Seedling stage resistance tests were conducted on 86 experimental wheat materials, including two susceptible checks and 10 near-isogenic lines (NILs)/differentials carrying known Lr genes. The resistance of these materials against 14 distinct Pt pathotypes was evaluated. The experiments were performed under controlled conditions in temperature-regulated growth chambers or greenhouses at two locations: the ICAR-Indian Institute of Wheat and Barley Research (IIWBR), Regional Station, Flowerdale, Shimla; and the National Phytotron Facility, ICAR-Indian Agricultural Research Institute (IARI), New Delhi.

Each test material was represented by 4–5 seeds sown in aluminum trays measuring 11 × 4 × 3 inches. The trays were filled with sterilized fine-loamy soil autoclaved at 60°C for 1 h to eliminate microbial contaminants. The soil was enriched with farmyard manure at a 3:1 ratio and supplemented with 5 g NPK fertilizer (12:32:16 ratio) to ensure optimal seedling growth. The seedlings were cultivated in a spore-free greenhouse environment maintained at a temperature of 25–30°C, relative humidity of 50–70%, and a 12-h photoperiod to simulate ideal growing conditions.

At the two-leaf stage, seven-day-old seedlings were inoculated using a spray gun. Urediospores of each Pt pathotype were suspended in light mineral oil (Soltrol 170^®, Chevron Phillips Chemicals Asia Pvt. Ltd., Singapore) at a concentration of 2–3 mg/mL, corresponding to 5–10 × 10³ spores/mL. This suspension ensured the uniform coverage and effective delivery of the inoculum. Following inoculation, the seedlings were incubated for 48 h in a humid chamber maintained at 20°C, 100% relative humidity, and a 12-h photoperiod to promote fungal infection and establishment. After the incubation period, the inoculated plants were transferred to a greenhouse environment maintained at a temperature of 20 ± 2°C, relative humidity of 80–90%, a 12-h photoperiod, and illumination intensity of approximately 15,000 lux (Bhardwaj et al., 2010). These controlled conditions ensured consistent disease development and facilitated accurate assessment of resistance responses in the test materials.

Fourteen days post-inoculation, the infection types (ITs) were assessed for each seedling using a modified Stakman scale (Stakman et al., 1962), as described by Bhardwaj et al. (2010). The ITs were categorized into distinct classes to differentiate resistance levels: ITs of 0 to “–” indicated no visible symptoms or hypersensitive flecks, reflecting a highly resistant response; ITs of 1 to 2 represented minute uredia with necrosis, indicating resistance; an IT of 2+ denoted small to medium uredia with chlorotic halos, indicating moderate resistance; IT 3 corresponded to large uredia with or without chlorosis and profuse sporulation, suggesting moderate susceptibility; and an IT of 3 + indicated the presence of both 3+ and 3+ pustules, characteristic of susceptibility to highly susceptible responses. Seedling tests were conducted three times under uniform conditions to ensure accuracy and resolve any inconsistencies, enabling the robust and reliable characterization of resistance phenotypes across the tested materials.

Field experiments were conducted at the Wheat Rust Pathology Experiment Farm, ICAR-Indian Agricultural Research Institute (IARI), New Delhi, during two consecutive rabi seasons (2020–21 and 2021–22). The objective was to assess and phenotype the same set of 86 test materials, along with two susceptibility checks, for adult plant resistance (APR) against leaf rust under artificially induced epiphytic conditions. The experimental farm is geographically located at 28°40′23″N latitude and 77°13′27″E longitude, at an elevation of 228.61 meters above sea level. This site falls within a semi-arid, subtropical climate zone characterized by clay loam and alluvial soil textures, making it ideal for wheat cultivation and disease screening. Additionally, this location serves as a well-established wheat rust resistance screening center in the Union Territory of the Delhi region, which is part of the North-Western Plain Zone (NWPZ) of India. Detailed meteorological data recorded on a standard weekly basis during the study period are presented in Supplementary Tables S3, S4. These data confirmed that environmental conditions were accurately correlated with disease development and resistance evaluations.

The field experiment was conducted using a randomized block design (RBD) with three replicates to ensure statistical robustness and minimize experimental error. Each wheat genotype was planted in small adjacent plots consisting of two rows, each 1 m in length, with a row spacing of 25 cm. The distance between individual plots was maintained at 50 cm to prevent cross-contamination and ensure uniformity in the disease assessment. Planting was carried out during the first week of December for both rabi seasons (2020–21 and 2021–22), adhering to the optimal sowing practices for wheat cultivation in the region. Standard agricultural practices, including timely irrigation, fertilization, and weed management, were strictly followed to maintain consistent crop growth and development. Irrigation channels were established between replicates to ensure uniform water distribution across the experimental site. Additionally, to create intense and continuous disease pressure under field conditions, the experimental plots were bordered with two rows of highly susceptible wheat cultivars, namely Local Red and A-9-30-1. These border rows served as a source of inoculum, facilitating the uniform spread of Pt pathotypes and ensuring the reliable evaluation of APR in the test materials.

