Advances in Genetics and Breeding of Grain Shape in Rice

Abstract

Grain shape is a critical determinant of rice yield, quality, and market value. Recent advances in molecular biology, genomics, and systems biology have revealed a complex regulatory network governing grain development, integrating genetic loci, plant hormone signaling, transcriptional regulation, protein ubiquitination, epigenetic modifications, and environmental cues. This review summarizes key genetic components such as QTLs, transcription factors, and hormone pathways—including auxin, cytokinin, gibberellin, brassinosteroids, and abscisic acid—that influence seed size through regulation of cell division, expansion, and nutrient allocation. The roles of the ubiquitin–proteasome system, miRNAs, lncRNAs, and chromatin remodeling are also discussed, highlighting their importance in fine-tuning grain development. Furthermore, we examine environmental factors that impact grain filling and size, including temperature, light, and nutrient availability. We also explore cutting-edge breeding strategies such as gene editing, functional marker development, and wild germplasm utilization, along with the integration of multi-omics platforms like RiceAtlas to enable intelligent and ecological zone-specific precision breeding. Finally, challenges such as pleiotropy and non-additive gene interactions are discussed, and future directions are proposed to enhance grain shape improvement for yield stability and food security.

1. Introduction

Competition in the global rice market has become increasingly intense, driving producers and breeders to focus on improving grain shape to meet consumer preferences. For example, slender grains are generally considered more appealing due to their superior cooking qualities [1]. Consequently, grain shape improvement has emerged as a crucial factor for enhancing rice market value. Modern consumers are also more conscious of nutritional and health benefits. Improving grain shape not only affects rice appearance and taste but also influences nutritional content, including starch type and concentration. Research indicates that grain shape significantly correlates with starch composition and digestibility [2]. Advances in molecular biology and genomics have enabled breeders to identify genes associated with grain shape more effectively, using techniques like gene editing and marker-assisted selection to achieve desired traits [3,4]. These technological advancements help breeders to quickly respond to evolving market demands for high-quality rice. Additionally, agricultural policies in various countries actively encourage the production of premium rice. Governments may provide subsidies or establish support programs to motivate farmers, thus promoting research and practical implementation of grain shape improvements [5]. In summary, growing global demand for high-quality rice directly drives research and practices aimed at enhancing grain shape, helping breeders to satisfy consumer preferences, boost market competitiveness, and improve nutritional value. This trend significantly benefits rice production efficiency, economic returns, global food security, and nutritional improvement.

Grain shape in rice refers to the integrated traits of mature kernels, including length, width, thickness, and the degree of chalkiness, which not only serve as key determinants of yield components but also strongly influence grain appearance and market quality. According to the internationally accepted classification based on length-to-width ratio (L/W), grains can be categorized as extra-long slender (length ≥ 7.5 mm, L/W ≥ 3.0), long (6.6–7.5 mm, L/W 2.5–3.0), medium (5.5–6.6 mm, L/W 2.0–2.5), and short/bold (4.0–5.5 mm, L/W ≤ 2.0) [6]. Based on the chalkiness pattern, five types are commonly distinguished: milky-white, white-core, white-belly, white-based, and white-back [7]. Regardless of grain type, nutrient distribution within rice kernels is highly heterogeneous. At the macroscopic level, brown rice can be divided into the hull, bran layer (pericarp + aleurone), embryo (mainly the scutellum), and endosperm. The bran layer and embryo are enriched in proteins, lipids, ash, and B vitamins, whereas the endosperm consists predominantly of starch and contains the lowest concentrations of other nutrients. Spatially, mineral elements such as P, K, Ca, Mn, Fe, and Zn are preferentially accumulated in the aleurone layer and embryo (especially the scutellum), while their concentrations are lowest in the central endosperm [8,9]. Elemental distribution also shows distinct patterns: Zn is concentrated in the plumule and radicle of the embryo; Fe is most abundant in the scutellum and aleurone layer; K exhibits high intensity in the scutellum; Ca is preferentially distributed in the hull and scutellum; and Mn is localized in both the hull and embryo. During grain filling, a substantial proportion of minerals is remobilized from vegetative tissues, with approximately 60% of grain K and 20% of P derived from remobilization, while Fe and Zn can also be redistributed from the culm and sheath to the grain under flag leaf deficiency [6]. Ultimately, rice grain shape is shaped by the combined action of major negative regulators, polygenic networks, maternal genetic background, and environmental factors. Dissecting these genes and their interaction networks not only provides insights into the evolutionary and domestication history of grain morphology but also offers precise targets for modern molecular design breeding.

Grain size significantly influences plant adaptability and agricultural yield. In agricultural contexts, optimal grain size directly contributes to reproductive success, nutritional value, and market appeal [10]. In natural environments, grain size impacts plant dispersal capabilities, germination rates, and seedling survival, collectively influencing plant population dynamics and evolutionary paths [11]. While larger grains generally correlate with higher planting efficiency and yield, excessively large grains may result in nutritional imbalances and decreased quality [12]. Thus, the development of grain size is governed by a complex interplay of genetic and environmental factors. Understanding these molecular mechanisms is essential for enhancing crop yield and adaptation.

The entire life cycle of rice comprises five distinct stages: seedling, tillering, heading, flowering, and maturity (Figure 1). Seed germination is a critical stage initiating plant life, requiring appropriate conditions such as adequate moisture, temperature, and oxygen. Upon water absorption, seeds swell and rupture their coats, activating stored nutrients in the endosperm to support embryonic growth. Increased enzyme activity stimulates cell division and elongation; embryonic roots extend downward, establishing the root system, while shoots grow upward to form stems and leaves. After early growth, rice seedlings enter the tillering phase, increasing plant numbers and yield potential. Subsequently, plants transition to reproductive growth, marked by heading—a pivotal developmental stage. Following heading, rice enters the grain-filling period, during which grains accumulate vital nutrients such as starch and proteins, directly influencing rice quality and yield. Grain filling is regulated by internal genetic factors and external environmental conditions until grain maturity, signaling the completion of the rice lifecycle and the beginning of harvest [13]. During grain filling, cell wall invertases, SWEETs, and SUTs function coordinately to mediate sucrose unloading and allocation to the endosperm. Invertases hydrolyze transported sucrose into monosaccharides, SWEETs facilitate their efflux along concentration gradients, and SUTs actively import residual sugars. This coordinated “source–flow–sink” mechanism is critical for sustaining assimilate flux, thereby determining starch deposition and final grain yield [14,15].

Grain size formation in rice is a complex biological process intricately regulated by genetic factors, hormone signaling pathways, epigenetic modifications, and environmental cues [16], offering valuable insights for the advancement of modern rice breeding technologies. This review comprehensively examines the roles of quantitative trait loci (QTLs), non-coding RNAs (including miRNAs and lncRNAs), plant hormones (such as auxin and gibberellin), the ubiquitin–proteasome system, and environmental variables (including temperature, light, and nutrient availability) in controlling grain development and final grain size. These factors collectively determine grain size through mechanisms involving cell division, cell expansion, and nutrient accumulation and transport. Future research integrating multi-omics and systems biology approaches will further clarify these regulatory networks, providing a robust scientific foundation for crop improvement.

2. Genetic Regulation Mechanisms of Rice Grain Shape

2.1. Hormone Signaling Pathways

Hormone signaling pathways play critical regulatory roles in rice growth and development, influencing cell division, cell expansion, and nutrient accumulation (Figure 2, Table 1), all processes directly affecting grain size [17,18].

