OsRAV1 Regulates Seed Vigor and Salt Tolerance During Germination in Rice

Abstract

Seed vigor is a complex trait encompassing seed germination, seedling emergence, growth, seed longevity, and stress tolerance, all are crucial for direct seeding in rice. Here, we report that the AP2/ERF transcription factor OsRAV1 (RELATED TO ABI3 AND VP1) positively regulates seed germination, vigor, and salt tolerance. Additionally, OsRAV1 was differently expressed in embryo and endosperm, with the OsRAV1 localized in the nucleus. Transcriptomic analysis revealed that OsRAV1 modulates seed vigor through plant hormone signal transduction and phenylpropanoid biosynthesis during germination. Haplotype analysis showed that rice varieties carrying Hap3 displayed enhanced salt tolerance during seed germination. These findings suggest that OsRAV1 is a potential target in breeding rice varieties with high seed vigor suitable for direct seeding cultivation.

Background

Rice (Oryza sativa) serves as a staple food for billions of people worldwide, highlighting the critical importance of producing and preserving rice seeds for food security. Seed vigor, encompassing traits like seed germination, seedling emergence and establishment, growth, seed longevity, and stress tolerance, is essential for successful direct seeding in rice (Zhao et al. 2021; Li et al. 2023). High-vigor seeds consistently exhibit rapid and uniform germination, robust seedling establishment, and resilience under stress conditions in agricultural production (He et al. 2019, 2020a; Zhao et al. 2021). However, seed vigor declines over time during storage, posing challenges to food production and genetic integrity (Yin et al. 2017).

Seed longevity, or seed aging, refers to a seed’s ability to remain viable in a dormant, dry state during prolonged storage. Factors such as high temperature and humidity during storage can diminish seed vitality, leading to reduced germination rates, abnormal seedling growth, and decreased tolerance to stress. Methods to study seed longevity include natural aging, where seeds deteriorate under ambient storage conditions, and artificially accelerated aging, involving controlled exposure to high temperature (around 42℃) and humidity (approximately 80%) over a specified duration (Zhou et al. 2024; Hay et al. 2018). Variability in seed responses among rice varieties underscores the need for refined methods in artificial aging studies; for instance, varieties like Kasalath can remain a 80% germination rate even after 18 days of artificial aging (Lee et al. 2019; Wang et al. 2022).

In field conditions, seed germination faces challenges from abiotic stresses such as drought, low temperature, and salinity, which particularly hinder rice seedling establishment (Zhao et al. 2021). Among these stresses, salt stress poses persistent obstacles, significantly impairing both seed germination and seedling’s growth (He et al. 2019). Thus, enhancing seed germination and seedling establishment under adverse conditions is a priority in rice breeding. Understanding the physiological and molecular mechanisms underlying abiotic stresses in seed vigor is crucial for safeguarding genetic resources, ensuring food security, and promoting consistent seedling establishment in simplified cultivation systems.

Recent researches have elucidated relationships between seed germination, longevity, and salt tolerance. For instance, precise regulation of gene expression levels, such as OsLOX2, ZmDREB2A, and ZmPIMT2, can balance regulation germination rate and seed longevity (Huang et al. 2014; Han et al. 2020; Zhang et al. 2023c). Moreover, overexpression of OsGLYI3 and ZmGOLS2 enhances both germination rate and salt tolerance during germination (Gu et al. 2016; Liu et al. 2022), whereas overexpression of OsLOX10 and ZmAGA1 may improve salt tolerance but compromise seed longevity (Zhang et al. 2021; Wang et al. 2023). Strategic use of allelic variations in gene promoters or coding regions, such as OsNAC3 and OsPK5, can modulate transcription levels or protein activity to optimize seed germination, longevity, and salt tolerance (Aung et al. 2023; Huang et al. 2024; Yamasani et al. 2023; Yang et al. 2022; G. Zhang et al. 2023).

APETALA2/ethylene response factors (AP2/ERFs) comprise a large transcription factors super-family in plants, characterized by one or two AP2/ERF domains crucial for DNA binding (Nakano et al. 2006; Sharoni et al. 2011). RNA-seq analyses highlight the involvement of AP2-domain-containing regulators in seed germination in rice (He et al. 2020b). Within this regulatory network, OsRAV1 interacts notably with OsbZIP23, known for its positively regulation of seed vigor (Wang et al. 2022). Previous studies suggest that OsRAV1 may play a role in salt tolerance regulation based on its induction in salt-tolerant varieties (Liu et al. 2024), implicating OsRAV1 in governing seed vigor from germination through early seedling stages.

