CRISPR/Cas9 Editing of the OsLOX3 Gene Enhances Rice Grain Weight and Seed Vigor

Abstract

Rice lipoxygenase 3 (OsLOX3) is a lipid hydroperoxidase found in rice embryos. Previous studies have reported that OsLOX3 is associated with seed quality and stress resistance, however, its relationship with grain shape and weight remains unknown. In this study, the first exon of OsLOX3 gene was edited in the indica rice variety GDR998 using CRISPR/Cas9 technology. Two homozygous mutants, Oslox3-1 (single-base deletion) and Oslox3-2 (single-base insertion) were identified among eight positive mutant plants from the T2 generation. The agronomic evaluation of genotypic OsLOX3 mutants showed significant increase in grain length, grain length-to-width ratio, 1000-grain weight, plant height, panicle length, and yield per plant compared with the wild type GDR998. The number of effective panicles and total grains per panicle did not significantly change. Further germination tests of seeds after three years of natural aging revealed that, compared with the control GDR998, the germination percentages of the mutants Oslox3-1 and Oslox3-2 increased significantly by 41.1% and 45.6%, respectively. These findings indicate that the knockout of OsLOX3 simultaneously improve grain weight and seed vigor, providing valuable germplasm resources for rice breeding targeting high-yield, improved seed longevity and rice quality.

1. Introduction

Rice (Oryza sativa L.) is one of the most important food crops worldwide, and increasing the rice yield per unit area is critical for ensuring food security. Rice yield is primarily determined by three components: number of grains per panicle, the number of effective panicles, and grain weight [1]. Among these, grain weight, which is closely influenced by grain length and width, constitutes a key trait targeted in molecular design breeding for high yield and quality. Investigating the genetic and molecular mechanisms controlling grain size can provide valuable insights for improving rice productivity and grain quality [2,3].

Grain shape in rice is a quantitative trait regulated by multiple genetic systems involving the embryo, endosperm, and maternal plant [4,5]. Several major signaling pathways are known to regulate grain size, including G protein signaling, the ubiquitin–proteasome pathway, mitogen-activated protein kinase (MAPK) cascades, phytohormone pathways, and transcriptional regulatory networks [6]. For example, GS3 is a well characterized negative regulator of grain length and weight, and represents a major quantitative trait locus (QTL) differentiating indica and japonica subspecies [7]. Similarly, DENSE AND ERECT PANICLE1 (DEP1) influences panicle architecture and grain size through G-protein signaling [8,9]. Within the MAPK pathway, OsMKK4/SMG1 promotes grain growth by enhancing cell proliferation in the spikelet hull [10], while OsMAPK6 positively regulates grain size through brassinosteroid (BR) mediated signaling [11]. The ubiquitination pathway also plays critical roles; GW2 encoding an E3 ubiquitin ligase, negatively regulates grain width by inhibiting cell division [12,13,14]. Phytohormones, including brassinosteroids (BRs), auxins (IAAs), and cytokinins (CTKs), further modulate grain development. For instance, BR biosynthesis genes such as BRASSINOSTEROID-DEFICIENT DWARF3 (BRD3) enhance grain yield, [15]. while OsARF6 (an auxin response factor) [16,17] and OsCKX1 (a cytokinin oxidase) act as negative regulators of grain size [18,19].

As the carrier of genetic information, seed quality profoundly influences crop establishment, growth, and ultimate yield. In southern regions, high temperature and humidity accelerate seed aging, often reducing germination rates to below 70% within a year compromising commercial value and yield stability. Thus, understanding the heredity and molecular mechanisms of seed longevity has become a central issue in seed biology. In the 1990s, Japanese researchers identified a rice variety named Daw Dam, which showed exceptional storability due to deficiency in lipoxygenase isoenzyme Lox-3. This trait was shown to be inherited recessively [20] Collaborations between Nanjing Agricultural University and Japanese scientists introduced the Lox-3 deficiency into the cultivar Koshihikari, generating the storage-tolerant line W017. The absence of lipoxygenase activity is believed to improve storability by reducing lipid peroxidation a key driver of seed deterioration [21].

With advances in gene editing, CRISPR/Cas9 technology has enabled the efficient development of improved germplasm. For example, Su et al. generated LOX3 knockout lines via Agrobacterium-mediated transformation and observed significantly improved germination vigor. However, this improvement came at the cost of reduced disease resistance, which was partially restored by exogenous jasmonic acid application. Their study did not fully investigate seed longevity or grain morphological traits [22].

