In this study, we selected ten wheat varieties from Shandong (‘Shannong 28,’ ‘Luyan 128,’ ‘Jimai 44,’ ‘Jimai 38,’ and ‘Yannong 1212’), Henan (‘Zhongmai 578,’ ‘Malan 1,’ and ‘Bainong 4199’), and Hebei (‘Zhongxinmai 998’ and ‘Gaoyou 5766’). In production, ‘Shannong 28,’ ‘Luyan 128,’ ‘Jimai 38,’ ‘Yannong 1212,’ ‘Malan 1,’ ‘Bainong 4199,’ and ‘Zhongxinmai 998’ are high-yielding varieties, and ‘Jimai 44,’ ‘Zhongmai 578,’ and ‘Gaoyou 5766’ are strong-gluten quality varieties.

A ⁶⁰Co-γ radiation source provided by the Shandong Irradiation Center (Jinan, Shandong Province, China) was used for the experiment. Thirteen radiation dose levels were set: 0, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1100, and 1200 Gy. For each wheat cultivar, 500 g of seeds was irradiated at each dose level. After irradiation, 200 seeds per cultivar (per dose) were sown in a greenhouse. The growth conditions were controlled at a constant temperature of 25 °C, 16/8 h light/dark cycle, light intensity of 300 μmol m^–2 s^–1, and relative humidity of 70%. After 14 days, germination and seedling growth were evaluated. Each treatment was conducted in triplicate. The formulas used to calculate the germination and normal growth rates are as follows:

Efficient mutagenesis is typically associated with a normal growth rate of 20-40%. Based on this criterion, 100 kg of seeds per cultivar was subjected to ⁶⁰Co-γ radiation mutagenesis for subsequent experiments. To balance the mutagenesis efficiency and population scale requirements, a gradient irradiation scheme centered on the optimal mutagenic dose, with ±50 Gy variations around this dose, was designed. Field sowing was performed at 225 kg ha^-1, covering 0.67 ha per cultivar in October 2021 (Supplementary Figure S1). Given the large size of the mutant population, we pooled the M_1:2 seeds separately for each cultivar in June 2022. The M₂ mutant population served as the material for subsequent phenotypic characterization.

In November 2022, Changqing experienced unseasonably warm conditions with an average temperature of 10 °C, followed by a cold wave from November 30 to December 3, during which temperatures decreased rapidly by 13-15 °C to a low of -5 °C. Severe frost damage occurs owing to the lack of cold acclimation in wheat under warm conditions. This natural frost event created a field-based selection environment for the identification of frost-tolerant mutants.

In October 2022, a mutant population (2 ha) was established in Kenli, Shandong Province. The experimental site featured saline-alkaline soil. The basic physicochemical properties of the soil are as follows: pH value ranges from 7.2 to 8.0; the soil ion composition includes Na⁺ content of 3.2-5.6 g kg^-1, CO₃^2- content of 0 g kg^-1, and HCO₃⁻ content of 1.6-1.9 g kg^-1. Phenotypic screening for salt tolerance was conducted during the seedling (March) and maturity (June) stages in 2023.

The sampling site for ROS and antioxidant enzyme determination is specified as the first true leaf of wheat seedlings, and the sampling time is 14 days after radiation treatment (consistent with the seedling growth evaluation time).

For O₂⁻ measurements, the samples were homogenized in a pre-chilled 50 mM potassium phosphate buffer (pH 7.8). The homogenate was mixed with 50 mM phosphate buffer (pH 7.8) and 10 mM hydroxylammonium chloride at a volume ratio of 1:1:2, incubated at 25 °C for 20 min, and subsequently mixed with twice its volume of ethyl ether. After absorbance quantification at 530 nm (Hao et al., 2018), the H₂O₂ content was determined following the method described by Mukherjee and Choudhuri (1983).

