HHM was crossed with PM-susceptible wheat varieties NingmaiZi119 (NMZ119, Figure 1 ) and Chinese Spring (CS, Figure 1 ). Populations of F₁, F₂, F₃, BC₂F₂, and BC₃F₁ were utilized for the genetic analysis of PM resistance in HHM. In addition to the susceptible parents, Heshangmai (HSM), Honghuaxiaomai (HHXM), Chancellor (CC), and Zhongzuo (ZZ) were also used as controls. The crosses HHM/HSM and HHM/HHXM were used for testing allelism. All of the genotypes used in this study were sourced from the crop germplasm bank of the Jiangsu Academy of Agriculture.

A total of 25 Bgt isolates collected from various regions of China were used in this study. Among them, Bgt isolate B03 was provided by Professor Hongxing Xu from the School of Life Sciences at Henan University and is the most prevalent isolate in the Nanjing area. Additionally, 18 isolates were obtained from a collection maintained at Hubei Academy of Agricultural Sciences in Wuhan (Zeng et al., 2014), while six isolates were sourced from Dr. Hongjie Li’s laboratory at the Chinese Academy of Agricultural Sciences in Beijing. Detailed information on the Bgt isolates is provided in Supplementary Table S1.

Assessments of PM response were conducted following the procedure outlined in Zeng et al. (2014). The test materials, resistant parent HHM and susceptible control Sumai 3 (S3), were randomly sown in 50-hole trays (5 cm × 5 cm) for seedling response assays. The trays were placed in an artificial climate chamber. The settings for the climate chamber were as follows: 14-h day/10-h night cycle, 22°C day/18°C night temperature, and 50% to 60% relative humidity.

At the one-leaf stage, the seedlings were dust-inoculated using sporulating plants infected with the Bgt isolate B03. The relative humidity of the climate chamber was increased to 80%–100% following the inoculation. The disease responses of HHM, HSM, and HHXM to various Bgt isolates were assessed using 25 isolates, with CC and ZZ serving as susceptible controls. Disease severity was evaluated when the susceptible controls S3, CS, and NMZ119 displayed full sporulation. Scoring of infection types (ITs) was carried out by utilizing a scale ranging from 0 to 4 (Fu et al., 2017), comprising six levels: 0, 0;, 1, 2, 3, and 4. The results were classified based on ITs, with ITs 0 to 2 categorized as resistant and ITs 3 to 4 labeled as susceptible.

To investigate the inheritance of adult plant PM resistance in HHM at the adult stage, in 2021 we inoculated and phenotyped 1,352 NMZ119/HHM F₂ plants and 820 HHM/CS F₂ plants that were susceptible at the seeding stage. Following this, we inoculated and phenotyped the BC₂F₂ and BC₃F₁ populations of NMZ119/HHM in 2023. All of these plants were grown in an isolated net house that contained no other wheat or Bgt isolates, thereby ensuring there was no contamination from any other Bgt isolates. To ensure uniform infection, S3 was planted around each F₂ plants. In mid-February, S3 was inoculated with Bgt isolate B03, and the disease responses were recorded during the flowering period in April when S3 was fully susceptible (IT = 8 or 9), using the 11-point scale of Sheng and Duan (1991)—that is, ITs 0, 0;, 1, 2, 3, 4, 5, 6, 7, 8, and 9. The entire plant is divided into nine equal sections from the bottom to the top, and the resistance level is determined based on the severity of diseased leaves (or spikes) at different heights. The levels are defined as follows: 0, immune—no disease spots present on the entire plant; 0;, near immune—leaves show dead spots but have no disease spots; 1, high resistance—a few lesions are present on the leaves in the first segment (lesions account for less than 2% of the total leaf area); 2, high resistance—a few lesions appear on the leaves in the second segment, with mild symptoms in the first segment (lesions account for 3%–10% of the leaf area); 3, moderate resistance—mild symptoms occur in the third leaf segment, moderate disease in the second leaf segment (lesions account for 11%–25% of the leaf area), and severe disease in the first leaf segment (lesions account for more than 25% of the leaf area); 4, moderate resistance—mild symptoms in the fourth leaf segment, with moderate to severe symptoms in the third segment and below; 5, moderate infection—mild symptoms in the fifth leaf segment, with moderate to severe symptoms in the fourth segment and below; 6, moderate infection—mild symptoms in the sixth leaf segment, with moderate to severe symptoms in the fifth segment and below; 7, highly sensitive—mild symptoms in the seventh leaf segment, with moderate to severe symptoms in the sixth segment and below. Occasionally, the flag leaves may show a few disease spots; 8, highly sensitive–mild or moderate disease affects the flag leaf, with moderate to severe disease present in the leaves below the flag leaf; 9, extremely sensitive—the entire leaf is severely affected, and the ear may show varying degrees of damage. ITs 0–4 were regarded as resistant, while ITs 5–9 were classified as susceptible.