In this study, we explore the regulator role of OsRAV1 in seed vigor. Overexpression of OsRAV1 significantly enhances seed vigor, germination, and salt tolerance, whereas disruption of OsRAV1 reduces seed vigor. RNA-Seq data indicate that OsRAV1 may modulate seed vigor through plant hormone signal transduction and phenylpropanoid biosynthesis pathways. Moreover, an elite haplotype of OsRAV1 found in the Geng accession version shows promise for seed germination improvement under salt stress, suggesting OsRAV1 as a potential target for genetic improvement of seed germination, vigor, and salt tolerance in rice, particularly for direct seeding applications in the future.

Results

OsRAV1 Expression Pattern in Different Rice Tissues and Subcellular Localization

To analyze the function of OsRAV1 in the growth and development of rice plant, we examined its expression profile in various tissues at the booting stage using RT-qPCR. The results demonstrate constitutive expression of OsRAV1 across different tissues, with higher levels in mature leaves and sheaths, and lower expression in newly formed organs. Expression was lower in young panicles but increased in later stages of panicle development (Fig. 1A). To validate these patterns, we constructed a vector with the GUS gene driven by the OsRAV1 promoter and used it for rice transformation. Importantly, GUS activity was the highest in panicle tissues and roots during the booting stage, indicating a crucial role of OsRAV1 in rice panicle developmental (Additional file 1: Fig. S1A. GUS activity was notably strong in roots and sheaths at 14d after germination (Additional file 1: Fig. S1B). Subcellular localization studies using an OsRAV1-green fluorescent protein (GFP) fusion construct under the CaMV 35 S promoter showed that OsRAV1 localized in the nucleus, confirmed by confocal microscopy in both Nicotiana benthamiana leaf epidermal cells and rice protoplasts (Fig. 1B).

Fig. 1

Expression patterns of OsRAV1 and its subcellular localization in rice. A Expression patterns of OsRAV1 in various developmental stages in rice by RT-qPCR. B Subcellular localization of OsRAV1 with GFP in tobacco leaves and rice protoplasts. Scale bars = 50 μm (tobacco leaves) or 10 μm (rice protoplast). C Expression of OsRAV1 in response to salt (NaCl). Rice seedlings (14 days old) were treated with 120 mM NaCl. RNA from shoots (leaf) were used for RT-qPCR, with OsActin as an internal control. The error bars indicate the SE based on three independent replicates

Next, we investigated OsRAV1 expression in response to various abiotic stress and hormone treatment. Our findings indicate that OsRAV1 expression is notably induced by salt (NaCl), ABA, and osmotic stress (PEG) (Fig. 1C; Additional file 1: Fig. S1C and D), suggesting its responsiveness to a broad range of abiotic stressors.

OsRAV1 Improves Germination Rate and Seed Vigor

To further explore OsRAV1 function, we generated overexpression (OE) lines by inserting the OsRAV1 coding region under the 35 S promoter. Increased OsRAV1 in the OE lines were confirmed by RT-qPCR (Additional file 1: Fig. S2A). Additionally, OsRAV1 knockout mutants (osrav1-1 and osrav1-2) were generated using CRISPR/Cas9 technology, resulting in premature termination of OsRAV1 and disruption of the AP2 and B3 domains (Additional file 1: Fig. S2B and C).

The OsRAV1 expression is upregulated and linked to gene networks related to seed vigor in Kasalath (Wang et al. 2022). Our germination assay showed that under control conditions, OE lines and Nipponbare (NIP) had similar germination rates, whereas osrav1 mutants showed lower germination rate than NIP (Fig. 2A and B). Notably, only OE lines maintained seed vigor after one year of indoor storage (S-1Y), with germination rates reaching 28–48% within 120 h (Fig. 2A and C). As storage time increased, seeds aged and lost germination ability. OsRAV1 appears to play a crucial role in enhancing seed vigor. Seed accelerated aging, a method for rapid seed vigor assessment, was employed in aging seeds of NIP, OE lines, and osrav1 mutants at 42℃ and 80% relative humidity for 11、14、17, and 20 days (Fig. 2D-G). With increasing aging time, initial germination rates of OE lines were higher than those of NIP, while osrav1 mutants almost completely lost their germination ability at 17 and 20 days.

