclidean distance (ED)-based analysis of SNP and InDel allele depths, we detected a prominent association signal on chromosome 2 A. The resistance-associated peak localized to a physical interval spanning 766.9-784.6 Mb (Fig. 2).

Fig. 2.

BSR-Seq analysis identifies a candidate interval on chromosome 2 A. a Manhattan plot of Euclidean distance (ED⁴) values for bulked segregant analysis across the wheat genome. The red dashed line indicates the significance threshold. b ED values along chromosome 2 A, highlighting a prominent peak near the distal region

Linkage mapping places PmYZ16 at the Pm4 locus and supports allelism with Pm4d/Pm4e

Polymorphic markers between the parents and bulks were subsequently genotyped in the mapping population, enabling construction of a linkage map and genetic localization of the resistance locus (Fig. 3a). PmYZ16 was flanked by two markers, Xwggc116 and Xwgrc763, at genetic distances of 0.2 cM and 0.5 cM, respectively. Notably, JS717×JS718, Xbcd1231, and Xwgrc1096 all co-segregated completely with the resistance phenotype in the YZ16 population (Fig. 3a). Among them, JS717×JS718 is a functional marker for the cloned Pm4 gene, providing particularly strong evidence that PmYZ16 resides at the Pm4 locus (Fig. 3b-c).

Fig. 3.

Genetic mapping and molecular characterization of PmYZ16. a Linkage map of chromosome arm 2AL showing the position of PmYZ16 (red) relative to flanking markers; genetic distances are shown in centiMorgans (cM). b Representative infection phenotypes of parents (PR, resistant parent; PS, susceptible parent) and selected segregants (R, resistant; S, susceptible). c PCR amplification patterns of the functional marker JS717×JS718 in parents and representative segregants. M, Trans2K Plus II; PR, resistant parent YZ16; PS, susceptible parent Mingxian 169; R, resistant individuals; S, susceptible individuals. The red arrow indicates the diagnostic polymorphic band associated with PmYZ16. d Protein domain architecture and amino-acid polymorphisms among Pm4 alleles. The schematic shows major domains (including the S_TKc domain and transmembrane region), and variant-specific amino-acid states for V1 and V2 are summarized, with PmYZ16 highlighted. Residues shown in grey are identical to those in Pm4a

To clarify the allelic relationship between PmYZ16 and previously reported Pm4 alleles, the candidate gene from YZ16 was amplified and subjected to comparative sequence analysis. Protein domain analysis confirmed that PmYZ16 encodes the typical Pm4-type protein, consistent with the structural features of known Pm4 alleles (Fig. 3d). Notably, the predicted amino-acid sequence of PmYZ16 was identical to that of Pm4d and Pm4e, indicating that YZ16 carries a Pm4 allele. However, the current evidence is insufficient to determine whether PmYZ16 corresponds precisely to Pm4d, Pm4e, or another highly similar Pm4 allele. Together, the linkage mapping and sequence comparison results demonstrate that powdery mildew resistance in YZ16 is conferred by a Pm4 allele located on chromosome arm 2AL.

Expression analysis of PmYZ16 splice variants after Bgt inoculation

Consistent with the reported mechanism of Pm4, in which both splice variants are required for resistance [23], the transcript levels of both PmYZ16_V1 and PmYZ16_V2 exhibited dynamic changes following inoculation with Bgt isolate E09 (Fig. 4). Consistent with the reported mechanism of Pm4, in which both splice variants are required for resistance [23], PmYZ16_V1 and PmYZ16_V2 showed dynamic but distinct temporal expression patterns after inoculation with Bgt isolate E09 (Fig. 4). No significant differences were detected for either splice variant at 0 h post inoculation (hpi). At 2 and 4 hpi, both PmYZ16_V1 and PmYZ16_V2 were significantly down-regulated in infected plants relative to the mock controls. At 8 hpi, both splice variants remained lower in infected plants, although the differences were not significant. At 12 hpi, PmYZ16_V1 was significantly up-regulated, whereas PmYZ16_V2 was also higher than the mock control but without a significant difference. At 18 hpi, both splice variants were significantly down-regulated. At 24 hpi, PmYZ16_V1 remained significantly down-regulated, while PmYZ16_V2 showed no significant change. At 36 hpi, PmYZ16_V1 tended to increase but without significance, whereas PmYZ16_V2 was significantly down-regulated. At 48 hpi, both splice variants were significantly down-regulated. At 72 hpi, PmYZ16_V1 showed a non-significant increase, whereas PmYZ16_V2 was significantly up-regulated. Taken together, these results indicate that the two splice variants responded dynamically to Bgt inoculation but differed in their temporal regulation.