2.1.1. Auxin

Auxin is one of the most critical plant hormones, playing a central role in cell division and expansion through its response factors, Auxin Response Factors (ARFs) [19]. The functions of auxin during plant development are mediated by complex signaling pathways. Particularly in crops such as rice, auxin signaling pathways regulate grain size via various genes and molecular mechanisms. These genes indirectly or directly influence grain size through their involvement in cell division, cell expansion, and endosperm development [20].

In rice, the genes OsARF11 and OsARF6 have been extensively studied and shown to positively regulate grain size. Activated ARF transcription factors bind to the promoters of target genes, promoting the expression of genes associated with cell wall synthesis and expansion, consequently increasing cell volume, grain filling rate, and final grain size [21,22,23,24]. Research indicates that OsARF11 mutants exhibit reduced cell proliferation and inhibited cell expansion during grain development, resulting in smaller grain size and weight [21,25].

OsIAA1 is an auxin-responsive factor in rice, whose expression is regulated by auxin concentration. Auxin binds to the intracellular receptor TIR1, which facilitates the degradation of OsIAA1, enabling ARF transcription factors to function [26]. Studies demonstrate that OsIAA1 negatively regulates the auxin signaling pathway, controlling cell division and expansion [27,28]. OsARF2, another member of the ARF family, also plays an essential role in auxin signaling. Knockout studies have shown that OsARF2 mutants produce smaller and developmentally limited grains, highlighting its crucial role in grain development and cellular expansion [29,30]. OsPIN1, a key auxin transporter in rice, regulates auxin distribution between different cellular regions, significantly influencing plant morphology, including grain size. Research has revealed that OsPIN1 plays essential roles during embryo, endosperm, and seed coat development, affecting final grain size [31].

Additionally, grain development in rice relies on auxin signaling and cell cycle regulation genes. Auxin increases grain size by upregulating the expression of OsCyclinD1 and OsCDKA, significantly enhancing cell division rates and cell numbers during grain development [20,32].

2.1.2. Cytokinin

Cytokinins are essential phytohormones that govern various aspects of plant development, notably influencing the cell division cycle and cellular proliferation [27,28,29,30]. The homeostasis of cytokinins is tightly regulated by cytokinin oxidase/dehydrogenase (CKX) enzymes, which catalyze the irreversible degradation of active cytokinins [31]. Through modulating cytokinin levels, CKXs indirectly affect grain development by influencing the rate and extent of cell division. A reduction in cytokinin concentration is generally associated with decreased cell proliferation, often resulting in smaller grain size, whereas elevated cytokinin levels tend to enhance cell division and expansion, thereby contributing to increased grain size [32,33,34].

Rice possesses several OsCKX genes, including OsCKX1, OsCKX2, and OsCKX3. The genes OsCKX1 and OsCKX2 are involved in cytokinin degradation in rice, regulating grain development by reducing cytokinin levels. Studies indicate that overexpression of OsCKX1 or OsCKX2 results in smaller grains, whereas suppression of these genes enhances grain size. OsCKX3 also participates in cytokinin degradation and plays a significant role in rice growth and development. Its expression is modulated by environmental factors, consequently affecting grain size and yield [33].

Rice Response Regulators (OsRRs) are pivotal factors in cytokinin signaling, representing one of the most important transcription factor families within the cytokinin signaling pathway. OsRR genes regulate grain development through cytokinin signaling interactions, activating downstream gene expression and thus promoting cell division and expansion, ultimately impacting grain size [30,34,35]. In rice, the OsRR family includes several members such as OsRR1, OsRR2, and OsRR3. OsRR1 negatively regulates cytokinin signaling; its overexpression leads to smaller grains, whereas knockout mutants enhance cytokinin signaling, resulting in larger grains [30]. OsRR2 and OsRR3 also participate in cytokinin response and signaling, activating downstream target gene expression to influence grain development and size [36]. Additionally, OsLOG encodes a cytokinin synthesis enzyme, contributing to cytokinin synthesis and affecting grain growth and size [37].

In interactions between cytokinin and auxin, cytokinin typically dominates the early stages of cell division, while auxin primarily regulates later stages, such as cell expansion and grain filling. The coordination of these hormones is vital for normal grain development [38,39,40].

2.1.3. Gibberellin

Gibberellins (GAs) are pivotal regulators of plant growth and development, with pronounced roles during seed development. One of their key functions is to modulate cell elongation by promoting cell wall loosening and extensibility [41,42,43]. During the grain filling stage, GAs facilitate the expression of genes encoding cell wall-modifying proteins, such as expansins and xyloglucan endotransglucosylase/hydrolases, which enhance cell wall plasticity. This, in turn, promotes cellular expansion and contributes to the determination of final grain size [44].

OsGA20ox1 is a critical gene in the gibberellin biosynthesis pathway. Negative regulatory factors in the GA signaling pathway, such as OsSLR1, are also essential for regulating grain expansion. Overexpression of OsGA20ox1 inhibits these negative regulatory factors, enhancing cell expansion and influencing final grain size [41,45,46]. The OsGA3ox2 gene is also involved in GA biosynthesis, regulating GA activity, which affects rice grain development. Elevated OsGA3ox2 expression enhances GA activity, resulting in increased grain size [42]. DELLA proteins, negative regulators within the GA signaling pathway, typically restrict cell division and expansion by suppressing plant growth. Multiple DELLA genes in rice are regulated by GA. GA promotes DELLA protein degradation, removing restrictions on grain size [43,44,45,47,48].

The synergistic interaction among auxin, cytokinin, and gibberellin forms a complex regulatory network, promoting seed development at different growth stages [49].

2.1.4. Abscisic Acid

Abscisic acid (ABA) is one of the primary plant hormones, extensively involved in regulating rice growth, particularly playing essential roles in seed development, germination, and drought responses [45,50,51,52]. OsABI5 is a crucial gene within the ABA signaling pathway, significantly influencing the ABA response. Studies have found that OsABI5 regulates hormone balance within seeds, affecting their final size [52,53,54]. OsNCED is a key enzyme in ABA biosynthesis in rice, responsible for ABA synthesis. Modulating OsNCED expression can alter ABA production levels, thus influencing rice grain size [55]. OsCYP707A, involved in ABA metabolism, facilitates ABA degradation. Research indicates that ABA levels are closely related to OsCYP707A activity, thereby affecting grain development and size [56].

2.1.5. Brassinosteroids

Brassinosteroids (BRs) are a class of plant hormones that play crucial roles in plant growth and development. Recent studies have demonstrated that brassinosteroids significantly influence not only the development of plant roots, stems, and leaves but also regulate seed size, shape, and quality [57,58,59]. Particularly in rice, brassinosteroids significantly affect grain size and weight by regulating processes such as cell division, cell expansion, and endosperm development [60,61,62].

BRI1 is a key receptor in the BR signaling pathway, responsible for perceiving brassinosteroids and initiating downstream signaling, thus influencing plant growth and development. BRI1 directly regulates grain size by modulating cell division and expansion. Studies have shown that rice grains are significantly smaller in BRI1 mutants, highlighting BRI1’s critical role in controlling grain size in rice [63,64]. OsBZR1, a transcription factor within the BR signaling pathway, enters the nucleus after activation by BRI1, where it regulates genes involved in seed development. Overexpression of OsBZR1 increases grain size, while its suppression significantly reduces grain size. These findings suggest that brassinosteroids promote rice grain development through OsBZR1 regulation [65,66]. OsGSK3, a kinase within the BR signaling pathway, regulates the cell cycle and proliferation. During BR signaling, OsGSK3 directly influences rice grain development by modulating cell growth and division. Functional loss of OsGSK3 leads to smaller rice grains and impaired embryo development [67]. Studies further indicate that OsGSK3 expression correlates closely with BR signaling strength. When the BR signaling pathway is activated, OsGSK3 enhances cell division and expansion, thereby promoting grain enlargement [68,69,70]. OsGSW3.2 controls grain size by inhibiting cell proliferation in rice husks. Knockout mutants of OsGSW3.2 exhibit significantly shortened and narrowed husk cells, resulting in increased cell numbers and larger grains. OsGSW3.2 interacts with OsGSK4, an important component of the BR signaling pathway. This interaction may regulate grain size by influencing cell cycles and proliferation through modulation of BR signaling [71].