Based on the identification criteria for both seedling and adult stages outlined above, the identified population is categorized into two groups: resistant and susceptible. By analyzing the ratio of resistant plants to susceptible plants, we can infer the potential genetic ratio. Chi-square tests were employed to determine whether the observed segregation ratios aligned with the expected ratios.

The procedure for DNA extraction used in this study was that outlined by Fu et al. (2017). F₃ families from the cross NMZ119/HHM were genotyped using mixtures of equal amounts of DNA from 10 randomly selected homozygous resistant families (the resistance pool) and equal amounts of DNA from 10 randomly selected homozygous susceptible families (the susceptibility pool). A 55K SNP chip was utilized to analyze these DNA samples along with DNA from the parents. This process aimed to detect SNP markers displaying consistent polymorphisms between the parental samples and between the two pools. When PmHHM was located within a specific physical interval, primers were designed to amplify specific regions within the interval based on CS RefSeq v2.1. Sequence variations between HHM and CS were utilized in developing various molecular markers, including SSR, KASP, CAPS, dCAPS, and SNP markers. The markers were developed on PrimerServer of WheatOmics 1.0 (http://202.194.139.32/PrimerServer/). PCR amplifications and detection of markers were carried out following the procedures detailed by Fu et al. (2017). For the detection of KASP markers, standard FAM tail (5′-GAAGGTGACCAAGTTCATGCT-3′) and HEX tail (5′-GAAGGTCGGAGTCAACGGATT-3′) sequences were incorporated at the 5′-ends of HHM allele (A)- and CS allele (B)-specific primers, respectively. The KASP marker reactions were carried out using a Hydrocycler-16 Water Bath Thermocycler (LGC Genomics, Hoddesdon, Herts, UK), with fluorescence detection performed utilizing a PHERAstar high-end microplate reader (LGC Genomics).

To conduct fine-mapping of PmHHM, markers flanking the gene from the initial mapping were employed to screen DNA from homozygous susceptible F₂ plants in the NMZ119/HHM and HHM/CS populations. Genotypic analysis of recombinants was carried out using newly developed SSR markers, dCAPS markers, KASP markers, and SNP markers to construct a saturated genetic map of the PmHHM region. To confirm the homozygous susceptibility of the recombinants, self-pollinated seeds (20–30 seeds) were utilized to assess PM resistance. Fine-mapping of PmHHM was accomplished by comparing the genotype–phenotype relationships of the recombinants. The sequences for all markers used in this study are listed in Supplementary Table S2 .

To anchor PmHHM on the CS genome, sequences of markers linked to the gene were aligned with CS RefSeq. v2.1. Markers linked to Pm61 (Sun et al., 2018), PmHSM (Fu et al., 2024), MlIW30 (Geng et al., 2016), QPm.tut-4A (Janáková et al., 2019), PmPBDH (Liang et al., 2022), MlNFS10 (Yin et al., 2021), PmHHXM (Xue et al., 2021), and PmXNM (Xue et al., 2024) were also aligned to determine their physical locations in the CS genome. The recombination rate was calculated using the ratio of genetic distance (cM) to physical distance (Mb).

The genomic sequence of the PmHHM region in CS (chromosome arm 4AL, 740–754 Mb) and that of 13 other sequenced wheat cultivars were obtained from the WheatOmics 1.0 (http://202.194.139.32). The CS sequence was compared with the corresponding sequences of the other cultivars using the MUMmer 3.0 package (Kurtz et al., 2004). The alignments were filtered based on identity (≥85%) and length (≥5,000 bp) criteria and visualized using the MUMmer 3.0 package. The candidate genes for PmHHM were predicted according to JBrowse in WheatOmics 1.0 (http://202.194.139.32/jbrowse-1.12.3-release/?data=Chinese_Spring2.1).

