Natural variation in the promoter of qRBG1/OsBZR5 underlies enhanced rice yield

Zhang, Qiuli; Wu, Renhong; Hong, Tao; Wang, Dachuan; Li, Qiaolong; Wu, Jiayi; Zhang, Han; Zhou, Kai; Yang, Hongxia; Zhang, Ting; Liu, JinXiang; Wang, Nan; Ling, Yinghua; Yang, Zhenglin; He, Guanghua; Zhao, Fangming

doi:10.1038/s41467-024-52928-9

Download PDF

Article
Open access
Published: 03 October 2024

Natural variation in the promoter of qRBG1/OsBZR5 underlies enhanced rice yield

Nature Communications volume 15, Article number: 8565 (2024) Cite this article

15k Accesses
9 Citations
1 Altmetric
Metrics details

Subjects

Abstract

Seed size, a key determinant of rice yield, is regulated by brassinosteroid (BR); however, the BR pathway in rice has not been fully elucidated. Here, we report the cloning and characterization of the quantitative trait locus Rice Big Grain 1 (qRBG1) from single-segment substitution line Z499. Our data show that qRBG1^Z is an unselected rare promoter variation that reduces qRBG1 expression to increase cell number and size, resulting in larger grains, whereas qRBG1 overexpression causes smaller grains in recipient Nipponbare. We demonstrate that qRBG1 encodes a non-canonical BES1 (Bri1-EMS-Suppressor1)/BZR1(Brassinazole-Resistant1) family member, OsBZR5, that regulates grain size upon phosphorylation by OsGSK2 (GSK3-like Kinase2) and binding to D2 (DWARF2) and OFP1 (Ovate-Family-Protein1) promoters. qRBG1 interacts with OsBZR1 to synergistically repress D2, and to antagonistically mediate OFP1 for grain size. Our results reveal a regulatory network controlling grain size via OsGSK2–qRBG1–OsBZR1–D2–OFP1 module, providing a target for improving rice yield.

Verification and dissection of one quantitative trait locus for grain size and weight on chromosome 1 in rice

Article Open access 14 September 2021

Reducing brassinosteroid signalling enhances grain yield in semi-dwarf wheat

Article Open access 26 April 2023

Comprehensive mapping and modelling of the rice regulome landscape unveils the regulatory architecture underlying complex traits

Article Open access 03 August 2024

Introduction

Improving grain yield in rice (Oryza sativa), a staple food crop worldwide, could help resolve food shortages caused by climate change and ever-increasing global population¹. Grain weight, a key determinant of yield typically used to evaluate grain size², is a polygenic trait for which numerous contributory genes have been identified^3,4. These include genes influencing cell proliferation and/or cell expansion in the grain hulls associated with multiple signaling pathways, including transcriptional regulation^5,6,7 and mitogen-activated protein kinase^8,9, G protein^10,11, ubiquitin-mediated proteasome^12,13 and plant hormone (e.g., brassinosteroid [BR]) pathways^14,15.

BRs are plant-specific steroids that regulate various aspects of plant growth and development¹⁶. Genetic variations affecting both BR biosynthesis and signaling pathway influence rice grain size; in rice, five BR biosynthesis-related genes (D2, DWARF4, D11, BRD1 and BRD2) have been isolated^{17,18,19,20,21}. BRs are first perceived by the membrane-localized receptor kinase BRI1 and the co-receptor BAK1, which initiate BR signaling and a cascade of cellular events²². This leads to inactivation of BIN2 and activation of two transcription factors (TFs), BZR1 and BES1, that regulate various downstream genes in Arabidopsis thaliana²². In rice, loss of OsBRI1 or OsBAK1 function yields BR-insensitive plants with small grains^23,24. OsGSK2, a homolog of Arabidopsis BIN2, affects cell proliferation and expansion to negatively control grain size by phosphorylating substrates including OsBZR1, GL2 and OFP1 etc.^{16,25,26,27,28,29,30,31}. The BES1/BZR1 family proteins are plant-specific TFs containing a conserved N-terminal region, the BES1 DNA-binding domain, that recognizes and binds E-box (CANNTG) and/or BRRE (CGTGC/TG) cis-acting elements in the promoter regions of thousands of target genes^32,33. The family contains at least six members in Arabidopsis and four (OsBZR1–OsBZR4) in rice³³. Overexpressing OsBZR1 increases grain length, width and weight³⁴, and the importance of OsBZR1–OsBZR4 in regulating rice seed size was further confirmed using single and higher-order knockout mutants of these genes³⁵. However, the BR pathway still requires further elucidation, and some constituent genes remain unidentified.

Here, we identified an unselected rare variation, qRBG1^Z, encoding a non-canonical BES1/BZR1 family member, OsBZR5, that negatively regulates rice grain size via the OsGSK2–qRBG1–OsBZR1–D2–OFP1 regulatory module. Our results help elucidate the molecular mechanisms by which the BR pathway regulates grain size and open the door to a new strategy for improving rice productivity.

Results

Large grains in the single-segment substitution line (SSSL) Z499 are caused by qRBG1 ^Z influencing both cell division and expansion

A chromosome segment substitution line (CSSL), Z1364, with three substitution segments (averaging 1.19 Mb) was previously developed from advanced backcrosses of Nipponbare (small grains) as the recipient and Xihui 18 (large grains) as the donor parent. By mapping the quantitative trait loci (QTLs) associated with large grains in a secondary F₂ population derived from Nipponbare/Z1364, qGW1 for grain width and qGWT1 for grain weight on chromosome 1 were identified in the RM6667 substitution segment on chromosome 1³⁶. A SSSL, Z499, containing the QTL associated with large grains was then developed by fine-mapping qGWT1 and applying simple sequence repeat (SSR) marker-assisted selection (MAS) to the progeny of a Nipponbare × Z1364 cross. Z499 harbors the substitution segment RM10198--RM1329–RM10224--RM6777 on chromosome 1, with the longest substitution length of 0.35 Mb, an estimated length of 0.20 Mb and the shortest length of 50-kb (Supplementary Fig. 1a, b).

Z499 is still different from Nipponbare in plant morphology (leaf angle, panicle bending) (Fig. 1a), However, the outstanding character is its long-wide-large grains (Fig. 1b). We analyzed the additive effects of the qRBG1 (qGL1, qGW1 and qGWT1), revealing that grain length (8.21 mm), grain width (3.64 mm) and 1000-grain weight (25.95 g) of Z499 were significantly increased by the qGL1 (additive effect = 0.73 mm), qGW1 (additive effect = 0.14 mm) and qGWT1 (additive effect = 3.27 g) QTLs, respectively, in comparison with Nipponbare (Fig. 1c–e). Increased grain size often influences rice quality³⁷; however, there were no significant differences between the genotypes in brown rice rate, chalkiness ratio or chalkiness degree (Supplementary Fig. 2a–f). Thus, the rice quality of Z499 was unaffected by the increases in grain length and width caused by qRBG1.