2.1.6. Ethylene

Ethylene is another critical hormone involved in plant growth and development, regulating various physiological processes, especially those related to grain size development [72,73,74,75]. The action of ethylene in plants begins with its binding to ethylene receptors.

OsACS1 is a key enzyme gene responsible for ethylene biosynthesis in rice, catalyzing the synthesis of 1-aminocyclopropane-1-carboxylate (ACC) and promoting ethylene production. Ethylene, a gaseous hormone, is vital for plant growth and development. Overexpression of OsACS1 can lead to excessive ethylene production, promoting seed maturation and development, potentially increasing grain size. Studies show enhanced ethylene synthesis promotes rice grain cell expansion, thereby increasing grain size [76,77]. OsETR1 is an ethylene receptor gene in rice, part of the ethylene receptor family. Ethylene binding to OsETR1 initiates downstream signaling pathways. OsETR1 perceives ethylene molecules and activates signaling cascades, leading to expression of downstream regulatory factors. Excessive ethylene through OsETR1 accelerates seed aging, affecting grain quality and size. Mutants exhibit incomplete seed development and smaller grains. Similarly, OsERS1, another member of the ethylene receptor family, participates in ethylene perception and signaling. Studies demonstrate OsERS1 impacts rice grain size by influencing cell division and expansion via ethylene signaling [78]. OsEIN2 serves as a mediator between ethylene receptors and downstream effector genes in ethylene signaling pathways. After ethylene binds to receptors, OsEIN2 transmits signals to nuclear transcription factors, activating or repressing downstream gene expression. Overexpression of OsEIN2 in rice enhances grain cell expansion, increasing overall grain size, whereas its mutants produce smaller grains [79,80]. OsCTR1, a negative regulator within the ethylene signaling pathway, suppresses excessive ethylene responses. By interacting with ethylene receptors, OsCTR1 moderates ethylene signaling. Studies indicate that loss of OsCTR1 function leads to exaggerated ethylene signaling, promoting cell expansion and resulting in larger grains [81,82].

Table 1. Hormone-related genes in rice and their functions.

Gene	Type	Function	Reference
OsARF2	Transcription factor	An auxin response factor that positively regulates grain size	[29,30]
OsARF11	Transcription factor	An auxin response factor that positively regulates grain size	[21,25]
OsIAA1	Member of the rice Aux/IAA family	An auxin response factor that negatively regulates the auxin signaling pathway	[26]
OsPIN1	Auxin transport protein	Determines rice grain size by “establishing and maintaining the auxin (IAA) concentration gradient between vascular bundles and endosperm”	[31]
OsCyclinD1	Cell cycle regulatory protein	Positively regulates grain size	[20,32]
OsCDKA	Cell cycle regulatory protein	Positively regulates grain size	[20,32]
OsCKX1	Member of the cytokinin oxidase/dehydrogenase gene family	Negatively regulates grain length and width by degrading active cytokinins	[33]
OsCKX2	Member of the cytokinin oxidase/dehydrogenase gene family	Negatively regulates grain length and width by degrading active cytokinins	[33]
OsCKX3	Member of the cytokinin oxidase/dehydrogenase gene family	Negatively regulates grain length and width by degrading active cytokinins	[33]
OsRR1	Member of the type-A cytokinin response regulator family	Negatively regulates grain size by repressing cytokinin signaling	[30]
OsRR2	Member of the type-A cytokinin response regulator family	Negatively regulates grain size by repressing cytokinin signaling	[36]
OsRR3	Member of the type-A cytokinin response regulator family	Negatively regulates grain size by repressing cytokinin signaling	[36]
OsLOG	Cytokinin activation/biosynthesis gene	Promotes cytokinin biosynthesis, thereby positively regulating grain size	[37]
OsGA20ox1	Key late-stage enzyme gene in gibberellin biosynthesis	Negatively regulates grain size by reducing gibberellin activity	[41,45,46]
OsGA3ox2	Key late-stage enzyme gene in gibberellin biosynthesis	Positively regulates grain size by enhancing gibberellin activity	[42]
OsABI5	bZIP transcription factor gene in the ABA signaling pathway	Negatively determines grain size by repressing the cell cycle, expansins, and sugar transport	[52,53,54]
OsNCED	Member of the 9-cis-epoxycarotenoid dioxygenase gene family	Positively regulates ABA biosynthesis, promoting cell differentiation and early grain filling	[55]
OsCYP707A	Member of the cytochrome P450 monooxygenase A-type family	Negatively regulates ABA biosynthesis, preventing premature termination of grain filling	[56]
OsBRI1	Key receptor in the brassinosteroid (BR) signaling pathway	Perceives brassinosteroids and initiates downstream signaling, thereby positively regulating grain size	[63,64]
OsBZR1	Core transcription factor in the BR signaling pathway	Functions with OsBRI1 to positively regulate grain size	[65,66]
OsGSK3	Key kinase in the BR signaling pathway	Positively regulates grain size by controlling cell growth and division	[67]
OsACS1	Key enzyme gene in ethylene biosynthesis	Positively regulates ethylene biosynthesis, thereby influencing grain size	[76,77]
OsETR1	Member of the ethylene receptor gene family	Perceives ethylene and activates signaling cascades, inducing downstream regulatory factors to control grain size	[78]
OsERS1	Member of the ethylene receptor gene family	Regulates cell division and expansion via ethylene signaling, thereby influencing grain size	[78]
OsEIN2	Core transducer in the ethylene signaling pathway	Transmits signals to nuclear transcription factors, activating or repressing downstream genes, thus positively regulating grain size	[79,80]
OsCTR1	Negative regulator RAF-like serine/threonine protein kinase gene in the ethylene signaling pathway	Suppresses abnormal responses triggered by excessive ethylene, thereby negatively regulating grain size	[81,82]

Table 1. Hormone-related genes in rice and their functions.

