Breeding method for overcoming sterility of rice interspecific and subspecific hybrids and retaining genetic variation

Abstract

This invention provides a rice breeding method that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation. The method includes: hybridization and F1 generation evaluation, a first backcross to obtain BC1F1, a second backcross to obtain BC2F1, self-pollination to obtain BC2F2 and initiating selection, and stabilization and purification. This invention, based on a "two backcrosses + one self-pollination" (BC2F2) strategy, effectively overcomes intersubspecific or interspecific hybrid sterility by clearly defining four key decision nodes and their corresponding fertility thresholds. Simultaneously, it obtains a breeding population with normal fertility (≥80%) and rich agronomical trait variation in the BC2F2 generation.

Description

Technical Field

[0001] This invention belongs to the field of rice breeding technology, specifically relating to a breeding method that overcomes the genetic variation of interspecific and intersubspecific hybrid sterility in rice. Background Technology

[0002] Asian cultivated rice is divided into two subspecies: indica and japonica, both exhibiting significant heterosis with a yield potential increase of over 15%. However, the F1 generation of intersubspecies hybrids typically shows semi-sterility (<50% spikelet fertility), severely limiting the utilization of heterosis. Similarly, interspecific hybridization between Asian cultivated rice and African cultivated rice (O. glaberrima) also faces severe heterosis barriers (Oka, 1988; Zhang et al., 2024).

[0003] In current breeding practices, backcrossing is a common method to overcome hybrid sterility, but it has the following technical drawbacks: blind selection of backcross generations, lack of clear guidance on the dynamic laws of fertility recovery, early selection (such as BC1F1) leads to incomplete fertility recovery, and late selection (such as BC3 and above) leads to a narrowing of the genetic background and loss of genetic diversity; it cannot balance fertility recovery and preservation of genetic variation: traditional backcross breeding often loses the excellent agronomic traits carried by the donor parents while pursuing fertility recovery; and it lacks standardized breeding decision points: breeders find it difficult to determine when to initiate self-pollination selection and when to conduct marker-assisted selection (MAS), resulting in low breeding efficiency.

[0004] While existing breeding methods employ a generational process of "hybridization + two backcrosses + self-pollination" (e.g., CN102524046A), these are primarily applied to disease resistance breeding and lack standardized decision-making nodes and fertility threshold systems for the complex trait of hybrid sterility. To overcome this specific technical problem of hybrid sterility, current technologies mainly employ a strategy of continuous backcrossing for 5-6 generations (e.g., CN116034868A), resulting in long breeding cycles and significant loss of genetic diversity. Therefore, a standardized backcross-self-pollination synergistic breeding method is urgently needed to overcome hybrid sterility, clearly defining key generational nodes to effectively overcome hybrid sterility while simultaneously preserving rich genetic variation. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a rice breeding method that overcomes interspecific and intersubspecific hybrid sterility and preserves genetic variation, in order to address the shortcomings of the prior art. This method is based on the rice breeding strategy of "two backcrosses + one self-cross" (BC2F2). By clarifying four key decision nodes and their corresponding fertility thresholds, it can effectively overcome intersubspecific or interspecific hybrid sterility and simultaneously obtain a breeding population with normal fertility (≥80%) and rich agronomic trait variation in the BC2F2 generation.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a rice breeding method that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, characterized in that the method comprises: S1, evaluation of hybridization and F1 generation: (Decision node 1: bottleneck period of hybrid sterility) Crossing was performed using the first parent as the female parent and the second parent as the male parent to obtain F1 generation seeds. The F1 generation seeds were planted to obtain F1 generation plants. The spikelet fertility of the F1 generation plants was investigated: spikelet fertility <40% was considered highly sterile, 40% ≤ spikelet fertility <60% was considered semi-sterile, 60% ≤ spikelet fertility <80% was considered moderately fertile, and spikelet fertility ≥80% was considered normal. It was confirmed that the F1 generation plants exhibited significant spikelet fertility impairment. S2, First submission yields BC1F1: (Decision Node 2: First submission evaluation period) Using the F1 generation plants obtained from S1 as the female parent and the first parent from S1 as the recurrent parent, backcrossing was performed to obtain BC1F1 seeds; A population of BC1F1 seeds was planted to obtain a BC1F1 plant population, and the spikelet fertility of the BC1F1 plant population was investigated. S3, Second backcross yields BC2F1: (Decision Node 3: Fertility Breakthrough Period) Plants with good overall traits and spikelet fertility restored to more than 50% in the BC1F1 plant population obtained in S2 were used as female parents, and backcrossing was performed with the first parent in S1 as the recurrent parent to obtain BC2F1 seeds. The population planted with the BC2F1 seeds was used to obtain the BC2F1 plant population. The spikelet fertility of the BC2F1 plant population was investigated. The investigation and analysis were conducted to determine whether the spikelet fertility was significantly improved compared with the BC1F1 plant population and was close to the spikelet fertility requirements for production. S4. Self-crossing yields BC2F2 and selection begins: (Decision Node 4: Self-crossing selection initiation period) Multiple independent plants were randomly selected from the BC2F1 plant population obtained from S3 for self-pollination to obtain seeds of the BC2F2 segregating population. After planting, the BC2F2 segregating population was obtained. The spikelet fertility and agronomic traits of the BC2F2 segregating population were investigated, and BC2F2 plants with spikelet fertility ≥80% and excellent agronomic traits were screened. S5. Stable purification: (subsequent breeding procedures) The BC2F2 single plants obtained from S4 were selected for multiple generations using the pedigree method or single seed propagation method to develop stable lines.