**Fig. 1: Phenotypes analysis of Nipponbare (Nip) and Z499.**

Scanning electron microscopy (SEM) observation showed that Z499 glumes contained significantly more cells than Nipponbare in both the longitudinal (7.8% more) and transverse directions (3.2% more), and the inner epidermal cells of Z499 were significantly longer (8.5%) and wider (5.8%) (Fig. 1f–j). The cell number and area in the outer parenchyma cell layer of Z499 were increased by 9.8% and 26.0%, respectively, relative to Nipponbare, as determined by paraffin section analysis (Fig. 1k–n). Moreover, Z499 endosperm fresh weight and dry weight were significantly higher than those of Nipponbare at 6d after fertilization; these traits peaked after 21 d were 32.1% and 28.6% greater than those of Nipponbare, respectively (Supplementary Fig. 3a–c). This indicated that the heavier grains of Z499 resulted from larger hulls, accelerated accumulation of dry matter, and accelerated grain filling. Finally, average grain yield per hectare over 3 years (2021–2023) was significantly (18.1%) higher for Z499 than for Nipponbare (3805.3 kg versus 3221.8 kg) (Fig. 1o; Supplementary Table 1).

Map-based cloning of qRBG1

For fine-mapping of qRBG1, we developed an F₃ population of 900 plants descended from a recombinant plant carrying a single heterologous genome segment containing the qRBG1 locus. We then fine-mapped qRBG1^Z to the 50-kb region RM1329–RM10224 on chromosome 1 using 220 large-grain (recessive) plants and two newly developed polymorphic SSR markers (Fig. 2a). This region contains seven annotated genes: Os01g0177150 (a non-protein-coding transcript), LOC_Os01g08160 (a MYB TF), Os01g0176751 (a hypothetical protein), LOC_Os01g08170 (an expressed protein), LOC_Os01g08180 (BES1/BZR1 homolog protein-like), LOC_Os01g08190 (a LEUNIG protein) and LOC_Os01g08200 (UBIQUITIN-SPECIFIC PROTEASE 14) (Fig. 2a; Supplementary Fig. 4a–e). All of the genes displayed variations in their coding or non-coding sequences (Fig. 2a; Supplementary Fig. 4a–e). We subjected six of them (excluding the non-protein-coding Os01g0177150) to a genetic complementation test by introducing a Nipponbare DNA fragment for each gene into Z499. Only LOC_Os01g08180 restored a small grain phenotype similar to that of Nipponbare when complemented (Fig. 2b; Supplementary Fig. 5a–d). Thus, LOC_Os01g08180 is the likely target gene of qRBG1.

LOC_Os01g08180 exhibits two 4-bp deletions, seven SNPs and a 1-bp promoter region insertion in Z499 compared with Nipponbare (Fig. 2a). We transformed the dominant coding sequence of LOC_Os01g08180 from Nipponbare driven by the Ubiquitin promoter into Z499 to produce transgenic plants. All positively transformed lines recovered a small grain phenotype similar to that of Nipponbare, and LOC_Os01g08180 transcript levels in the T₀ transgenic plants were significantly higher than in Nipponbare or Z499 plants (Fig. 2b–f). Next, we selected two target sites in the qRBG1 coding sequence to generate qRBG1 knockout mutants using CRISPR/Cas9 in the Nipponbare background (Fig. 2a). Mutant lines with 20-bp (cr-qrbg1-1) and 1-bp (cr-qrbg1-2) deletions causing reading frame shifts and premature termination displayed significant increases in grain length, grain width and 1000-grain weight relative to Nipponbare, along with a significant decrease in qRBG1 transcript level (Fig. 2b–f). RNA interference (RNAi) lines generated in Nipponbare showed significantly lower qRBG1 expression and larger grains compared to Nipponbare, resulting from a significant increase in hull cell number and size (Fig. 2b–f; Supplementary Fig. 6a–k). qRBG1 overexpression (OE) transgenic lines exhibited significantly increased qRBG1 expression and formed small grains due to smaller and less abundant hull cells (Fig. 2b–f; Supplementary Fig. 7a–k). Moreover, by analyzing the promoter activity of qRBG1^N and qRBG1^Z in rice protoplasts, the result exhibits that the promoter of qRBG1^Z has lower LUC activity than that of qRBG1^N (Supplementary Fig. 8a–b). Together, these data suggest that the promoter variations of LOC_Os01g08180 underlie the qRBG1^Z variation in Z499 that causes its large-grain phenotype.

qRBG1 encodes a non-canonical BES1/BZR1 family member OsBZR5

By aligning multiple sequences related to qRBG1, a BES1/BZR1-like protein, we determined that the C terminus exhibited obvious differences from the other four BES1/BZR1 family proteins reported in rice (OsBZR1–OsBZR4) (Supplementary Fig. 9)^34,35. The N-terminal sequence of qRBG1 had greater similarity with OsBZR1–OsBZR4, sharing 26.7%, 29.4%, 25.7%, and 21.9% amino acid sequence identity, respectively (Supplementary Fig. 9). Phylogenetically, qRBG1 is located in an independent branch relatively distant from them (Fig. 3a; Supplementary Table 2), and it contains two conserved motifs (BES1 DNA-binding domain and EAR motif), in contrast to five in OsBZR1–OsBZR4 (BES1 DNA-binding domain, β-hairpin, PEST motif, 14-3-3 binding site and EAR motif)^38,39,40,41 (Fig. 3a; Supplementary Fig. 9). The qRBG1 promoter contains four seed-specific regulatory elements (RY elements) (Supplementary Fig. 10). Overexpression of OsBZR1 increases grain size, whereas higher-order mutants of OsBZR1–OsBZR4 have decreased grain size^34,35. In contrast, reduced expression of qRBG1 in Z499 increases its grain size (Fig. 2b–f). These results indicate that qRBG1, though differing substantially from OsBZR1–OsBZR4, constitutes a non-canonical BES1/BZR1 gene family member involved in regulating rice grain size, which we named OsBZR5.