Fig. 4.

Expression dynamics of two alternative transcripts of PmYZ16 after Bgt inoculation. Relative expression levels of PmYZ16_V1 and PmYZ16_V2 in mock-treated and Bgt-inoculated samples across a time course (0–72 h). Bars indicate mean ± SD from three biological replicates (three technical replicates per biological replicate). Asterisks denote significant differences between mock and infected samples at the same time point (P < 0.05; P < 0.01, Student’s t-test)

Differential expression and functional enrichment analyses of resistant and susceptible bulks

To gain additional biological insight into the resistance response associated with YZ16, differential expression analysis was performed based on the BSR-Seq data. A total of 20,070 differentially expressed genes (DEGs) were identified between the resistant and susceptible bulks, including 13,210 upregulated and 6,860 downregulated genes in the resistant bulk relative to the susceptible bulk (Fig. 5, Table S1).

Fig. 5.

Volcano plot of differentially expressed genes (DEGs) identified between the resistant and susceptible bulks derived from the YZ16 × Mingxian 169 population. Each dot represents one gene. Red dots indicate significantly up-regulated genes, blue dots indicate significantly down-regulated genes, and grey dots indicate non-significant genes. The vertical dashed lines indicate the fold-change cutoff, and the horizontal dashed line indicates the adjusted P value threshold

Gene Ontology (GO) enrichment analysis showed that the DEGs were significantly enriched in terms related to translation, ribosome, peptide biosynthetic process, peptide metabolic process, and structural constituent of ribosome (Fig. 6, Table S2). In addition to these highly represented categories, several defense- and stress-associated GO terms were also significantly enriched, including glutathione metabolic process, response to biotic stimulus, defense response, hormone-mediated signaling pathway, and phenylpropanoid metabolic/biosynthetic process.

Fig. 6.

Top 20 significantly enriched Gene Ontology (GO) terms of the differentially expressed genes (DEGs) identified between the resistant and susceptible bulks derived from the YZ16 × Mingxian 169 population. Bar length represents the rich factor, and bar color indicates the adjusted P value. Red asterisks indicate significantly enriched GO terms

Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis further showed significant enrichment in pathways such as Ribosome, Glutathione metabolism, Phenylpropanoid biosynthesis, Phenylalanine metabolism, Plant–pathogen interaction, Plant hormone signal transduction, and MAPK signaling pathway – plant (Fig. 7, Table S3). Additional enriched pathways related to secondary metabolism and lipid-associated defense responses, including alpha-Linolenic acid metabolism and Benzoxazinoid biosynthesis, were also detected. Together, these results provide supportive biological context for the resistance response in YZ16, although they do not by themselves identify the causal gene or establish the underlying mechanism.

Fig. 7.

Top 20 significantly enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways of the differentially expressed genes (DEGs) identified between the resistant and susceptible bulks derived from the YZ16 × Mingxian 169 population. Bar length represents the rich factor, and bar color indicates the adjusted P value. Red asterisks indicate significantly enriched pathways

Agronomic performance of YZ16 is broadly comparable to Jimai 22 across two seasons

Across both seasons, yield-related traits, including grain yield per plant (GYPP), spike number per plant (SNPP), fertile spikelet number per spike (FSN), thousand-kernel weight (TKW), kernel number per spike (KNS), and spike length (SL), showed generally comparable performance between YZ16 and Jimai 22, although SL differed significantly between the two genotypes in 2023–2024 (same letter grouping within each season; Fig. 8). In contrast, plant height (PH) differed significantly between YZ16 and Jimai 22 in both seasons. However, YZ16 was taller than Jimai 22 in one season but shorter in the other (Fig. 8). Collectively, these results indicate that YZ16 combines strong powdery mildew resistance with generally competitive agronomic performance.

Fig. 8.