Notably, no previous study had comprehensively examined whether OsLOX3 knockout simultaneously enhances seed vigor and grain morphology. Our study aimed to address this gap by using CRISPR/Cas9 to generate OsLOX3 knockout lines and evaluating their impacts on both seed storability and grain traits. We demonstrate that OsLOX3 mutation not only improves seed longevity likely through reduced lipid peroxidation, but also increases grain dimensions and weight, providing valuable germplasm for breeding high-yield, high-quality rice with improved storage tolerance.

2. Materials and Methods

2.1. Experimental Materials

The indica rice variety GDR998 was used as the recipient material for transformation with the recombinant vector. Genetic transformation of the recombinant pYL CRISPR/Cas9-LOX3 expression vector was performed using Agrobacterium-mediated transformation.

The transgenic rice materials were cultivated at the Baiyun Experimental Base of the Guangdong Academy of Agricultural Sciences (113°26′22″ E, 23°23′15″ N). Conventional field management practices adapted to the local temperature and light conditions were implemented. Completely randomized design was used.

The rice transformation recipient material GDR998 is preserved by the Agricultural Biotechnology Research Center of the Guangdong Academy of Agricultural Sciences, and the gene editing vector was provided by Dr. Liu Qinjian.

2.2. Construction of CRISPR/Cas9 Editing Vector and Generation of Transgenic Plants

The specificity of gene editing depends on the specificity of the gRNA. The essential PAM sequence for gene editing was selected, and target sites were designed using the online CRISPR target site design tool E-CRISPR http://www.e-crisp.org/ (accessed on 21 December 2019), for the construction of a single-target knockout vector. The primers for target site design are listed in

Table 1. The primer sequences used in this study.

Usage	Primer Name	Primer Sequence (5′-3′)
		Forward	Reverse
Vector constructs	OsLOX3-cas9	cagGCTCCTCGACAACGTCCATG	aacCATGGACGTTGTCGAGGAGC
Hygromycin detection	hpt-t	GATGTTGGCGACCTCGTATTGG	CGTGCTTTCAGCTTCGATGTAGGAG
Cas9 detection	Cas9	CATCCAGAAAGCCCAGGTGT	GTTCCTGGTCCACGTACATA
Target sequencing	OsLOX3 T0	AGTTCTTCCCCATCCATTG	CCTTGATTCTTTCTACATAGCA

, and were synthesized by Tsingke Biotechnology (Guangzhou, China) Co., Ltd. The constructed vector was sent to Biorun Biosciences Co., Ltd. (Wuhan, China). for genetic transformation, and ultimately, T0 generation OsLOX3 knockout transgenic plants were obtained.

2.3. Detection of Transgenic-Positive Rice Lines and Screening of Mutant Plants

PCR amplification identification was performed on knockout transgenic plants. Total DNA was extracted from leaves of T0 generation transgenic rice plants using the CTAB method. 2 × Taq Master Mix and hygromycin detection primers hpt-t/F and hpt-t/R were used, with their sequences shown in Table 1. PCR amplification was conducted; the amplification products were subjected to 1% agarose gel electrophoresis The amplification results corresponding to each plant were observed to determine whether a band of approximately 600 bp was produced. If the target band was present, it indicated that the knockout vector had been integrated into the chromosome of the corresponding plant template DNA. Mutation detection primers OsLOX3 T0-F/OsLOX3 T0-R were designed using Primer Premier 5 at the upstream and downstream regions of the target site sequence. PCR reactions were performed using 2 × Gold Mix (green) and primers OsLOX3 T0-F/OsLOX3 T0-R according to the following system—25 μL PCR reaction system: 0.5 μL DNA template, 22.5 μL 2 × Gold Mix, 1 μL OsLOX3 T0-F, 1 μL OsLOX3 T0-R. PCR products ware ligated into the T19 simple vector, and transformed into Escherichia coli DH5α. Ten independent monoclonal colonies were selected and sent to Tsingke Biotechnology (Shanghai, China) Co., Ltd. for duplex sequencing. The sequencing results of mutants and wild type (WT) were aligned using SnapGene software 3.2.1.