Expression pattern and subcellular localization of qRBG1

qRT-PCR analysis showed that qRBG1 was ubiquitously expressed in the roots, stems, leaves, sheaths and panicles, especially 0.5-cm young panicles, of Nipponbare (Fig. 3b). When GUS expression was driven by the qRBG1 promoter (proqRBG1:GUS), we detected GUS activity in the roots, stems, leaves, sheaths, spikelet hulls and young panicles of Nipponbare (Fig. 3c–i). Although qRBG1/OsBZR5 is ubiquitously expressed, while displaying higher expression in panicle, thus we only pay attention to the direct grain variation. An in situ hybridization in the young panicles revealed qRBG1 signals in the primordia of lateral organs such as the lemma, palea, floral meristem, stamen, carpel, rudimentary glume and sterile lemma (Fig. 3j–l). These data indicated that qRBG1 mainly influences grain development. Furthermore, three experiments using a qRBG1-GFP fusion protein revealed that the qRBG1 protein is localized to the nucleus (Fig. 3m, n; Supplementary Fig. 11).

OsGSK2 phosphorylates qRBG1 to reduce its stability

qRBG1 is homologous to OsBZR1 (Fig. 3a; Supplementary Fig. 9), the core TF in BR signaling that is phosphorylated by OsGSK2 to regulate downstream BR signaling²⁷. We wondered if qRBG1 interacts with, and is phosphorylated by, OsGSK2. A yeast two-hybrid (Y2H) assay showed that qRBG1 interacted with OsGSK2 (Fig. 4a; Supplementary Fig. 12), which we further verified through bimolecular fluorescence complementation (BiFC) assays in Nicotiana benthamiana leaves and rice protoplasts (Fig. 4b, c) and a glutathione S‐transferase (GST) pull-down assay in vitro (Fig. 4d). The data suggested that OsGSK2 interacts with qRBG1 and that the latter might be a substrate of the former. To explore this hypothesis, we performed a kinase assay in vitro with GST-tagged OsGSK2 (GST-OsGSK2) and His-tagged qRBG1 (His-qRBG1), showing that OsGSK2 phosphorylated qRBG1 (Fig. 4e). As OsGSK2 regulates the stability and activity of many TFs by phosphorylating them⁴², we wondered whether OsGSK2 mediated qRBG1 stability. Considering BR inhibit OsGSK2 activity⁴³, we detected whether BR influenced qRBG1 stability. Our data showed that an exogenous 2,4-epiBL treatment promoted the accumulation of qRBG1-GFP protein in the transgenic plants (Fig. 4f). We also detected a significant accumulation of qRBG1-GFP following treatment with the GSK3-like kinase activity inhibitor bikinin (Fig. 4g). Taken together, these results revealed that OsGSK2 interacts with and phosphorylates qRBG1, reducing its stability; thus, qRBG1 is a substrate of OsGSK2. Finally, reduced OsGSK2 expression increased hull cell number and size, resulting in big grains, whereas its overexpression led to smaller grains (Supplementary Fig. 13a–h). Thus, the regulation of grain size by OsGSK2 was consistent with that of qRBG1, further indicating that both genes regulate grain size through a common pathway.

**Fig. 4: OsGSK2 phosphorylates qRBG1 and decreases qRBG1 protein stability.**

qRBG1 and OsBZR1 can form either homo- or hetero-dimers

Since qRBG1/OsBZR5, lack three important conserved motifs, may compete with OsBZR1, for homo- or hetero-dimer formation, to suppress their activity. We first demonstrated that qRBG1 lacks auto-activation ability using a Y2H experiment (Supplementary Fig. 14a). Y2H screening assay of a young panicle cDNA library showed that OsBZR1 interacts with qRBG1 (Fig. 5a; Supplementary Fig. 12; Supplementary Table 3). The interaction was further confirmed through BiFC assays in both N. benthamiana leaves and rice protoplasts and through an in vitro GST pull-down assay (Fig. 5b–d). Besides, qRBG1 and OsBZR1 also interacted with themselves in yeast cells (Fig. 5a). BiFC assays further confirmed their interactions (Fig. 5b). The results suggested that qRBG1 and OsBZR1 might function by forming homo- or hetero-dimers.

**Fig. 5: The interaction assays of both qRBG1 and OsBZR1.**

qRBG1 acts genetically upstream of D2 and together with OsBZR1 synergistically mediates D2 expression

qRBG1, a TF without self-activation activity, might function as a transcriptional inhibitor (Fig. 3a; Supplementary Fig. 14a). A dual-luciferase reporter system analysis showed that the positive control (VP16) had significantly higher luciferase (LUC) activity than the VP16-qRBG1 fusion protein (Supplementary Fig. 14b, c), indicating that qRBG1 might function as a transcriptional repressor.

To identify target genes of qRBG1, we performed qRT-PCR analysis, showing that D2 was significantly upregulated in Z499 and downregulated in the qRBG1-OE transgenic lines compared with Nipponbare (Fig. 6a). D2 was previously reported to positively regulate grain size as a gene associated with BR biosynthesis¹⁷. A dual-luciferase result showed that qRBG1 inhibited D2 expression (Fig. 6b, c). BES1/BZR1 proteins regulate the expression of many genes via their BES1 DNA-binding domain, which recognizes and binds cis-acting E-box (CANNTG) and/or BRRE (CGTGC/TG) elements in target promoter regions. Similarly, qRBG1 contains a BES1 DNA-binding domain, and the D2 promoter contains three E-box and three BRRE elements (Fig. 6d); therefore, we explored whether qRBG1 binds these promoter elements to regulate D2 expression. In vitro electrophoretic mobility shift assays (EMSA) demonstrated that qRBG1 binds the E-box, but not BRRE, elements in the D2 promoter (Fig. 6e, f). A microscale thermophoresis (MST) assay showed that these E-box elements caused concentration-dependent effects on the fluorescently labeled qRBG1 (Fig. 6g), implying specific binding between those elements and qRBG1. Chromatin immunoprecipitation (ChIP) assay exhibited that qRBG1 could indeed bind stably to the E-box elements of D2 promoter (Supplementary Fig. 14d). Taken together, these results indicated that qRBG1 directly represses D2 expression by binding to E-box elements in its promoter.

**Fig. 6: qRBG1 represses D2 expression by directly binding to E-box element of D2 promoter.**

To further explore the genetic relationship between qRBG1 and D2, we produced two D2 knockout mutants using CRISPR/Cas9 in Nipponbare and Z499. The Nipponbare D2 knockout line, cr-d2, had a 1-bp insertion and smaller grains than the wild type (Fig. 6h–k; Supplementary Fig. 14e). The Z499 D2 knockout line (carrying abnormal qRBG1), Z499/cr-d2 (a double mutant), had significantly smaller grains than Z499 (large grains) or Nipponbare (small grains) (Fig. 6h–k). These results suggest that qRBG1 functions upstream of D2 to control grain size.

