https://orcid.org/0009-0004-1954-4750
Figures
Abstract
Grain shape is a critical factor that directly influences rice yield and quality, however, the molecular mechanisms underlying the regulation of grain shape development remains elusive. In this study, we characterized an oval-shaped grain mutant, osg, from the rice radiation mutagenesis mutant library via 60Co-γ ray irradiation. Compared to ZH11, the osg mutant exhibited decreased grain length and thousand-grain weight but increased grain width and thickness, while displaying significantly reduced plant height, tiller number, pollen viability, and seed setting rate. Map-based cloning revealed that OsSRS3, a known regulator of rice grain size and fertility, was identified as the candidate gene for OSG. The loss-of-function mutant in OsSRS3 exhibit abnormal phenotypes similar to that in osg. Comparative transcriptome sequencing of young panicles from ZH11 and osg showed that up-regulated genes were predominantly enriched in pathways related to plant hormone signal transduction and MAPK signaling, whereas down-regulated genes were mostly associated with starch and sucrose metabolism. Further analysis of differentially expressed genes (DEGs) within the pathways revealed that multiple DEGs in osg, such as OsJAZ11, OsUgp1 and OsUgp2, have been functionally characterized, and their mutant phenotypes align with those observed in osg. These findings provide a crucial foundation for elucidating the molecular mechanisms by which the SRS3 gene regulates grain size and fertility in rice.
Citation: Zhao S, Zhang S, Xu W, Fang S, Liu M, Wang L, et al. (2026) Fine mapping and transcriptomics reveal OSG function in regulation of grain size and pollen fertility in rice (Oryza sativa). PLoS One 21(1): e0338401. https://doi.org/10.1371/journal.pone.0338401
Editor: Muhammad Abdul Rehman Rashid, Government College University Faisalabad, PAKISTAN
Received: March 4, 2025; Accepted: November 21, 2025; Published: January 8, 2026
Copyright: © 2026 Zhao et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This study was supported by the Natural Science Foundation of Anhui Province (2308085MC70) awarded to HZ; the key project of the Education Department of Anhui Province (2023AH051850) awarded to HZ; the Excellent Scientific Research and Innovation team of the Education Department of Anhui Province (2022AH010087) awarded to HZ; the Key Discipline Construction Fund for Crop Science of Anhui Science and Technology University (No. XK-XJGF001) awarded to HZ; the Scientific and Technological Innovation Team of Anhui University of Science and Technology (2023KJCXTD001) awarded to HZ; the Talent Introduction Start-up Fund Project of Anhui University of Science and Technology (NXYJ202104) awarded to HZ; the Hubei Provincinal Natural Science Foundation-General Project (2019CFB625) awarded to SW; the National College Student Innovation and Entrepreneurship Training Program Project of China (202410879064, 202310879076) awarded to XZ and SZ; the College Student Innovation and Entrepreneurship Training Program Project of Anhui Province (S202310879168, S202410879163) awarded to LW and BC.The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Rice (Oryza sativa) is one of the most important food crops, playing a crucial role in agricultural production. Increasing rice yield and quality remains an urgent challenge in rice breeding research [1]. Rice yield is directly influenced by effective panicle number, grains per panicle, and thousand-grain weight [2]. Effective panicle number depends on tiller number, tiller angle, and plant height; grains per panicle depend on panicle length, branches per panicle, and fertility; thousand-grain weight is directly affected by grain shape and size, which also influence the appearance quality of milled rice [3]. Therefore, studying the genetics and molecular mechanisms underlying rice grain shape can aid in genetic improvement of yield and quality traits, which is significant for developing high-yield and high-quality rice varieties.
With the development of functional genomics and molecular biology technologies, an increasing number of grain shape-related genes have been mapped and cloned. Previous studies have shown that rice grain shape is mainly regulated by lemma cell development and involves genetic pathways such as the ubiquitin-proteasome pathway, G-protein signaling pathway, mitogen-activated protein kinase (MAPK) signaling pathway, plant hormone signal transduction pathways, and transcription factor regulation pathways [4]. Grain Size 3 (GS3) was the first cloned grain shape gene, primarily controlling grain length. GS3 encodes a transmembrane protein with a domain regulating organ size (OSR), which negatively regulates grain length, while the von Willebrand factor type C (VWFC) domain inhibits the effect of OSR on grain length within a certain range [5]. Grain Width 2 (GW2) was the first cloned gene controlling grain width and is the primary regulator of grain width. GW2 encodes an E3 ubiquitin ligase involved in degrading proteins that promote cell division, thereby controlling grain width [6]. The seed width on chromosome 5 (GW5), located on chromosome 5, is a major quantitative trait locus (QTL) controlling grain width and weight. This gene promotes cell division in the lemma and increases grain width and weight [7,8]. Subsequent studies have found that GW5 inhibits the activity of transcription factors in the brassinosteroid (BR) signaling pathway, affecting the expression of downstream BR response gene [9]. Grain Weight 8 (GW8) positively regulates grain width by influencing genes related to lemma cell cycle [10]. Grain Weight 7 (GW7), located on chromosome 7, regulates the TONNEAU1 recruitment motif sequence protein and alters grain length-to-width ratio by influencing cell division patterns [11]. Grain size 2 (GS2) regulates grain size by promoting cell division and expansion; GS2 encodes a growth-regulating factor OsGRF4 that changes the cell life cycle, thus altering the grain length-to-width ratio [12].
Rice grain shape is closely related to rice yield and quality. Despite the discovery and cloning of numerous QTLs and related genes controlling grain shape, the genetic regulatory network of rice grain shape remains incompletely understood. There is still a need to discover new regulatory genes to fully understand the molecular network regulating grain shape. In this study, we screened for a short-panicle and oval-grain mutant osg from the japonica variety ZH11 using radiation mutagenesis. We conducted agronomic trait surveys, histological phenotype analyses, fine mapping to identify the mutated gene, and performed transcriptome analysis between wild-type and osg. Our aim is to provide a new theoretical basis for elucidating the molecular network regulating rice grain morphology and to lay an important genetic resource foundation for molecular design breeding in rice.
