Loss of GW5 function is involved in the unique grain shape of “Tanpo”, a Japanese landrace rice

Abstract

“Tanpo”, a Japanese rice landrace widely cultivated approximately 120 years ago in Akita Prefecture, exhibits a shorter, wider, thicker, and heavier grain compared to Akitakomachi. Microscopic analysis has revealed that the epidermal cells of Tanpo spikelet hulls are narrower and shorter, with an increased number of cells in the grain width direction, thus resulting in a distinctive grain shape. In a genetic analysis of an F₂ population derived from a cross between Tanpo and Akitakomachi, the Tanpo GW5 allele was found to determine the grain shape in a recessive manner. The GW5 allele in Tanpo is a loss-of-function allele because it generates a stop codon immediately after the start codon with a 100-bp deletion within the first exon. Because the GW5 protein suppresses glycogen synthase kinase 2 (GSK2), a negative regulator of brassinosteroid (BR) signaling, GW5 deficiency in Tanpo results in reduced BR signaling. As a result, the expansion of epidermal cells was suppressed, while the radial cell division was promoted, which led to thicker and shorter spikelet hulls and, ultimately, the characteristic grain shape of Tanpo. The identification of this unique allele in the Tanpo landrace provides a valuable resource for breeding new rice varieties with unique grain characteristics.

Introduction

Rice landraces are genetically diverse and exhibit a large variation in morphology (Hour et al. 2020, Kobayashi et al. 2006). These landraces possess unique characteristics and alleles that have been lost in modern cultivars; thus, they are valuable genetic resources for breeding (Jackson 1997, Yamaguchi et al. 2007). In Japan, rice landraces have adapted to the specific climatic conditions of each region and were cultivated until approximately 120 years ago.

Various landraces have been documented in Akita Prefecture, northern Japan, including “Tanpo”, which was widely cultivated in the northern region (Akimoto 1995, Saito 2006). We cultivated the Tanpo landrace, which had been preserved in the Genebank, using modern rice cultivation techniques and observed that the harvested brown rice grains were short and round in shape (Supplemental Fig. 1).

Numerous quantitative trait loci (QTLs) associated with the rice grain shape have been identified (Huang et al. 2013), and several genes encoding these QTLs have been isolated (Duan et al. 2017, Fan et al. 2006, Ishimaru et al. 2013, Li et al. 2011, Liu et al. 2017, Song et al. 2007, Wang et al. 2012). These genes regulate grain shape via networks of phytohormone processes and multiple signaling pathways (Chen et al. 2021, Li et al. 2022).

GW5 (qSW5) has been identified as a major QTL responsible for the differences in grain width and weight between Japonica and Indica rice (Shomura et al. 2008, Wan et al. 2008, Weng et al. 2008), and it has been shown to encode a calmodulin-binding protein (Duan et al. 2017, Liu et al. 2017). A 1212-bp deletion in the promoter region of Japonica rice reduces the expression of GW5, which results in increased grain width (Duan et al. 2017, Liu et al. 2017). Furthermore, the GW5 protein, localized in the plasma membrane, regulates brassinosteroid (BR) signaling by suppressing glycogen synthase kinase 2 (GSK2), a negative regulator of BR signaling, thereby determining the grain shape (Liu et al. 2017).

In this study, we investigated the role of the GW5 gene in the characteristic grain shape of the Tanpo landrace and proposed a mechanism by which the Tanpo GW5 allele influences the grain shape.

Materials and Methods

Plant material

We used the Japanese landrace Tanpo (JP7283, maintained at the NARO Gene Bank), the Japanese cultivar Akitakomachi, and their F₂ population. Forty-two of each of the Tanpo and Akitakomachi cultivars, along with 116 individuals from the F₂ population, were transplanted into an experimental paddy field at the Akita Agricultural Experiment Station on May 30, 2023, with a planting density of 20.7 plants m^–2. Fertilizer was applied before planting, which consisted of: N, 0.6 kg a^–1; P₂O₅, 0.6 kg a^–1; and K₂O, 0.6 kg a^–1. No additional fertilizer was provided.