[0007] Preferably, the combination of the first parent and the second parent in S1 is a combination of indica rice and japonica rice subspecies or an interspecific combination of Asian cultivated rice and African cultivated rice.

[0008] Preferably, in the combination of indica and japonica rice subspecies, the maternal parent is temperate japonica rice Dianjingyou 1, and the paternal parent is indica rice Minghui 63; in the interspecific combination of Asian cultivated rice and African cultivated rice, the maternal parent is African cultivated rice germplasm Tog5681, and the paternal parent is Asian cultivated rice variety IR64.

[0009] Preferably, when the maternal parent is temperate japonica rice Dianjingyou 1 and the paternal parent is indica rice Minghui 63, in S1 the spikelet fertility of the F1 generation plants is confirmed to be semi-sterile, with a spikelet fertility of 48.26%±9.15%; in S2 the spikelet fertility of the BC1F1 plant population is 58.63%±8.41%; and in S3 the spikelet fertility of the BC2F1 plant population is 73.08%±6.58%, which is significantly higher than the average spikelet fertility of the BC1F1 plant population.

[0010] Preferably, when the maternal parent is the African cultivated rice germplasm Tog5681 and the paternal parent is the Asian cultivated rice variety IR64, in S1, the spikelet fertility of the F1 generation plants is confirmed to be 0%; in S2, the average spikelet fertility of the BC1F1 plant population is 15%, with a variation range of 0% to 40%; in S3, the average spikelet fertility of the BC2F1 plant population is 40%, with a variation range of 5% to 70%, and is significantly higher than the average spikelet fertility of the BC1F1 plant population.

[0011] Preferably, the screening of agronomic traits in S4 includes spikelet fertility, number of grains per spike, and number of effective spikes.

[0012] Preferably, when the maternal parent is temperate japonica rice Dianjingyou 1 and the paternal parent is indica rice Minghui 63, the screening threshold criteria for excellent agronomic traits are: spikelet fertility > 88.15%, number of grains per spike > 150.44, number of effective spikes > 11.37, grain length-width ratio > 3.22, and plant height < 93.63 cm.

[0013] Preferably, when the maternal parent is African cultivated rice germplasm Tog5681 and the paternal parent is Asian cultivated rice variety IR64, the screening threshold criteria for excellent agronomic traits are: spikelet fertility ≥80%, number of grains per spike >130, number of effective spikes >13, and number of secondary branches >20.

[0014] Compared with the prior art, the present invention has the following advantages: 1. Wide applicability of the invention: The invention is applicable to both indica-japonica subspecies hybridization and Asian-African rice hybridization, and has universality.

[0015] 2. The invention has a significant effect on fertility restoration: Through the strategy of "two backcrosses + one self-cross", the invention can gradually improve the fertility of interspecific and intersubspecific hybrid offspring to a level close to normal, effectively overcoming the obstacle of hybrid sterility.