Gene	Type	Function	Reference
OsARF2	Transcription factor	An auxin response factor that positively regulates grain size	[29,30]
OsARF11	Transcription factor	An auxin response factor that positively regulates grain size	[21,25]
OsIAA1	Member of the rice Aux/IAA family	An auxin response factor that negatively regulates the auxin signaling pathway	[26]
OsPIN1	Auxin transport protein	Determines rice grain size by “establishing and maintaining the auxin (IAA) concentration gradient between vascular bundles and endosperm”	[31]
OsCyclinD1	Cell cycle regulatory protein	Positively regulates grain size	[20,32]
OsCDKA	Cell cycle regulatory protein	Positively regulates grain size	[20,32]
OsCKX1	Member of the cytokinin oxidase/dehydrogenase gene family	Negatively regulates grain length and width by degrading active cytokinins	[33]
OsCKX2	Member of the cytokinin oxidase/dehydrogenase gene family	Negatively regulates grain length and width by degrading active cytokinins	[33]
OsCKX3	Member of the cytokinin oxidase/dehydrogenase gene family	Negatively regulates grain length and width by degrading active cytokinins	[33]
OsRR1	Member of the type-A cytokinin response regulator family	Negatively regulates grain size by repressing cytokinin signaling	[30]
OsRR2	Member of the type-A cytokinin response regulator family	Negatively regulates grain size by repressing cytokinin signaling	[36]
OsRR3	Member of the type-A cytokinin response regulator family	Negatively regulates grain size by repressing cytokinin signaling	[36]
OsLOG	Cytokinin activation/biosynthesis gene	Promotes cytokinin biosynthesis, thereby positively regulating grain size	[37]
OsGA20ox1	Key late-stage enzyme gene in gibberellin biosynthesis	Negatively regulates grain size by reducing gibberellin activity	[41,45,46]
OsGA3ox2	Key late-stage enzyme gene in gibberellin biosynthesis	Positively regulates grain size by enhancing gibberellin activity	[42]
OsABI5	bZIP transcription factor gene in the ABA signaling pathway	Negatively determines grain size by repressing the cell cycle, expansins, and sugar transport	[52,53,54]
OsNCED	Member of the 9-cis-epoxycarotenoid dioxygenase gene family	Positively regulates ABA biosynthesis, promoting cell differentiation and early grain filling	[55]
OsCYP707A	Member of the cytochrome P450 monooxygenase A-type family	Negatively regulates ABA biosynthesis, preventing premature termination of grain filling	[56]
OsBRI1	Key receptor in the brassinosteroid (BR) signaling pathway	Perceives brassinosteroids and initiates downstream signaling, thereby positively regulating grain size	[63,64]
OsBZR1	Core transcription factor in the BR signaling pathway	Functions with OsBRI1 to positively regulate grain size	[65,66]
OsGSK3	Key kinase in the BR signaling pathway	Positively regulates grain size by controlling cell growth and division	[67]
OsACS1	Key enzyme gene in ethylene biosynthesis	Positively regulates ethylene biosynthesis, thereby influencing grain size	[76,77]
OsETR1	Member of the ethylene receptor gene family	Perceives ethylene and activates signaling cascades, inducing downstream regulatory factors to control grain size	[78]
OsERS1	Member of the ethylene receptor gene family	Regulates cell division and expansion via ethylene signaling, thereby influencing grain size	[78]
OsEIN2	Core transducer in the ethylene signaling pathway	Transmits signals to nuclear transcription factors, activating or repressing downstream genes, thus positively regulating grain size	[79,80]
OsCTR1	Negative regulator RAF-like serine/threonine protein kinase gene in the ethylene signaling pathway	Suppresses abnormal responses triggered by excessive ethylene, thereby negatively regulating grain size	[81,82]

2.2. Transcriptional Regulatory Pathways

Transcription factors play central roles in plant growth and development, particularly in regulating grain size. Research indicates that transcription factors regulate grain size primarily through several pathways, including the SPL family, WRKY family, and NAC family, among others (Figure 3).

The SPL family, a class of plant-specific transcription factors, plays key roles in plant growth and development, particularly in regulating grain size [83]. OsSPL12 regulates grain width in rice by binding to the promoter region of the GW5 gene [84]. OsSPL13, another plant-specific transcription factor, positively regulates rice hull width by increasing hull cell size, thereby positively impacting grain length and yield [85]. OsSPL14 is a key transcription factor in rice, negatively regulating grain size by influencing cell division and expansion through auxin-related metabolism and signaling pathways, affecting grain length and weight [86,87,88]. OsSPL16, an SBP-domain transcription factor, regulates grain width in rice. Increased expression of the GW7 gene is associated with more slender grains, while OsSPL16 directly binds to and suppresses GW7 promoter expression, thus modulating hull width [89,90]. OsSPL18 influences grain width and thickness by directly affecting cell proliferation during rice hull development [91].

The WRKY family is another highly conserved transcription factor family in plants, extensively involved in responses to environmental stress and developmental regulation [92]. Specifically, WRKY53 plays crucial roles in regulating grain size and plant structure by interacting with the brassinosteroid (BR) and mitogen-activated protein kinase (MAPK) signaling pathways [93]. OsWRKY78, specifically in rice, promotes grain enlargement by regulating genes associated with cell division and expansion. Studies suggest OsWRKY78 interacts with auxin signaling pathways, regulating early cell proliferation and differentiation [94].

The NAC family is another plant-specific transcription factor family involved in stress responses and developmental processes [95,96]. OsNAC23 indirectly regulates plant growth and development, including grain filling and yield formation, by influencing sugar utilization and allocation [97]. OsNAC024 and OsNAC025 directly bind to the promoters of the GW2, GW5, and D11 genes, known regulators of rice grain weight. OsMED15a enhances OsNAC024 binding to these promoters through its KIX domain, potentially boosting transcriptional activity. In OsMED15a RNAi transgenic rice, expression of GW2, GW5, and D11 is downregulated, resulting in shorter and wider seeds [98]. SS1/ONAC025, an ER- and nucleus-localized transcriptional repressor, regulates seed size by influencing cell elongation [99]. Compared to the wild-type, osnac129 mutants exhibit increased grain length, grain weight, and plant height, while overexpression lines show reduced grain width, grain weight, and plant height, indicating OsNAC129 negatively regulates these traits. OsNAC129 modulates rice grain size and starch synthesis by repressing expression of genes like OsPGL1 and OsPGL2; its functional loss alters starch content, affecting grain quality [100].

Additionally, numerous other transcription factors significantly influence grain size. GS2 (OsGRF4) directly binds to the SUG1 promoter, activating its expression. SUG1 interacts with transcription factors such as OsBZR1, OsMADS56, and OsSPL13, integrating gibberellin (GA) and brassinosteroid (BR) signaling pathways to promote cell expansion and regulate grain size [101]. GS2 interacts directly with IPA1 to co-regulate downstream DEP1 expression, affecting rice panicle type and grain number [102]. DWARF AND LOW-TILLERING (DLT), a GRAS family member, acts as a positive regulator in the BR signaling pathway, interacting with WG3 to influence plant architecture and grain size [103]. OVATE family proteins, including OFP3 and OFP8, are transcription factors in the BR signaling pathway, regulating leaf angle and grain size [104,105].

In rice, editing OsSPL16 resulted in mutant plants exhibiting abnormal carbohydrate allocation, poor grain filling, and reduced grain size [106]. RGA1 regulates rice grain size through multiple mechanisms, including involvement in G-protein signaling, regulation of cell proliferation and expansion, interactions with other grain-size-related genes, influencing grain filling, and responding to environmental and hormonal signals. RGA1 controls grain size, rice quality, and seed germination in minimal rice plants [107,108]. OsbZIP47, a bZIP transcription factor, has its transcriptional activation inhibited by WG1. GW2-mediated WG1 degradation relieves WG1’s repression, activating OsbZIP47 and restricting grain growth [109]. OsMYB73 regulates rice grain size by controlling endosperm storage substance accumulation and auxin biosynthesis signaling pathways. In cooperation with OsNF-YB1, OsMYB73 binds promoters of downstream genes (e.g., OsISA2, OsLTPL36, OsYUC11), regulating starch, lipid biosynthesis, and auxin synthesis, thus impacting endosperm development and grain chalkiness. OsMYB73 mutations increase grain length but cause chalkiness, while its overexpression reduces grain size and increases chalkiness, highlighting its crucial role in regulating grain size and quality [110]. OsZHD1 and OsZHD2, zinc finger homeodomain transcription factors in rice, may influence grain shape by directly binding to promoters of target genes [111].

These transcription factor families are involved in complex regulatory networks controlling rice grain size through various signaling pathways and interactions. They finely tune multiple aspects of grain development in response to plant hormones, environmental factors, and developmental signals [81].