To artificially induce epiphytotic conditions in the field, a balanced mixture of the four most prevalent Pt pathotypes (12–5, 77–5, 77–9, and 104–2) representing all major virulence groups was utilized. The urediospore suspension was prepared by suspending spore dust in water supplemented with a few drops of Tween-20 to ensure uniform dispersal and adherence to the plant surfaces. Inoculation of the infector rows, consisting of highly susceptible wheat cultivars, was performed during the last week of January each year. The suspension was applied evenly onto susceptible infector rows using a hand sprayer in the evening to maximize the infection efficiency under favorable environmental conditions. Following inoculation, the foliar canopy was kept moist by applying fine droplets of water until nightfall, ensuring prolonged leaf wetness to facilitate fungal penetration and establishment. This procedure ensured consistent and intense disease pressure across the experimental plots, enabling reliable evaluation of APR in the test materials.

Slow-rusting APR was evaluated by analyzing the host response and quantifying various epidemiological parameters, including the coefficient of infection (CI), final disease severity (FDS), relative area under the disease progress curve (rAUDPC), and apparent infection rate (r). Disease severity and ITs were assessed six times at seven-day intervals between February 24th and March 30th during each study year (2020–21 and 2021–22). Disease scoring commenced when susceptibility checks exhibited 25–30% severity across all replicates for each genotype/plot. Disease severity was scored using the modified Cobb scale (Peterson et al., 1948), whereas adult plant infection types were categorized as follows: R (resistant), MR (moderately resistant), MS (moderately susceptible), or S (susceptible) (Roelfs et al., 1992). The CI value was calculated by multiplying the disease severity by a constant factor assigned to each infection type: immune = 0, R = 0.2, MR = 0.4, M = 0.6, MS = 0.8, and S = 1 (Stubs et al., 1986). AUDPC and rAUDPC were estimated from multiple disease severity measurements using the established methodologies described by Milus (1986). Briefly, the AUDPC was calculated as the sum of the trapezoidal areas under the disease progress curve, whereas the rAUDPC was derived by normalizing the AUDPC of each genotype against that of the susceptible check and expressed as a percentage. The apparent infection rate (r) was determined using the method outlined by Vanderplank (1963), which quantifies the rate of disease development over time based on disease scores recorded at successive time points.

The 86 test materials were categorized into three distinct levels of adult plant resistance (APR): high, moderate, and low, based on their disease response and the values of key epidemiological parameters, including CI, FDS, rAUDPC, and r. These parameters were evaluated under field conditions to assess the durability and effectiveness of slow-rusting resistance for each genotype. Categorization was performed using standardized criteria outlined in Supplementary Table S5 (Singh et al., 2020), which provided clear thresholds for classifying genotypes into the respective APR categories.

Genomic DNA was isolated from fresh leaves of individual plants at the 3–4 leaf stage using the cetyltrimethylammonium bromide (CTAB) extraction method for all 86 wheat materials (Doyle and Doyle, 1990). The concentration and purity of the extracted DNA were assessed by 0.8% agarose gel electrophoresis, with lambda uncut DNA serving as a standard. The quality and quantity of the DNA samples were confirmed using a NanoDrop™ Lite Spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, United States). The DNA samples were subsequently diluted to a working concentration of 25 ng/μL and stored at −20°C for further analysis. To detect and confirm the presence or absence of specific Lr genes, three gene-based molecular markers closely linked to known Lr genes (Lr10, Lr24, and Lr34) were used (Table 2). Polymerase chain reaction (PCR) was performed for each target gene using the corresponding molecular markers under optimized conditions. The amplified PCR products were separated on 4% MetaPhor agarose gels at a constant voltage of 80 volts for 2 h to ensure clear resolution of the bands. This molecular analysis facilitated the precise identification of Lr genes in the test materials, providing critical insights into their genetic resistance profiles.