Fig. 2

OsRAV1 positively regulates seed vigor under aging conditions. A Images of germination. B Newly harvested seeds. C The seeds after stored in room conditions for one year (S 1Y). C-G The seeds aged under 42 ℃ and 80% RH for various period as indicated

Given the lengthy process of traditional aging method (high temperature and humidity), we developed an improved method using seeds soaked in saturated NaCl solution and stored in darkness at 42℃ (Fig. 3). For germination, the seeds were firstly rinsed in tap water, then imbibed in distilled water for 72 h at 4℃ and followed by placing on moistened filter paper at 28℃ under alternating dark and light (12 h/12 h) conditions. Germination rates decreased rapidly with aging time, with a majority of seeds losing germination ability by 9 days (Fig. 3H). Germination rates (36 h, 48 h, and 120 h) and other indices of OE lines were significantly higher than those of NIP at aging times of 5 and 7 days (Fig. 3I-L), confirming that OsRAV1 positively regulates seed vigor.

Fig. 3

OsRAV1 positively regulates seed vigor. A-H The germination rate after aging treatment. The seeds aged in saturated NaCl solution at 42℃ for various length as indicated. I-K The germination rates at 24, 36, and 48 h. L Germination index. Different small letters indicate significant differences at p ≤ 0.05 within the comparison group

OsRAV1 Expression Patterns in Seeds at Various Germination Stages

To further comprehend OsRAV1’s physiological role, we examined its expression patterns during seed germination and aging stages using RT-qPCR and GUS activity assays (Fig. 4). OsRAV1 transcript were prominently detected the imbibition phase (1–3 days), decreasing significantly in early germination stages (Fig. 4B). Further analysis revealed significant OsRAV1 expression in both embryo and endosperm during imbibition, with higher expression in the embryo (Fig. 4C). To exclude low-temperature effects on OsRAV1 expression patterns, we examined transcript levels during germination without prior imbibition. OsRAV1 transcripts were highly expressed during early germination (3 and 6 h), with higher levels in the embryo, though the gap narrowed over time (Fig. 4D). During aging treatments, OsRAV1 expression in the endosperm increased significantly, surpassing levels observed in the embryo. This pattern resembled that found in dry seeds (Fig. 4E), suggesting distinct roles for OsRAV1 during seed germination and storage conditions.

Fig. 4

OsRAV1 expression and GUS activity were observed in germination stages. A Histochemical GUS staining reporting OsRAV1 activity during seed germination. I-1d, I-2d, and I-3d refer to seeds imbibed in distilled water at 4℃ for 1, 2, and 3 days, respectively. G-1d, G-2d, and G-3d indicate seeds germination in distilled water at 28℃ for 1, 2, and 3 days after imbibed, respectively. BOsRAV1 expression in whole seeds. COsRAV1 expression in embryo and endosperm during imbibition (4℃). DOsRAV1 expression in embryo and endosperm during germination (28℃), without prior imbibition (4℃). EOsRAV1 expression in embryo and endosperm under aging

OsRAV1 Improve Seed Vigor and Direct Seeding under Salinity Stress

Salt stress severely limits rice production globally, particularly under direct seeding conditions. Under control conditions, OE lines exhibited significantly higher germination rates than NIP at 24 h, but this trend reversed at other time points (Fig. 5A and B). As NaCl concentration increased, NIP germination rates gradually decreased compared to the OE lines, particularly evident at 200mM NaCl (Fig. 5C, D, F, and G). Conversely, osrav1 mutants showed greater sensitivity to increasing NaCl concentration compared to NIP, with significantly reduced germination rates (Fig. 5).

Fig. 5

OsRAV1 enhances initial germination rate under salt stress. A Images of seed germination. B-D The germination rate under mock (control treatment), 120, and 200 mM NaCl. E-G The initial germination rates (24, 36, and 48 h) under 0, 120, and 200 mM NaCl. Different small letters indicate significant differences at p ≤ 0.05 within the comparison group

To assess OsRAV1’s utility in direct seeding, we compared seedling establishment and growth between NIP, OE lines, and osrav1 mutants directly sown in soil (Fig. 6). Under normal conditions, no significant differences were observed in seedling percentage between NIP and OE lines at 7 days after sowing (Fig. 6B). However, OE lines exhibited significantly longer roots and greater root dry weights than NIP. In contrast, seedling percentage, shoot length, shoot dry weight, root length, and root dry weight were significantly reduced in osrav1 mutants compared to NIP (Fig. 6). Under 0.5% NaCl stress, OE lines showed significantly higher seedling establishment and growth than NIP. osrav1 mutants exhibited heightened sensitivity to increasing NaCl concentrations, ultimately losing the ability to germinate.