Morphological analyses of the rice grains

Mature rice plants were harvested on October 1, 2023, and dried in a well-aerated room for one month. Brown rice grains with a thickness of ≥1.9 mm were selected using a 1.9-mm sieve (Fuji Kinzoku Co., Ltd.). Approximately 200 grains from each sample were measured for grain length, width, and thickness using a Rice Analyzer (RGQI 20A; Satake Co., Ltd.). The moisture content of the rice grains was determined using a digital moisture meter (PB-1D2; Kett Electric Laboratory Co., Ltd.), and then the grain weight was converted to 15% moisture content.

Electron microscope analysis of the epidermal cells of spikelet hulls

Spikelet hulls of the Tanpo and Akitakomachi cultivars were sampled two weeks after heading for electron microscope analysis. Four spikelet hulls of average size were selected, which were then coated with a multi-coater VES-10 (Vacuum Devices) and observed for epidermal cell morphology using the Miniscope TM3030Plus (Hitachi High-Tech Corporation). The length and width of the surface cells were measured from the photographed images.

DNA extraction

Leaf tips from both parental varieties and F₂ individuals were sampled 28 days after transplanting and stored at –20°C. Extraction buffer (0.1 M Tris-HCl [pH 8.0], 10 mM EDTA [pH 8.0], and 1 M KCl) was added to each sample with a stainless-steel ball (φ3 mm), and DNA was extracted using a multi-bead shocker (MB1021; Yasui Kikai Corporation) at 1600 rpm for 40 s. DNA was precipitated from 100 μL of the supernatant by adding an equal volume of 2-propanol. The DNA pellet was washed with 70% ethanol, air-dried, and suspended in 50 μL of 1/10 TE. The extracted DNA was used for PCR analysis.

PCR

PCR was performed using Premier EX (Takara Bio Inc.) with the primer pairs that are listed in Supplemental Tables 1 and 2. The PCR conditions were an initial denaturation of 1 min at 94°C, followed by 35 cycles of 10 s at 98°C, 15 s at 55°C, and 1 min at 68°C, using a PCR9700 thermocycler (Thermo Fisher Scientific Inc.). Polymorphisms in the PCR products were detected by electrophoresis using 1% UltraPure agarose (Thermo Fisher Scientific Inc.) in 1× TAE buffer (Supplemental Tables 1, 2).

Sequencing of the GW5 gene

The entire region of the GW5 gene was amplified from the Tanpo and Akitakomachi cultivars by PCR using the GW5-F1 and GW5-R4 primers (Supplemental Table 2). The amplified DNA fragments were purified using a QIAquick PCR Purification Kit (QIAGEN), followed by sequencing using an ABI PRISM^® 3100 Genetic Analyzer (Thermo Fisher Scientific Inc.). The nucleotide sequence data for the Tanpo GW5 gene has been deposited in the DDBJ database under the accession number LC848690.

RNA extraction and RT-qPCR

Six young inflorescences (approximately 1 cm in length) were sampled from each of the Tanpo and Akitakomachi cultivars and stored at –80°C. Total RNA was extracted using the TRIzol^TM Plus RNA Purification Kit (Invitrogen), and RNA concentration was measured using a NanoDrop^TM One spectrophotometer (Thermo Fisher Scientific Inc.). Complementary DNA was synthesized from 50 ng of total RNA using a ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO). RT-qPCR was performed using a THUNDERBIRD Next SYBR qPCR Mix (TOYOBO) with the primers that are listed in Supplemental Table 3 (Liu et al. 2017). Real-time PCR was performed with an initial denaturation at 95°C for 30 s, followed by 40 cycles of 5 s at 95°C and 30 s at 60°C for 30 s, using a Bio-Rad CFX96 Real-Time System (Bio-Rad). Melting curve analysis was performed from 65°C to 95°C. Data analysis was performed using Bio-Rad CFX Manager 3.1 (Bio-Rad), and the relative gene expression levels were calculated relative to the ubiquitin gene (Os03g0234200) using the 2-^ΔΔCt method.