A marked increase in disease pressure was observed during both rabi seasons, with susceptible checks exhibiting the highest severity levels. In 2020–21, susceptible checks reached 95–100% FDS, reflecting extreme susceptibility. Adult plant infection responses revealed diverse resistance profiles: 18 genotypes showed ‘R-TR’ (resistant to moderately resistant), 20 had ‘TMR-MR’ (moderately resistant), and 21 displayed ‘MS’ (moderately susceptible) reactions (Figure 3A and Supplementary Table S3). Genotypes with ‘MS’ or ‘MR’ responses are likely to carry genes for all-stage or slow-rusting adult plant resistance (APR) (Singh et al., 2005). Based on the FDS thresholds, genotypes were categorized into high (1–20% FDS), moderate (21–40%), and low (41–60%) APR levels (Singh et al., 2020). Notably, 66 genotypes exhibited high APR, 14 showed moderate APR, and two had low APR (Figure 3B and Supplementary Table S7). In 2021–22, disease patterns mirrored the previous season, although the severity slightly declined (Figure 3B and Supplementary Table S7). Susceptible checks reached 90–95% FDS. Three genotypes (NP 111, NP 12 and HI 1500) improved from ‘TR’ to ‘R’ ITs, while two (GW 10 and MP 4010) shifted from ‘MS’ to ‘S’ (susceptible), highlighting the variability in resistance stability.

The CI integrates both host response and FDS to quantify resistance levels in adult plants. Genotypes were classified into four categories based on CI values: high APR (CI ≤ 20), moderate APR (CI ≤ 40), low APR (CI ≤ 60), and high susceptibility (CI > 60), following established thresholds (Ali et al., 2009; Singh et al., 2020). In the 2020–21 growing season, susceptible checks exhibited extreme disease severity (95–100% FDS), resulting in correspondingly high CI values (Figure 3C and Supplementary Table S7). Of the 86 genotypes evaluated, 69 showed high adult plant resistance (CI ≤ 20), 11 exhibited moderate APR (CI ≤ 40), three displayed low APR (CI ≤ 60), and three were classified as highly susceptible (CI > 60). In the 2021–22 season, the CI trends remained consistent with those observed in the previous year, although disease pressure was slightly reduced (Figure 3C and Supplementary Table S7). Susceptible checks reached 90–95% FDS, yet the overall categorization of genotypes remained stable across both seasons. This consistency underscores the reliability of CI as a robust metric for evaluating APR under variable environmental conditions and reinforces its utility in resistance breeding programs.

The rAUDPC quantifies disease progression over time, with lower values indicating slower epidemic development. Genotypes were categorized into high (≤30% of susceptibility check), moderate (≤40%), and low (≤60%) APR levels (Schachermayr et al., 1997; Tong et al., 2024). In 2020–21, 78 genotypes, including 18 with ‘R-TR’ infection types (ITs), exhibited rAUDPC values ≤30%, reflecting high slow-rusting APR (Figure 3D and Supplementary Table S7). These genotypes likely harbor APR genes, as their low rAUDPC values contrast with late-season ITs, suggesting delayed disease progression (Parlevliet, 1988). In 2021–22, the results remained consistent, with the rAUDPC values aligned with the previous season (Figure 3D and Supplementary Table S7). This stability highlights rAUDPC as a robust metric for evaluating durable resistance, particularly for identifying genotypes with slow-epidemic traits that are critical for long-term disease management.

It measures disease progression over time, with lower values indicating a slower epidemic spread. In 2020–21, all genotypes exhibited r-values lower than those of susceptible checks (0.37–0.38). Notably, 63 genotypes showed minimal disease progression (r < 0.05), reflecting high APR (Figure 3E and Supplementary Table S7). Susceptible (S) genotypes had the highest r-values (0.21–0.23), whereas moderately susceptible (MS) genotypes displayed the lowest (0.02–0.08). High APR genotypes were defined as those with r < 0.05, moderate APR as 0.06–0.09, aligning with other resistance metrics (Ali et al., 2007; Singh et al., 2020). In 2021–22, the trends mirrored the prior season, with consistent r-value categorizations (Figure 3E and Supplementary Table S7). This consistency reinforces the reliability of r-values for identifying genotypes with durable resistance, particularly those exhibiting slow epidemic progression that is critical for APR breeding strategies.

The 86 test genotypes and two susceptible controls were categorized into three APR levels (high, moderate, and low) based on the adult plant infection type and integrated analysis of slow-rusting components (CI, rAUDPC, FDS, and r-values). Across both the 2020–21 and 2021–22 seasons, 64 genotypes consistently exhibited high APR, 16 showed moderate APR, and 2 displayed low APR (Figure 4 and Table 4).