Fig. 6

Seedling growth of direct seeding in soil. A Seedling establishment and seedling growth under normal conditions or salinity stress at 7 days after seeding. B-F The seedling percentage, shoot lengths, root lengths, shoot dry weight, and root dry weight of OE, osrav1, and NIP. Different small letters indicate significant differences at p ≤ 0.05 within the comparison group

Association of OsRAV1 Haplotype with Seed Germination under Salinity Stress

To investigate whether OsRAV1 allelic variation affects seed germination under salinity stress, the 1 kb upstream region of OsRAV1 were analyzed using SNP and saline-alkali tolerance germination data from 603 accessions (McCouch et al. 2016; Zhang et al., 2023) (Fig. 7A; Additional file 3: Table. S2). Analysis of SNP correlation with germination identified four haplotypes (n > 15 accession) of OsRAV1. Hap3 showed the highest mean relative germination rate (RGR), suggesting it as the favorable haplotype for saline-alkali tolerance, predominantly found in Geng accessions (Fig. 7B and C).

Fig. 7

Haplotype of OsRAV1 associated with germination under salt stress. A Haplotypes in the 1 kb promoter region upstream of OsRAV1 CDS. B Distribution of relative germination rate (RGR) for the four OsRAV1 haplotypes. C Frequency of four OsRAV1 haplotypes in accessions. Different small letters indicate significant differences at p ≤ 0.05 within the comparison group

RNA-seq and Differential Expression Analysis

Given the significant difference in germination rates between osrav1 mutants and OE lines, and the sustained higher germination rates in OE lines, we performed RNA-seq to elucidate the molecular pathways underlying OsRAV1-regulated seed germination. Post-imbibition seeds of NIP, OE lines, and osrav1 mutants at initial germination stages (0 and 24 h) were sequenced (RNA-seq), with 30 cDNA libraries constructed representing three biological replicates for each of NIP, OE1, OE2, osrav1-1, and osrav1-2. Principal component analysis (PCA) revealed close overlap among biological replicates of OE lines and osrav1 mutants, respectively (Fig. 8A). DESeq (differential gene expression analysis) identified DEGs based on |log2FoldChange| ≥ 1 and P value ≤ 0.05 criteria. Venn diagram analysis revealed 306 genes (0H2) differentially expressed in osrav1_0H and OE_0H, while osrav1_24H and OE_24H shared 528 genes (24H2). Notably, 4124 DEGs (24H1) were exclusively present in osrav1_24H (Fig. 8B), indicating profound transcriptome changes at 24 h of germination, particularly affected by absent of OsRAV1 expression.

Fig. 8

Transcriptome analysis of DEGs regulated by OsRAV1. A Principal component analysis (PCA) based on the expressed genes. B Venn diagrams of the DEGs in OE and osrav1 seeds compared with NIP. 0 H and 24 H refer to the germination stages at 0 and 24 h, respectively. 0H1: DEGs exclusive to osrav1; 0H2: DEGs common between osrav1 and OE; 0H3: DEGs exclusive to OE in germinated seeds (0 h). 24H1: DEGs exclusive to osrav1; 24H2: DEGs common between osrav1 and OE; 24H3: DEGs exclusive to OE in germinated seeds (24 h). C-H KEGG enrichment analysis of DEGs exclusive to osrav1 (C and F) or OE (E and H), or common in both (D and G)

Gene ontology (GO) analysis of 0H1, 0H2, 0H3, 24H1, 24H2, 24H3 revealed enrichment in catalytic activity, metabolic process, response to stimuli, and stress (Additional file 1: Fig. S3). KEGG analysis highlighted enrichment in protein processing in endoplasmic reticulum and phenylpropanoid biosynthesis in 0H1 and 0H2 comparisons (Fig. 8C and D). However, the 0H3 exhibits significant enrichment in phenylpropanoid biosynthesis, rather than protein processing in the endoplasmic reticulum (Fig. 8E), indicating distinct seed status between OE lines and osrav1 mutants, formed during seed development and storage, lasted to even after a 3-day imbibition. Additional evidences are DEGs in 24H1 and 24H3 enriched in plant hormone signal transduction, whereas not for 24H2 (Fig. 8F-H; Additional file 1: Fig. S4A). These suggested that OsRAV1 regulates seed vigor (balance) through different hormonal pathways, in addition to crucial regulation of phenylpropanoid biosynthesis in the germination stage of rice seeds (Fig. 8F-H; Additional file 1: Fig. S4B).