Materials and methods
Experimental materials and growing conditions
The oval-shaped grain mutant osg was derived from the japonica variety Zhonghua 11 (ZH11) through 60Co-γ ray irradiation mutagenesis. After several generations of selection, a mutant with stable heritable phenotype was obtained. The mutant was crossed with two indica varieties, HJX74 (Hua Jing Xian 74) and 9311, to obtain F1 hybrids, and self-crossed to produce F2 segregating populations. The plants with the osg mutant phenotype were selected from the F2 population for gene mapping. All materials were grown in the experimental paddy field at Fengyang Campus, Anhui Science and Technology University (32°52’ 30’‘ N, 117°33’ 15’‘ E) under conventional planting and water-fertilizer management.
Statistical analysis of major agronomic traits
At maturity, 20 plants each of wild-type ZH11 and mutant osg were selected for observation and statistical analysis of plant height and tiller number. Main panicles were harvested to measure panicle length, primary branch number, total grain number per panicle, filled grain number, seed setting rate, grain length, width, and thousand-grain weight. Each group had three biological replicates, with the average value taken from 20 individual plants. Data analysis and graph creation were performed using GraphPad Prism v8.0.2. Paired t-tests were utilized to evaluate the significance of differences between groups.
Observation and analysis of lemma cells in osg
During the heading stage, mature spikelets from mutant osg and wild-type ZH11 were selected. Samples were fixed overnight at 4°C in 2.5% glutaraldehyde, dehydrated in a series of ethanol concentrations (30%, 50%, 70%, 85%, 95%, and 100%), replaced with isoamyl acetate, dried at critical point, coated with gold, and observed and photographed under a scanning electron microscope. Cell size and number were calculated using Image J v1.54g.
For paraffin sectioning, samples were fixed in FAA solution (70% ethanol: glacial acetic acid: formaldehyde = 90:5:5) for 48 hours, dehydrated in a series of ethanol concentrations (70%, 85%, 95%, and 100%), and immersed in Paraplast Plus. Sections were cut using a microtome, stained with toluidine blue, and observed under a panoramic high-resolution slide scanning system, with cell sizes measured using Image J v1.54g.
Observation of fertility in osg
Mature panicles from wild-type ZH11 and osg were selected during the heading stage. Pollen grains were removed from the anthers and stained with a 1% I2-KI solution (0.6% KI, 0.3% I2, w/w) for viability analysis. Pollen fertility was observed and photographed using an Eclipse E600561 microscope.
Mature ovules were fixed in FAA solution for 48 hours, washed twice in 50% ethanol, and transferred to 70% ethanol. Adequate numbers of ovules were dissected under a dissection microscope to remove the ovaries, then stained with 2% potassium aluminum sulfate for 20 minutes, followed by staining with 10 mg/L eosin B (dissolved in 4% sucrose solution) for 12 hours and counterstained with 2% potassium aluminum sulfate for 20 minutes. Samples were dehydrated in a series of ethanol concentrations (30%, 50%, 70%, 80%, 90%, and 100%), placed in a 1:1 mixture of absolute ethanol and methyl salicylate for 1 hour, and finally cleared in methyl salicylate for over 1 hour. Observations and photography were conducted using a Leica SP2 laser scanning confocal microscope.
Fine mapping
Using F2 segregating populations obtained by crossing osg with HJX74 and 9311, leaf samples from individual plants with the osg phenotype were collected, and DNA was extracted using the cetyltrimethylammonium bromide (CTAB) method. InDel markers were designed based on insertions or deletions between ZH11 and HJX74/9311 reference genomes using Primer 3 line website (https://primer3.ut.ee/), synthesized by Universal Biosciences Co., Ltd. (Chuzhou, China)(S1 Table). PCR reactions were performed using a LongGene instrument. Reaction conditions included initial denaturation at 95°C for 15 seconds, annealing at 55°C for 15 seconds, extension at 72°C for 15 seconds, 35 cycles, and final extension at 72°C for 5 minutes. PCR products were analyzed using 4% agarose gel electrophoresis.
In addition, 20 WT and 20 osg mutant plants from the F2 population were pooled into two pools, and SNP chip detection was performed by Wuhan Shuanglvchuang Biotechnology Research Institute Co., Ltd. using a Green Super Rice 40K high-density rice gene chip containing 44,263 SNPs. Allelic differences between the two pools and their parents were analyzed to locate the genomic segment associated with the phenotype, thereby localizing the OSG gene.
Candidate gene analysis
Based on the results of InDel markers and SNP chip detection, all genes within the mapped interval were annotated using the rice genome annotation database (http://rice.plantbiology.msu.edu/). Functional studies of genes within the interval were reviewed, comparing reported mutant phenotypes to identify candidates with phenotypes similar to osg. Exon-intron structures of candidate genes were used to design PCR primers, and PCR product sequencing was used to analyze and verify the sequence of candidate genes in osg.
SRS3 gene loss-of-function mutants were purchased from Bigene Biotech Co., Ltd. (Jiangsu, China). Primers were designed according to target site positions, and PCR product sequencing was used to confirm the sequence mutation status of the SRS3 gene in T2 generation. Agronomic traits such as plant height, tiller number, panicle length, and grain shape were observed in mature plants of srs3.
Transcriptome and qRT-PCR analysis
In the early stage of booting, young panicles (0–3 cm) were collected from ZH11 and osg, with three biological replicates of 0.3 g each, including at least five plants. Samples were frozen in liquid nitrogen and sent to Wuhan Biorun Biotechnology Co., Ltd. for RNA-seq (https://plant.biorun.com/. Wuhan, China) using the Illumina NovaSeq 6000 platform. The clean reads were aligned to the rice reference genome (MSU7) using the software Hisat2 with the default parameters [13]. Differential expression analysis was performed using DESeq2, with filtering criteria of Padj < 0.05 and log2|Fold Change| > 1. Up-regulated and down-regulated genes were subjected to GO and KEGG enrichment analysis.