[0016] 3. The present invention has abundant genetic variation: Unlike traditional backcrossing (BC3 and above) which leads to excessive homozygosity in the genetic background, the present invention simultaneously retains abundant genetic variation in the BC2F2 generation, providing sufficient genetic variation material for the aggregation of traits such as high yield, high quality, and stress resistance.

[0017] 4. The invention has clear breeding nodes and is highly operable: The invention clarifies four key decision nodes, solves the technical problem of "when to choose" in backcross breeding, avoids the blindness of breeding work, and significantly shortens the breeding cycle.

[0018] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] Figure 1 This is the dynamics of spikelet fertility recovery and multiple comparison results of different backcross generations in Example 1 of the present invention.

[0020] Figure 2 This is the dynamics of spikelet fertility recovery and multiple comparison results of different backcross generations in Example 1 of the present invention. The red dashed line represents the mean line of Dianjingyou 1, and the green dashed line represents the mean line of Minghui 63.

[0021] Figure 3 It refers to the proportion of plants in the BC2F2 population of Example 1 of this invention that have better traits than the recurrent parents. Detailed Implementation

[0022] Example 1

[0023] This example illustrates interspecific hybridization breeding between indica and japonica rice (Dianjingyou 1 × Minghui 63). Materials: The maternal parent was the temperate japonica rice variety "Dianjingyou 1" (recurrent parent), and the paternal parent was the indica rice variety "Minghui 63" (donor parent). All of the above materials were provided by the Institute of Food Crops, Yunnan Academy of Agricultural Sciences.

[0024] The rice breeding method described in this embodiment overcomes intersubspecific hybrid sterility in rice while preserving genetic variation. The method is as follows: S1, evaluation of hybridization and F1 generation: (Decision node 1: bottleneck period of hybrid sterility) Crossing was performed using Dianjingyou 1 as the female parent and Minghui 63 as the male parent to obtain F1 generation seeds. These F1 seeds were planted to obtain F1 generation plants. The spikelet fertility of the F1 generation plants was investigated and found to be 48.26% ± 9.15%, confirming that the F1 generation plants exhibited semi-sterility, a decrease of 42.89 percentage points compared to the average fertility of the parents (91.15%). Figure 1 (F1 data points in the data); This stage focuses on assessing the compatibility between parents, without fertility selection; Spike fertility <40% is considered high sterility, 40% ≤ spikelet fertility <60% is considered semi-sterile, 60% ≤ spikelet fertility <80% is considered moderately fertile, and spikelet fertility ≥80% is considered normal. S2, First submission yields BC1F1: (Decision Node 2: First submission evaluation period) Using F1 plants obtained from S1 as the female parent and Dianjingyou 1 from S1 as the recurrent parent, backcrossing was performed to obtain BC1F1 seeds. A population planted with these BC1F1 seeds was then used to obtain BC1F1 plants. The spikelet fertility of these BC1F1 plants was investigated and found to be 58.63% ± 8.41%, which was not significantly higher than that of F1 (P = 0.170), remaining at a semi-sterile level. Figure 1 Based on the BC1F1 data points in the data set, it was determined that backcrossing was necessary, and no agronomic trait selection was performed in this generation. S3, Second backcross yields BC2F1: (Decision Node 3: Fertility Breakthrough Period) Using BC1F1 plants from S2 with good overall traits and spikelet fertility restored to over 50%, the female parent was used. Dianjingyou 1 from S1 was used as the recurrent parent for backcrossing to obtain BC2F1 seeds. These BC2F1 seeds were planted to obtain a BC2F1 plant population, and spikelet fertility was investigated. A fertility breakthrough was achieved when the average spikelet fertility of the population increased to over 70%, and was significantly higher than the average spikelet fertility of the BC1F1 plant population. In this example, after planting BC2F1, the spikelet fertility of the BC2F1 plant population reached 73.08% ± 6.58%, a significant increase of 14.45 percentage points compared to the BC1F1 plant population (P = 0.001), reaching the fertility breakthrough period (greater than 70%). Figure 1 (The BC2F1 data points in the data can be further selected through self-crossing); S4. Self-crossing yields BC2F2 and selection begins: (Decision Node 4: Self-crossing selection initiation period) Five independent BC2F1 plants were randomly selected from the BC2F1 population obtained from S3 for self-pollination to obtain seeds of the BC2F2 segregating population, namely five independent BC2F2 lines (numbered 21J2E2111~21J2E2115), totaling 366 plants. After planting, the BC2F2 segregating population was obtained. The spikelet fertility and agronomic traits (spikelet fertility, number of grains per spike, number of effective spikes, grain length-width ratio, and plant height) of the BC2F2 segregating population were investigated and compared with the recurrent parent "Dianjingyou 1". BC2F2 plants with spikelet fertility ≥80% and excellent agronomic traits were screened. Figure 2 Parental statistics (mean ± standard deviation): Spike fertility: Dianjingyou 1 93.33%±4.09%, Minghui 63 89.03%±4.22%; Number of grains per ear: Dianjingyou 1 139.33±24.41, Minghui 63 159.03±17.44; Effective spikelets: Dianjingyou 1 10.40±2.13, Minghui 63 9.50±2.59; Grain length-to-width ratio: Dianjingyou No. 1 3.19±0.1, Minghui 63 3.2±0.07; Plant height: Dianjingyou No. 1 102.47±5.37cm, Minghui 63 89.03±4.22cm; Sample size of each BC2F2 population: 21J2E2111 75 plants, 21J2E2112 66 plants, 21J2E2113 68 plants, 21J2E2114 78 plants, 21J2E2115 79 plants, totaling 366 plants.