Venn diagram analysis showed that 5525 genes were differentially expressed in 24 h germinated seeds compared to 0 h (Fig. 9A). To discover pathways that confer to the seed’s ability to germinate rapidly, we performed a cluster analysis of the genes. 2640 genes were higher expressed in 24 h germinated seeds compared to 0 h and then manually classified into two clusters (C1 and C2) (Fig. 9B). The genes of cluster C1 were down-regulated in osrav1 mutants, while those in cluster C2 were up-regulated in OE lines. Both C1 and C2 were mainly enriched in similar GO terms: response to stimulus, response to abiotic stimulus, and response to stress (Fig. 9C and D). KEGG analysis showed that C1 genes were involved in phenylpropanoid biosynthesis and plant hormone signal transduction, whereas C2 genes were significantly enriched in protein processing in the endoplasmic reticulum (Fig. 9E and F). MapMan analysis revealed that DEGs involved in abiotic stress, antioxidant, signaling, transcription factors, and defence pathways were active during seed germination (Additional file 1: Fig. S5). Further studying the functions of these genes will shed light on the molecular mechanisms of seed germination.

Fig. 9

Transcriptome analysis of DEGs in seeds germinating at 0 and 24 h. A Venn diagrams of DEGs in osrav1, OE, and NIP comparisons. B Expression profile of common DEGs (5525). OE: OE1 and OE2; osrav1: osrav1-1 and osrav1-2; with three biological replicates each. Clustering was drawn based on normalized FPKM. C, D Top 20 GO terms of cluster 1 (C1) and cluster 2 (C2). E, F KEGG enrichment analysis of C1 and C2

Discussion

Seed vigor is a complex trait encompassing seed storability, germination, seedling emergence and growth, and stress tolerance, influenced by various factors such as reserves accumulation and seed dormancy during development, seed deterioration during storage, and seed treatments and stress conditions during germination in rice. Developing new varieties with high seed vigor is crucial for direct seeding of rice. To achieve this goal, understanding the complex regulatory network of seed germination is prerequisite. Previously, AP2/ERF transcription factors have been reported to involve the balance of ABA and GA, seed vigor and stress-related processes in plants. ABI4 directly binds to NCED6 and GA2ox7, regulating ABA synthesis and GA degradation by controlling their transcription levels, which in turn regulates seed dormancy (Shu et al. 2013, 2016). Moreover, salt-induced ABI4 enhances RbohD expression, promotes ROS accumulation, thus results in cell membrane damage, and, consequently, a decrease in seed vigor (Luo et al. 2021). OsAP2-39 activates OsNCED1 transcription to enhance ABA synthesis, promoting GA degradation through the inhibition of OsEUI accumulation, therefore, enhances seed dormancy in rice (Yaish et al. 2010; Shu et al. 2013). Additionally, OsSAE1 promotes seed germination and salt tolerance in seedlings by suppressing OsABI5 expression. In this study, OsRAV1 positively regulated seed vigor, including initial germination rate and seedling establishment under salt stress (Figs. 5 and 6). Furthermore, transcriptome and metabolome analysis of seed aging treatments revealed that various AP2/ERF transcription factors likely serve as pivotal regulators within the complex network that governing seed vigor. Among them, OsRAV1 is highly induced and associated with positive regulation of OsbZIP23 (Wang et al. 2022). Under both natural and accelerated aging treatment, OE lines exhibited higher germination rates compared to NIP, while osrav1 mutants showed the opposite trend (Figs. 2 and 3). In plants, the growth-defense trade-off is commonplace. Readjustment of growth and adaptation to various stresses are crucial for survival under adverse environments. Previous studies have revealed associations among seed germination, longevity, and salt tolerance during germination phase. OsRAV1 is a candidate gene that positively regulates seed vigor and salt stress resistance.

Previous studies have shown that the indica (Xian) variety have higher seed vigor than japonica (Geng) varieties. To determine whether OsRAV1 contributes to the differences of seed vigor between Xian and Geng rice, allelic diversity analysis of OsRAV1 was conducted using 603 rice accessions, including both Xian and Geng accessions (Mansueto et al. 2017; Zhang et al. 2023a). Analysis of SNP data revealed that the germination rate of hap3 was significantly higher than that of other haplotypes under salt stress (Fig. 7B). Interestingly, Hap3 was predominantly found in Geng accessions and absent in Xian accessions. The genetic determinants of crop phenotypic diversity often involve natural mutations in promoters or coding sequences, which can modify transcriptional expression or protein functionality. We speculate that SNPs in OsRAV1 could affect its expression, and further investigation is needed to confirm its potential as a candidate gene for enhancing seed vigor, particularly in Geng accessions.