2.3. Multigenic Interactions and Regulatory Networks

2.3.1. Multigenic Regulatory Networks

Seed size is a critical agronomic trait in plant growth and development, intricately regulated by numerous quantitative trait loci (QTL). Extensive QTL studies have demonstrated that seed size across various crops results from the collaborative actions of multiple genetic loci (Figure 3) [112,113,114]. In rice, several key genes such as GW2, GS3, qGL3, and GW5 significantly influence seed development and grain shape [115].

GW2 functions as a negative regulator through ubiquitination, affecting cell division and expansion by reducing cellular proliferation, consequently decreasing grain width and weight [109,116]. GS3 regulates grain shape by modulating glume cell morphology, with specific alleles leading to elongated but narrower glume cells [117,118]. qGL3, an essential QTL in rice, encodes a putative protein phosphatase containing kelch-repeat domains homologous to BSU1, a positive regulator of brassinosteroid (BR) signaling in Arabidopsis [68]. GW5 negatively regulates rice grain width, with loss-of-function mutations causing increased grain width and weight [119]. qSW5 and GW2 positively regulate GS3 expression, although qSW5 expression can be suppressed by GW2-mediated transcriptional inhibition [120]. GIF1 expression is positively regulated by qSW5 and negatively regulated by GW2 and GS3. In natural rice populations, an epistatic interaction exists between qSW5 and GS3, where qSW5’s effect on seed length is masked by GS3 alleles, and conversely, GS3’s effect on seed width is masked by qSW5 alleles. In the Kasalath cultivar, reduced ORF1 expression at the qSW5 locus increases grain width [121]. Variations in qGL3 alleles influence its capacity to dephosphorylate OsGSK3, thus regulating grain length and yield. Loss-of-function mutations in GS3 and qGL3 markedly enhance grain length, with qGL3 exerting a stronger effect [122].

OsMADS1 affects grain size by regulating glume cell proliferation; loss of OsMADS1 function reduces cell numbers and sizes, leading to smaller grains. OsMADS1 directly controls several downstream genes, such as OsMADS17 and OsMADS55, pivotal in grain development. Additionally, interaction with OsNF-YC1 significantly enhances the transcriptional activation activity of OsMADS1. GS3 and DEP1, both Gγ subunits, directly interact with the conserved keratin-like domain of OsMADS1, acting as co-factors to enhance its transcriptional activity, thereby synergistically regulating grain size and shape [123].

GW5 interacts with WRKY53, potentially suppressing WRKY53-induced activation of SGW5, negatively regulating grain width. SGW5 positively regulates grain width by influencing spikelet hull cell division and size. The WRKY53 transcription factor binds to the W-box motif in the SGW5 promoter, modulating its expression and grain width. GW5, WRKY53, and SGW5 form a regulatory module affecting grain width, with GW5 suppressing SGW5 expression and consequently reducing grain width. Conversely, deletion of the WRKY53 binding site enhances SGW5 expression and grain width [124].

qGL3 interacts with OsGSK3, modulating its phosphorylation state and influencing OsBZR1 phosphorylation and subcellular localization. Variations among qGL3 alleles alter their ability to dephosphorylate OsGSK3, thereby impacting grain length and yield [68]. Moreover, GW5 promotes seed length and width by regulating grain expansion and increasing cell volume [119].

The TGW2 gene encodes CELL NUMBER REGULATOR 1 (OsCNR1). Overexpression of TGW2 results in reduced grain width and thousand-grain weight, whereas CRISPR/Cas9-mediated knockout of TGW2 enhances these traits. TGW2 interacts with KRP1, a cell cycle regulatory protein, whose overexpression similarly reduces grain width and thousand-grain weight [125].

Integrating QTL models with genome-wide association studies (GWAS) has further elucidated the complex regulatory networks between QTLs and genes, providing fundamental insights essential for crop yield and quality improvement [126].

2.3.2. Conservation and Functional Divergence of Genes Across Species

Although genes regulating seed size vary across crops, there is notable genetic conservation. For instance, similar genes such as GS3 and GW2 control comparable processes of cell proliferation and expansion in both rice and wheat [116,118]. In rice, GS3 negatively regulates cell division rates and endosperm development, and its function is conserved across other cereal crops [127]. GW2 in rice and wheat modulates cell wall formation, influencing grain size in both species [128,129].

Despite this conservation, gene expression patterns and functional mechanisms may differ among species. For example, TaSPL13 in wheat predominantly regulates carbohydrate allocation, whereas SPL genes in rice primarily control endosperm expansion [85,130]. Cross-species analyses of gene function provide valuable insights for breeding strategies, especially the identification and utilization of conserved genes. These findings lay theoretical foundations for gene editing and molecular marker-assisted breeding [131,132]. Advancements in genomics will further clarify functional divergences of conserved genes across species, offering promising opportunities for enhancing seed size and crop productivity.

2.4. Mechanisms of Protein Metabolism and Ubiquitination Regulation

2.4.1. Functions of the Ubiquitin–Proteasome System (UPS) in Seed Development

The ubiquitin–proteasome system (UPS) plays a central role in maintaining protein homeostasis in plant cells by selectively degrading misfolded, damaged, or regulatory proteins [133]. This tightly controlled proteolytic pathway is especially important during seed development, where it modulates key signaling and metabolic pathways that ultimately affect seed size and growth dynamics [134,135]. Ubiquitination proceeds through a sequential cascade involving E1 activating enzymes, E2 conjugating enzymes, and E3 ubiquitin ligases, the latter conferring substrate specificity by recognizing and tagging target proteins for degradation via the 26S proteasome [136]. The functional diversity and large number of E3 ligases in plants allow for highly selective and dynamic regulation of protein turnover during developmental processes.

In rice, the E3 ligase GW2 negatively regulates seed size by ubiquitinating proteins involved in cell division and expansion. Mutations in GW2 enhance cellular proliferation rates, resulting in increased grain size, highlighting its negative regulatory role [116]. GW2 directly influences endosperm cell proliferation through the targeted degradation of proteins involved in cellular expansion processes [137]. GW9 has been identified as a ubiquitination target of GW2; together, they regulate rice grain dimensions [138]. Another E3 ligase, OsPUB9, interacts with GSK2, preventing its own degradation while promoting degradation of GSK2. Additionally, OsPUB9 interacts with OsOFP8, stabilizing itself but inducing degradation of OsOFP8, collaboratively influencing rice grain size [139]. Furthermore, OsDA1, an E3 ligase family member, significantly impacts seed development by regulating endosperm cell size and proliferation during grain filling [140]. A similar DA1 E3 ligase identified in Arabidopsis also controls seed development through ubiquitination-mediated protein turnover [133,140]. This regulatory mechanism shows high conservation across various crops, including rice, wheat, and soybean, indicating the broad applicability of ubiquitination systems in controlling seed size [134,135]. Moreover, ubiquitination enables plants to adapt to environmental stresses, such as elevated temperatures or nutrient deficiencies, by accelerating the degradation of stress-responsive proteins, thus stabilizing seed development and yield under adverse conditions [136].

2.4.2. Molecular Mechanisms Linking Protein Degradation and Signal Transduction

Beyond its canonical role in protein turnover, ubiquitination serves as a key regulatory mechanism in hormone signaling pathways that govern seed development. By modulating the stability and activity of signaling components, the ubiquitin–proteasome system (UPS) enables precise temporal and spatial control of hormone responses. In the brassinosteroid (BR) signaling pathway, for example, the transcription factor OsBZR1, which regulates endosperm and embryonic cell expansion, is subject to ubiquitin-mediated degradation. This regulation prevents excessive cell enlargement and contributes to energy homeostasis during seed development [71]. In rice, OsUBC13, a ubiquitin-conjugating enzyme, has been shown to influence cytokinin signaling by modulating the turnover of key pathway components, thereby coordinating endosperm cell proliferation and optimizing grain filling processes [141,142].