For qRT-PCR (quantitative real time PCR), total RNA was extracted from young panicles (0–3 cm) collected from WT and osg using TRIzol reagent (Invitrogen), following the manufacturer’s protocols. The total RNA was reverse transcribed into cDNA using the HiScript III RT SuperMix kit (Vazyme, China) according to the instruction manual. The cDNA was diluted and used as the template to perform qRT-PCR with SYBR Green Ⅰ PCR Master Mix system (TOYOBO, Japan). The reaction condition was as follows: 95°C for 2 min, 40 cycles of 95°C for 15 s, 56°C for 15 s, and 72°C for 20 s. The 2-ΔCT method was employed to quantify the relative expression levels of genes. The Actin gene in rice was used as an internal control. The gene-specific primers of DEGs were listed in S1 Table.
Results
Phenotypic analysis of the osg mutant
To understand the genetic control of rice fertility and grain size, we developed a set of mutants with sterility or/and altered grain appearance. The osg mutant was isolated from the M3 generation of the rice cultivar ‘Zhonghua 11’ subjected to mutagenesis with 60Co-γ ray irradiation. During the maturation stage of rice, we conducted the agronomic trait survey comparing the wild-type ZH11 with the osg mutant. Our analysis revealed that, compared to ZH11, the plant height and tiller number in osg were significantly reduced by 27.03% and 7.55%, respectively (Fig 1A, 1D, 1E). Additionally, the panicle length of the osg mutant decreased by 25.92%, and the seed setting rate was markedly lower at only 50.2%; however, there was no noticeable difference in the number of primary branches or grains per panicle (Fig 1B, 1C, 1F, 1G, 1H, 1I).
(A) Observations of plant architecture in ZH11 and osg at maturity, scale bar = 20 cm; (B) Observations of panicle morphology in ZH11 and osg at maturity, scale bar = 5 cm; (C) Observations of individual panicle yield in ZH11 and osg, scale bar = 1 cm; (D) Plant height; (E) Tiller number; (F) Panicle length; (G) Number of primary branches; (H) Number of grains per panicle; (I) Seed setting rate. n = 20, * indicates P < 0.05; ** indicates P < 0.01. Significance of the results was assessed using a Student’s t-test, two tailed.
Regarding grain morphology, the grain length in osg was significantly shortened by 34.46% compared to ZH11 (Fig 2A, 2D, 2G). Conversely, both grain width and thickness showed significant increases of 14.22% and 12.77%, respectively (Fig 2B, 2C, 2E, 2F, 2H, 2I). Despite these dimensional changes, the thousand-grain weight in osg was notably reduced by 22.63% (Fig 2J).
(A) Observations of grain length in ZH11 and osg, scale bar = 1 cm; (B) Observations of grain width in ZH11 and osg, scale bar = 1 cm; (C) Observations of grain thickness in ZH11 and osg, scale bar = 1 cm; (D) Observations of dehulled grain length in ZH11 and osg, scale bar = 1 cm; (E) Observations of dehulled grain width in ZH11 and osg, scale bar = 1 cm; (F) Observations of dehulled grain thickness in ZH11 and osg, scale bar = 1 cm; (G) Grain length in ZH11 and osg; (H) Grain width in ZH11 and osg; (I) Grain thickness in ZH11 and osg; J) Thousand-grain weight in ZH11 and osg. n = 20, ** indicates P < 0.01. Significance of the data was assessed using a Student’s t-test, two tailed.
Observation of caryopsis lemma cells in the osg mutant
Observations of cross-sections of caryopses from both the wild type and the osg mutant revealed that, compared to ZH11, the osg mutant exhibited an increase in the number of cells within the parenchyma, along with a significant increase in cell area (Fig 3A-3E). Scanning electron microscopy (SEM) observations of the outer epidermis of lemmas from ZH11 and osg showed that the contours and arrangement of lemma epidermal cells were clearly distinguishable between ZH11 and osg (Fig 3F-3G). Statistical analysis of lemma cell length indicated that the lemma cells in osg were significantly shorter than those in ZH11, while the cell width was notably wider (Fig 3H-3I). In summary, the mutation in the OSG gene primarily affects the regulation of lemma cell length and width.
(A) Observations of lemmas from ZH11 and osg, scale bar = 1 mm; (B) Cross-sectional observations of lemmas from ZH11 and osg, scale bar = 1 mm; (C) Magnified view of lemma cross-sections from ZH11 and osg, scale bar = 100 μm; (D) Number of cells in the outer parenchyma of the lemma, n = 10; (E) Cell area in the outer parenchyma of the lemma, n = 10; (F) Scanning electron microscopy (SEM) images of mature caryopsis lemmas from ZH11 and osg, scale bar = 100 μm; (G) SEM images of mature caryopsis lemmas from ZH11 and osg, scale bar = 100 μm; (H) Length of outer epidermal cells of the lemma, n = 20; (I) Width of outer epidermal cells of the lemma, n = 20. ** indicates P < 0.01. Significance of the data was assessed using a Student’s t-test, two tailed.
Fertility observation of the osg mutant
In addition to the significant changes in grain morphology, the osg mutant exhibited a markedly reduced seed setting rate, with only approximately 50% (Fig 1J). To determine whether the decrease in seed setting rate was associated with pollen fertility, we observed the viability of mature pollen grains from ZH11 and osg using a 1% I2-KI staining solution. In the wild-type ZH11, pollen grains were plump and stained blackish-brown (Fig 4A), indicating high viability. In contrast, over 50% of the pollen grains from the osg mutant appeared yellowish-brown, with only a minority staining blackish-brown (Fig 4B and 4C), suggesting significantly reduced pollen fertility in the osg mutant.
(A) Staining observation of mature pollen grains from ZH11, scale bar = 100 μm; (B) Staining observation of mature pollen grains from osg, scale bar = 100 μm; (C) Pollen viability comparison between ZH11 and osg, n = 10, ** indicates P < 0.01, significance of the results was assessed using a Student’s t-test, two tailed; (D) Observation of mature ovules from ZH11; (E-F) Observation of mature ovules from osg.