Omics analyses such as transcriptome, proteome, and metabolome analyses, have been employed to elucidate seed germination process in several crop species. Previous studies have identified a variety of biological processes involved in rice seed germination, including cell-wall metabolism, nucleotide degradation, amino acid biosynthesis, stress-response pathways, and more (Zhao et al. 2020b; Yang et al. 2020). However, these results were obtained using dry seeds, which inevitably undergo imbibition. Water uptake by mature dry seeds occurs in three phases: rapid phase (phase I), plateau phase (phase II), and a second rapid phase of water uptake (phase III). Seed imbibition and germination are overlapping processes; during phase II and phase III, the radicle protrudes through the seed coat (Zhao et al. 2020b; He et al. 2020b). It is equally important to understand genes responses from imbibed seeds to seedling establishment (radicle protrusion included).

In this study, we analyzed gene expression changes during the initial germination stage using RNA-Seq. Protein processing in the endoplasmic reticulum was mainly observed in 0H1 and 0H2, while phenylpropanoid biosynthesis was enriched in 0H3, with no significant protein processing pathways observed in the endoplasmic reticulum (Fig. 8C-E). These results indicate that OE lines achieved germination earlier, as confirmed by KEGG analyses of 24 h germinated seeds. DEGs in 24H1, 24H2, and 24H3 were all enriched in phenylpropanoid biosynthesis (Fig. 8F-H). Many pathways involved in seed germination are directly or indirectly related to phenylpropanoid biosynthesis (Foley et al. 2013). Phenylpropanoid biosynthesis continues to be enriched during seed germination and seedling establishment stages (Yang et al. 2020), and changes in phenylpropanoids-related metabolites contents have been observed in rice seeds under low-temperature stress(Yang et al. 2019). Phenolic acids, a major component of phenylpropanoids, play critical roles in defense against reactive oxygen species (ROS). In the Bidens pilosa L, Bruceine D inhibits seed germination by suppressing phenylpropanoid biosynthesis and inducing ROS accumulation (Tong et al. 2022). Lignin, a crucial phenylpropanoid compounds, acts as an inducible physical barrier that enhances plant seeds resistance to environmental stress during germination (Rao and Dixon 2017). SiMYB16 enhances salt tolerance in rice by regulating lignin biosynthesis (Yu et al. 2023). Interestingly, a recent report reveals that bZIP23-PER1A–mediated detoxification pathway enhances seed vigor in rice (Wang et al. 2022). The expression of OsbZIP23 and PER1A were significantly upregulated in osrav1 mutants, suggesting that the H₂O₂ content might be elevated (Additional file 1. Fig. S4C). We speculate that OsRAV1 enhances seed vigor through phenylpropane pathway.

ABA-inhibited seed germination is regulated by a core regulatory mechanism (Zhao et al. 2020a). Interestingly, DEGs unique to 24H1 and 24H3 were significantly enriched in plant hormone signal transduction, whereas shared DEGs in 24H2 were not (Fig. 8F-H; Additional file 1: Fig. S4A), indicating that OsRAV1 regulates seed vigor through different hormonal pathways. Several genes in the ABA signaling core are clearly regulated by OsRAV1, such as OsAPKs, OsBZIPs, OsABIs, and OsPYL. The expressions of OsbZIP23, OsbZIP72, OsABI5, and OsPP2C51, all were significantly upregulated in osrav1 seeds (Additional file 1: Fig. S4C), have essential role in ABA signaling, stress tolerance or seed vigor in rice (Bhatnagar et al. 2017; Wang et al. 2020, 2022; Li et al. 2022). On the contrary, the expressions of auxin related genes were significantly reduced in osrav1 seeds (Additional file 1: Fig. S4C), indicating that OsRAV1 affects seed germination through the crosstalk of auxin and ABA. Further studies should focus on interactions of OsRAV1 with upstream components of the auxin and ABA signaling pathway.