Considering that qRBG1 interacts with OsBZR1 and D2 was the downstream target gene of qRBG1, we suspected qRBG1 and OsBZR1 might form a complex co-regulating D2 expression. To test this, we first performed an EMSA, revealing that qRBG1 and OsBZR1 both bind the D2 promoter. Notably, qRBG1 binds only the E-box element, whereas OsBZR1 binds both the E-box and BRRE elements (Fig. 6d–f and Fig. 7a, b). Next, using a LUC reporter gene driven by the D2 promoter and qRBG1 and OsBZR1 as effectors, we observed that qRBG1 or OsBZR1 alone repressed the transcription activity of the D2 promoter, and its transcription was further decreased when qRBG1 or OsBZR1 were co-expressed in a dual-luciferase reporter system (Fig. 7c, d). These results indicated that qRBG1 and OsBZR1 form a complex that synergistically inhibits D2 expression.

**Fig. 7: qRBG1 and OsBZR1 synergistically repress D2 expression and qRBG1 may directly inhibit *OFP1* expression.**

qRBG1 and OsBZR1 antagonistically mediate OFP1 expression

OsBZR1 has dual roles in inhibiting BR biosynthesis and activating BR signaling^9,26. We wondered if qRBG1 and OsBZR1 co-mediate BR signaling to antagonistically control grain size. Considering that OFP1, as the downstream target gene of OsBZR1, and OsBZR1 positively regulate grain morphology in BR signaling²⁶, we conducted qRT-PCR analysis, which showed that OFP1 was significantly downregulated in the qRBG1-OE transgenic line and upregulated in Z499 and Osbzr1-D (OsBZR1 gain-of-function mutant) compared with Nipponbare (Fig. 7e). The OFP1 promoter contains three potential E-box elements (Fig. 7f) that qRBG1 might bind (Fig. 6d–g; Supplementary Fig. 14d). As expected, qRBG1 directly bound the OFP1 promoter in a yeast one-hybrid (Y1H) assay (Fig. 7g), and could bind stably to the E-box element of OFP1 promoter by ChIP-qPCR and EMSA (Fig. 7h, i). In addition, In vitro EMSA demonstrated that qRBG1 and OsBZR1 competitively bind the E-box in the OFP1 promoter (Fig. 7i–k). Thus, qRBG1 and OsBZR1 antagonistically mediate OFP1 expression in BR signaling pathway.

The qRBG1 ^Z variation in Z499 is rare despite abundant natural promoter variation in qRBG1 between indica and japonica

To further explore the effect of qRBG1 promoter variation on grain size and the evolutionary pattern of qRBG1 during rice domestication, we mainly identified nine haplotypes of qRBG1 with variation in the promoter region (2-kb upstream of the transcription start site) using data from the Rice Variation Map (Fig. 8a). The qRBG1^Z haplotype in Z499, resulting in long, wide and large grains, was not seen in any of the 2604 varieties in the database and was therefore considered a rare variation. Z499 is a SSSL in the Nipponbare genetic background, which is a japonica-type indica-compatible line. HapV except for the C at the –64,461^th base, Hap VI except for the C at the –64,461^th base and no deletion of “ACAC”, and Hap IV except for the C at the –64,461^th base and no deletion of “ACAC” and “TGAG”, were similar to qRBG1^Z. However, these haplotypes account for 72.0% of 1321 indica varieties, and all had long, narrow grains, not wide grains (Fig. 8a, b). In addition, Hap III with 3 SNP and no deletion of “ACAC” from qRBG1^Z also displayed indica grain shape, accounting for 25% of indica varieties. Hap I varieties, including Nipponbare, account for 25.0% of the 739 japonica varieties, 20.0% of Aus varieties and 4.0% and 8.0% of intermediate-type and Aro varieties; Hap II except for 1 deletion of “ACAC”, and Hap IX except for 2 SNP and 1 deletion of “ACAC” and Hap VIII except for 2 SNP were similar to Nipponbare, accounting for 73% of japonica varieties. While Hap I and Hap II exhibited short-wide type, whereas Hap VIII and Hap IX displayed long-wide type. Hap VII, showing more differences from both Nipponbare or qRBG1^Z, accounted for 18.0%, 11.0%, 19.0% and 1.0% of Aus, intermediate-type, Aro and indica cultivars (Fig. 8a, b). In addition, there are the other 16 haplotypes with fewer cultivars listed in supplementary Fig. 15. Thus, the qRBG1 promoter has acquired abundant variation during indica and japonica subspecies evolution and artificial domestication selection. In particular, the qRBG1^Z variation in Z499 is rare and has not been selected in previous breeding efforts, suggesting that it could have great potential in breeding of high-yielding rice cultivars.

Fig. 8: Natural variation and haplotype analysis of *qRBG1.*

Discussion

BR play important roles in controlling grain size, leaf angle and other plant developmental characteristics¹⁶. The BES1/BZR1 family proteins in the BR pathway are plant-specific TFs containing a non-canonical basic helix-loop-helix (bHLH) domain, the BES1 DNA-binding domain³³. In Arabidopsis, this family has been widely studied, with a particular focus on BZR1 and BES1³³. In rice, OsBZR1–OsBZR4 have been described³⁵. Here, we characterized qRBG1, encoding OsBZR5, a previously uncharacterized member of this family. qRBG1 possesses a BES1 DNA-binding domain (Fig. 3a) that can directly bind E-box elements in the D2 promoter (Fig. 6d–g; Supplementary Fig. 14d). qRBG1 is distinct from OsBZR1–OsBZR4 and located on an independent phylogenetic branch due to its different C-terminal region and low homology with these proteins (26.7%, 29.4%, 25.7% and 21.9% amino acid sequence similarity with OsBZR1–OsBZR4, respectively; Supplementary Fig. 9). qRBG1 contains only two relatively conserved motif, BES1 and EAR motif, whereas OsBZR1–OsBZR4 contain five each (Fig. 3a), reflecting likely functional differences between qRBG1 and the other four rice BES1/BZR1 family members. OsBZR1–OsBZR4 positively regulate grain size^34,35, whereas our research shows that the Nipponbare qRBG1 allele (qRBG1^N) has an opposite effect, negatively regulating grain size (Fig. 1b–e). Thus, we classify qRBG1 as a non-canonical BES1/BZR1 family member, OsBZR5, that might play a crucial role in regulating grain size via the BR pathway.