[0025] (1) Spike fertility: Dynamics of spikelet fertility recovery and multiple comparisons of different backcross generations showed that the fertility of the F1 generation was 48.26%, the BC1F1 generation was 58.63% (no significant difference from F1), the BC2F1 generation was 73.08% (significantly higher than the previous two generations), and the BC2F2 generation was 81.77% (no significant difference from the donor parent Minghui 63). The average fertility of the BC2F2 population was close to the normal level, and the fertility of each population was mostly normally distributed, with a variation range of 63.2% to 94.8%. Among them, the 21J2E2115 population was most concentrated in the high fertility area (>88%), and 58.23% of the individual plants had fertility exceeding that of the recurrent parent ( Figure 2(Middle AE). From F1 to BC2F2, fertility shows a "slow-fast-stable" recovery trend. A single backcross (BC1F1) has limited improvement in fertility, while self-pollination after the second backcross (BC2F2) can achieve a leap in fertility. This is consistent with the genetic mechanism of the gradual homozygosity of heterozygous male-sterile loci (such as S5, Sa, Sb, etc.): most individual plants in BC2F1 still have 1-2 heterozygous male-sterile loci, and after self-pollination, homozygous fertile types appear, thus significantly improving the fertility of the population. Therefore, BC2F2 is the optimal generation to initiate fertility selection.

[0026] (2) Number of grains per ear: The coefficient of variation was 18%–25%, making it the most variable among all traits. The proportion of off-parents (with the upper limit of 95% confidence of 150.44 grains per ear for one-sided recurrent parent as the threshold) was 85.33%, 98.48%, 100%, 98.72%, and 50.63% in the five populations, respectively, with the 21J2E2113 population reaching 100%. Figure 2 (China FJ).

[0027] (3) Number of effective spikes: The 21J2E2115 population performed best, with a significant peak in the range of 12-15 spikes / plant, and an over-parent ratio of 55.70%; the peak of other populations was between 8-10 spikes / plant, close to the level of the recurrent parent. Figure 2 KO in the middle).

[0028] (4) Grain length-to-width ratio: Significant differentiation among populations. The proportion of overparents (>3.22) in the 21J2E2115 population reached 82.28%, while that in the 21J2E2111 and 21J2E2112 populations was only 22.67% and 9.09%, respectively, reflecting the differential segregation of indica and japonica grain type genes. Figure 2 (China PT).

[0029] (5) Plant height: The range of variation was 85–145 cm. Short plants (<93.63 cm, i.e., the 95% confidence limit of the recurrent parent) had a low frequency in all populations (0%–5.06%), and were mostly concentrated in the 90–100 cm range. Figure 2 (China UY).