In most studies, rice seed vigor factors are mainly considered at the seed germination stage, with limited consideration of other stages. Seed storability, defined as the ability to remain viable during storage, is an important agronomic and physiological characteristic. Compared to natural aging, accelerated aging accelerates seed deterioration using high temperature and high humidity, significantly reducing processing time (Hay et al. 2018). In this study, seed germination rates of OE lines remained high under accelerated aging treatment (Fig. 2). In the accelerated aging method, seeds were suspended above either a saturated NaCl solution or water to generate a high-humidity environment (Hay et al. 2018; Wang et al. 2022). Would soaking seeds in an saturated NaCl solution achieve similar results? As expected, with increased aging time (9d), seeds lost their ability to germinate. This method represents another potential accelerated aging method for testing seed vigor, without requiring humidity control.

The endosperm is the major tissue for seed reserve accumulation in cereal grains, serving as a nutrient source for seed germination and subsequent seedling establishment. Recent evidence suggests that mature endosperm is essential for seedlings establishment; its removal impedes seedling cuticle formation, resulting in nonviable seedlings with developmental defects. This outcome is not due to lack of nourishment in the embryo (De Giorgi et al. 2021). Transcriptome analysis of embryo and endosperm at 0, 1, 3, 6, 9, 12, and 24 h post-imbibition revealed that differentially responsive genes are specific to embryo and endosperm. Embryo-specific DEGs are enriched in plant hormone signal transduction pathways, while endosperm-specific DEGs are enriched in biosynthesis pathways of phenylalanine, tyrosine, and tryptophan (Zhang et al., 2023). In this study, we observed distinct expression patterns of OsRAV1 between embryo and endosperm, highlighting tissue-specific regulatory roles. High OsRAV1 expression in endosperm of both dry seeds and seeds subjected to aging treatment suggests a significant role in stress resistance (storage). Upon water absorption at right temperature and initiation of germination, OsRAV1 expression levels in endosperm decrease gradually, while there is a significant increase (up to 9-fold in G-3 h) in embryos (Fig. 4D). At low temperature (4℃), OsRAV1 expression in embryo significantly increases (approximately 119-fold in I-3d), underscoring its crucial role in germination, particularly under challenging conditions (Fig. 4C). We speculate that OsRAV1 may regulate signaling communication between endosperm and embryo, playing a critical role in seed vigor and germination.

Conclusion

In conclusion, we have validated the AP2/ERF TF gene OsRAV1 is essential for seed vigor in rice. The Hap3 of OsRAV1 is a beneficial variant that increases rice seed germination under salt stress, and overexpression of OsRAV1 improves seed germination, vigor, and salt resistance, potentially making a significant contribution to the direct seeding cultivation of rice. The RNA-Seq results demonstrated that OsRAV1 may regulate seed germination through plant hormone signal transduction and phenylpropanoid biosynthesis. Furthermore, immersing seeds in a saturated NaCl solution at elevated temperature is an effective method for significantly accelerating the aging process.

Materials and Methods

Rice Varieties and Culture

The japonica rice variety Nipponbare (NIP) served as a wild-type plant and recipient for genetic transformation. Rice plants were cultivated at the farm of Yangzhou University, Jiangsu, China, during the rice seasons (May to October). Seed preparation and seedling nursery followed local practices as previously described (Gao et al. 2024). Seeds were harvested at their maturity stage and dried at 42℃ for 7 days to break seed dormancy.

Generation of Transgenic Plants

To generate OsRAV1 overexpression lines (OE), the coding region was amplified from NIP first-strand cDNA and inserted into the binary vector p1301UbiNOS under the maize ubiquitin promoter. Knockout osrav1 mutants were generated using CRISPR/Cas9 technology, targeting the sequence 5’-GCTCGTCCACGGACAAGCTGAGG-3’ of OsRAV1 in the pC1300-Ubi: Cas9 vector (Miao et al. 2020; Gu et al. 2022). The 2.0-kb OsRAV1 promoter sequence from NIP genomic DNA was inserted into the pCAMBIA1301 vector to drive expression of the GUS-encoding gene. All constructs were transformed into Agrobacterium tumefaciens EHA105 cells via chemical transformation and then subsequently into plants using A. tumefaciens-based method (Hiei et al. 1994). Primers used for constructing recombinant vectors are listed in Additional file 2: Table S1.