Statistical analysis

Differences between the Tanpo and Akitakomachi cultivars were evaluated using the Student’s t-test with the data analysis module in Microsoft Office Excel 2019. Furthermore, differences in grain length, width, thickness, and weight among the GW5 genotypes of the F₂ population were determined by Tukey’s test using the R package 4.0.0. Statistical significance was established at P < 0.05.

Results

Features of Tanpo grains and spikelet hulls

The grains of Tanpo were shorter and wider than those of Akitakomachi (Supplemental Fig. 1). The grain length of Akitakomachi was 5.38 mm, whereas that of Tanpo was significantly shorter at 4.86 mm. Conversely, Tanpo grains were significantly wider (2.92 mm) and thicker (2.21 mm) than the Akitakomachi grains (2.77 mm and 2.07 mm, respectively). Similarly, the grain weight was also significantly higher in Tanpo (26.32 mg) compared to Akitakomachi (23.04 mg) (Fig. 1, Table 1). Thus, Tanpo grains were shorter, wider, thicker, and heavier than those of Akitakomachi.

Fig. 1.

Comparison of the grain shape between the Tanpo (left) and Akitakomachi (right) cultivars from three angles. Scale bar = 2 mm.

Table 1.Comparison of the grain shape of Tanpo and Akitakomachi

Variety	Grain length (mm)	Grain width (mm)	Grain thickness (mm)	Grain weight (mg)
Akitakomachi	5.38 ± 0.20**	2.77 ± 0.08**	2.07 ± 0.08**	23.0 ± 0.35**
Tanpo	4.86 ± 0.27	2.92 ± 0.11	2.21 ± 0.08	26.3 ± 0.59

Mean ± SD (n = 14)

Double asterisks indicate significant difference at P < 0.01.

Because rice grains develop within fully formed spikelet hulls, which constrain the grain shape, we analyzed the epidermal cells of the spikelet hulls using electron microscopy (Fig. 2A, 2B). Epidermal cells in Tanpo hulls were significantly shorter (59.0 μm) and narrower (50.2 μm) than those of Akitakomachi (77.2 μm and 62.3 μm, respectively) (Fig. 2C, 2D). Although Tanpo spikelet hulls were narrower than Akitakomachi hulls, they were wider, which suggests a greater number of epidermal cell rows in the Tanpo cultivar (Supplemental Fig. 2). Therefore, the distinctive grain shape of Tanpo is likely due to the combination of limited cell expansion and the increased number of cell rows in their spikelet hulls.

Fig. 2.

Comparison of epidermal cells of the spikelet hulls of Tanpo and Akitakomachi. A (Tanpo) and B (Akitakomachi) show the scanning electron microscope analysis of the epidermal cells of the spikelet hulls after two weeks of heading. Scale bar = 2 μm. C and D show the cell width and cell length, respectively. Black and white bars represent the Tanpo and Akitakomachi cultivars, respectively. Values represent the mean ± standard deviation (SD) (n = 4). Double asterisks indicate significant differences between Tanpo and Akitakomachi as determined using the Student’s t-test (**P < 0.01).

Linkage analysis of the GW5 gene and grain shape in the F₂ population

The GW5 gene has previously been demonstrated to control rice grain width and weight (Duan et al. 2017, Liu et al. 2017). To investigate whether the GW5 gene was involved in the characteristic grain shape of Tanpo, we analyzed the linkage between the GW5 genotype and grain traits in an F₂ population between the Tanpo and Akitakomachi cultivars.

F₂ individuals that were homozygous for the Tanpo GW5 allele exhibited significantly wider, thicker, and heavier grains compared to those homozygous for the Akitakomachi allele, with intermediate phenotypes observed in heterozygotes (Fig. 3A–3C). These results indicate that the Tanpo GW5 allele is recessive and has an additive effect on both grain width and thickness. However, the grain length of F₂ individuals homozygous for the Tanpo GW5 allele was significantly shorter than that of heterozygotes and Akitakomachi homozygotes (Fig. 3D). A slightly shorter grain length was observed in heterozygous compared to the Akitakomachi homozygous, which suggests that the GW5 allele from Tanpo contributes to the shorter grain length.

Fig. 3.