The activities of the antioxidant enzymes SOD (Wang et al., 2020) and CAT (Wang et al., 2022b) were determined following previously reported methods. SOD activity assay: The reaction system contains 50 mM phosphate buffer (pH 7.8), 13 mM methionine, 75 μM nitroblue tetrazolium (NBT), 10 μM EDTA-Na₂, and 2 μM riboflavin. The reaction is carried out at 25 °C under a light intensity of 4000 lx for 20 min, and the absorbance is measured at 560 nm. One enzyme activity unit (U) is defined as the amount of enzyme required to inhibit NBT photoreduction by 50%; CAT activity assay: The reaction system contains 50 mM phosphate buffer (pH 7.0) and 20 mM H₂O₂. The reaction is performed at 25 °C, and the change in absorbance (ΔA₂₄₀) is measured at 240 nm. One enzyme activity unit (U) is defined as the amount of enzyme that decomposes 1 μmol of H₂O₂ per minute. All samples were analyzed using a UV-2550 spectrophotometer (Shimadzu, Kyoto, Japan).

In a preliminary experiment, ten wheat varieties from Shandong, Henan, and Hebei were exposed to different doses of ⁶⁰Co-γ radiation. Seedling growth was evaluated 14 days after the seeds had germinated. The results showed a dose-dependent reduction in the normal growth rate, with significant varietal differences in radiosensitivity among the genotypes (Figure 1). At irradiation doses ≤400 Gy, all varieties grew normally. When the dose reached ≥600 Gy, the normal growth rates of radiation-sensitive varieties, including Shannong 28, Luyan 128, Jimai 44, Zhongmai 578, Malan 1, and Zhongxinmai 998, significantly decreased, whereas those of highly tolerant varieties, including Jimai 38, Yannong 1212, Bainong 4199, and Gaoyou 5766, only showed a marked decline at doses ≥900 Gy. Subsequently, the irradiation intensity that yielded a 20–40% normal growth rate was selected for the follow-up experiments. The respective ⁶⁰Co-γ radiation doses and corresponding normal growth rates of the varieties were as follows: Shannong 28 (600 Gy, 28.8%), Luyan 128 (700 Gy, 26.6%), Jimai 44 (700 Gy, 25.4%), Jimai 38 (900 Gy, 27.5%), Yannong 1212 (1000 Gy, 33.7%), Zhongmai 578 (600 Gy, 25.5%), Malan 1 (600 Gy, 29.5%), Bainong 4199 (900 Gy, 31.6%), Zhongxinmai 998 (700 Gy, 22.9%), and Gaoyou 5766 (1100 Gy, 35.5%) (Table 1).

ROS accumulation and antioxidant enzyme responses were analyzed in the irradiated wheat seedlings. At doses ≤400 Gy, the H₂O₂ content (Figure 2A) and O₂⁻ generation rate (Figure 2B) remained consistently low, with minor fluctuations, whereas SOD (Figure 3A) and CAT (Figure 3B) activities increased steadily with increasing doses. At ≥600 Gy, radiation-sensitive cultivars, including Shannong 28, Luyan 128, Jimai 44, Zhongmai 578, Malan 1, and Zhongxinmai 998, exhibited a sharp increase in H₂O₂ and O₂⁻; the SOD and CAT activities peaked at 300–400 Gy, then declined progressively. In contrast, highly tolerant cultivars, such as Jimai 38, Yannong 1212, Bainong 4199, and Gaoyou 5766, exhibited a significant increase in H₂O₂ and O₂⁻ at doses ≥900 Gy. Correspondingly, SOD and CAT activities were maximized at 500–700 Gy, followed by a steady decrease in activity.

In this study, a 10,350,000 ⁶⁰Co-γ irradiated M₂ mutant population was established, comprising the wheat varieties ‘Shannong 28’(1,010,000 lines), ‘Luyan 128’(1,420,000 lines), ‘Jimai 44’(1,730,000 lines), ‘Jimai 38’(1,080,000 lines), ‘Yannong 1212’(750,000 lines), ‘Zhongmai 578’(1,200,000 lines), ‘Malan 1’(940,000 lines), ‘Bainong 4199’(890,000 lines), ‘Zhongxinmai 998’(920,000 lines), and ‘Gaoyou 5766’(410,000 lines). The M₂ generation exhibited the highest frequency of induced mutations, making it a critical genetic resource for wheat improvement. Currently, M₂ bulk seed pools have been distributed to collaborative research programs to facilitate the screening of germplasm with desirable traits, including resistance to rust, powdery mildew, and drought tolerance.