Seed Accelerated Aging

Seed accelerated aging followed established protocols with some modifications (Delouche and Baskin 1973; Wang et al. 2022). Seeds were subjected to saturated NaCl solution and aged in darkness at 42℃. We prepare ed the saturated NaCl solution and completed the aging process of the seeds in a 42℃ oven. First, we prepared an oversaturated NaCl solution in a glass bottle and placed it in the oven for 24 h, until undissolved crystals appeared at the bottom of the bottle, indicating the solution was saturated. Next, we transferred the solution to a 2.5 L conical flask and filled it to the brim to ensure it could submerge the seeds. We then placed the dry seeds in a mesh bag and immersed them in the solution. To prevent the seeds from floating, we blocked the mouth of the flask with a mesh bag and sealed it with a glass stopper. Prior to germination, seeds were rinsed with tap water and imbibed in distilled water at 4℃ for 3 days.

Seed Germination and Direct Seeding

For germination assays, twenty seeds per replicate of NIP, OE lines (OE1 and OE2), and osrav1 mutants (osrav1-1 and osrav1-2) were placed in 9 cm diameter Petri dishes with 12 mL distilled water at 28℃ for 5 days. NaCl concentrations 120 and 200mM were chosen when exposed to salt stress in Petri dishes. Germination was recorded when radicle protruded the seed coat (1 mm), germination index was calculated as ∑ (Gt/t), where Gt is the number of the germinated seeds on day t (He et al. 2019). Additionally, thirty seeds per replicate were sown at a depth of approximately 1 cm in soil at 28℃ for 7 days, with NaCl concentration of 0%, 0.25% and 0.5% (w/w). Root and shoot lengths and dry weight were measured. Three replications were conducted.

Expression Analysis and Histochemical Analysis of GUS Activity

Total RNA was extracted from dry seeds, seeds subjected to aging (3, 5, and 7 days), seeds undergoing imbibition (1–3 days), germinating seeds (1–4 days, after 3 days imbibition ), germinating seeds (3, 6, 12, 24, 48, and 72 h, without pre-imbibition), and various tissues (leaves, nodes, sheaths, and young panicles at the booting stage) using an RNA simple Total RNA kit (Tiangen, Beijing, China). Reverse transcrition into cDNA was performed using the Fast King One-step RT-qPCR kit (Tiangen, Beijing, China). RT-qPCR analysis was conducted as previously described (Liu et al. 2024). Primers used for RT-qPCR are listed in Additional file 2: Table S1. GUS activity was detected using the GUSblue kit (Huayueyang, Beijing, China) and samples were destained with pure ethanol prior to examination.

Subcellular Localization

The OsRAV1 coding sequence was PCR-amplified and cloned into pAN580 and pHG vectors to obtain pAN580-OsRAV1 and pHG-OsRAV1 constructs, respectively. Transient expression assays in rice protoplasts were performed as described previously (Chen et al. 2006; Gu et al. 2022). Agrobacterium tumefaciens GV3101 cells transformed with pHG empty vector or pHG-OsRAV1 constructs were used to infiltrate healthy Nicotiana benthamiana leaves. Fluorescent signals were visualized using a Zeiss LSM 710 laser scanning confocal microscope. The primers used for constructing recombinant vectors are listed in Additional file 2: Table S1.

Haplotype Analyses

Relative germination rates (ratio of saline-alkali stress to control) of 603 accessions of rice were obtain from a previous study (Zhang et al., 2023) (Additional file 3: Table S2). SNPs within the 1 kb region upstream of OsRAV1 were downloaded from the Rice SNP-Seek database (https://snpseek.irri.org/) (Mansueto et al. 2017). Haplotype analysis focused on variants present in at least 15 accessions.

RNA-seq and Analysis

Thirty tissue samples representing three OsRAV1 levels (NIP, OE1, OE2, osrav1-1, and osrav1-2) and two germination stages (0 and 24 h) with three biological replicates each were collected. Seeds (25 per replicate) were frozen in liquid N₂ and stored at -80℃ until RNA extraction. RNA extraction, sequencing, and data analysis were conducted as previously described (Zhang et al. 2019). Differentially expressed genes (DEGs) (P_adj<0.05 and fold change ≥ 2) were identified using DEGseq2 software. GO enrichment and KEGG pathway analysis were performed using Personal Gene Cloud (https://www.genescloud.cn). Heatmaps were generated using TBtools software with row scaling set to normalized (Chen et al. 2020).

Statistical Analysis

IBM SPSS Statistics version 26.0 software was used for statistical analysis. Differences were considered significant at p ≤ 0.05. Figures were prepared using OriginPro 2016 (OriginLab, Northampton, Massachusetts, USA) or Excel 2019.