For T1 or T2 single plants of mutant lines, PCR amplification was performed using primers hpt-F/hpt-R and Cas9-F/Cas9-R. Plants in which neither pair of primers amplified the target fragment of the vector were identified as mutant lines without transgenic sequences. Simultaneously, we detected the mutation status of T1 or T2 plants using target site amplification primers.

2.4. Detection of mRNA Expression Levels in Plants

Seeds from T1 generation homozygous lines lacking transgenic elements were harvested and used to cultivate T2 generation plants. Homozygous mutant and WT seeds were germinated in an incubator at 25–28 °C under a 14 h light/10 h dark photoperiod for 14 days. Young leaves were collected for RNA extraction. The RNA A260/280 ratio ranged from 1.8 to 2.1. Quantitative real-time PCR (qRT-PCR) was conducted with three biological and technical replicates to evaluate the expression level of the OsLOX3 gene in both mutant and WT plants. The qRT-PCR reaction system and conditions followed the instructions provided with the SYBR^® Premix Ex Taq™ II (Tli RNaseH Plus) kit from TaKaRa (catalog number RR820Q, Dalian, China). The rice Ubiquitin gene, which maintains stable expression under various treatments, was used as an internal reference. The primers used for OsLOX3 gene amplification were as follows: forward primer: 5′-TCCTGCACCAAGATTGCCTT-3′; reverse primer: 5′-ATGTGGGCTTGGGTCCATTT-3′. Primers were designed for quantitative analysis. Quantitative PCR was carried out using a LightCycler 480 instrument (Roche, Mannheim, Germany). The thermal cycling conditions were as follows: 95 °C for 30 s, followed by 45 cycles of 95 °C for 5 s and 68 °C for 30 s. After completion of the qRT-PCR reactions, relative expression levels were analyzed using the 2^−ΔΔCt method, with the WT used as the reference for normalization.

2.5. Natural Aging Treatment and Seed Germination Experiment

WT, Oslox-1, and Oslox-2 rice seeds were placed in a dry environment at an average room temperature of 25 °C for natural aging over a period of 3 years, followed by germination experiments. Clean disposable Petri dishes and germination paper were prepared in advance. Uniform and plump seeds from the WT, Oslox-1, and Oslox-2 lines were selected. The aged seeds were placed in 9 cm diameter Petri dishes lined with two layers of germination paper, with 30 seeds per dish, and 10 mL of distilled water was added, with no additional water supplied during cultivation. The dishes were randomly placed in an incubator for 7 days for germination at a temperature of 25–28 °C, under a 14 h light/10 h dark photoperiod. Seed germination was defined as the emergence of the radicle breaking through the seed coat (visible white tip). The number of germinated seeds was recorded daily. Three biological replicates were performed for each treatment. After germination, 10 seedlings were randomly selected from each replicate to measure seedling length, root length, fresh weight, and dry weight. Germination percentage, seeding percentage, germination potential, germination index, and vigor index were calculated.

Germination percentage (%) = number of germinated seeds/number of tested seeds × 100

Seeding percentage (%) = number of normal seedlings/number of tested seeds × 100

Germination Potential (%) = number of germinated seeds (3d)/number of tested seeds × 100

Germination Index (GI) = ∑Gt/Dt (Gt represents the number of seeds germinated per day, Dt represents the number of days for seed germination, ∑ indicates the total sum)

Vigor Index (VI) = GI × S (S represents the average root length)

Percentage data were arcsine-transformed prior to analysis. Data were tested for normality and homogeneity of variance, and the results confirmed that the data met the assumptions of normal distribution and equal variance. Statistical analyses, including difference and correlation analyses, were conducted using IBM SPSS Statistics 27.0. One-way analysis of variance (ANOVA) was used to evaluate differences among groups, followed by post hoc multiple comparisons using LSD, Tukey’s HSD, and Waller–Duncan tests. The significance level for post hoc tests was set at p < 0.05.

2.6. Agronomic Trait Investigation of Mutant Plants

Based on laboratory phenotypic analysis of rice seeds, the phenotypes of seeds from both recipient and gene-edited materials were analyzed using a multispectral seed vitality analyzer (Videometer Lab 4) (Videometer A/S Company, Copenhagen, Denmark; purchased from Beijing Biopute Technology Co., Ltd., Beijing, China). Measurements were taken for average grain length (AGL), average grain width (AGW), and grain length-to-width ratio in mature and intact seeds. For each line, 300 grains were measured for both grain length and grain width.