Although qRBG1/OsBZR5 belongs to BES1/BZR1 family protein, it seems opposite in function to OsBZR1 for regulating grain size. Due to qRBG1, lack of three important conserved motifs, may compete with OsBZR1, for DNA binding (Fig. 7k) or homo- or hetero-dimer formation (Fig. 5), to suppress their activity. Especially qRBG1 only keeps the BES1 DNA-binding domain and EAR motif (a well-known transcriptional repression-associated motif) (Fig. 3a; Supplementary Fig. 9). In our study, qRBG1 inhibits D2 expression (a BR synthesis gene) by binding the E-box elements of its promoter (Fig. 6a–k), whereas OsBZR1 inhibits D2 expression by binding both the E-box and BRRE promoter elements (Fig. 7a–b). qRBG1 interacts with OsBZR1 to synergistically inhibit D2 expression in BR synthesis (Figs. 5a–d and 7c–d). Previous studies have shown that OsBZR1 plays dual roles in feedback inhibition BR synthesis and activating BR signaling^9,26. Our data show that qRBG1 directly represses the expression of OFP1 (a key gene for regulating grain size) in BR signaling (Fig. 7e–i), while OsBZR1 directly activates OFP1 expression (Fig. 7e, j). Moreover, our results also show that qRBG1 and OsBZR1 competitively bind the E-box in the OFP1 promoter (Fig. 7k), suggesting that qRBG1 act antagonistically to OsBZR1 in regulating OFP1 expression. Based on these results, the plausible model should be that qRBG1/OsBZR5 and OsBZR1 synergistically repress D2 expression in BR synthesis pathway (Figs. 6a–k and 7c, d) but antagonistically mediate OFP1 expression in BR signaling pathway to regulate grain size (Fig. 7e, k). The model can then be verified by a series of experiments. Firstly, the BR sensitivity test of the Nipponbare, qRBG1-related plant (qRBG1-OE) and OsBZR1 gain-of-function mutant (Osbzr1-D) showed that qRBG1-OE is insensitive to exogenous 2,4-epiBL, while Osbzr1-D is sensitive to exogenous 2,4-epiBL compared with Nipponbare (Supplementary Fig. 16). The result indicates that qRBG1 had an opposite function to OsBZR1 in BR signaling. Then, we comprehensively analyzed the expression of BR synthesis-related genes and BR signaling-related genes in young panicles and shoots of Nipponbare, qRBG1-RNAi, qRBG1-OE and Osbzr1-D plants. The results showed that the other BR synthesis-related genes (DWARF4, D11, BRD1 and OsCPD2) were significantly downregulated in young panicles and shoots of qRBG1-OE and Osbzr1-D plants, while upregulated in young panicles and shoots of qRBG1-RNAi compared with Nipponbare; whereas some BR signaling-related genes (OsBZR1, OsBU1, OsBSK2 and OsILI1) were significantly downregulated in young panicles and shoots of qRBG1-OE plants, while upregulated in young panicles and shoots of qRBG1-RNAi and Osbzr1-D compared with Nipponbare (Supplementary Figs. 17 and 18). These results suggest that qRBG1 and OsBZR1 synergistically repress BR synthesis, whereas qRBG1 had an antagonistic function to OsBZR1 in BR signaling pathway. Finally, we measured the BR content in Nipponbare, Z499, qRBG1-OE, qRBG1-RNAi and Osbzr1-D plants by HPLC-electrospray ionization tandem mass spectrometry system (HPLC-MS/MS). The results showed that BR content in qRBG1-OE was significantly lower than that in Z499 and qRBG1-RNAi (Supplementary Fig. 19), suggesting that BR content change is consistent with the qRBG1 expression. All these results support that qRBG1/OsBZR5 negatively regulates while OsBZR1 positively regulates rice grain size. Similarly, OsLIC1 and OsBZR1 antagonize each other in controlling BR-mediated leaf bending⁴⁴.

Based on these findings, we propose a working model for how rice grain size is regulated via the OsGSK2–qRBG1–OsBZR1–D2–OFP1 pathway (Fig. 9a, b). OsGSK2 as a negative regulator have been reported by phosphorylating the substrates of BR signaling, including OsBZR1, GL2, OFP1, DLT, OFP3, OFP8, SG2, OML4, RLA1, OsLIC and OsNAC016^{16,25,26,27,28,29,30,31,44,45,46}. In Nipponbare plants without BR, OsGSK2 phosphorylates normal qRBG1 to reduce qRBG1^N protein abundance (Fig. 4e–h), and the transcriptional activity of the qRBG1^N–OsBZR1 complex is inhibited. In Nipponbare plants with BR, OsGSK2 cannot phosphorylate qRBG1^N, resulting in the release and stabilization of qRBG1^N that can then form a complex with OsBZR1 to synergistically repress D2 expression. qRBG1^N also inhibits OFP1 expression in BR signaling, resulting in small grains similar to the those of d2 mutant. WRKY53 and OsBZR1 are known to synergistically repress D2 to regulate BR signaling⁹, indicating that qRBG1 or WRKY53 might function in parallel with OsBZR1 in the BR pathway. In Z499, due to the promoter variation of qRBG1, abnormal qRBG1^Z cannot inhibit D2 expression regardless of the presence or absence of BR, and OFP1 must be more strongly expressed to further activate BR signaling, resulting in large grains. This model clearly illustrates how qRBG1 negatively regulates rice grain size.

**Fig. 9: A proposed working model for *qRBG1* in the regulation of grain size.**

It is worth mentioning, although qRBG1 is ubiquitously expressed, which may have an effect on the other tissues of Z499 (Fig. 1a), such as leaf angle etc., we mainly pay attention to the direct grain variation of qRBG1. The other phenotypes are very complex due to the other 5 different genes existed between Z499 and Nipponbare (Fig. 1a; Supplementary Fig. 4a–e). Just as increased BR content usually lead to increased leaf angle^27,44, which seems inconsistent with the slightly erect leaves observed in Z499 (Fig. 1a). This phenotypic complexity may be due to another gene (LOC_Os01g08160) encoding a MYB TF, which is located in the same substitution interval as qRBG1 in Z499. This was confirmed by the genetic transformation of LOC_Os01g08160^N into Z499 (Supplementary Fig. 20), although this preliminary finding requires further research to explain how the leaf angle and compact plant phenotype of Z499 are regulated by both MYB and qRBG1^Z–OsBZR1–D2.