Chi-square (χ ²) tests were conducted to assess genetic ratios. The genetic distances in the preliminary map were calculated using MAPmarker 3.0, with a LOD value set at 3.0. The recombination rate of the fine-map was calculated using the formula r = (2f1 + f2)/2N, where f1 represents the number of homozygous recombinants, f2 is the number of heterozygous recombinants, N is the number of screened individual plants, and the recombination value was converted to map distance using the formula d = {ln (1 + 2r)/(1 - 2r)}/4. The resulting genetic and physical maps were visualized using MapDrawV2.1 (Liu and Meng, 2003).

HHM showed high resistance to each of the 25 Bgt isolates ( Table 1 ; Figure 1 ), indicating that it has highly effective, broad-spectrum resistance at the seedling stage. The F₁ plants from both NMZ119/HHM (IT = 1, Figure 1 ) and HHM/CS (IT = 1, Figure 1 ) plants were resistant, indicating that the resistance is dominant. The segregation data ( Table 2 ) showed clear evidence for segregation of a single gene in both crosses. The resistance allele in HHM is temporarily named PmHHM.

HHM was also resistant to PM at the adult stage (IT 1–2). We identified 1,352 susceptible NMZ119/HHM F₂ plants and 820 susceptible HHM/CS F₂ plants that exhibited susceptibility (IT = 7–8) during the adult stage, consistent with the results of seedling evaluation. Based on the genomic composition of these plants and a sufficiently large population sample, we concluded that PmHHM is responsible for mediating adult resistance. To further validate the results, phenotyping of adult plants from the two backcross populations was conducted, and the results completely agreed with those obtained at the seedling stage ( Table 2 ). The bimodal distribution observed ( Supplementary Figure S1 ) again confirmed that PmHHM not only mediates resistance to PM at the seedling stage but also confers resistance at the adult stage.

Analyzing HHM, NMZ119, and the resistant and susceptible pools using the wheat 55K SNP array identified 1,194 polymorphic SNPs distributed across all the 21 wheat chromosomes ( Supplementary Figure S2a ). Among the SNPs, 231 were found on chromosome 4A (19.35%), 212 on chromosome 6D (17.76%), and 192 on chromosome 5A (16.08%). Further analysis revealed enrichment in the 700–750-Mb region on the chromosome arm 4AL ( Supplementary Figure S2b ).

Further analysis of the 700–750-Mb region on chromosome arm 4AL showed that Xgwm160 and Xwmc313 were polymorphic between HHM and NMZ119 as well as between the R and the S pools. However, they were not polymorphic between HHM and CS. Therefore, these two markers were genotyped in 154 families from the NMZ119/HHM cross and 156 families from the reciprocal HHM/NMZ119 cross. PmHHM was mapped between Xgwm160 and Xwmc313. For fine-mapping PmHHM, 15 newly developed SSR markers based on CS RefSeq v2.1 were analyzed. Markers Xmp1503, Xmp1512, Xmp1445, and Xmp1446 were polymorphic among HHM, NMZ119, and two pools ( Supplementary Figure S3 ), whereas Xmp1888, Xmp1863, Xmp1864, Xmp1970, and Xmp1564 were polymorphic among HHM, NMZ119, CS, and their corresponding pools ( Supplementary Figure S3 ). PmHHM was further mapped within the marker intervals of Xmp1512–Xwmc313 ( Figure 2A ) and Xmp1970–Xmp1863 ( Figure 2B ) between these two populations, respectively.

When different populations were used for mapping, clear differences in the recombination rates were noticed among them. To investigate this phenomenon, we conducted a detailed analysis of the recombination rates in gene regions, distal-telomere regions, and proximal-telomere regions ( Table 3 ). Our results showed that (1) the recombination rates of the NMZ119 population were lower than those of the CS population and (2) the recombination rate of the targeted region was lower than those on either side in the NMZ119 population.