[0030] Multi-trait aggregation analysis: Using four traits (spikelet fertility ≥87.56%, number of grains per spike >150.44, effective spikes >11.37, grain length-width ratio >3.22) as screening criteria, 8 plants (2.19%) with excellent performance in all four traits were obtained from 366 plants, all from populations 21J2E2114 and 21J2E2115. No plants meeting all five traits (including plant height <93.63cm) were found in this population, indicating that short stature and high yield / quality traits are still independently segregated in the BC2F2 generation and aggregation needs to be achieved through a larger population or MAS. Figure 3In practical breeding, it is recommended to adopt a "stepwise aggregation" strategy: first, select individual plants with normal spikelet fertility and 3-4 excellent target traits from the BC2F2 population, and then break the linkage burden through subsequent self-pollination generations or MAS to achieve the aggregation of multiple traits.

[0031] S5. Stable purification: The BC2F2 single plants selected from S4 were continuously self-crossed for 3-4 generations using the pedigree method until the agronomic traits were uniform, thus developing a stable line.

[0032] This embodiment clearly defines the backcross generation node and fertility threshold system. In the prior art, the selection of backcross generations is blind and lacks quantitative guidance. This invention systematically proposes four key decision nodes for the first time, each corresponding to a clearly defined fertility threshold range: Node 1 (F1 generation), fertility <60% (hybrid sterility bottleneck period); Node 2 (BC1F1 generation), fertility 50%–70% (continued backcrossing period); Node 3 (BC2F1 generation), fertility 70%–75% (fertility breakthrough period, MAS can be initiated); Node 4 (BC2F2 generation), screening for single plants with fertility ≥80% (self-pollination selection initiation period). The standardized strategy of "two backcrosses + one self-pollination" (BC2F2) differs from the traditional practice of multiple backcrosses (such as BC3 and above) or immediate self-pollination after a single backcross. This invention explicitly limits the self-pollination of BC2F1 to obtain BC2F2, and simultaneously achieves fertility recovery and genetic variation preservation in this generation. The technical balance of simultaneously achieving fertility recovery and genetic variation preservation is crucial. Existing technologies often suffer from trade-offs, with premature self-pollination leading to incomplete fertility recovery and delayed self-pollination resulting in loss of genetic diversity. This invention achieves fertility ≥80% in the BC2F2 generation while maintaining a grain number variation coefficient of 18%–25%, with an over-parent ratio of 50%–100%. The timing of MAS intervention for hybrid sterility loci is clearly defined, limited to the BC2F1 / BC2F2 generation (fertility breakthrough period) combined with marker-assisted selection. Homozygous genotypes of recurrent parents are screened for hybrid sterility loci such as S5, Sa, Sb, Sc, Sd, and Se / pf12 / RHS12. Quantitative agronomic trait screening standards provide specific and operable screening thresholds (based on recurrent parental statistics): spikelet fertility ≥87.56%, grain number >150.44, effective spike number >11.37, grain length-to-width ratio >3.22, and plant height <93.63 cm. Its applicability transcends interspecific or subspecific hybridization, possessing universality, whereas existing technologies are mostly limited to single types.

[0033] This embodiment demonstrates significant fertility recovery: F1 generation fertility was approximately 48% → BC2F2 generation average fertility was 81.77%, with over 60% of individual plant spikelets recovering fertility to over 80%, reaching a maximum of 94.8%; rich genetic variation was preserved: abundant variation in grain number per spike, with a coefficient of variation of 18%–25%, with over 80% of most populations exceeding the parental number of grains per spike, and some populations reaching 100%; the proportion of effective spikes exceeding the parental number in some populations reached 55.70%; significant differentiation in grain length-width ratio among populations, reflecting the segregation of indica and japonica grain type genes; plant height, effective spikes, and other traits still maintained significant segregation, providing a material basis for multi-trait aggregation; clear breeding nodes and standardized operation: four decision nodes combined with specific fertility thresholds solved the technical problem of "when to backcross, when to selfcross, and when to select," avoiding the blindness and experience dependence of traditional backcross breeding, and significantly improving the efficiency and quality of distant hybridization breeding; clear potential for multi-trait aggregation: In a large BC2F2 population (366 plants), 2.19% of plants exhibited superior performance in all four traits, demonstrating the feasibility of multi-trait aggregation and providing a clear technical path for achieving aggregation of short-stalked, high-yielding, and high-quality traits through larger populations or MAS. The invention has broad applicability and cross-species validation: it was validated in interspecific hybridization between indica and japonica rice (Yunnan Japonica 1 × Minghui 63) and also in interspecific hybridization between Asian and African rice (IR64 × CG14). The invention provides synergistic validation through theory and practice: it explains the genetic mechanism underlying the restoration of fertility in hybrid sterility genes (gradual homozygosity of hybrid sterility loci) and the rich genetic variation and segregation beyond parents achieved through self-pollination of beneficial alleles from different agronomical trait loci of both parents. Furthermore, it provides a reproducible standardized process for breeding practice, quantitatively demonstrating the superiority of this invention.