Agronomic performance of YZ16 compared with the wheat cultivar Jimai 22 across two seasons. Boxplots show comparisons between YZ16 and Jimai 22 for grain yield per plant (GYPP), spike number per plant (SNPP), fertile spikelets per spike (FSN), thousand-kernel weight (TKW), kernel number per spike (KNS), spike length (SL), and plant height (PH) evaluated in 2022–2023 and 2023–2024. Different letters indicate significant differences between genotypes within the same season (P < 0.05)

Discussion

Genetic identification and confirmation of PmYZ16 as the Pm4 allele on chromosome arm 2AL

In this study, powdery mildew resistance in the common wheat line YZ16 was genetically and molecularly characterized through the integration of inheritance analysis, BSR-Seq, linkage mapping, and allele validation. The resistance segregated as a single dominant gene. BSR-Seq analysis initially mapped the resistance locus to the distal region of chromosome arm 2AL, spanning approximately 766.9–784.6 Mb. This region has repeatedly been reported as a hotspot for major powdery mildew resistance loci, including the well-characterized Pm4 locus, in both tetraploid and hexaploidy wheat backgrounds [23]. Disease resistance loci in wheat are frequently enriched toward distal chromosome regions that are relatively recombination-active, which can facilitate both the evolution of resistance diversity and genetic mapping resolution [24]. Linkage analysis with polymorphic markers revealed complete co-segregation between resistance and the diagnostic functional marker JS717×JS718, which has been validated for tracking Pm4 alleles in independent populations [25, 26]. Sequencing further showed that the encoded protein sequence is identical to that of Pm4d/Pm4e, confirming that the resistance in YZ16 corresponds to a Pm4 allele.

The Pm4 locus as a conserved and versatile powdery mildew resistance locus in wheat

Pm4 is one of the best- characterized multi-allelic powdery mildew resistance loci in wheat, with alleles reported across diverse genetic backgrounds. Multiple Pm4 alleles, including Pm4a, Pm4b, Pm4c, and Pm4d, have been identified in diverse wheat genetic backgrounds, highlighting both the evolutionary conservation and functional importance of this locus [25–31]. Cloning of Pm4 revealed an atypical resistance mechanism: Pm4 encodes a kinase fusion protein and relies on constitutive alternative splicing to generate two transcript isoforms required for full resistance [23]. The functional diversity reported at the Pm4 locus further suggests that coding-sequence identity alone may not fully explain phenotypic behavior. Although Pm4d and Pm4e are typically dominant, the recessive allele pmXMM [31] was reported to encode the same predicted protein sequence, implying that variation in resistance expression at this locus may also be shaped by regulatory context rather than by protein sequence alone. Similarly, distinct Pm4-related materials identified from different wheat backgrounds, including GR18-1 [25], DR88 [26], and YAV249 [29], have shown variation in resistance spectrum, inheritance behavior, or transcript processing despite high sequence similarity. These observations raise the possibility that non-coding variation, differential transcript regulation, or splice-variant dynamics may contribute to the distinct functional outputs of Pm4-related alleles in different genetic backgrounds. This expands the repertoire of molecular architectures contributing to cereal disease resistance beyond canonical NLR receptors [23]. Notably, recent work further extended the relevance of Pm4 beyond Bgt, demonstrating that Pm4 can also confer resistance to wheat blast through recognition of the blast effector Avr-Rmg8, with allelic differences affecting tissue specificity [32]. Moreover, virulence on Pm4-mediated resistance has been shown to be determined by two divergent Bgt effectors, supporting the idea that Pm4 durability and performance are closely tied to pathogen effector composition and suppressor dynamics [33]. Together, these insights place Pm4 within a well-supported mechanistic framework and strengthen its value as a deployable resistance component.