To further investigate the fertility of ovules in osg, we dissected ovaries from rice plants at maturity and used whole-mount staining and laser scanning confocal microscopy to observe the ovules of both ZH11 and osg. Our results showed that, compared to ZH11, some of the ovaries in osg developed normally, with clearly visible polar nuclei and embryo sac structures (Fig 4D and 4E). However, in other mature ovules of osg, complete embryo sac structures could not be observed (Fig 4D and 4F), indicating abnormal development in some of the ovules within the osg mutant, which likely contributes to its reduced fertility and seed setting rate.
Map-based cloning of the OSG gene
To localize the OSG gene, we utilized an F2 population derived from a cross between the osg mutant and HJX74. We selected the individual plants with the mutant phenotype and performed linkage analysis using InDel markers designed for the target locus of osg. The results indicated that primers M1 and M12 on chromosome 5 were linked to the mutant phenotype of osg (Fig 5A). Further InDel markers were designed between M1 and M12 based on the differences in the ZH11 and HJX74 genomes, and the target gene was narrowed down to the region between markers M3 and M10 on chromosome 5. Subsequently, we crossed osg with 9311 to obtain another F2 population. By designing more densely spaced polymorphic markers within the previously defined interval and analyzing plants with the mutant phenotype, we further narrowed the target region to between M14 and M17 (Fig 5B), corresponding to approximately 3.1 ~ 4.4 Mb region on chromosome 5.
(A) Preliminary localization of the OSG gene using InDel markers M1 and M12 on chromosome 5; (B) Fine mapping of the OSG gene using newly designed InDel markers between M3 and M10, further narrowed down to between M14 and M17; (C) Genotype analysis of parents and F2 pools using SNP chip data; parent osg appears gray (homozygous), parent NIP appears red (homozygous), indicating homozygous genotypes for both parents; (D) SNP analysis at differential loci showing mostly heterozygous genotypes in the WT pool; (E) SNP analysis at differential loci showing homozygous mutant genotype (AA) in the mutant pool; (F) Amplification abnormalities observed in exons 8, 9, 10 and 11 of the SRS3 gene in osg; (G) Identification of a homozygous individual plant srs3-6 with a single-base deletion (T) leading to loss of function in the SRS3 gene; (H) Observation of grain morphology in srs3-6, scale bar = 1 cm.
Additionally, we constructed an F2 population by crossing osg with Nipponbare (NIP) and used SNP chips with bulked segregant analysis (BSA) to locate the OSG gene. The SNP chip results showed that the genotype of parent osg appeared gray, while that of parent NIP appeared red, indicating homozygous genotypes for both parents (Fig 5C). In the F2 segregating population, most genotypes in WT and osg mutant pools were heterozygous; however, a segment on chromosome 5 showed differences between the two pools. Comparison of genomic differences revealed that the WT pool was predominantly heterozygous (AB) (Fig 5D), whereas the genotype of the mutant pool matched the osg genotype (AA) (Fig 5E). Based on BSA principles and SNPs within this differential segment, the OSG gene was limited to approximately 3.1 ~ 3.6 Mb region on chromosome 5.
Analysis of candidate genes within the precise positioning segment
To identify candidate genes for osg, we screened all genes present in the localization region. According to the rice genome annotation database (http://rice.uga.edu/), there are 33 functional genes within the localization interval (S2 Table), among which SRS3 (LOC_Os05g06280) has been reported to have a mutantion phenotype similar to that of osg. To verify whether the osg mutant phenotype is due to a mutation in SRS3, we amplified and sequenced the exonic regions of the SRS3 gene in osg. The results showed that most exons of SRS3 in osg could be normally amplified and were consistent with the wild-type sequence, except for exons 8, 9, 10, and 11, which failed to amplify (Fig 5F, S1 Fig), indicating an abnormality in the coding sequence of SRS3 in osg.
To further confirm whether the osg phenotype is caused by a mutation in SRS3, we obtained a functional knockout mutant of SRS3 and analyzed the target site in 12 individual plants. One of the plants, designated srs3–6, was identified as a homozygous mutant with a single-base deletion (T) at the target site, leading to changes in the protein sequence and loss of function of SRS3 (Fig 5G, S1 Fig). Observations of grain morphology in srs3–6 showed similarities to osg, with significantly shorter grain length and increased grain width and thickness compared to the wild-type ZH11 (Fig 5H). These data collectively suggest that the abnormal phenotype observed in osg is due to a sequence variation in SRS3 (LOC_Os05g06280), thereby identifying LOC_Os05g06280 as the encoding gene for OSG.
Transcriptome analysis of the osg mutant
To further understand the transcriptomic changes in osg, we conducted RNA-Seq analysis on young panicles (0–3 cm) from ZH11 and osg. The three biological replicates from ZH11 were named WT-1, WT-2, WT-3, while those from osg were named osg-1, osg-2, osg-3, for a total of six samples. The data analysis from the RNA sequencing showed that the correlation coefficients between the wild-type samples and the mutant samples were all above 0.92, indicating satisfactory consistency (Fig 6A). A total of 2560 differentially expressed genes (DEGs) were identified between ZH11 and osg, among which 2305 genes were significantly up-regulated and 255 genes were significantly down-regulated in osg (Fig 6B), with the number of up-regulated genes notably exceeding that of down-regulated genes (S3 Table, S2 Fig).
(A) Correlation analysis between transcriptome samples; (B) Volcano plot of differentially expressed genes (DEGs); (C) GO enrichment analysis of up-regulated DEGs; (D) GO enrichment analysis of down-regulated DEGs; (E) KEGG pathway analysis of up-regulated DEGs; (F) KEGG pathway analysis of down-regulated DEGs.
GO enrichment analysis of the DEGs revealed that the up-regulated genes were significantly enriched in biological processes such as response to abiotic stimulus, response to organic substance, and response to oxygen-containing compound (Fig 6C). In contrast, the down-regulated genes were significantly enriched in biological processes like hydrolase activity, nuclease activity, and hydrolysis of nucleic acid phosphodiester bonds (Fig 6D).