[0034] The fertility recovery effect of this embodiment is significant: the present invention can gradually increase the fertility of interspecific hybrid offspring from about 48% in the F1 generation to about 82% in the BC2F2 generation through the strategy of "two backcrosses + one self-cross". Among them, more than 60% of the individual plants have a fertility of ≥80%, which is close to the normal level, effectively overcoming the obstacle of hybrid sterility.

[0035] This embodiment exhibits abundant genetic variation: In the BC2F2 generation, this invention ensures the restoration of spikelet fertility while simultaneously preserving rich genetic variation. Taking indica-japonica hybridization as an example, the coefficient of variation for the number of grains per spike in the BC2F2 population reaches 18%–25%, with most populations having a higher proportion of grains per spike exceeding the parent's number than 85%, and some populations reaching 100%. Traits such as the number of effective spikes and the length-to-width ratio of grains show significant differences among populations, providing ample genetic material for the aggregation of traits such as high yield, high quality, and stress resistance.

[0036] This embodiment clearly defines the breeding nodes and is highly operable: the invention clarifies four key decision nodes and their corresponding fertility threshold ranges (F1 generation <60%, BC1F1 generation 50%~70%, BC2F1 generation 70%~75%, BC2F2 generation screening ≥80%), solving the technical problem of "when to select" in backcross breeding and avoiding the blindness of breeding work.

[0037] Example 2

[0038] This example illustrates interspecific hybridization breeding between Asian and African cultivated rice (O. sativa × O. glaberrima). Materials: The maternal parent was the African cultivated rice germplasm Tog5681 (O. glaberrima). Although Tog5681 had poor agronomic traits (few effective panicles, low number of grains per panicle, easy lodging, and easy grain shattering) after long-term domestication, it had good tolerance to a variety of abiotic and biotic stresses (such as drought, extreme temperature, rice blast, etc.).

[0039] The male parent is the Asian cultivated rice variety IR64 (O. sativa, recurrent parent). IR64 is a high-yielding and widely adaptable indica rice variety developed by the International Rice Research Institute (IRRI). It is widely planted in irrigated rice areas of Africa, but it has poor disease resistance and stress resistance.