Box plot representing the grain characteristics among the GW5 genotypes in F₂ individuals of a cross between Tanpo and Akitakomachi. Panels A, B, C, and D show the grain width, thickness, length, and weight, respectively. The letters T, H, and A in each plot indicate homozygous for the Tanpo allele, heterozygous alleles, and homozygous for the Akitakomachi allele, respectively. The average of each box plot is indicated with an X and connected by a line. Box plots with different letters (a, b, and c) indicate statistically significant differences using the Tukey test (P < 0.05).

Analysis of the Tanpo GW5 gene

Previous studies have revealed that a 1212-bp deletion in the promoter region of the GW5 gene in Japonica rice reduces GW5 expression, thus resulting in wider grains (Liu et al. 2017). We analyzed the promoter region of the GW5 gene in Tanpo and found that there was no deletion, which suggests a promoter structure similar to that of Indica rice (Supplemental Table 1, Fig. 4A, 4B). The GW5 gene expression was significantly higher in the young panicles of Tanpo compared to Akitakomachi (Fig. 4C), indicating that the promoter region alone does not account for the characteristic grain shape of Tanpo.

Fig. 4.

Analysis of the GW5 gene in Tanpo and Akitakomachi. Panel A shows the GW5 gene structure, including the promoter regions of Tanpo and Akitakomachi. Arrows indicate the primers used to analyze the gene structure. Panel B shows the amplified DNA fragments of the promoter regions of Tanpo (T) and Akitakomachi (A). Panel C shows a comparison of the relative expression of the GW5 gene in the young inflorescences. White and black bars represent Akitakomachi and Tanpo, respectively. Values represent the mean ± standard error (SE) (n = 4). Single asterisks indicate significant differences between Tanpo and Akitakomachi as determined using the Student’s t-test (P < 0.05). Panel D shows the amplified DNA fragments of the entire GW5 gene region from Tanpo (T) and Akitakomachi (A).

The entire region of the GW5 gene from Tanpo was shorter than that of Akitakomachi, which suggests that deletion had occurred within the Tanpo GW5 gene (Fig. 4A, 4D). Further analysis revealed a 100-bp deletion within the first exon of the GW5 gene in Tanpo. Moreover, a single nucleotide substitution generated a stop codon immediately after the start codon, thus resulting in a loss-of-function allele (Supplemental Fig. 3). GW5 protein deficiency in Tanpo likely contributes to the increased grain width and thickness observed compared to common Japonica varieties with reduced GW5 expression.

Expression profile of BR-related genes

GW5 has been shown to function as a positive regulator of BR signaling, thereby affecting grain width and weight (Liu et al. 2017). We analyzed the expression profiles of BR biosynthetic genes (D2, DWARF, and CPD) and BR signaling genes (BRI1, GSK2, BZR1, and BU1) in the young inflorescences of Tanpo and Akitakomachi cultivars.

While the expression of the BR biosynthesis gene CPD was downregulated in Tanpo, the expression of another BR biosynthesis gene, DWARF, was upregulated (Fig. 5). Considering the multistep nature of BR biosynthesis, the loss of GW5 function may not affect the overall BR biosynthesis in the young inflorescences of Tanpo. However, we observed a significant upregulation of GSK2, a negative regulator of BR signaling, in Tanpo compared to Akitakomachi (Fig. 5). In addition, the expression of other BR signaling genes, BRI1, BZR1, and BU1, were also higher in Tanpo, which suggests that the loss of GW5 function affected BR signaling in the young Tanpo inflorescences.

Fig. 5.

RT-qPCR analysis of genes related to brassinosteroid (BR) biosynthesis (D2, Dwarf, and CPD) and BR signaling (BRI1, GSK2, BZR1, and BU1) in the young inflorescences. Black and white bars represent Tanpo and Akitakomachi, respectively. Values represent the mean ± standard error (SE) (n = 4). Single asterisks indicate significant differences between Tanpo and Akitakomachi as determined using the Student’s t-test (*P < 0.05).