KEGG pathway analysis indicated that the up-regulated genes were significantly enriched in pathways such as plant hormone signal transduction, MAPK signaling pathway, and phenylpropanoid biosynthesis (Fig 6E, S4 Table, S3-S5 Figs). Meanwhile, the down-regulated genes were significantly enriched in pathways like starch and sucrose metabolism and oxidative phosphorylation (Fig 6F, S5 Table, S6 Fig). These results suggest that the mutation in OSG primarily affects pathways related to plant hormone signal transduction, MAPK signaling, and carbohydrate synthesis metabolism (S6-S9 Tables).
Analysis of differentially expressed genes in KEGG enriched pathways
To further clarify which other genes are affected by the mutation in OSG and how these changes lead to the mutant phenotype, we conducted an analysis of pathways enriched with DEGs using the KEGG database.
The up-regulated DEGs were primarily enriched in plant hormone signal transduction pathways with a total of 44 genes, involved including auxin, cytokinin, gibberellin, brassinosteroid (BR), abscisic acid (ABA), ethylene, jasmonic acid (JA), and salicylic acid signaling processes. Through in-depth analysis of these pathways, we found that among the auxin, cytokinin, abscisic acid, and ethylene signaling pathways, 12, 3, 6, and 6 genes were significantly up-regulated respectively, with most genes showing an increase of more than 2.5-fold (S6 Table, S3 and S4 Figs). In the gibberellin signaling pathway, a total of 12 genes were significantly up-regulated, involving the GID, DELLA, and PIF3 families. In the gibberellin signaling pathway, a total of 12 genes were up-regulated, involving the GID, DELLA, and PIF3 families. Among these, 8 genes belong to the GID family, 3 genes belong to the DELLA family, and 1 gene belongs to the PIF3 family. Specifically, Os07g44890, SLRL1 (LOC_Os01g45860), and OsPIL16 (LOC_Os05g04740) showed significant up-regulation in the mutant compared to ZH11, with more than a 2-fold increase. It has been reported that the OsPIL16 acts as a negative regulator of grain length in rice, and silencing of OsPIL16 increases the grain length [14]. This suggests that OSG genes may co-participate with OsPIL16 in the regulation of rice grain length.
In the jasmonic acid signaling pathway, several families are involved in signal transduction, including the JAR1 family, COI1 family, JAZ family, and MYC2 family. Analysis revealed that 12 genes were up-regulated, all belonging to the JAZ family. These include OsJAZ9 (LOC_Os03g08310), OsJAZ11 (LOC_Os03g08320), OsJAZ10 (LOC_Os03g08330), OsJAZ6 (LOC_Os03g28940), OsJAZ1 (LOC_Os04g55920), OsJAZ7 (LOC_Os07g42370), OsJAZ3 (LOC_Os08g33160), OsJAZ4 (LOC_Os09g23660), OsJAZ8 (LOC_Os09g26780), OsJAZ13 (LOC_Os10g25230), OsJAZ14 (LOC_Os10g25250), and OsJAZ12 (LOC_Os10g25290). Compared to ZH11, these genes showed varying levels of increased expression in osg, with OsJAZ11, OsJAZ7, OsJAZ8, OsJAZ13, and OsJAZ12 being significantly up-regulated by more than 2.5-fold (Fig 7C and 7D). Expression analysis indicated that OsJAZ11 is preferentially expressed at multiple stages during rice panicle development (Fig 7E). Previous studies have shown that overexpression of OsJAZ11 leads to reduced panicle length and increased grain width in rice plants, resulting in significant decreases in yield and seed setting rate [15]. These findings suggest that OSG genes may co-regulate rice grain width, panicle length, and seed setting rate together with the JAZ11 gene within the jasmonic acid signaling pathway.
(A) Schematic of the gibberellin signaling pathway; (B) Analysis of differentially expressed genes in the gibberellin signaling pathway; (C) Schematic of the jasmonic acid signaling pathway; (D) Analysis of differentially expressed genes in the jasmonic acid signaling pathway; (E) Expression analysis of OsJAZ11 at various stages of rice development. * indicates up-regulated DEGs.
The down-regulated differentially expressed genes are mainly enriched in the oxidative phosphorylation pathway and the starch and sucrose metabolism pathway (S8 and S9 Tables). More down-regulated genes are enriched in the starch and sucrose metabolism pathway, with a total of 7 genes showing decreased expression in this pathway (S8 Table). These genes include Os01g34890, OsTPP1 (LOC_Os02g44230), OsUgp1 (LOC_Os09g38030), OsUgp2 (LOC_Os02g02560), OsSSSⅢα (LOC_Os08g09230), OsBEⅡα (LOC_Os04g33460), and OsISA1 (LOC_Os08g40930). Compared to ZH11, most of these genes showed reduced expression levels in osg, decreasing to half or even lower (Fig 8A). These genes play crucial roles in plant carbohydrate metabolism. Expression analysis revealed that OsUgp2 is specifically expressed during the formation stage of mature rice pollen (P6) (Fig 8B and 8C). Previous studies have shown that silencing of OsUgp1 and OsUgp2 leads to pollen abortion in rice [16–18], suggesting that OSG gene might regulate rice pollen development by controlling the expression of OsUgp1 and OsUgp2.
(A) Analysis of differentially expressed genes in the gibberellin signaling pathway; (B-C) Expression analysis of the OsUgp2 gene at various stages of rice development. In Panel B, YS indicates 7-day-old Seedling, YR indicates Root of a 7-day-old Seedling, YL indicates young leaf, ML represents Mature leaf, SAM, P1-P6 represent inflorescence (P1, up to 3 cm) (P2, 3-5 cm) (P3, 5-10 cm) (P4, 10-15 cm) (P5, 15-22 cm) (P6, 22-30 cm), St indicates stigma, Ov represents Ovary, S1-S5 indicate Seed (S1, 0-2 dap) (S2, 3-4 dap) (S3, 5-10 dap) (S4, 11-20 dap) (S5, 21-29 dap). * indicates down-regulated DEGs.