[0040] The rice breeding method in this embodiment that overcomes interspecific hybrid sterility in Osage while preserving genetic variation is as follows: S1, evaluation of hybridization and F1 generation: (Decision node 1, bottleneck period of hybrid sterility) Tog5681 was used as the female parent and IR64 as the male parent to obtain F1 generation seeds. The F1 generation seeds were planted to obtain F1 generation plants. The spikelet fertility of the F1 generation plants was investigated. The results showed that the spikelet fertility was 0%. At this stage, it was confirmed that there was a serious hybrid sterility phenomenon among interspecific hybrids. Fertility selection was not carried out. The combining ability of disease resistance and stress resistance traits was evaluated. S2, First submission yields BC1F1: (Decision node 2, first submission evaluation period) Using F1 plants obtained from S1 as the female parent and recurrent parent IR64 from S1 as the male parent, backcrossing was performed to obtain BC1F1 seeds. A population of these BC1F1 seeds was then planted to obtain BC1F1 plants. The spikelet fertility of the BC1F1 plants was investigated. The average spikelet fertility of the BC1F1 plant population was 15%, with a variation range of 0%–40%. Some BC1F1 plants showed better agronomic performance than the recurrent parent IR64, indicating richer variation after one backcross. However, the overall fertility of the population remained at a low level, indicating the need for continued backcrossing. No individual plant selection was performed in this generation. S3, Second backcross yields BC2F1: Plants with good overall traits and spikelet fertility restored to over 50% in the BC1F1 plant population obtained from S2 were used as the female parent, and backcrossed with the recurrent parent IR64 from S1 as the male parent to obtain BC2F1 seeds. A population of BC2F1 plants was obtained by planting these BC2F1 seeds, and spikelet fertility was investigated. When the average spikelet fertility of the BC2F1 plant population increased to 40%, with a variation range of 5%–70%, and some individual plants achieved fertility of over 60%, and significantly improved compared to the average spikelet fertility of the BC1F1 plant population, the fertility breakthrough period was reached. In conjunction with marker-assisted selection (MAS), single plants with homozygous genotypes of the recurrent parent and good overall traits were screened for self-pollination at major hybrid sterility gene loci such as S1, as used for the next step of self-pollination selection. S4. Self-pollination yields BC2F2 and initiates selection: From the BC2F1 plant population obtained from S3, several independent plants with good spikelet fertility were randomly selected for self-pollination to obtain seeds of the BC2F2 segregating population. A large population was constructed, and after planting, the BC2F2 segregating population was obtained. The spikelet fertility and agronomic traits (spikelet fertility, number of grains per spike, number of effective spikes, and number of secondary branches) of the BC2F2 segregating population were investigated, and BC2F2 plants with spikelet fertility ≥80% and excellent agronomic traits were screened. (1) Spikelet fertility: The spikelet fertility variation ranged from 0% to 90% in the BC2F2 segregating population, showing a continuous distribution. The proportion of fertile (>70% fertility) plants varied significantly among different families, ranging from 30% to 100%. Some individual plants had spikelet fertility exceeding 80%, reaching a level comparable to IR64 (IR64 had spikelet fertility ≥80% in both early and late rice). This indicates that through two backcrosses followed by self-pollination, individuals with fully restored fertility could be screened in the BC2F2 generation.

[0041] (2) Number of grains per ear: The number of grains per ear varies greatly in the BC2F2 segregating population. Some individual plants have 10% to 30% more grains per ear than the recurrent parent IR64. Among them, one line has an average number of 168 grains per ear, while IR64 has 108 grains per ear.

[0042] (3) Number of effective panicles: The number of effective panicles in the BC2F2 segregating population showed rich variation. Tog5681 had a lower number of effective panicles (early rice did not head due to photosensitivity, and late rice had about 5 to 8 panicles per plant), while IR64 had a higher number of effective panicles (about 10 to 15 panicles per plant). Some individual plants in the BC2F2 population had more effective panicles than IR64.

[0043] (4) Number of secondary branches: The number of secondary branches is a key factor determining the number of grains per ear. Tog5681 has few secondary branches, while IR64 has a moderate number. In the BC2F2 segregating population, some individual plants have significantly more secondary branches than their parents, with the line WAS131 being the most prominent, reaching 25-30 per ear. This significant segregation beyond the parents' line provides new genetic resources for high-yield breeding.

[0044] In summary, the screening threshold criteria for excellent agronomic traits in this embodiment are: spikelet fertility ≥80%, number of grains per spike >130, number of effective spikes >13, and number of secondary branches >20.

[0045] The experiment verified that the "two backcrosses + one self-cross" (BC2F2) strategy in interspecific hybridization can yield a breeding population with normal spikelet fertility (≥80%) and rich agronomic trait variation in the BC2F2 generation. The key traits of the BC2F2 segregating population, such as the number of grains per spike, the number of effective spikes, and the number of secondary branches, all showed extensive segregation beyond the parents, providing sufficient genetic material for subsequent breeding selection.

[0046] S5. The BC2F2 single plants selected in S4 are self-crossed for 3-4 generations using the pedigree method or selected for multiple generations using the single seed method to develop stable lines.

[0047] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any way. Any simple modifications, alterations, and equivalent changes made to the above embodiments based on the inventive essence shall still fall within the protection scope of the present invention.