In the field, statistical analyses were performed on plant height, number of effective panicles, panicle length, number of primary branches, number of secondary branches, number of grains per panicle, thousand-grain weight, and theoretical yield per plant for 20 mature-stage plants of both cultivated WT and CRISPR/Cas9 gene-edited lines. Thousand-grain weight was measured in five replicates, with 1000 grains per replicate. Data were analyzed using IBM SPSS Statistics 27.0 for both difference and correlation analyses. Homogeneity of variance and one-way ANOVA were used to assess differences among groups. Post hoc multiple comparisons were conducted using LSD, Tukey’s HSD, and Waller–Duncan tests, with the significance threshold set at p < 0.05.

3. Results

3.1. OsLOX3 Target Design and CRISPR/Cas9-LOX3 Vector Construction

The rice lipoxygenase gene OsLOX3 (LOC_Os03g49260), retrieved from the RGAP database https://www.ricedata.cn/gene/ (accessed on 17 May 2020), contains 4284 base pairs (bp) and comprises nine exons and eight introns, located on chromosome 3 of rice. The open reading frame (ORF) is 2592 bp, encoding a polypeptide of 864 amino acids, with an isoelectric point of 6.88 and a molecular weight of 97.5 kDa. The target site was designed within the first exon to induce frameshift mutations that effectively disrupt the coding function of the gene. (Figure 1a). The protein domain of the OsLOX3 gene was analyzed using the online protein structure domain identification and annotation tool SMART https://smart.embl.de/ (accessed on 24 May 2022). The results showed that OsLOX3 contains a protein domain LH2 (Figure 1b).

3.2. Obtaining Mutant Plants with Two Different OsLOX3 Genetic Mutation Types

To screen T2 generation OsLOX3 mutant rice plants for homozygous edits we PCR-amplified a 348 bp fragment encompassing the target site and its flanking regions, then performed sequencing analysis. Two types of homozygous alleles were identified: Oslox3-1 exhibited a single-base deletion mutation at the target site, and Oslox3-2 exhibited a single-base insertion mutation at the target site (Figure 1c). Both types of mutations result in a frameshift in the encoded amino acid sequence. PCR with hygromycin-specific primers confirmed that no hygromycin gene residues were detected in the DNA of OsLOX3 mutant plants (Figure 1d). The expression levels of the OsLOX3 gene in mutants and the WT were measured using RT-qPCR. The relative expression levels of the gene-edited lines Oslox3-1 and Oslox3-2 compared to the WT were (0.72 ± 0.03, 0.36 ± 0.10), indicating that the expression of the OsLOX3 mutants was significantly reduced (Figure 1e).

3.3. Seed Vigor of OsLOX3 Mutants Was Significantly Improved

Germination experiments were conducted using seeds of WT and OsLOX3 knockout homozygous mutant lines (Oslox3-1 and Oslox3-2) following natural aging (3 years of storage at room temperature). The results indicated that the mutant lines displayed significantly increased germination percentages (Oslox3-1: 45.56 ± 3.14%; Oslox3-2: 50.00 ± 2.72%) and seedling percentages (Oslox3-1: 36.67 ± 2.72%; Oslox3-2: 45.56 ± 1.57%) compared to the WT (4.44 ± 1.57% and 2.22 ± 1.57%, respectively). Moreover, Oslox3-1 and Oslox3-2 showed markedly enhanced germination potential, germination index, vigor index, seeding length, root length, fresh weight, and dry weight values compared to the WT, demonstrating that the knockout of OsLOX3 significantly improves seed storability and vigor (Figure 2).

3.4. The Grain Length and Grain Length-to-Width Ratio of OsLOX3 Mutants Were Significantly Increased

Grain morphology was evaluated in the T2 generation of OsLOX3 knockout lines and WT plants. Notable changes in both grain length and width were observed in the mutants. Compared to the WT (9.5 ± 0.02 mm), the grain lengths of Oslox3-1 and Oslox3-2 were significantly increased to 10.6 ± 0.01 mm and 10.7 ± 0.01 mm, corresponding to increases of 11.2 ± 0.1% and 11.7 ± 0.4%, respectively (Figure 3a). Moreover, the grain width of the mutants also exhibited significant increases of 3.1 ± 0.7% and 3.9 ± 0.3% compared to WT (Figure 3b). Consequently, the grain length-to-width ratios were elevated by 7.8 ± 0.8% and 7.5 ± 0.2% in Oslox3-1 and Oslox3-2, respectively (Figure 3c). These findings demonstrate that OsLOX3 functions as a negative regulator of grain development in rice.