The two molecular markers flanking the targeted locus (Xmp1512 and Xmp1445) were used to analyze 1,352 susceptible plants from the NMZ119/HHM population. Xmp1512 identified 12 heterozygous lines, and Xmp1445 identified nine ( Supplementary Table S3 ). Analyzing 820 susceptible lines from the population of HHM/CS using Xmp1970 and Xmp1863 identified 14 and 3 heterozygous lines, respectively ( Figure 3A ). Further saturation of the genetic map by genotyping existing and newly developed markers against the NMZ119/HHM population narrowed down the location of PmHHM in the NMZ119/HHM population to a 0.0026-cM region, with a physical distance of 4.46 Mb between markers Xmp1567 and Xmp1444 ( Figure 2C ). In the HHM/CS population, PmHHM was mapped to a 0.0031-cM genetic interval of 187.4 kb (CS RefSeq v2.1) between Xmp1998 and Xmp1990 ( Figures 2D , 3B ). The lower recombination rate in the NMZ119 population and the higher recombination rate in the CS population were also apparent in the fine-mapping analysis ( Table 3 ).

Several genes conferring PM resistance were found to be located in the region of PmHHM. They include Pm61, PmHSM, MlIW30, QPm.tut-4A, PmPBDH, MlNFS10, PmHHXM, and PmXNM. However, we could only be able to obtain seeds of the HSM and HHXM from the eight wheat differentials for resistance spectral analysis at the seeding stage and for allelic testing. As shown in Table 1 , HHM was resistant to all 25 Bgt isolates, whereas HSM was susceptible to two of them (WH-11 and 39-19), and HHXM was susceptible to only one of them (46-25). Assuming that a single resistance gene was responsible for PM resistance in each of these lines, these results indicated that these genotypes possess different genes. To ascertain the resistance spectrum of PmHHM, 10 randomly chosen homozygous resistant F_2:3 lines from NMZ119/HHM were inoculated with the 25 Bgt isolates. All 10 F_2:3 lines were homogeneously resistant to all of the 25 isolates, indicating that PmHHM conferred a broad-spectrum resistance to this disease.

A total of 13 F₁ and 1,801 F₂ plants from the population of HHM/HHXM and 10 F₁ and 1,204 F₂ plants from HHM/HSM were all resistant to isolate B03 ( Table 2 ). Under the conditions of a chi-square test with a P-value of 0.05, the maximum genetic distances from PmHHM to PmHHXM and from PmHHM to PmHSM are 0.0043 and 0.0064 cM, respectively, indicating that PmHHM, PmHHXM, and PmHSM are likely allelic or closely linked to each other. Although we did not obtain the other six materials, we anchored all nine genes on the CS genome and constructed physical location maps of these genes ( Figure 4 ). Based on the physical location of the genes on the chromosome, we determined that PmHHM differs from Pm61, PmPBDH, MlNFS10, MlIW30, and QPm.tut-4A and is allelic or closely linked to PmHHXM, PmXNM, and PmHSM.

Due to differences in recombination rates between populations and issues with dominant markers ( Supplementary Figure S3 ) between NMZ119 and HHM/CS, we suspect that there are differences in presence/absence variations (PAVs) in the PmHHM region between NMZ119 and HHM/CS. For conformation, we conducted a synteny analysis of the 740–754-Mb region among CS and 13 other common wheat genome sequences ( Figure 5 ). As anticipated, a strong level of synteny was identified among Jagger, CDC Landmark, and CDC Stanley in the vicinity of PmHHM. However, the synteny between CS and the remaining 10 cultivars was disrupted due to significant sequence divergence in the specified region ( Figure 5 ). These findings provide further support for our hypothesis that PAVs exist among different wheat varieties in the PmHHM gene region.

The targeted locus was mapped within an interval of 187.4 kb containing four high-confidence genes and seven low-confidence ones. Three of the high-confidence genes were of the NLR type ( Figure 3B ). Among the genomes analyzed, these genes were present in Jagger, CDC Landmark, CDC Stanley, and CS. However, they were missing in other genotypes, including Kenong 9204, Julius, Kariega, Renan, ArinaLrFor, LongReach Lancer, Mace, SY Mattis, Norin 61, and Zang 1817 ( Figure 3B ).