To confirm the results of RNA-Seq analysis, we selected several up-regulated or down-regulated DEGs for qRT-PCR analysis on young panicles (0–3 cm). The results showed that four genes (SLRL1, OsPIL16, JAZ10, JAZ11) were significantly up-regulated and two genes (OsUgp1, and OsUgp2) were significantly down-regulated in the osg mutant (S7 Fig), which were consistent with the transcriptomic results.
Discussion
SRS3 gene involvement in rice fertility and grain development
Grain size is a crucial agronomic trait that determines rice yield and is an important target for genetic improvement in high-yield breeding. Therefore, identifying grain-size-related mutants in rice and elucidating the regulatory mechanisms of grain size are of great significance for high-yield rice breeding. This study obtained a mutant osg with stable inheritable phenotypes through mutation induced by 60Co-γ rays. Observaion of agricultural traits on osg revealed significantly reduced plant height and tiller number; notably shorter panicle length, no significant changes in primary branch number or grains per panicle, but a significant decrease in seed setting rate due to markedly reduced pollen viability. Regarding grain morphology, the grain length of osg was significantly reduced while grain width and thickness were significantly increased, leading to a significant decrease in thousand-grain weight.
Through map-based cloning, the OSG gene was localized to the 3.1–3.6 Mb (500kb) region on chromosome 5 (Fig 5A-5E). Analysis of all genes within this interval found that the coding region of the SRS3 gene could not be normally amplified in the osg mutant, and this was verified through the identification and analysis of gene-edited mutants of SRS3 (Fig 5F-5H). Studies have shown that the SRS3 gene is a member of the Kinesin 13 subfamily, encoding a protein composed of 819 amino acid residues with a motor domain and a coiled-coil structure [19]. The srs3 mutant exhibits reduced plant height, shortened panicle length, significantly decreased grain length, and significantly increased grain width and thickness. This gene regulates grain length by controlling lemma cell length [19–21]. These results indicate that the sequence variation of the SRS3 gene in the osg mutant led to its phenotypic variation, confirming SRS3 as the gene encoded by OSG.
These data showed that the loss of function of the SRS3 gene led to developmental abnormalities in rice plant height, panicle length, pollen fertility, and grain morphology, indicating that the SRS3 gene plays a critical regulatory role in these processes. However, the downstream regulated genes and molecular regulatory mechanisms remain unclear.
Function of kinesins in plant growth and development
The Kinesin-13 family comprises microtubule-binding proteins containing a conserved kinesin motor domain in their mid-region amino acid sequences. Kinesin-13s are motor proteins involved in microtubule (MT) depolymerization, which is crucial for regulating MT dynamics during mitosis, particularly chromosome segregation [22]. Kinesin-13A and Kinesin-13B are two types of internal-motor kinesins [23]. Kinesin-13A colocalizes with Golgi bodies in Arabidopsis cells, suggesting it is a special type of internal-motor kinesin [24]. In Arabidopsis, the Kinesin-13A protein preferentially localizes to cortical microtubules in secondary cell walls, preventing cell wall deposition via MT depolymerization, which is essential for the establishment of secondary cell walls in Arabidopsis [25]. In rice, OsKinesin13A is an active MT-depolymerizing enzyme mainly distributed in vesicles derived from the Golgi apparatus and localized on microtubules. Mutations in OsKinesin13A lead to poorly developed plants with smaller grains due to altered MT orientation, resulting in changed cellulose microfibril direction and cell elongation [20–22]. Research has also shown that BHS1 (SRS3) encodes the kinesin Kinesin-13A, which regulates grain length downstream of BRI, negatively regulating BR signaling and rice architecture [26].
In this study, the identified osg is a novel allele variant of SRS3. The mutation in OSG leads to aberrant coding of the motor protein Kinesin-13A, resulting in shortened grain length and increased grain width in rice. Combined with RNA-seq analysis, it was found that the abnormal expression of the OSG gene caused significant changes in the expression of numerous other genes, affecting pathways such as auxin signaling, gibberellin signaling, jasmonic acid signaling, and brassinosteroid (BR) signaling. These findings provide theoretical insights into the molecular regulatory network of Kinesin-13A in rice.
Possible mechanisms of SRS3 regulation on rice grain size and fertility
RNA-seq analysis showed that mutations in the OSG gene resulted in differential expression changes in several genes. Most up-regulated genes were enriched in pathways such as plant hormone signal transduction, while most down-regulated genes were enriched in starch and sucrose metabolism pathways.
Plant hormone signal transduction participates in almost all stages of plant growth and development. As the main component of rice yield, rice grains are influenced by multiple hormones during their formation and development [27]. Analysis of differentially expressed genes revealed that the OsPIL16 gene was up-regulated in gibberellin signaling pathways. Previous studies indicate that APG, encoded by this gene, acts as an antagonistic interaction factor of PGL1, inhibiting the increase in rice grain size by suppressing the expression of the PGL1 gene [14]. Additionally, multiple JAZ family genes in the jasmonic acid signaling pathway were up-regulated in the osg mutant, including OsJAZ11 and OsJAZ10. According to previous research, OsJAZ11 coordinates the expression of OsGW7 and MADS genes, regulating grain size and floret development [28,29], whereas overexpression of OsJAZ10 affects grain width and thickness [30].
Sucrose metabolism plays a key role in plant development, stress response, and yield formation. The synthesis and degradation of sugars drive the entire process of plant growth and development, with sucrose acting as a signaling molecule that regulates related gene expression and interacts with other genes, hormones, and defense signals [31]. Through the analysis of down-regulated genes enriched in the starch and sucrose metabolism pathway, we identified OsUgp2, primarily expressed in pollen and developmentally regulated. Silencing of OsUgp2 leads to the inability to form mature pollen, and pollen fertility correlates with the amount of OsUgp2 mRNA [17], suggesting that the reduced pollen fertility in the osg mutant may be associated with the down-regulation of OsUgp2 expression.