The mutagenic efficiency of ⁶⁰Co-γ radiation is determined by the synergistic interplay of multiple factors. Irradiation parameters, including the total dose, dose rate, and exposure mode, which collectively determine the extent of genetic damage and mutation frequency, should be precisely regulated (Lowe et al., 2022; Sikder et al., 2013). The inherent biological properties of irradiated materials are equally important because significant variations in radiosensitivity exist across species and cultivars. Concurrently, mutagenic efficiency is modulated by critical physiological parameters, such as seed moisture content, developmental stage, and oxygen availability (Duarte et al., 2023; Khare et al., 2025).

Early studies primarily focused on the median lethal dose (LD50), defined as the ⁶⁰Co-γ radiation dose that kills 50% of treated seeds (Xiong et al., 2023; Yang et al., 2014). The LD50 criterion performs reasonably well in practice, as it consistently generates populations with a sufficient percentage of mutations. In the present study, we determined the optimal ⁶⁰Co-γ dose, under which the normal growth rate of 20-40% under optimal germination conditions (Figure 1; Table 1).

In previous studies, irradiating dry wheat seeds with 200-400Gy of ⁶⁰Co-γ radiation could achieve the half-lethal dose and obtain high mutation efficiency. This approach facilitates the selection of wheat germplasm with enhanced disease resistance, stress tolerance, high yield, and improved quality for breeding (Cheng et al., 2008; Wang et al., 2019; Xiong et al., 2023; Zhu et al., 2010). Although a rigorous dose-gradient experiment was conducted in the current study, the required mutagenic dose (with a normal growth rate of 20-40%) was significantly higher than that reported 200-400Gy previously. However, a repeated irradiation experiment was conducted on seeds of Jimai 38 in 2023, with a moisture content of 13.0% and a post-harvest storage period of 3 months. The radiation dose corresponding to a normal growth rate of 28.3% was 1100 Gy, suggested that the higher optimal radiation dose was due to the decay of the ⁶⁰Co-γ radiation source over time. Nevertheless, germplasm screening was effective.

Plants exposed to ⁶⁰Co-γ radiation undergo a rapid surge in intracellular ROS, a phenomenon known as “oxidative burst” (Ksas et al., 2024). This occurs via two primary mechanisms: first, the direct radiolysis of water molecules generates ROS, such as hydroxyl radicals, O₂⁻, and H₂O₂ (Wojtaszek, 1997); second, the disruption of intracellular electron transport chains in mitochondria and chloroplasts enhances electron leakage and subsequent ROS formation (Asada, 1999; Robertson et al., 1995).

Excessive ROS accumulation causes significant cellular damage by inducing membrane lipid peroxidation, compromising membrane integrity (Langebartels et al., 2002), promoting oxidative protein denaturation (Sood, 2025), and causing DNA strand breaks (Roldán-Arjona and Ariza, 2009). Plants activate endogenous antioxidant enzyme systems to mitigate oxidative stress (Cannea and Padiglia, 2025; Zheng et al., 2025).

In the present study, the antioxidant enzyme activity exhibited a characteristic biphasic response to increasing radiation intensity, with an initial enhancement followed by a gradual decline. Under low-dose irradiation, elevated enzymatic activity was essential for preserving cellular redox homeostasis by effectively scavenging ROS. However, beyond a critical radiation threshold, the compromised antioxidant system proved insufficient to neutralize excessive ROS accumulation, ultimately triggering an oxidative burst in wheat seedlings (Figures 2, 3). This redox imbalance caused growth arrest and seedling mortality. Notably, ROS accumulation and antioxidant enzyme activity exhibited a clear dose-dependent relationship with radiation intensity.

The ⁶⁰Co-γ irradiation mutant library systematically constructed in this study comprised elite cultivars from Shandong, Henan, and Hebei, with 10,350,000 lines. The library offers extensive genetic diversity and supports multiple breeding objectives. For breeding applications, we identified mutants with improved freezing tolerance, saline-alkali resistance, yield potential, and quality traits (Figures 4-6). These mutants are currently being used in crossing programs to introduce desirable traits into advanced breeding lines. The mutant library provides valuable resources for gene cloning and mechanistic studies of stress tolerance, yield, and quality. In genomic research, exome capture sequencing enables the compilation of a genotype-phenotype association database to support the targeted selection of favorable mutations. Integrating these mutant resources into molecular design breeding frameworks will facilitate systematic trait pyramiding and accelerate the development of elite cultivars with superior agronomic performance.