Discussion

Role of the GW5 gene in determining Tanpo grain shape

The Japanese landrace rice Tanpo was widely cultivated in Akita Prefecture during the Meiji period and exhibits unique grain characteristics, including a shorter, wider, thicker, and heavier shape compared to modern cultivars. Microscopic analysis revealed that the epidermal cells of the Tanpo spikelet hulls were significantly smaller in both length and width than those of Akitakomachi. Because the spikelet hulls were wider in the Tanpo cultivar than in Akitakomachi, the number of cells in the grain width direction in the spikelet hulls was increased in Tanpo. The spikelet hull plays a predominant role in determining the grain size and shape as a container during grain filling (Li and Li 2016, Li et al. 2018). For instance, rice sts-1 mutants exhibited short grains because of reductions in both cell length and cell number in the longitudinal direction of the spikelet hulls (Abe et al. 2010). Therefore, the distinctive grain shape of Tanpo is likely attributed to a combination of the reduced cell length and increased cell number in the grain width direction of the spikelet hulls.

The GW5 gene, encoding a calmodulin-binding protein, was identified as a major QTL on chromosome 5 that regulates grain width and weight (Duan et al. 2017, Liu et al. 2017). The Tanpo GW5 allele additively increased the grain width, thickness, and weight while decreasing the grain length in the F₂ population derived from a cross between Tanpo and Akitakomachi. As the shape of the F₃ grains was strongly influenced by the shape of the parental F₂ spikelet hulls, the grain shape of the F₂ individuals reflected their respective F₂ genotypes. These results indicate that the Tanpo GW5 allele is recessive and contributes to the shorter, wider, thicker, and heavier grain characteristics.

The Tanpo GW5 gene has a loss of function because of a nonsense mutation and a 100-bp deletion within the first exon. Previous studies have shown that reduced GW5 gene expression in Japonica varieties results in wider grains (Duan et al. 2017, Liu et al. 2017). The GW5 protein deficiency in Tanpo, resulting from the loss-of-function allele, likely contributes to the further increase in grain width and thickness compared to common Japonica varieties with reduced GW5 expression.

GW5 protein physically interacts with and suppresses GSK2, a positive regulator of BR signaling (Liu et al. 2017). GSK2 restricts cell expansion in spikelet hulls, thus resulting in longer and wider grains (Che et al. 2016, Lyu et al. 2020, Zhang 2020). In Tanpo, the loss of GW5 function prevents GSK2 suppression, which results in restricted epidermal cell expansion in spikelet hulls and smaller cells.

The overexpression of GW5 has been shown to reduce the number of epidermal cells in the grain width direction on the spikelet hulls (Liu et al. 2017), while knockout of GSE5 (GW5) results in narrower, more numerous epidermal cells (Duan et al. 2017). These findings suggest that GW5 (GSE5) controls grain width by regulating cell proliferation in the spikelet hulls (Duan et al. 2017). In Tanpo, the loss of GW5 function likely promotes the formation of shorter and broader spikelet hulls by increasing the cell number in the grain width direction while decreasing the cell size.

BR signaling and grain shape in Tanpo

Fig. 6 summarizes the regulation of grain shape by the Tanpo GW5 allele via BR signaling, based on the model proposed by Liu et al. (2017). The GW5 protein is localized to the plasma membrane and suppresses the kinase activity of GSK2, a key component of BR signaling (Liu et al. 2017). GSK2 phosphorylates the transcription factor BZR1, inhibiting its accumulation in the nucleus and ultimately affecting BR-responsive gene expression (Tong et al. 2012, Zhang et al. 2014). In rice, GW5 acts positively and upstream of GSK2 in the BR signaling pathway (Liu et al. 2017). In Tanpo, GSK2 function was not suppressed by GW5 because of its loss of function. Moreover, GSK2 expression was upregulated, suggesting that BR signaling is suppressed in the Tanpo cultivar. To compensate for this, the expression of BU1, a BR receptor, and BZR1 and BRI1, transcription factors for BR-responsive genes, may be upregulated in the young Tanpo inflorescences.BR plays a crucial role in regulating various agronomic traits in rice, including grain development (Chen et al. 2021, Xiong et al. 2022). For example, BR-deficient rice mutants, such as ebisu dwarf (d2), exhibit shortened grains (Hong et al. 2003). Therefore, the reduction in BR signaling in the young Tanpo inflorescences likely resulted in smaller cells while increasing cell proliferation in the grain width direction in the spikelet hulls (Fig. 6).