In auxin signaling, the UPS precisely regulates the activity of Auxin Response Factors (ARFs), orchestrating cell division and differentiation [29]. TIR1 and AFB proteins function as auxin receptors, interacting with ubiquitin ligases to rapidly degrade AUX/IAA repressors, thereby activating ARFs and promoting dynamic adjustments of auxin signaling during different stages of seed development [26,143]. This ensures precise modulation of cell proliferation and expansion, tailored to the developmental needs of seeds.

Moreover, ubiquitination is crucial in plant responses to environmental stress. Under drought or saline conditions, plants rapidly degrade non-essential stress-response proteins via ubiquitination, conserving energy and improving survival chances [144]. Studies in rice seeds demonstrate that the UPS facilitates degradation of negative regulators in stress response pathways, optimizing energy utilization efficiency, enhancing seed viability, and increasing final yields [145,146]. These adaptive responses illustrate the versatile roles of the UPS during seed development under fluctuating environmental conditions.

In summary, protein metabolism and ubiquitination significantly influence seed size regulation. Selective protein degradation mediated by the ubiquitin–proteasome system (UPS) precisely modulates essential proteins involved in cell proliferation, division, and expansion, integrated with hormone signaling pathways to form complex, multi-layered regulatory networks [136]. Exploiting the adaptability and specificity of UPS pathways presents promising avenues for crop improvement, potentially leading to enhanced seed yield and quality through targeted modifications of ubiquitination mechanisms.

2.5. Epigenetic and Non-Coding RNA Regulatory Mechanisms

2.5.1. Epigenetic Modifications in Seed Development

Epigenetic modifications, including DNA methylation, histone modification, and chromatin remodeling, are among the most critical molecular mechanisms influencing seed size and development in plants (Table 2). DNA methylation, a chemical modification of DNA, alters gene expression without changing the underlying DNA sequence [147]. As a pivotal epigenetic regulatory mechanism, DNA methylation influences the expression of genes essential for plant growth and development. In rice, the gene OsDDM1b modulates gene activation and silencing by regulating genomic DNA methylation status. Loss-of-function mutations in OsDDM1b significantly reduce overall DNA methylation levels, leading to activation of several genes that negatively regulate cell division, ultimately resulting in reduced seed size [148]. Moreover, DNA methylation is closely associated with environmental adaptability and seed development. Under adverse environmental conditions, changes in methylation patterns can rebalance gene expression to ensure proper seed development despite environmental stress [148].

Histone modification represents another crucial epigenetic mechanism, with histone acetylation and methylation playing particularly important roles in regulating gene expression [141,149]. Histone modifications, mediated by specific enzymes, include methylation, acetylation, phosphorylation, adenylation, ubiquitination, and ADP-ribosylation [149]. OsHDT701, an important histone deacetylase in rice, regulates gene acetylation levels during grain development to synchronize cell division and germination. Loss of OsHDT701 activity leads to premature endosperm cell expansion and division, resulting in poor grain filling, highlighting the significance of histone deacetylation in grain development [142].

Epigenetic modifications are critically involved during the seed filling phase, especially the histone marks H3K27me3 and H3K4me3, commonly associated with gene activation states, significantly influencing transcription of grain development genes. Genes closely associated with grain development exhibit elevated H3K27me3 modification levels during seed filling [150,151]. H3K4me3, an activating histone methylation mark, enhances transcriptional activity, thereby promoting grain growth and filling. Activation of these genes ensures both grain expansion and nutrient accumulation, substantially impacting crop yield and quality [152]. A deeper understanding of the regulatory roles of H3K27me3 and H3K4me3 in grain development will facilitate genetic improvement and molecular breeding strategies aimed at producing higher-yielding and superior-quality crop varieties.

2.5.2. Regulatory Roles and Biological Significance of miRNAs and lncRNAs

Non-coding RNAs, particularly microRNAs (miRNAs) and long non-coding RNAs (lncRNAs), significantly contribute to gene expression regulation and phenotypic determination during seed development in plants [23,153,154,155]. miRNAs finely regulate gene expression at the post-transcriptional level by binding target mRNAs and inhibiting their translation [156]. For example, the miRNA156 family regulates seed size by targeting members of the SPL gene family, influencing cell division and expansion [88,157]. SPL genes play critical roles in auxin signaling pathways, and miRNA156 controls plant responses to auxin by modulating SPL expression, thereby regulating the rate of cell expansion [87,88].

Compared to miRNAs, lncRNAs exhibit greater structural complexity and employ diverse regulatory mechanisms, including transcriptional, pre-transcriptional, post-transcriptional, and epigenetic controls. They interact with protein complexes and modulate chromatin structures, thereby regulating gene expression at multiple levels [158,159]. For instance, in rice, the lncRNA TCONS_00023703 is highly expressed at crucial stages of seed development, precisely regulating transcriptional activity of specific genes involved in cell division and proliferation during grain filling by interacting with chromatin remodeling proteins [155]. This precise control mechanism decisively impacts seed development, influencing seed size, shape, and nutrient accumulation, thereby significantly affecting crop yield and quality [159]. Understanding lncRNA-mediated regulation provides valuable insights and potential strategies for crop improvement and enhanced adaptability.

2.5.3. Synergistic Effects of Non-Coding RNAs and Epigenetic Modifications

Non-coding RNAs, including miRNAs and lncRNAs, act synergistically with epigenetic modifications to regulate gene expression. Specific lncRNAs can interact with histone-modifying enzymes, such as H3K27 methyltransferases, to regulate chromatin states in particular genomic regions at specific developmental stages, thereby determining gene activation or silencing [160]. This mechanism is crucial for regulating gene expression during cellular differentiation and tissue development. Similarly, miRNAs indirectly influence traits like seed size by targeting the expression of epigenetic regulators. By binding and destabilizing or inhibiting translation of specific mRNAs, miRNAs reduce the levels of epigenetic factors, thus adding complexity to their regulatory roles [156,161,162,163].

2.5.4. Role of Chromatin Remodeling in Seed Size Regulation

Chromatin remodeling is a fundamental epigenetic mechanism controlling gene expression by altering chromatin accessibility and openness. Chromatin remodelers, such as the SWI/SNF complex, play essential roles in seed development by modifying chromatin states at targeted genomic regions, thereby regulating the expression of seed-specific genes [164]. An open chromatin structure facilitates the transcription of genes related to grain filling and cell division, whereas a condensed chromatin state suppresses unnecessary gene activation [165,166,167].

Table 2. Epigenetic and non-coding RNA factors in rice and their roles in grain size regulation.

Gene	Type	Function	Reference
OsDDM1b	Chromatin remodeling enzyme gene of the WI2/SNF2 family	Loss-of-function mutation significantly reduces overall DNA methylation, activating multiple genes that negatively regulate cell division, ultimately leading to smaller grain size	[148]
OsHDT701	Histone deacetylase	Loss of activity causes premature expansion and division of endosperm cells, resulting in poor grain filling	[142]
H3K27me3	Trimethylation of lysine 27 on histone H3	Acts as an epigenetic “brake”; its level and distribution determine the balance between cell proliferation and grain filling, directly regulating grain size	[150,151]
H3K4me3	Trimethylation of lysine 4 on histone H3	Enhances transcriptional activity, promoting grain growth and filling	[152]
miRNA156	MicroRNA (miRNA)	Regulates SPL expression to control plant responses to auxin, thereby influencing grain size	[87,88]
lncRNA	Long non-coding RNA	Interacts with chromatin remodeling proteins to precisely regulate the transcription of specific genes involved in cell division and proliferation during grain filling	[155]
SWI/SNF complex	Chromatin remodeling factor	Open chromatin structure promotes transcription of genes related to grain filling and cell division	[164]

Table 2. Epigenetic and non-coding RNA factors in rice and their roles in grain size regulation.