Transcriptomic responses associated with Pm4-mediated resistance in YZ16

Consistent with the regulatory complexity reported at the Pm4 locus, we observed dynamic regulation of the two splice variants, PmYZ16_V1 and PmYZ16_V2, following Bgt inoculation. Although both variants responded dynamically to infection, their temporal expression patterns were not identical, in line with the coordinated but differentiated splice-variant activity reported for Pm4-mediated resistance [23]. To further characterize the resistance response associated with YZ16, we performed DEG-based GO and KEGG enrichment analyses using the resistant and susceptible bulks. In addition to the strong enrichment of translation- and ribosome-related categories, many DEGs were also assigned to GO terms and KEGG pathways commonly associated with plant defense, including plant–pathogen interaction, plant hormone signal transduction, MAPK signaling, glutathione metabolism, and phenylpropanoid biosynthesis. Such enrichment patterns are frequently observed in wheat transcriptomic studies under fungal stress and are generally regarded as reflecting extensive defense-associated transcriptional reprogramming during effective resistance responses [34]. In particular, the enrichment of pathways related to immune signaling, redox regulation, and secondary metabolism is consistent with the view that successful resistance involves coordinated activation of recognition, signal transduction, and downstream defense outputs, including stress adaptation and reinforcement-related processes reported in wheat [35]. Because these enrichment results were derived from bulked RNA-seq samples generated primarily for mapping, they should be interpreted as supportive rather than causal evidence. Nevertheless, they provide useful biological context for the resistance response in YZ16 and may help guide future functional analyses.

Breeding utility of YZ16 carrying a Pm4 allele

Resistance genes introgressed from wild relatives are often associated with linkage drag caused by large alien chromosomal segments, which can impair agronomic performance and hinder their effective deployment in breeding programs [36–38]. By contrast, resistance alleles already present in adapted wheat backgrounds are generally more accessible in practical breeding. From this perspective, YZ16 should not be regarded merely as another Pm4 allele, but rather as part of a broader set of Pm4-related resistance resources identified from diverse wheat germplasm. Representative examples include the common wheat breeding line GR18-1 [25], the durum wheat accession DR88 [26], the synthetic hexaploid wheat line YAV249 [29], and the Chinese landrace Xiaomaomai [31]. Together, these materials show that the Pm4 locus has been repeatedly identified in distinct germplasm types, each offering different opportunities and constraints for breeding use. GR18-1 represents an elite common wheat resource with resistance at both seedling and adult stages and a favorable agronomic background, making it readily usable in conventional breeding. In contrast, DR88 provides a durum wheat source with broader and stronger resistance, although its utilization generally requires transfer into adapted common wheat backgrounds. YAV249 highlights the value of synthetic hexaploid wheat as a bridge resource, as its Pm4-related resistance was successfully introduced into a commercial wheat background through marker-assisted selection, leading to the development of an advanced line with excellent agronomic performance. Xiaomaomai further suggests that highly similar, or even identical, Pm4 haplotypes may exhibit different genetic behavior in distinct genetic backgrounds.

Against this background, the significance of YZ16 lies not only in its molecular similarity to Pm4d/Pm4e, but also in its value as a deployable resistance source in an adapted common wheat background. In the present study, YZ16 conferred resistance at both seedling and adult stages, exhibited a relatively broad resistance spectrum, and maintained generally competitive agronomic performance relative to Jimai 22 over two growing seasons. These features distinguish YZ16 from more distant donor types and support its practical use in marker-assisted breeding. In addition, pyramiding resistance genes with different recognition specificities and molecular mechanisms is widely recognized as an effective strategy for achieving durable disease resistance in wheat [39–41]. In this context, YZ16 represents a useful Pm4-based donor for resistance pyramiding in common wheat breeding. At the same time, comparisons among reported Pm4-related materials indicate that this locus encompasses substantial diversity in resistance spectrum, developmental-stage stability, and breeding applicability across different germplasm categories. Because these comparisons are derived from independent studies using different isolate panels and experimental conditions, they should be interpreted as providing comparative breeding context rather than a strict ranking of Pm4 donors. Nevertheless, together with the co-segregating functional marker JS717×JS718, our results support YZ16 as a practically useful donor for improving powdery mildew resistance in common wheat.

Conclusion

Our results showed that powdery mildew resistance in the common wheat line YZ16 is conferred by a Pm4 allele on chromosome arm 2AL and can be efficiently tracked using the co-segregating functional marker JS717×JS718. Together with its stable resistance at both seedling and adult stages and its generally competitive agronomic performance under the tested conditions, YZ16 represents a useful donor for marker-assisted deployment and resistance gene pyramiding in common wheat breeding. However, the transcriptomic enrichment results presented here provide supportive rather than causal evidence, and broader multi-environment evaluation will be needed to further assess the breeding potential of this material.