In summary, we speculate that the mutation in SRS3 affects multiple genes such as OsPIL16, OsJAZ11, OsJAZ10, and OsUgp2 involved in plant hormone signal transduction and starch and sucrose metabolism pathways, thereby collectively influencing the plant height, grain morphology, pollen fertility, and other traits of the osg mutant. These findings help understand the interrelationships between metabolic pathways and pathways during rice grain development and provide theoretical references for further analyzing and confirming downstream regulated genes of OSG, offering insights into the molecular regulatory networks of rice grain morphology and pollen fertility.
Supporting information
S1 Fig. Primer design of the SRS3 gene and mutant identification.
ATG and TAG represent the start codon and stop codon, respectively. The arrows (F1-18, R1-18) indicate the location of the primers for gene sequence analysis. The red vertical line represents the target site of gene editing. The primers of F9 and R9 were used for identifying the CRISPR/Cas9-mediated mutations in SRS3 in the T2 generation. The black letters indicate the DNA sequences with WT in the target site. The sequencing peak plot of the target site sequence is represented by the curve. The dotted lines indicate nucleotide deletions, and dark blue letters indicate the amino acids, and the asterisk (*) represents the stop codon generated in advance.
https://doi.org/10.1371/journal.pone.0338401.s001
(TIF)
S2 Fig. RNA-Seq analysis between ZH11 and osg.
(A) Gene expression analysis of ZH11 and osg samples; (B) Heat map of differentially expressed genes (DEG); (C-D) Some pathway classes in which the differentially expressed up-regulated (C) and down-regulated (D) genes were significantly enriched, respectively.
https://doi.org/10.1371/journal.pone.0338401.s002
(TIF)
S3 Fig. The heatmap of up-regulated genes in the auxin signaling pathway.
(A) Auxin signal transduction pathway diagram; (B-F) Heat map of up-regulated DEGs from AUX (C), ARF (C), IAA (D), GH3 (E), SAUR (F) family in RNA-seq between osg and ZH11.
https://doi.org/10.1371/journal.pone.0338401.s003
(TIF)
S4 Fig. The heatmap of up-regulated genes in the cytokinine, ehtylene, brassinosteroid, abscisic acid, salicylic acid signaling pathway.
(A-E) The diagram of biosynthesis or signal transduction pathway and heat map of zeatin biosynthesis (A), cysteine and methionine metabolism (B), brassinosteroid biosyuthesis (C), ABA signaling (D), and salicylic acid signaling pathway (E).
https://doi.org/10.1371/journal.pone.0338401.s004
(TIF)
S5 Fig. The heatmap of up-regulated genes in the MAPK signaling pathway.
(A) The diagram of MAPK signal transduction pathway in plants; (B-G) Heat map of up-regulated DEGs from EPF/EPFLs (B), SERKs (C), TMM (D), ER/ERL (E), MAP3Ks (F), MAP2Ks (G), MAPKs (H) families, respectively.
https://doi.org/10.1371/journal.pone.0338401.s005
(TIF)
S6 Fig. The heatmap of down-regulated DEGs in the oxidative phosphorylation pathway.
(A) The diagram of oxidative phosphorylation pathway process; (B-D) Heat map of down-regulated DEGs for NADH dehydrogenase (B), cytochrome c oxidase (C), ATP synthesis (D), respectively.
https://doi.org/10.1371/journal.pone.0338401.s006
(TIF)
S7 Fig. qRT-PCR analysis of DEGs in WT and osg. The samples for qRT-PCR analysis are young panicles from WT and osg; Values are means ± SD (standard deviations) of three independent biological replicates.
Two asterisks (**, P < 0.01) indicate extremely significant differences between WT and osg lines as determined by Student’s t-test.
https://doi.org/10.1371/journal.pone.0338401.s007
(TIF)
S2 Table. Screening of candidate genes in the localization region.
https://doi.org/10.1371/journal.pone.0338401.s009
(XLSX)
S3 Table. Down and Up-regulated DEGs between osg and ZH11 in RNA-Seq analysis.
https://doi.org/10.1371/journal.pone.0338401.s010
(XLSX)
S4 Table. KEGG enrichment pathways of up-regulated DEGs in osg vs ZH11.
https://doi.org/10.1371/journal.pone.0338401.s011
(XLSX)
S5 Table. KEGG enrichment pathways of down-regulated DEGs in osg vs ZH11.
https://doi.org/10.1371/journal.pone.0338401.s012
(XLSX)
S6 Table. Up-regulated genes in the plant hormone signaling pathway.
https://doi.org/10.1371/journal.pone.0338401.s013
(XLSX)
S7 Table. Up-regulated Genes in the MAPK signaling pathway.
https://doi.org/10.1371/journal.pone.0338401.s014
(XLSX)
S8 Table. Down DEGs in the starch and sucrose metabolism pathway.
https://doi.org/10.1371/journal.pone.0338401.s015
(XLSX)
S9 Table. Down DEGs in the oxidative phosphorylation pathway.
https://doi.org/10.1371/journal.pone.0338401.s016
(XLSX)
References
- 1. Chen WF, Xu ZJ, Zhang WZ, Zhang LB, Yang SR. Creation of new plant type and breeding of super high-yielding rice. Acta Agronomica Sinica. 2001;665–72, 681–2.
- 2. Song X-J, Ashikari M. Toward an optimum return from crop plants. Rice. 2008;1(2):135–43.
- 3. Xing Y, Zhang Q. Genetic and molecular bases of rice yield. Annu Rev Plant Biol. 2010;61:421–42. pmid:20192739
- 4. Kang YH, Fang Y, Li QL, Du M, Zhang CH. Research progress on rice grain shape genes and their breeding applications. Hybrid Rice. 2022;37:7–10.
- 5. Mao H, Sun S, Yao J, Wang C, Yu S, Xu C, et al. Linking differential domain functions of the GS3 protein to natural variation of grain size in rice. Proc Natl Acad Sci U S A. 2010;107(45):19579–84. pmid:20974950
- 6.
Sun XC. Genetic effects of rice grain shape genes GS3, qGL3 and GW2. Tibet University; 2020.