3.5. The Thousand-Grain Weight and Theoretical Yield of OsLOX3 Mutants Were Significantly Increased

To investigate the effects of the OsLOX3 gene on rice growth and development, key agronomic traits were systematically evaluated in OsLOX3 mutant lines (Oslox3-1 and Oslox3-2), using the WT as a control. These traits included plant height, the effective tiller number, the length of the main panicle, the number of grains per panicle, the primary branch number per panicle, the secondary branch number per panicle, the thousand-grain weight and the theoretical yield per plant. The results demonstrated that, compared to the WT, the mutants exhibited significantly increased plant height, with increments of 5.5% ± 0.3% and 7.3% ± 1.3%, respectively (Figure 4a), and significantly longer panicles, with increases of 11.1% ± 1.5% and 13.5% ± 1.2% (Figure 4c). The number of primary branches and secondary branches increased by 8.2% ± 2.2% to 11.7% ± 3.8% and 2.1% ± 0.6% to 8.5% ± 1.1%, respectively (Figure 4h,i). The thousand-grain weight was elevated by 10.3% ± 0.4% to 17.7% ± 1.2% (Figure 4k), and the theoretical yield per plant was significantly enhanced by 8.4% ± 0.8% to 16.6% ± 0.4% (Figure 4j). No significant differences were observed between the mutants and WT in the number of effective tiller or grains per panicle (Figure 4b,e). These findings indicate that mutation of OsLOX3 significantly regulates plant height, panicle architecture, and yield-related traits in rice.

4. Discussion

With the advancement of breeding technologies, CRISPR/Cas9 enables the precise and efficient generation of homozygous mutants with stable and improved traits, significantly shortening the genetic improvement cycle [23,24]. In this study, we utilized CRISPR/Cas9-mediated single-target editing to mutagenize OsLOX3 and isolated two distinct homozygous mutant types from the T2 generation: a single-base deletion mutant and a single-base insertion mutant. The two mutant alleles by CRISPR editing are caused by the occurrence of transcode mutation, which leads to the premature termination of protein synthesis and loss of function, resulting in the enlargement of seed cells and an increase of thousand-grain weight, which may be the key factor for the significant enhancement of seed vigor and seed longevity. CRISPR/Cas9-edited T0 lines typically produce four genotypic classes: homozygous mutants, heterozygous mutants, biallelic mutants, and non-mutants. Since homozygous mutants are less frequent, careful screening and progeny segregation analysis from heterozygous parents are essential [25].

In this study, sequencing analysis was first performed on the OsLOX3 editing target sites of T1 generation plants derived from twelve T0-generation positive transgenic lines. Eight types of heterozygous mutant plants were screened from the thirty T1 generation, and subsequently, two types of homozygous mutant plants, namely Oslox3-1 and Oslox3-2, were obtained from the T2 generation. In addition, the detection of the hygromycin resistance gene in the two homozygous OsLOX3 mutant plants obtained from the T2 generation revealed no residual hygromycin gene in the mutants, indicating that the mutant plants were free of vector DNA residues. Therefore, they can be used for hybrid breeding with other conventional varieties or for biological function analysis. The use of CRISPR/Cas9 editing technology enables the precise and efficient molecular design breeding, which can reserve high-quality gene resources to ensure high yield, stable yield, and the sustainable development of rice, and is of great significance for increasing rice yield per unit of cultivated land [26,27].