Notably, all genotypes classified as high/moderate APR were susceptible to one or more Pt pathotypes at the seedling stage, confirming the presence of non-race-specific, slow-rusting APR genes expressed exclusively at the adult plant stage. This distinction underscores the utility of these genotypes for durable resistance breeding, as APR genes often confer broad-spectrum long-lasting protection against dynamic pathogen population.

To evaluate the relationships between slow-rusting adult plant resistance (APR) metrics, correlations among key epidemiological parameters—coefficient of infection (CI), final disease severity (FDS), relative area under the disease progress curve (rAUDPC), and disease progress rate (r)—were analyzed, with CI considered the primary parameter because of its comprehensive representation of disease intensity (Singh et al., 2020; Ali et al., 2009). In the 2020–21 growing season, CI exhibited strong positive correlations with FDS (R² = 0.9838), rAUDPC (R² = 0.9781), and r-values (R² = 0.9649), indicating that these components collectively reflect the disease progression dynamics in wheat genotypes (Figure 5). Similar correlations were observed during the 2021–22 season, reinforcing the stability and consistency of these relationships across the different growing environments (Figure 5). The persistent correlation patterns highlight the reliability of the CI as a composite metric for assessing slow-rusting APR. The integration of multiple disease parameters into a unified evaluation framework is visually depicted through a scatter plot matrix (Figure 6) and chord diagram (Figure 7), which facilitates the interpretation of complex relationships among metrics and enables streamlined resistance evaluation in breeding programs. By consolidating multiple disease parameters into a single reliable metric, such as CI, breeders can more effectively identify genotypes with durable resistance, contributing to the development of climate-resilient wheat varieties. This approach not only enhances the precision of resistance screening but also supports the efficient deployment of APR genes in future breeding efforts.

The Lr10-specific molecular marker is a dominant marker, producing a 289 bp amplicon, confirming the presence of the resistance gene. Among the 86 wheat genotypes screened, 24 (27.9%) tested positive for Lr10 (Supplementary Table S9 and Supplementary Figure X2). This subset of genotypes harboring Lr10 highlights its prevalence and potential utility in resistance breeding.

The molecular marker Xbarc71, tightly linked to the Lr24 resistance gene, generated a 100 bp amplicon, indicating the presence of the functional allele, while the 120 bp fragment indicates the recessive allele. This marker behaves as a codominant marker, positioned 2.4 cM from the Lr24 locus, making it a reliable tool for marker-assisted selection. Screening of the 85 wheat genotypes revealed that 16 (18.8%) carried Lr24 (Supplementary Table S9 and Supplementary Figure X2). These findings validate Xbarc71 as an effective marker for identifying Lr24-carrying genotypes, which are critical for enhancing leaf rust resistance in wheat breeding initiatives.

Lr34 is a non-race-specific APR gene that plays a critical role in imparting durable resistance to leaf rust in wheat. Amplification of Lr34 yielded two distinct band sizes: 150 bp (presence of the resistant allele, Lr34) and 229 bp (presence of the recessive allele). In certain cases, both bands appeared, indicating heterozygosity (Lagudah et al., 2006). The positive control ‘Baxter’ consistently amplified the 150 bp band, confirming the presence of Lr34. Of the 86 genotypes tested, 33 (38.4%) were found to carry the 150 bp band, thereby confirming the presence of the Lr34 gene (Supplementary Table S9 and Supplementary Figure X3).

Several of the tested genotypes carried combinations of two or more Lr genes. Three genotypes (UP 1109, HS 1138-6-4, and K 8962) were identified as carriers of all three resistance genes (Lr10, Lr24, and Lr34). These genotypes exhibited consistently low final disease severity (FDS ≤ 15%), coefficient of infection (CI ≤ 15), and relative area under the disease progress curve (rAUDPC ≤ 25%) across both years of field evaluation, indicating strong slow-rusting resistance. Other dual-gene combinations, Lr24 + Lr34 (C 286 and WR 544), Lr10 + Lr34 (D 134, HP 1102, HUW 12, IWP 72, UP 2121, DL 153–2, PBW 299, HP 1731, HS 365, and HW 1085) and Lr10 + Lr24 (K 8020 and C 285) also showed improved resistance compared to single-gene carriers. These genotypes demonstrated reduced disease progression and lower AUDPC values than those carrying only one of the genes. The apparent infection rate (r) was also significantly reduced in lines with multiple Lr genes, indicating slower epidemic development and improved resistance stability over time. These results highlight the importance of gene pyramiding in enhancing the durability and effectiveness of resistance, particularly under high disease pressure. The integration of the major seedling resistance genes (Lr10 and Lr24) with durable APR genes (Lr34) provides a robust defense mechanism against evolving pathogen races.