Fig. 6.

Proposed model for the regulation of grain shape in the Tanpo cultivar based on the model report by Liu et al. (2017). GW5 represses GSK2, which controls BZR1 via phosphorylation. BZR1 induces the expression of brassinosteroid (BR)-responsive genes, which leads to normal cell growth in the spikelet hulls of Akitakomachi. However, in Tanpo, GSK2 is not suppressed because of the absence of the GW5 protein, which strongly suppresses BR signaling and results in the suppression of cell expansion but enhances cell division in the grain width direction of the spikelet hulls.

BR has been shown to suppress cytokinin responses, which induce the periclinal division of vascular cells in the Arabidopsis root apical meristem (Ohashi-Ito et al. 2023). The reduced BR signaling in Tanpo spikelet hulls may have led to a lack of cytokinin response suppression, potentially resulting in thicker grains via cytokinin-induced periclinal division within the grains.

Breeding using the gw5 allele of Tanpo

The non-functional gw5 allele in Tanpo results in wider and thicker but shorter grains. Genome editing to create a loss-of-function GSE5 (GW5) allele in an Indica variety has been shown to result in wider grains without reducing the grain length (Duan et al. 2017). This suggests that the genetic background influences the grain shape, as numerous genes are involved in regulating the grain size and shape (Chen et al. 2021, Li et al. 2022). The gw5 allele, a natural variation found in the landrace Tanpo, could be used to breed varieties with larger and heavier grains by increasing the width and thickness, in addition to length.

Thicker rice grains in Koshihikari, a Japanese good-tasting variety, are known to have a higher amylose and lower protein content, which contributes to the improved taste (Kuroda and Matsumoto 1996). In addition, grain thickness has a positive effect on the taste quality (He et al. 2022). These findings suggest that the Tanpo gw5 allele could also contribute to improving the taste quality by increasing the grain thickness.

Tanpo exhibits undesirable traits, such as excessive plant height (over 120 cm) and poor appearance quality of brown rice. Rice mutants with reduced brassinosteroid biosynthesis are known to exhibit decreased plant height (Hong et al. 2003), suggesting that the Tanpo gw5 allele, which suppresses brassinosteroid signaling, does not contribute to increased plant height in Tanpo. Therefore, the gw5 allele in Tanpo can be specifically utilized to enhance seed morphology. By integrating the Tanpo gw5 allele into the genetic background of a leading variety with a superior brown rice quality through marker-assisted selection targeting the 100-bp deletion, Tanpo can serve as a valuable genetic resource. The gw5 allele in Tanpo represents a novel natural mutation that promotes large grain size and holds potential for high-yielding breeding programs in the future.

Author Contribution Statement

KS and HA designed the research. RN and SM developed the materials. MI, KC, and RN performed data collection and analysis. MI and KU performed microscopy analysis. MI and AW performed RT-qPCR analysis. MI, HA, and KS wrote the manuscript. All authors commented on the manuscript.

 Acknowledgments

We thank the NARO Genebank for providing the Japanese landrace Tanpo (JP7283) seeds.