SSSLs can be valuable for the introgression of beneficial genes from distantly related species into cultivars and for the dissection of QTLs for complex traits. Because each SSSL carries only one single substituted segment in the genetic background of a recipient parent, they facilitate breeding-by-design strategies in which target genes are pyramided to develop new lines⁴⁷. Here, we characterized the SSSL-Z499, which carries a substitution segment from indica Xihui 18 in the japonica Nipponbare genetic background. Z499, harboring an unselected rare allele qRBG1^Z, produced long, wide and large grains, whereas plants with the same qRBG1 haplotype as Nipponbare (Hap I) displayed the short, wide grains typical of japonica rice. Although Hap IV, HapV, and Hap VI are closer to qRBG1^Z, their grains are the long, narrow type found in indica rice (Fig. 8a, b). The unselected rare allele qRBG1^Z therefore has great potential for improving rice yield, as validated by the 18.1% higher average yield per hectare (3805.3 kg) we observed for Z499 compared to Nipponbare (3221.8 kg) over 3 years (2021–2023) (Fig. 1o). Unlike other yield-related genes in rice, such as Ghd7 gene⁴⁸, qRBG1 increases grain yield not by delaying the heading date. Z499, which produced large grains, had the same heading date as Nipponbare (Fig. 1a) and similar grain quality traits, such as chalky grain rate and chalkiness degree (Supplementary Fig. 2e, f). Thus, qRBG1^Z is an attractive target for improving grain yield of japonica cultivars, as it enhances rice yield without negatively affecting quality. Moreover, as a japonica-type, indica-compatible SSSL, Z499 could be used to breed excellent rice cultivars with larger heterosis through crossing with a series of SSSLs containing other favorable alleles, such as qSSD5 and qSSD8, using MAS⁴⁹. In addition, Z499 carrying qRBG1^Z could also be transformed a big grain strong restorer line by pyramiding with 4 well-developed SSSLs-Nipponbare carrying strong restorer genes Rf1 (Chr.1), Rf2 (Chr.2), Rf3 (Chr.1) and Rf4 (Chr.10)^50,51,52,53 from indica restorer line Xihui 18. Then, the transformed restorer line with qRBG1^Z and Rf1-4 genes can be combined with various Japponica-type sterile lines to breed excellent hybrid rice cultivars. As a key regulator of grain size, qRBG1 will facilitate breeding by design to improve grain yield in rice.

Methods

Plant materials

The main materials were a single-segment substitution line (SSSL) Z499 harboring qRBG1 and its recipient parent Nipponbare. Z499 was developed from the progeny of a fine-mapping population (F_2:3) derived from a recombinant individual of qRBG1 in the F₂ population from the crossing of Niponbare with Z1364 (Supplementary Fig. 1a). Z1364 was a chromosome segment substitution line (CSSL) carrying three substitution segments (1.19 Mb of average substitution length) from an indica restorer line Xihui 18 in the genetic background of japonica rice Nipponbare, which was developed from advanced backcrosses of Nipponbare (small-grains) as the recipient and Xihui 18 (large grains) as the donor parent³⁶. In addition, the other materials used in the study included knockout transgenic lines, RNAi transgenic lines, overexpression transgenic lines, proqRBG1:GUS transgenic and complementary transgenic lines of qRBG1, knockout transgenic lines of D2 in the Nipponbare and Z499 background, gain-of-function mutant of OsBZR1 in Nipponbare genetic background (Osbzr1-D), as well as overexpression transgenic lines and RNAi transgenic lines of OsGSK2 gene²⁷.

Field experiment

The experiments were conducted on the experimental farm of Southwest University, Beibei, Chongqing (at 106° east longitude and 29° north latitude), China, in Spring (from March to July) of 2021, 2022 and 2023. The SSSL-Z499 and its recipient parent Nipponbare (Nip) were grown in all three years. In each experiment, the seeds were sown in a seedling bed and seedlings were transplanted to each plot in 35 days, with a space of 16.67 cm between hills and 26.67 cm between rows. Each plot consisted of 10 rows of long with 10 hills, and there was one plant per hill. All plots were arranged in a randomized complete block design with three replications. All transgenic lines of qRBG1 were grown on some experimental years with the same planting size, each line for 30 plants. The management of the field experiments was in accordance with Chongqing local standard practices. At maturity, grain length, grain width and 1000-grains weight were measured for each of the 10 hills from the middle of each plot, and yield of each plot was measured and transformed into yield per hectare (ha). Finally, the average value of the 10 hills for each grain size trait of all transgenic lines and the yield per plot of Nipponbare and Z499 were used as the raw data for further statistical analysis.

Identification of QTL for grain size in SSSL-Z499

Since Z499 carrying only one single substitution segment from donor Xihui 18 in the genetic background of Nipponbare, we gave the hypothesis (H₀) that no QTL for grain size existed in the substitution segment of Z499. When the P value was less than 0.05 according to two-tailed Student’s t test with Niponbare by Z499 in GraphPad Prism 8.0.2, we denied the hypothesis and considered that a QTL for a certain trait existed in Z499. According to the genetic model, P₀ = μ₀ + ε for Nipponbare and P_i = μ₀ + a + ε for Z499 carrying a specific QTL, where P₀ and P₁ represented the phenotypic value of any plant in a plot of Niponbare and Z499. μ₀ represented the mean value of the Nipponbare population, a represented the additive effect of the QTL from substitution segment of Xihui 18 donor in Z499, whose positive effect indicated increasing phenotypic value and negative one showed decreasing phenotypic value in substitution lines, and ε represented residual error. Thus, the additive effect of the QTL was calculated as half the difference between the mean phenotypic values of the SSSL and Nipponbare (the other half was estimated from environment)⁴⁹.

Scanning electron microscopy analysis

Spikelet hulls from Nipponbare, Z499, qRBG1-RNAi and qRBG1-OE transgenic plants were collected at the heading stage. The size of epidermal cells in spikelet hulls were investigated using the SU3500 scanning electron microscope (Hitachi, Tokyo, Japan). Image J was used to measure cell length, cell width and cell number. Comparisons were made by two-tailed Student’s t test or one-way ANOVA and Duncan multiple comparisons test (P < 0.05) in statistical software GraphPad Prism 8.0.2.

Histological analysis

At the heading stage, the spikelet hulls of Nipponbare, Z499, qRBG1-RNAi and qRBG1-OE plants were used for paraffin section. Sections were imaged using a Nikon Eclipse E600 microscope (Nikon, Tokyo, Japan). Image J was used to measure cell number and cell area.

RNA isolation and qRT-PCR analysis

Rice total RNA from the roots, stems, leaves, leaf sheaths, and developing panicles were extracted using the KK Fast Plant Total RNA Kit (Zhuangmeng, Beijing, China), and the first-strand cDNA was synthesized from 1 µg of purified total RNA in a 20 µL reaction volume with the SuperScript III Reverse Transcriptase Kit (Invitrogen, Shanghai, China). The cDNA after dilution with ddH₂O was used as a template for qRT-PCR analysis. qRT-PCR analysis was performed using the CFX Connect™ Real-Time System (Bio-Rad, Berkeley, CA, USA) and a SYBR premix Ex Taq II Kit (TaKaRa, Kyoto, China), and ACTIN was used as the endogenous control. All samples were analyzed in the three biological and technological replicates, and all data were performed with Eppendorf real-time PCR detection system and software. The primers used are listed in Supplementary Data 1.