Materials and methods

Plant materials

YZ16 is a common wheat breeding line provided by Prof. Pengtao Ma, Yantai University, and adapted to the Huang–Huai wheat region of China. For genetic analysis and mapping, YZ16 was crossed with the susceptible cultivar Mingxian 169 to generate F₁ plants, an F₂ population (n = 298) and corresponding F_2:3 families (n = 291) for genetic analysis and mapping. Mingxian 169 was included in each assay as the susceptible control and spreader.

Powdery mildew isolates and resistance evaluation at seedling and adult-plant stages

Seedling-stage resistance was evaluated using Bgt isolate E09 for inheritance analysis. Ten additional isolates (F02, F06, F07, F16, F19, F28, E20, E21, E23-1 and E32), kindly provided by Prof. Pengtao Ma, Yantai University, were used to further assess the resistance response of YZ16 to multiple isolates. For each isolate, ten seeds of YZ16 and ten seeds of Mingxian 169 were separately sown in 72-cell trays (approximately 540 × 280 mm, with individual cells ~ 36–38 mm wide and ~ 45 mm deep), with Mingxian 169 distributed randomly as the susceptible control. Trays for different isolates were maintained in strictly separated spaces to prevent cross-infection. Plants were grown at > 60% relative humidity with a 14 h light/10 h dark photoperiod at 24 °C/18°C. Seedlings at the two-leaf stage were inoculated with fresh conidiospores, and inoculation was repeated on the following two days to ensure sufficient inoculation for phenotypic evaluation. Infection types were scored on a 0–4 scale when the susceptible control developed abundant sporulation; IT 0, 0;, 1 and 2 were classified as resistant and IT 3 and 4 as susceptible [42]. Three independent inoculation experiments were performed, and consistent infection phenotypes were observed across all three experiments.

Adult-plant resistance was evaluated under field conditions at the Jinan Experimental Station of the Crop Research Institute, Shandong Academy of Agricultural Sciences, Jinan, Shandong, China, during the 2022–2023 and 2023–2024 cropping seasons. Plants were inoculated at the jointing stage with a mixture of Bgt isolates, together with naturally occurring field populations collected from northern China.YZ16 was planted in three replicated plots arranged in a randomized complete block design (RCBD), with each plot consisting of three rows (1.5 m length; 20 cm row spacing). Mingxian 169 was planted around the plots to facilitate disease spread. Adult-plant reactions were scored at heading and milk stages using a 0–9 disease severity scale, where scores 0–4 were considered resistant and 5–9 susceptible [43].

Genetic analysis

The inheritance of resistance to Bgt isolate E09 was evaluated using the parental lines, F₁ plants, an F₂ population, and the corresponding F_2:3 families. Segregation ratios in F₂ (resistant: susceptible) and F_2:3 (homozygous resistant: segregating: homozygous susceptible) were tested against expected Mendelian ratios (3:1 and 1:2:1, respectively) using chi-square (χ²) tests.

BSR-Seq analysis and candidate interval identification

To rapidly localize the powdery mildew resistance locus in YZ16, two extreme bulks were constructed from the F₂ population by pooling 40 resistant F₂ individuals (R bulk) and 40 susceptible F₂ individuals (S bulk) with equal contributions per plant. Library construction, sequencing, and primary bioinformatic analyses were conducted by Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). Paired-end libraries (average insert size ~ 350 bp) were sequenced on an Illumina NovaSeq X Plus platform (PE150). Raw reads were filtered using fastp to remove adaptor sequences and low-quality reads [44]. Clean reads were aligned to the IWGSC Chinese Spring RefSeq v2.1 assembly [45] using HISAT2 [46]. Variant calling (SNPs and InDels) was performed following the GATK Best Practices workflow [47], and variant effects were annotated using SnpEff [48].

Bulk association analysis was conducted using the ED algorithm, in which ED values were calculated at each biallelic variant site based on allele depths between the two bulks. To reduce background noise, ED values were transformed using a fourth-power operation (ED⁴) and smoothed with a sliding-window approach (1 Mb window with a 50 kb step). Candidate genomic intervals were defined using the genome-wide 99.5th percentile of the smoothed ED⁴ distribution as the significance threshold, and only regions containing ≥ 10 variants exceeding the threshold were retained as associated intervals.