Literature Cited

•  Abe,  Y.,  K.  Mieda,  T.  Ando,  I.  Kono,  M.  Yano,  H.  Kitano and  Y.  Iwasaki (2010) The SMALL AND ROUND SEED1 (SRS1/DEP2) gene is involved in the regulation of seed size in rice. Genes Genet Syst 85: 327–339.
•  Akimoto,  H. (1995) The transition of rice strains from the premodern to modern period in the Tohoku area—In the case of the Akita region—. The Economic review of Toyo University 20: 1–30 (in Japanese).
•  Che,  R.,  H.  Tong,  B.  Shi,  Y.  Liu,  S.  Fang,  D.  Liu,  Y.  Xiao,  B.  Hu,  L.  Liu,  H.  Wang et al. (2016) Control of grain size and rice yield by GL2-mediated brassinosteroid responses. Nat Plants 2: 15195.
•  Chen,  K.,  A.  Łyskowski,  Ł.  Jaremko and  M.  Jaremko (2021) Genetic and molecular factors determining grain weight in rice. Front Plant Sci 12: 605799.
•  Duan,  P.,  J.  Xu,  D.  Zeng,  B.  Zhang,  M.  Geng,  G.  Zhang,  K.  Huang,  L.  Huang,  R.  Xu,  S.  Ge et al. (2017) Natural variation in the promoter of GSE5 contributes to grain size diversity in rice. Mol Plant 10: 685–694.
•  Fan,  C.,  Y.  Xing,  H.  Mao,  T.  Lu,  B.  Han,  C.  Xu,  X.  Li and  Q.  Zhang (2006) GS3, a major QTL for grain length and weight and minor QTL for grain width and thickness in rice, encodes a putative transmembrane protein. Theor Appl Genet 112: 1164–1171.
•  He,  B.,  C.  Li,  X.  Pan,  Z.  Wang,  J.  Zhao,  J.  Wang and  D.  Chen (2022) Effects of grain thickness on yield and quality of rice varieties in Jilin province. Food Sci Technol 42: e81422.
•  Hong,  Z.,  M.  Ueguchi-Tanaka,  K.  Umemura,  S.  Uozu,  S.  Fujioka,  S.  Takatsuto,  S.  Yoshida,  M.  Ashikari,  H.  Kitano and  M.  Matsuoka (2003) A rice brassinosteroid-deficient mutant, ebisu dwarf (d2), is caused by a loss of function of a new member of cytochrome P450. Plant Cell 15: 2900–2910.
•  Hour,  A.,  W.  Hsieh,  S.  Chang,  Y.  Wu,  H.  Chin and  Y.  Lin (2020) Genetic diversity of landraces and improved varieties of rice (Oryza sativa L.) in Taiwan. Rice (NY) 13: 82.
•  Huang,  R.,  L.  Jiang,  J.  Zheng,  T.  Wang,  H.  Wang,  Y.  Huang and  Z.  Hong (2013) Genetic bases of rice grain shape: so many genes, so little known. Trends Plant Sci 18: 218–226.
•  Ishimaru,  K.,  N.  Hirotsu,  Y.  Madoka,  N.  Murakami,  N.  Hara,  H.  Onodera,  T.  Kashiwagi,  K.  Ujiie,  B.  Shimizu,  A.  Onishi et al. (2013) Loss of function of the IAA-glucose hydrolase gene TGW6 enhances rice grain weight and increases yield. Nat Genet 45: 707–711.
•  Jackson,  M.T. (1997) Conservation of rice genetic resources: the role of the International Rice Genebank at IRRI. Plant Mol Biol 35: 61–67.
•  Kobayashi,  A.,  K.  Ebana,  S.  Fukuoka and  T.  Nagamine (2006) Microsatellite markers revealed the genetic diversity of an old Japanese rice landrace ‘Echizen’. Genet Resour Crop Evol 53: 499–506.
•  Kuroda,  A. and  N.  Matsumoto (1996) Effect of grain thickness and quality on physicochemical properties of rice. Hokuriku Journal of Crop Science 31: 11–12 (in Japanese).
•  Li,  N. and  Y.  Li (2016) Signaling pathways of seed size control in plants. Curr Opin Plant Biol 33: 23–32.
•  Li,  N.,  R.  Xu,  P.  Duan and  Y.  Li (2018) Control of grain size in rice. Plant Reprod 31: 237–251.