Claims

1. A rice breeding method that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, characterized in that, The method is as follows: S1, crossbreeding and F1 generation evaluation: Crossing was performed using the first parent as the female parent and the second parent as the male parent to obtain F1 generation seeds; the F1 generation seeds were planted to obtain F1 generation plants, and the spikelet fertility of the F1 generation plants was investigated. Spikelet fertility <40% was considered high sterility, 40% ≤ spikelet fertility <60% was considered semi-sterile, 60% ≤ spikelet fertility <80% was considered moderately fertile, and spikelet fertility ≥80% was considered normal. S2, First backcross yields BC1F1: Using the F1 generation plants obtained from S1 as the female parent and the first parent from S1 as the recurrent parent, backcrossing was performed to obtain BC1F1 seeds; A population of BC1F1 seeds was planted to obtain a BC1F1 plant population, and the spikelet fertility of the BC1F1 plant population was investigated. S3, Second backcross yields BC2F1: The plants with good overall traits and spikelet fertility restored to more than 50% in the BC1F1 plant population obtained in S2 were used as the female parent, and the first parent in S1 was used as the recurrent parent for backcrossing to obtain BC2F1 seeds. The population planted with the BC2F1 seeds was used to obtain a BC2F1 plant population, and the spikelet fertility of the BC2F1 plant population was investigated. S4. Self-pollination yields BC2F2 and selection is initiated: Multiple independent plants were randomly selected from the BC2F1 plant population obtained from S3 for self-pollination to obtain seeds of the BC2F2 segregating population. After planting, the BC2F2 segregating population was obtained. The spikelet fertility and agronomic traits of the BC2F2 segregating population were investigated, and BC2F2 plants with spikelet fertility ≥80% and excellent agronomic traits were screened. S5. Stable purification: The BC2F2 single plants obtained from S4 were selected for multiple generations using the pedigree method or single seed propagation method to develop stable lines.

2. The rice breeding method according to claim 1, which overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, is characterized in that... In S1, the combination of the first parent and the second parent is either a combination of indica rice and japonica rice subspecies or an interspecific combination of Asian cultivated rice and African cultivated rice.

3. A rice breeding method according to claim 2 that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, characterized in that, In the combination of indica and japonica rice subspecies, the maternal parent is the temperate japonica rice Dianjingyou 1, and the paternal parent is the indica rice Minghui 63; in the combination of Asian cultivated rice and African cultivated rice, the maternal parent is the African cultivated rice germplasm Tog5681, and the paternal parent is the Asian cultivated rice variety IR64.

4. A rice breeding method according to claim 3 that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, characterized in that, When the maternal parent is the temperate japonica rice Dianjingyou 1 and the paternal parent is the indica rice Minghui 63, in S1, the spikelet fertility of the F1 generation plants is confirmed to be semi-sterile, with a spikelet fertility of 48.26%±9.15%; in S2, the spikelet fertility of the BC1F1 plant population is 58.63%±8.41%; and in S3, the spikelet fertility of the BC2F1 plant population is 73.08%±6.58%, which is higher than the average spikelet fertility of the BC1F1 plant population.

5. A rice breeding method according to claim 3 that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, characterized in that, When the maternal parent is the African cultivated rice germplasm Tog5681 and the paternal parent is the Asian cultivated rice variety IR64, in S1, the spikelet fertility of the F1 generation plants is confirmed to be 0%; in S2, the average spikelet fertility of the BC1F1 plant population is 15%, with a variation range of 0% to 40%; in S3, the average spikelet fertility of the BC2F1 plant population is 40%, with a variation range of 5% to 70%, and is higher than the average spikelet fertility of the BC1F1 plant population.

6. A rice breeding method according to claim 1 that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, characterized in that, The screening of agronomic traits described in S4 includes spikelet fertility, number of grains per spike, and number of effective spikes.

7. A rice breeding method according to claim 3 that overcomes interspecific and intersubspecific hybrid sterility in rice while preserving genetic variation, characterized in that, When the maternal parent is temperate japonica rice Dianjingyou 1 and the paternal parent is indica rice Minghui 63, the screening threshold standards for excellent agronomic traits are: spikelet fertility > 88.15%, number of grains per spike > 150.44, number of effective spikes > 11.37, grain length-width ratio > 3.22, and plant height < 93.63 cm.

8. A rice breeding method according to claim 3 that overcomes interspecific and intersubspecific hybrid sterility while preserving genetic variation, characterized in that, When the maternal parent is the African cultivated rice germplasm Tog5681 and the paternal parent is the Asian cultivated rice variety IR64, the screening threshold criteria for excellent agronomic traits are: spikelet fertility ≥80%, number of grains per spike >130, number of effective spikes >13, and number of secondary branches >20.

Images