Multiple sequence alignment and phylogenetic analysis

Protein sequences from rice and Arabidopsis were obtained using BLAST servers in gramene (https://ensembl.gramene.org/) or NCBI. The motif and domain structures were drawn using the TBtools software⁵⁴. The sequence of upstream 2-kb of ATG was used as the assumed promoter region, the cis-acting elements were identified and retrieved in PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/), and the visual display was realized by TBtools⁵⁵. The results of protein sequence alignment obtained through ClustalW were drawn by Jalview software. A phylogenetic tree was constructed based on the result of protein sequence alignment through ClustalW, in MEGA 5.0 software⁵⁶. The phylogenetic tree was built using the maximum likelihood method based on the Jones, Taylor, Thornton matrix-based model with the lowest Bayesian information criteria scores⁵⁶. Each node’s bootstrap support values from 1000 replicates were contained next to the branches.

Vector construction and rice transformation

To generate the complementation construct, the coding sequence of qRBG1 was amplified from Nipponbare and cloned into the digested PTCK303 vector containing Ubiquitin promoter. The genomic fragment containing the LOC_Os01g08160, Os01g0176751, LOC_Os01g08170, LOC_Os01g08190 and LOC_Os01g08200 were cloned into the binary vector pCAMBIA1301, respectively. These recombinant plasmids were respectively introduced into Z499 using the Agrobacterium tumefaciens-mediated transformation method as described previously⁵⁷. For the overexpression, the CDS of qRBG1 and the GFP sequence fragment were cloned into the binary vector PTCK303 vector. To create the CRISPR-Cas9 (cr-qrbg1) genome-edited lines, two targets of qRBG1 respectively were inserted into sgRNA construct, then the fragment with targets and sgRNA was efficiently annealed and connected into vector pYLCRISPR/Cas9 binary vectors. To generate the RNAi construct, a 256 bp qRBG1 complementary DNA was amplified and cloned into the vector pTCK303 to get the intermediate vector. A 2923 bp (before start codon) promoter sequence of qRBG1 was amplified from Nipponbare and ligated to the digested pCAMBIA1301 vector to drive GUS expression. The above four recombinant plasmids were introduced into the Nipponbare using the Agrobacterium tumefaciens-mediated transformation method, respectively. Likewise, For the CRISPR-Cas9 of D2, we chose one target of D2 to construct recombinant vector, and the recombinant plasmid was respectively introduced into Nipponbare and Z499 background. All primers used are listed in Supplementary Data 1.

Promoter activity assays in rice protoplasts

To explore whether the sequence polymorphisms in the promoter region of qRBG1 underlie different expression levels, two sequences (2,069 kb) upstream of the start codon of qRBG1 from Nipponbare or Z499 were fused upstream of the luciferase (LUC) gene. Then, qRBG1^N-LUC and qRBG1^Z-LUC were used to analyze promoter activity of qRBG1 promoter from Nipponbare and Z499 in the rice protoplasts. The rice protoplasts were isolated following the previously described method⁵⁸. The DLR assay system was used to measure the relative luciferase (LUC) activity using the GloMax 20-20 luminometer (Promega, Madison, WI, USA)⁵⁸.

Transient expression assays in rice protoplasts

To identify the subcellular localization of qRBG1, the full-length the protein coding region without the stop codon was amplified and cloned into the expression cassette 35S-GFP-NOS (pAN580) to create the qRBG1-GFP fusion expression vectors. Then, GFP and qRBG1-GFP were used to analyze the subcellular localizations of these proteins in the rice protoplasts. The rice protoplasts were isolated following the previously described method⁵⁸. The GFP and RFP fluorescence signals were observed with a confocal laser scanning microscope (LSM800; Zeiss, Germany). For subcellular localization in N. benthamiana leaves, the full-length coding sequence of qRBG1 was cloned into pCAMBIA1305 vector. The primers used are listed in Supplementary Data 1.

In situ hybridization

Specific fragment of qRBG1 probe was amplified and labeled using a DIG RNA Labeling Kit (Roche, Basel, Switzerland). Pretreatment of sections, hybridization, and immunological detection was performed following the previously described method⁵⁹. The primers used are listed in Supplementary Data 1.

BR sensitivity assay

To explore the role of qRBG1/OsBZR5 and OsBZR1 in BR response, lamina inclination experiments were performed. For the lamina inclination experiments, we soaked the sterilized seeds for 3 days in a chamber kept at 37 °C, and then seeds with the same degree of germination were then transferred to grown for 2 d in a chamber kept at 28 °C with a light/dark cycle of 16/8 h. Ethanol (1 L) containing 0, 0.1, 1, 10 or 20 μmol of 2, 4-epibrassinolide (2,4-epiBL) (MCE, HY-N0273) was spotted onto the tip of the lamina of 6-d-old seedlings following the previously described method⁶⁰. The treated seedlings were observed after 3 days of growth at 28 °C, and then the inclination angles of the lamina joint of the second leaf were measured using Image J software.

High-performance liquid chromatography-mass spectrometry (HPLC-MS)

To determine the BR level of qRBG1-related and OsBZR1-related transgenic plants, Nipponbare, qRBG1-related and OsBZR1-related plants from 0.6 g young panicles were subjected to HPLC-MS analysis. These samples were fully ground into powder, and then HPLC-MS analysis was carried out at Shimadzu LC-30AD high performance liquid chromatograph (Shimadzu Company, Japan) and SCIEX Triple Quad™ 6500 + LC-MS/MS (AB, USA), as described previously²⁸. The data were obtained using the MultiQuant^TM 3.0.3 software. Three biological repeats were used for statistical analysis. Statistical tests were performed by two-tailed Student’s t test.

Yeast two-hybrid assay (Y2H)

Yeast two-hybrid assays were performed with the Y2HGold yeast strain the Matchmaker Gold Yeast Two-Hybrid (Y2HGold) System (Clontech, Mountain View, CA, USA) following the manufacturer’s instructions. For self-activation analysis of qRBG1, the full-length coding sequence of qRBG1 was cloned into pGBKT7 vector, and then the pGBKT7-qRBG1 was transformed into yeast Y2HGold. pGBKT7-LF1 protein⁵⁷ and the empty pGBKT7 vector were separately used as the positive and negative control. Transformants were selected on plates lacking Trp (SD/-Trp) and lacking Trp, Ade and His (SD/-Trp/-Ade/-His). The pGADT7 and pGBKT7 (Clontech) yeast expression vectors were used for plasmid constructions. The pGADT7-53 and pGADT7-lam were used as positive and negative controls, respectively. Co-transformed colonies were selected using double dropout plates lacking Trp and Leu (SD/-Trp/-Leu). Protein interactions were analysized by measuring the growth of colonies on QDO plates lacking Trp, Leu, His and Ade (SD/-Trp/-Leu/-His/-Ade) and QDO plates lacking SD/-Trp/-Leu/-His/-Ade with X-a-gal (Coolaber, SL0640). The primers are listed in Supplementary Data 1.