Gene	Type	Function	Reference
OsDDM1b	Chromatin remodeling enzyme gene of the WI2/SNF2 family	Loss-of-function mutation significantly reduces overall DNA methylation, activating multiple genes that negatively regulate cell division, ultimately leading to smaller grain size	[148]
OsHDT701	Histone deacetylase	Loss of activity causes premature expansion and division of endosperm cells, resulting in poor grain filling	[142]
H3K27me3	Trimethylation of lysine 27 on histone H3	Acts as an epigenetic “brake”; its level and distribution determine the balance between cell proliferation and grain filling, directly regulating grain size	[150,151]
H3K4me3	Trimethylation of lysine 4 on histone H3	Enhances transcriptional activity, promoting grain growth and filling	[152]
miRNA156	MicroRNA (miRNA)	Regulates SPL expression to control plant responses to auxin, thereby influencing grain size	[87,88]
lncRNA	Long non-coding RNA	Interacts with chromatin remodeling proteins to precisely regulate the transcription of specific genes involved in cell division and proliferation during grain filling	[155]
SWI/SNF complex	Chromatin remodeling factor	Open chromatin structure promotes transcription of genes related to grain filling and cell division	[164]

2.6. Environmental Influences on Seed Development and Adaptive Mechanisms

2.6.1. Impact of High-Temperature Stress, Light, and Nutrients on Seed Size

External environmental factors, such as temperature, light conditions, and nutrient availability, significantly influence plant growth and development, directly affecting seed filling rates and ultimate seed size. High-temperature stress, especially during grain filling, notably hampers seed development, often resulting in poorly developed grains [168]. Elevated temperatures accelerate water evaporation from leaves and seeds, disrupting overall plant water balance. Studies indicate that temperatures exceeding 35 °C inhibit starch synthesis in rice grains, significantly slowing endosperm cell expansion, reducing grain filling, and consequently decreasing seed size [169]. Additionally, oxidative stress induced by high temperature leads to lipid peroxidation and cellular membrane damage, inhibiting endosperm cell division and enlargement [168,169,170]. To mitigate high-temperature stress, plants activate protective mechanisms such as synthesizing heat-shock proteins (HSPs) to stabilize protein structures and maintain cellular functions [171]. These proteins improve plant tolerance to heat stress, maintaining seed size under challenging conditions.

Light and photoperiod are key environmental cues that profoundly influence seed development by modulating hormonal balance and metabolic activity. Variations in day length and light intensity affect the biosynthesis and signaling of plant hormones such as auxin, gibberellins, and cytokinins, thereby shaping developmental outcomes in response to light stimuli [34,172,173]. Under favorable photoperiodic conditions, enhanced photosynthetic activity leads to increased carbohydrate production, which supports seed growth and filling. For example, extended day lengths have been shown to promote carbohydrate accumulation, contributing to improved grain filling and increased seed size and weight [174]. In addition, light regulates the expression of transcription factors involved in seed development, including members of the phytochrome-interacting factor (OsPIF) family. Under long-day conditions, OsPIFs activate downstream genes associated with endosperm development and nutrient accumulation, ultimately enhancing seed filling efficiency and grain enlargement [175].

Nutrient availability, particularly of nitrogen (N), phosphorus (P), and potassium (K), also plays a critical role in regulating seed development and grain filling [176]. Nitrogen is central to protein synthesis and cellular proliferation and is known to stimulate cytokinin production, which accelerates cell division and expansion during seed filling, thereby increasing final seed size [177,178]. Phosphorus, essential for nucleic acid synthesis and energy transfer, is a key component of ATP and phospholipids. Its deficiency leads to impaired metabolic activity and reduced seed filling, often resulting in smaller seeds [179]. Potassium contributes to osmotic regulation, enzyme activation, and the maintenance of photosynthetic and assimilate transport processes. Adequate potassium supply has been associated with improved endosperm development and higher grain weight due to its role in facilitating water and nutrient movement within the developing seed [180,181].

2.6.2. Effects of Local Environmental Conditions on Seed Development

In addition to broad environmental factors such as temperature, light, and nutrients, localized micro-environmental conditions, including soil moisture and plant density, significantly influence seed development. Localized soil moisture profoundly affects plant water and nutrient absorption. Adequate moisture supply is essential for cell expansion and grain filling, particularly during periods of high-temperature stress or drought, when water deficits cause cellular dehydration and impede seed enlargement [182,183]. Studies have shown that moderate irrigation significantly increases grain filling rates and seed plumpness in crops such as rice and wheat, whereas drought stress severely restricts endosperm cell expansion and decreases grain filling [184,185]. Additionally, localized moisture availability impacts enzymatic activities and protein synthesis efficiency, further influencing seed development quality and size [182].

Plant density and competition from neighboring vegetation also indirectly influence seed size. Higher planting densities intensify competition among plants for limited resources, especially light and nutrients, reducing photosynthetic efficiency and limiting seed filling [172,186]. Increased planting density lowers light availability and air circulation, reducing leaf photosynthesis and thus constraining seed expansion [187]. Furthermore, dense neighboring vegetation can restrict root growth and nutrient uptake, indirectly impacting seed size [188]. Adjusting plant spacing and density can improve the local growth environment, thus enhancing seed development potential and optimizing seed size [189].

2.6.3. Agricultural Implications of Regulating Seed Size

In modern agriculture, seed size significantly influences crop yield, seedling vigor, and market value. Understanding molecular mechanisms controlling seed size enhances germination rates and seedling vitality, thereby directly improving plant stress tolerance and disease resistance. Larger seeds typically store greater reserves of nutrients, including carbohydrates, proteins, and lipids, providing essential energy and structural materials for seedlings to better cope with adverse conditions such as drought, salinity, and poor soil fertility during early growth stages [173]. For instance, seed size directly affects germination and emergence rates in rice, while larger seeds in legumes boost seedling growth rates and resilience to early environmental stressors [10,174].

Genetic regulation of seed size can improve yield per unit area and enhance nutritional and commercial value. Key genes such as OsGS3 and OsGW2 significantly influence seed size traits in rice and other crops. OsGS3 primarily controls seed length, whereas OsGW2 modulates seed width and thickness through regulation of cell division [174]. Mutants of these genes are widely employed in crop breeding programs to substantially enhance seed size and productivity. Furthermore, gene-editing technologies such as CRISPR enable precise modification of seed size-regulating genes, accelerating the selection of superior crop varieties. For example, mutating or knocking out OsGS3 significantly increases grain length, thereby boosting grain weight and yield per area [3].

Understanding molecular control of seed size also facilitates optimization of seed nutritional composition. Adjusting carbohydrate, protein, and lipid metabolism during seed filling significantly enhances nutritional value, such as increased protein content, improved starch quality, and elevated micronutrient concentrations [175]. In high-protein crops like soybean, increased seed size boosts protein content and essential amino acid profiles. In cereals, modulating endosperm development enhances starch accumulation and nutritional quality [176,177]. Moreover, regulating genes associated with seed size and related metabolic pathways increases mineral nutrients (e.g., iron, zinc) in seeds, enhancing nutritional quality, food security, and market adaptability [178,179].