Molecular marker analysis and gene mapping

Based on the candidate interval identified by BSR-Seq, previously reported markers linked to the corresponding candidate region and/or the Pm4 locus were first tested for polymorphism between parents and bulks [25, 26]. Polymorphic markers were subsequently genotyped across the F₂ population for linkage analysis. Genetic linkage analysis was performed using MAPMAKER/EXP 3.0 [49], and genetic distances were calculated using the Kosambi mapping function [50].

Allele validation of PmYZ16

To determine the allelic relationship between the resistance gene in YZ16 and known Pm4 alleles, the candidate gene was amplified from YZ16 using published primers for Pm4 allele cloning [23], followed by Sanger sequencing. Sequence alignments were performed against published Pm4 alleles to confirm allelic identity.

Expression analysis by quantitative real-time PCR

YZ16 seedlings were inoculated with Bgt isolate E09, and young leaves were sampled at 0, 2, 4, 8, 12, 18, 24, 36, 48 and 72 hpi, with mock-inoculated samples collected in parallel. For each time point, three independent biological replicates were collected. Total RNA was extracted using TRIzol reagent (Invitrogen, USA). First-strand cDNA was synthesized using the FastKing RT Kit (With gDNase) (TIANGEN Biotech, Beijing, China). Quantitative real-time PCR (qRT-PCR) was performed using TB Green Premix Ex Taq II (Takara Bio, Shiga, Japan) according to the manufacturers’ instructions. Each biological replicate was analyzed with three technical replicates. Relative transcript abundance was calculated using the 2^⁻ΔΔCT method with TaActin as the internal reference. Data are presented as mean ± SD. The primer sequences used for qRT-PCR are provided in Table S4.

Differential expression and functional enrichment analyses of the resistant and susceptible bulks

To gain additional biological insight into the resistance response associated with YZ16, differential expression analysis was performed between the resistant bulk and the susceptible bulk using the BSR-Seq data. Genes with |log2FC| ≥ 1 and P adjust < 0.001 were defined as DEGs and used for downstream enrichment analyses. GO enrichment analysis was conducted across the three GO domains, including Biological Process (BP), Molecular Function (MF), and Cellular Component (CC), using the GO database [51]. KEGG pathway enrichment was performed using the KEGG database [52, 53]. Enrichment significance was evaluated using a hypergeometric test, and multiple testing correction was applied using the Benjamini–Hochberg method [54]. Terms/pathways were considered significantly enriched at adjusted P values < 0.05.

Agronomic traits evaluation of YZ16 for wheat powdery mildew resistance breeding

Field trials were conducted over two consecutive growing seasons (2022–2023 and 2023–2024) at Luancheng Agro-ecosystem Experimental Station (37° 53′ 15″ N, 114° 40′ 47″ E) to evaluate the agronomic performance of YZ16 in comparison with the widely grown cultivar Jimai 22. The experiment followed a randomized complete block design with three replicates. Each plot consisted of four rows (1.5 m in length; 0.25 m row spacing), and standard local agronomic practices were applied throughout both seasons.

At maturity, 20 representative plants from the center rows of each plot were harvested for trait measurements. The following agronomic traits were recorded: GYPP (g), SNPP, FSN, TKW (g), KNS, SL (cm), and PH (cm). For each trait, differences between YZ16 and Jimai 22 were evaluated using one-way ANOVA with genotype as the fixed factor, followed by Tukey’s multiple comparisons test. Significance was declared at P < 0.05. In the boxplots, different letters above boxes.

Authors' contributions

HY and CL conceived and supervised the study. SZ analyzed the data and wrote the manuscript. SZ and WG performed the molecular experiments. WG constructed the genetic populations. WG, XW, LG, JH and SJ performed the field experiments. SZ collected the data. WG evaluated the phenotype. HY and CL revised the writing of the article. All authors have approved the final manuscript.

Funding

This research was financially supported by Natural Science Foundation of Guangxi Province (2025GXNSFBA069184), Wheat Industry Technology System of Shandong Province (SDAIT-01-01), and Key Research and Development Program of Shandong Province (2025LZGC011).

Data availability

All the data generated or analyzed during the current study were included in the manuscript and its additional files. The raw data is available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.