- 7. Zhang X-F, Yang C-Y, Lin H-X, Wang J-W, Xue H-W. Rice SPL12 coevolved with GW5 to determine grain shape. Sci Bull (Beijing). 2021;66(23):2353–7. pmid:36654120
- 8. Yi CD, Li W, Wang DR, Jiang W, Wang Y, Zhou Y. Effect of 1212-bp indel variation in rice GW5 gene on grain shape. Jiangsu J Agric Sci. 2016;32:1201–5.
- 9. Liu J, Chen J, Zheng X, Wu F, Lin Q, Heng Y, et al. GW5 acts in the brassinosteroid signalling pathway to regulate grain width and weight in rice. Nat Plants. 2017;3:17043. pmid:28394310
- 10. Wang S, Wu K, Yuan Q, Liu X, Liu Z, Lin X, et al. Control of grain size, shape and quality by OsSPL16 in rice. Nat Genet. 2012;44(8):950–4. pmid:22729225
- 11. Wang S, Li S, Liu Q, Wu K, Zhang J, Wang S, et al. The OsSPL16-GW7 regulatory module determines grain shape and simultaneously improves rice yield and grain quality. Nat Genet. 2015;47(8):949–54. pmid:26147620
- 12. Hu J, Wang Y, Fang Y, Zeng L, Xu J, Yu H, et al. A rare allele of GS2 enhances grain size and grain yield in rice. Mol Plant. 2015;8(10):1455–65. pmid:26187814
- 13. Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 2019;37(8):907–15. pmid:31375807
- 14. Heang D, Sassa H. Antagonistic actions of HLH/bHLH proteins are involved in grain length and weight in rice. PLoS One. 2012;7(2):e31325. pmid:22363621
- 15. Mehra P, Pandey BK, Verma L, Prusty A, Singh AP, Sharma S, et al. OsJAZ11 regulates spikelet and seed development in rice. Plant Direct. 2022;6(5):e401. pmid:35582630
- 16. Mu H, Ke J, Liu W, Zhuang C, Yip W. UDP-glucose pyrophosphorylase2 (OsUgp2), a pollen-preferential gene in rice, plays a critical role in starch accumulation during pollen maturation. Sci Bull. 2009;54(2):234–43.
- 17. Chen R, Zhao X, Shao Z, Wei Z, Wang Y, Zhu L, et al. Rice UDP-glucose pyrophosphorylase1 is essential for pollen callose deposition and its cosuppression results in a new type of thermosensitive genic male sterility. Plant Cell. 2007;19(3):847–61. pmid:17400897
- 18. Wang B, Fang R, Zhang J, Han J, Chen F, He F, et al. Rice LecRK5 phosphorylates a UGPase to regulate callose biosynthesis during pollen development. J Exp Bot. 2020;71(14):4033–41. pmid:32270203
- 19. Kitagawa K, Kurinami S, Oki K, Abe Y, Ando T, Kono I, et al. A novel kinesin 13 protein regulating rice seed length. Plant Cell Physiol. 2010;51(8):1315–29. pmid:20587735
- 20. Wu T, Shen Y, Zheng M, Yang C, Chen Y, Feng Z, et al. Gene SGL, encoding a kinesin-like protein with transactivation activity, is involved in grain length and plant height in rice. Plant Cell Rep. 2014;33(2):235–44. pmid:24170341
- 21. Deng ZY, Liu LT, Li T, Yan S, Kuang BJ, Huang SJ, et al. OsKinesin-13A is an active microtubule depolymerase involved in glume length regulation via affecting cell elongation. Sci Rep. 2015;5:9457. pmid:25807460
- 22. Ems-McClung SC, Walczak CE. Kinesin-13s in mitosis: key players in the spatial and temporal organization of spindle microtubules. Semin Cell Dev Biol. 2010;21(3):276–82. pmid:20109574
- 23. Reddy AS, Day IS. Kinesins in the Arabidopsis genome: a comparative analysis among eukaryotes. BMC Genomics. 2001;2:2. pmid:11472632
- 24. Lu L, Lee Y-RJ, Pan R, Maloof JN, Liu B. An internal motor kinesin is associated with the Golgi apparatus and plays a role in trichome morphogenesis in Arabidopsis. Mol Biol Cell. 2005;16(2):811–23. pmid:15574882
- 25. Oda Y, Fukuda H. Rho of plant GTPase signaling regulates the behavior of Arabidopsis kinesin-13A to establish secondary cell wall patterns. Plant Cell. 2013;25(11):4439–50. pmid:24280391
- 26. Zhang YL, Dong GJ, Chen F, Xiong EH, Liu HJ, Jiang YH, et al. The kinesin-13 protein BR HYPERSENSITIVE1 is a negative brassinosteroid signaling component regulating rice growth and development. TAG Theor Appl Genet Theor Angew Genet. 2022; 135: 1751–66.
- 27. Wen C-K, Zhao Y, Ruan Y-L. Editorial: hormonal control of important agronomic traits. Front Plant Sci. 2018;9:1504. pmid:30386360
- 28. Chongloi GL, Prakash S, Vijayraghavan U. Regulation of meristem maintenance and organ identity during rice reproductive development. J Exp Bot. 2019;70(6):1719–36. pmid:30753578
- 29. He Y, Hong G, Zhang H, Tan X, Li L, Kong Y, et al. The OsGSK2 kinase integrates brassinosteroid and jasmonic acid signaling by interacting with OsJAZ4. Plant Cell. 2020;32(9):2806–22. pmid:32586913
- 30. Hakata M, Kuroda M, Ohsumi A, Hirose T, Nakamura H, Muramatsu M, et al. Overexpression of a rice TIFY gene increases grain size through enhanced accumulation of carbohydrates in the stem. Biosci Biotechnol Biochem. 2012;76(11):2129–34. pmid:23132589
- 31. Feng YL, Yin F, Xu K, Jia XY, Zhou S, Ma C. Role of sucrose metabolism and signal transduction in plant development and stress response. J Nucl Agric Sci. 2021;35:2044–55.