Amid global climate change, knowledge of seed size regulation provides potential solutions to environmental challenges such as extreme climates and resource scarcity. Increasing seed size enhances crop resilience, significantly boosting yields per unit area and reducing dependence on fertilizers and irrigation [10,179]. Under drought or saline conditions, larger seeds exhibit enhanced drought and salt tolerance at seedling stages, substantially improving survival rates and productivity, especially under limited water or nutrient availability [10,180] (Figure 4). Looking ahead, integrating genomic editing and conventional breeding strategies to optimize seed size will increasingly support agricultural productivity and global food security.

3. Application of Grain Shape Research in Rice Breeding

3.1. Molecular Regulatory Networks Lay the Foundation for Precision Breeding

Major progress in rice grain shape research has come from the elucidation of several key regulatory pathways, particularly BR signaling, the ubiquitin–proteasome system, and cell cycle regulation. In 2025, Wan Jianmin’s team identified SRG, a gene encoding a kinesin-domain protein, as a positive regulator of grain size through the BR pathway. OsWRKY53 activates SRG transcription under BR induction, while OsBSK3 interacts with and phosphorylates the SRG protein, enhancing its stability and promoting microtubule organization. This leads to increased spikelet cell proliferation and extended grain length. Overexpression of SRG significantly increased thousand-grain weight, demonstrating its potential in yield improvement [181].

Additionally, Li Yunhai’s group proposed a GS2–SUG1 module involved in the integration of multiple hormonal pathways. The transcription factor GS2/OsGRF4 not only regulates grain size but also activates SUG1, a plant-specific DEP2/SRS1/EP2/OsRELA-like protein. SUG1 functions as a molecular hub by interacting with OsBZR1 (BR receptor), OsMADS56 (flowering regulator), and OsSPL13 (panicle development), thus coordinating BR, GA, and auxin signaling to regulate grain development. Introduction of the SUG1Hap2 allele from indica rice into japonica backgrounds led to significant improvement in grain size and yield [101]. This result offers insight into inter-subspecific differences and provides a valuable allele for breeding programs.

3.2. Gene-Based Breeding Strategies and Germplasm Innovation

Functional characterization of grain shape genes has enabled various breeding strategies, including gene editing, use of wild germplasm, and marker-assisted selection (MAS). In 2025, Wang Lan and Liu Xiangdong’s team cloned OsGSW3.2 from a wild rice accession in Suixi, Guangdong. OsGSW3.2 exhibits pleiotropic effects, regulating grain size, awn development, and chlorophyll biosynthesis. It interacts with OsGSK4, a negative regulator of BR signaling. CRISPR/Cas9 knockout lines exhibited larger grains, while overexpression lines showed reduced grain size. Importantly, natural variation in wild rice for this gene includes favorable alleles that enhance grain weight [71]. This finding emphasizes the value of wild rice for modern molecular breeding and offers new insight into grain shape selection during domestication.

Although gene editing is a powerful tool for grain shape improvement, balancing yield and grain quality remains a challenge. Recent studies on OsANK3, encoding an ankyrin repeat protein, revealed dual roles in grain development. OsANK3 regulates spikelet cell proliferation via cell cycle genes and also modulates starch metabolism by affecting both biosynthetic (e.g., OsGBSSI, OsAGPL1) and hydrolytic (e.g., OsAmy1A, OsAmy3B) genes. Knockout mutants showed increased plant height, grain size, and thousand-grain weight, but also higher chalkiness and reduced starch quality [190]. This highlights the need to optimize grain filling traits, possibly through spatiotemporal regulation or targeted editing.

Marker-assisted breeding based on functional polymorphisms has also made notable progress. Wan Jianmin’s group identified SMG4, a MATE family transporter affecting grain shape by regulating spikelet cell division. Multiple alleles of SMG4 were found, some strongly associated with grain size. Functional markers developed from these variants have been implemented in high-yielding japonica breeding programs. Similarly, molecular markers for SUG1Hap2 and WLGhap.B have shown strong application potential in enhancing grain selection efficiency [191].

4. Perspectives

4.1. Multi-Omics Platforms Promote Intelligent Breeding of Grain Shape

As functional genomics advances, multi-omics integration and smart breeding platforms are becoming central to modern rice breeding. In 2025, the Crop Science Institute of the Chinese Academy of Agricultural Sciences released RiceAtlas, an integrated omics platform covering 6044 rice varieties from five major ecological regions. The platform includes high-depth resequencing data and phenotypic data from 212 trials across 19 locations. Using RiceAtlas, researchers have identified 3131 loci associated with traits such as heading date, yield, and stress response [192].

RiceAtlas has driven significant progress in grain shape research by providing a high-density variant map that enables tracking of grain shape-related alleles across different ecological regions, supporting the online design of trait-associated modules, and facilitating rapid gene mapping and functional validation. Leveraging this platform, researchers successfully improved the grain size of the japonica variety “Suijing 4” through precise introgression of favorable alleles, achieving enhancement of grain traits without compromising stress resistance [192]. Furthermore, RiceAtlas revealed ecotypic variation in grain shape preferences—long slender grains (L/W > 3.0) in indica regions and short round grains (L/W ~1.8–2.2) in japonica areas—reflecting both consumer preferences and environmental adaptations. Based on these patterns, RiceAtlas enables ecological zone-specific gene combination recommendations, supporting regionally optimized precision breeding [192].

4.2. Challenges and Future Outlook

4.2.1. Challenges

Despite substantial advances in elucidating the genetic control of grain traits, several challenges persist in the practical application of these genes in breeding programs. One of the major obstacles is the pleiotropic nature of many key regulatory genes. For example, OsGSW3.2 simultaneously influences grain size, awn development, and chlorophyll content, complicating targeted trait improvement [76]. Similarly, mutations in OsANK3 have been reported to enhance yield potential, but at the cost of reduced grain quality [190]. These examples highlight the difficulty of uncoupling desirable and undesirable phenotypes. To overcome this, refined genetic strategies—such as the use of tissue-specific or inducible promoters, as well as temporally controlled gene-editing tools like conditionally activated CRISPR/Cas systems—are increasingly being explored to achieve precise trait regulation while minimizing adverse trade-offs.

Another limitation is the lack of understanding of non-additive effects. For example, the OsMAPK6–GW6a and CLG1–GW6a modules exhibit complex interactions that are difficult to model [193]. Predictive tools integrating structural biology and artificial intelligence will be necessary to improve the reliability of multigene design.

4.2.2. Future Outlook

Future directions in grain shape research are expected to focus on the deep mining of wild rice resources to uncover novel alleles and broaden the genetic base [194], the development of epigenetic tools for stable and heritable trait regulation such as grain shape “memory” systems derived from modules like GW6a–HDR3 [195], and the integration of intelligent breeding platforms such as RiceAtlas with machine learning algorithms to achieve holistic optimization of yield, grain shape, and quality traits across diverse environments [192].

5. Conclusions

In conclusion, rice grain shape research has transitioned from single-gene analysis to complex regulatory networks and from empirical breeding to precision design. Building on this foundation, 726 cutting-edge technologies are being systematically incorporated into breeding pipelines: CRISPR/Cas9-based editors enable single-base precise substitutions and multiplexed site stacking; molecular module design platforms allow functional modules such as GL7, GW8, and GW5 to be combined in a “Lego-like” manner according to yield–quality–resistance requirements; over 3000 wild and local rice accessions have been deeply analyzed through pan-genomics and phenomics, providing a vast reservoir of allelic variation to broaden the genetic base; and AI-driven phenotype–genotype–environment big data prediction systems compress the traditional 8–10-year breeding cycle into just 3–4 years. Through integration of gene editing, molecular modules, wild germplasm, and omics platforms, grain shape breeding is progressing toward coordinated improvement of yield and quality. As functional genomics and synthetic biology converge, precision grain design will play a crucial role in ensuring global food security.