•  Li,  P.,  Y.  Chen,  J.  Lu,  C.  Zhang,  Q.  Liu and  Q.  Li (2022) Genes and their molecular functions determining seed structure, components, and quality of rice. Rice (NY) 15: 18.
•  Li,  Y.,  C.  Fan,  Y.  Xing,  Y.  Jiang,  L.  Luo,  L.  Sun,  D.  Shao,  C.  Xu,  X.  Li,  J.  Xiao et al. (2011) Natural variation in GS5 plays an important role in regulating grain size and yield in rice. Nat Genet 43: 1266–1269.
•  Liu,  J.,  J.  Chen,  X.  Zheng,  F.  Wu,  Q.  Lin,  Y.  Heng,  P.  Tian,  Z.  Cheng,  X.  Yu,  K.  Zhou et al. (2017) GW5 acts in the brassinosteroid signalling pathway to regulate grain width and weight in rice. Nat Plants 3: 17043.
•  Lyu,  J.,  D.  Wang,  P.  Duan,  Y.  Liu,  K.  Huang,  D.  Zeng,  L.  Zhang,  G.  Dong,  Y.  Li,  R.  Xu et al. (2020) Control of grain size and weight by the GSK2-LARGE1/OML4 pathway in rice. Plant Cell 32: 1905–1918.
•  Ohashi-Ito,  K.,  K.  Iwamoto,  A.  Yamagami,  T.  Nakano and  H.  Fukuda (2023) HD-ZIP Ⅲ-dependent local promotion of brassinosteroid synthesis suppresses vascular cell division in Arabidopsis root apical meristem. Proc Natl Acad Sci USA 120: e2216632120.
•  Saito,  S. (2006) Footprints of rice farmers—A brief history of breeding in Akita prefecture. Topics Sasaki, Daisen, 145–168 (in Japanese).
•  Shomura,  A.,  T.  Izawa,  K.  Ebana,  T.  Ebitani,  H.  Kanegae,  S.  Konishi and  M.  Yano (2008) Deletion in a gene associated with grain size increased yields during rice domestication. Nat Genet 40: 1023–1028.
•  Song,  X.,  W.  Huang,  M.  Shi,  M.  Zhu and  H.  Lin (2007) A QTL for rice grain width and weight encodes a previously unknown RING-type E3 ubiquitin ligase. Nat Genet 39: 623–630.
•  Tong,  H.,  L.  Liu,  Y.  Jin,  L.  Du,  Y.  Yin,  Q.  Qian,  L.  Zhu and  C.  Chu (2012) DWARF AND LOW-TILLERING acts as a direct downstream target of a GSK3/SHAGGY-like kinase to mediate brassinosteroid responses in rice. Plant Cell 24: 2562–2577.
•  Wan,  X.,  J.  Weng,  H.  Zhai,  J.  Wang,  C.  Lei,  X.  Liu,  T.  Guo,  L.  Jiang,  N.  Su and  J.  Wan (2008) Quantitative trait loci (QTL) analysis for rice grain width and fine mapping of an identified QTL allele gw-5 in a recombination hotspot region on chromosome 5. Genetics 179: 2239–2252.
•  Wang,  S.,  K.  Wu,  Q.  Yuan,  X.  Liu,  Z.  Liu,  X.  Lin,  R.  Zeng,  H.  Zhu,  G.  Dong,  Q.  Qian et al. (2012) Control of grain size, shape and quality by OsSPL16 in rice. Nat Genet 44: 950–954.
•  Weng,  J.,  S.  Gu,  X.  Wan,  H.  Gao,  T.  Guo,  N.  Su,  C.  Lei,  X.  Zhang,  Z.  Cheng,  X.  Guo et al. (2008) Isolation and initial characterization of GW5, a major QTL associated with rice grain width and weight. Cell Res 18: 1199–1209.
•  Xiong,  M.,  G.  Feng,  Q.  Gao,  C.  Zhang,  Q.  Li and  Q.  Liu (2022) Brassinosteroid regulation in rice seed biology. Seed Biol 1: 1–9.
•  Yamaguchi,  T.,  T.  Ebitani,  M.  Omoteno and  Y.  Yamamoto (2007) Japanese native red rice cultivar, ‘Awa-akamai’, a genetic resource for the improvement of germination in soil at low temperatures. Breed Res 9: 103–110 (in Japanese with English summary).
•  Zhang,  C.,  M.  Bai and  K.  Chong (2014) Brassinosteroid-mediated regulation of agronomic traits in rice. Plant Cell Rep. 33: 683–696.
•  Zhang,  T. (2020) When less is more: GSK2-OML4 module negatively regulates grain size in rice. Plant Cell 32: 1781.