The lipoxygenase (LOX) gene family is widely distributed in plants that regulating growth and development, seed germination, defense responses to biotic and abiotic stresses, wounding, fruit ripening, senescence, cell death, and the biosynthesis of the stress-responsive plant hormones jasmonic acid and abscisic acid [28]. Wang Ren et al. conducted localization and cloning studies on LOX-deficient genes and successfully cloned the encoding genes OsLOX1 and OsLOX3 [29]. Wu Yuejin et al. used RNA interference of the r9-LOX1 gene to obtain transgenic plants, which effectively improved the storage characteristics of rice and delayed the aging and deterioration of rice grains. It was also found that Lox-1 and Lox-2 may be key genes affecting seed viability [30]. Studies by Mou et al. demonstrated that the CRISPR/Cas9 knockout of the lipoxygenase gene OsLOX1 delayed the loss of seed vigor and quality, and that the knockout of OsLOX1 was beneficial for prolonging seed vigor [31]. However, more recent studies have shown that OsLOX1 positively regulates rice seed vigor and drought stress. The knockout of OsLOX1 reduced the longevity of rice seeds, increased levels of H₂O₂ and MDA, and decreased the activities of antioxidant enzymes superoxide dismutase and catalase compared to the wild type in a study [32]. Research by Wang et al. indicated that the knockout of the LOX10 mutant extended seed longevity while the overexpression of LOX10 enhanced the tolerance of rice seedlings to saline–alkaline stress [33].

Other studies have shown that the deletion of the LOX3 gene has been proven to improve rice seed quality but reduces resistance to rice epidemic disease and drought [34,35]. To investigate these opposing effects, Su et al. introduced the LOX3 knockout gene construct into rice using Agrobacterium tumefaciens-mediated transformation. The results showed that the germination percentage of the LOX3 knockout gene construct was faster than that of the WT. There were no significant differences in plant height, the number of tillers, and 100-grain weight among the three knockout rice lines compared with the WT, indicating that knockout of LOX3 accelerated rice seed germination without affecting plant height, number of tillers, or 100-grain weight [22]. This is partially contrasts with our results likely, maybe mainly due to differences in genetic background (their japonica vs. our indica), mutation type, and environmental conditions; there may also be other factors that need further study.

In this study, a preliminary investigation was conducted on the agronomic traits of two homozygous OsLOX3 mutant plants. It not only confirmed that the knockout of LOX3 can improve seed vigor, but also revealed, for the first time that OsLOX3 loss of function also significantly improves grain dimensions, grain weight, and overall yield, without compromising panicle number. This suggests a previously unrecognized role for OsLOX3 in regulating grain development and filling, highlighting its dual impact on both quality and yield traits. The results indicated that the OsLOX3 mutation caused an increase in the grain length-to-width ratio, appearance quality and yield. It is speculated that OsLOX3 may play a role in controlling the formation of rice grain cells and promoting grain filling.

As Lipoxygenases lipoxygenases (LOXs) are involved in jasmonic acid (JA) and abscisic acid (ABA) biosynthesis, ROS homeostasis, and cell wall modification [28], we propose that impaired JA/ABA signaling and altered ROS scavenging in the OsLOX3 mutants may lead to enhanced grain filling and cell expansion possibly through cell wall loosening. Moreover, the improved seed longevity and vigor observed here align with the role of LOXs in aging and oxidative stress [36].

These findings position OsLOX3 as a promising candidate for developing high-yield, climate-resilient rice varieties. By leveraging CRISPR-based editing, OsLOX3 alleles can be rapidly introgressed into elite lines adapted to diverse agroecological zones, supporting the sustainable intensification of rice production under changing climatic conditions. The mutants obtained here provide valuable genetic resources and theoretical support for breeding high-yield, high-quality rice varieties. However, although OsLOX3 knockout enhanced yield and seed quality, potential trade-offs concerning stress resilience, such as reduced resistance to pathogens or drought as reported previously [34,35], must be systematically evaluated before deployment in breeding programs. Furthermore, as grain shape is a quantitative trait governed by embryo, endosperm, and maternal interactions [4,5], the molecular mechanisms underlying OsLOX3 function require further investigation.

5. Conclusions

Using CRISPR/Cas9 editing technology, two distinct genotypes of homozygous OsLOX3 mutant plants were obtained. After three years of natural aging, the germination percentage of the OsLOX3 mutant plants was significantly increased. Additionally, the grain length, grain length-to-width ratio, and thousand-grain weight of the mutants were significantly improved. Compared to the wild type, the plant height, panicle length, number of primary branches, and theoretical yield per plant of the mutant plants were significantly increased. This was the first study to demonstrate that OsLOX3 knockout can simultaneously enhance both grain yield-related traits and seed longevity in rice. These improvements offer significant potential for breeding programs aimed at developing high-yield rice varieties with extended storage stability, which is crucial for reducing post-harvest losses and ensuring food security, especially in regions prone to high temperature and humidity.