Transient expression in N. benthamiana leaves

Agrobacterium (strain GV3101) including the target protein plasmid was grown in YEB medium with antibiotic selection to OD₆₀₀ = 0.5–0.6. Cells were suspended to OD₆₀₀ = 0.7 in MES buffer (10 mM MgCl₂, 10 mM MES, 150 mM AS; pH 5.6) and kept in the dark for 2-3 h before inoculation. After transformation 48 h under dark condition, GFP fluorescence signals were observed in leaves with a confocal laser scanning microscope. The primers are listed in Supplementary Data 1.

Bimolecular fluorescence complementation assay

The BiFC experiments were performed in N. benthamiana leaves as previously described method⁵⁸. The full-length CDS of qRBG1 and OsBZR1 were respectively fused with the sequence coding the C-terminal fragment of the yellow fluorescent protein (cYFP) in the pXY104 vector. The full-length CDS of qRBG1, OsBZR1 and OsGSK2 were fused with the sequence coding the N-terminal portion of the yellow fluorescent protein (nYFP) in the pXY106 vector, respectively. All these fused expression vectors were co-transformed into the Agrobacterium (strain GV3101). The pair cYFP and nYFP was used as a negative control. A. tumefaciens strains containing the different plasmids were mixed in a 1:1 ratio, and then transformed into N. benthamiana leaves. The BiFC signals were detected with a confocal laser scanning microscope. The BiFC experiments were performed as previously described method in rice protoplasts⁶¹. The full coding sequence of qRBG1 was cloned into the pAN580-cYFP (2 × 35S-cYFP-Nos) vector. The full coding sequences of OsBZR1 and OsGSK2 were respectively cloned into the pAN580-nYFP (2 × 35S-nYFP-Nos) vector. Some groups were used as the negative controls, and then these recombinant vectors were co-transformed into rice protoplasts. After overnight culture at 28 °C, the YFP fluorescence signal was visualized via an LSM 800 confocal laser scanning microscope. The primers are listed in Supplementary Data 1.

Protein purification and pull-down assay in vitro

The pull-down experiments were performed as previously described method⁵⁹. The full-length CDS sequences of qRBG1 and OsGSK2 were separately cloned into the pGEX-4T-1 vector, and the full-length coding sequences of OsBZR1 and qRBG1 were cloned into the pET32a vector, respectively. GST, GST-qRBG1, His-OsBZR1, GST-OsGSK2 and His-qRBG1 fusion proteins were respectively expressed in Escherichia coli (DE3). The fusion proteins GST, GST-qRBG1 and GST-OsGSK2 were induced by 1 mM IPTG at 37 °C and bound by GST beads (GE Healthcare, Chicago, IL, USA), and His-OsBZR1 and His-qRBG1 fusion protein were induced by 0.5 mM IPTG at 16 °C. His-OsBZR1 fusion protein was purified according to the manufacturer’s protocol (New England Biolabs, Ipswich, MA, USA). The prey and bait proteins were detected using an anti-His (Cat #K200060M; Solarbio; 1:2000 dilution) and anti-GST (Cat #LF305; epizyme; 1:5000 dilution) antibodies, and the anti-mouse HRP secondary antibody (epizyme, Cat #LF101, 1:5000 dilution). The primers used are listed in Supplementary Data 1.

Phosphorylation assay in vitro

In vitro phosphorylation assay was performed according to the previously described method⁴⁶. Phosphorylation assay used 0.1 mg GST-OsGSK2 and 0.5 mg His-qRBG1 recombinant protein. Phosphorylation reaction buffer contained 25 mM Tris (pH 7.4), 12 mM MgCl₂, 1 mM DTT and 1 mM ATP. The reaction buffer contained 25 mM Tris (pH 7.4), 12 mM MgCl₂ and 1 mM DTT was used as the control. The reaction was incubated at 37 °C for 5 h and boiled with 5 × SDS loading buffer and the reaction products were separated by 10% (w/v) SDS-PAGE with or without 50 μm Phos-tag. Subsequently, the anti-His antibody was used to detect phosphorylation signal.

Transcriptional activity assay

The transcriptional activity of qRBG1 was analyzed with the dual-luciferase reporter (DLR) assay system in rice protoplasts. The full-length CDS of qRBG1 was fused to the VP16 DNA-binding domain (BD) driven by the 35S promoter. The transcriptional activator VP16 was used as the control. The VP16 and VP16-qRBG1 effectors were respectively transiently expressed in rice protoplasts. The promoter of D2 (1598 bp) was amplified and cloned into pGreen II 0800-LUC double-reporter vector. The full-length CDS of qRBG1 and OsBZR1 were respectively amplified and cloned into the pAN580 vector with two CaMV35S promoters to generate the 2 × 35Spro::qRBG1 and 2 × 35Spro::OsBZR1 constructs. The DLR assay system was used to measure the relative luciferase (LUC) activity using the GloMax 20-20 luminometer (Promega, Madison, WI, USA)⁵⁸. The primers are listed in Supplementary Data 1.

ChIP-qPCR analysis

ChIP-qPCR assay was performed according to described method⁴⁶. The qRBG1-OE plants from 2-week-old seedlings were subjected to ChIP analysis. Chromatin was isolated from 1 g 2-week-old seedlings. The DNA-protein complex was immunoprecipitated with Immunoglobulin G (IgG) antibody (Abmart, B30010, 1 μg) or anti-GFP antibody (Proteintech, 50430-2-AP, 1 μg). The IgG antibody was used as the negative control. The immunoprecipitated DNA fragments were detected by qPCR with gene-specific primers, respectively. A minimum of three biological repeats (1 g of sample), each with three technical repeats, were used for statistical analysis. Primers used are listed in Supplementary Data 1.

Electrophoretic mobility shift assays (EMSA)

The CDS sequences of qRBG1 and OsBZR1 were separately amplified and cloned into the pET32a expression vector, and then all fusion vectors were expressed in the DE3 strain, respectively. His-qRBG1 and His-OsBZR1 recombinant proteins were purified according to the manufacturer’s protocol (New England Biolabs, Ipswich, MA, USA). All probes were labeled with biotin at the 5’ end for the electrophoretic mobility shift assays (EMSA) and synthesized by TSING KE (Hangzhou). Specific binding was confirmed by competition experiments with a 100-fold excess of unlabeled, identical oligonucleotides. EMSA analysis was performed and the bands were detected according to the instructions of the EMSA Kit (Beyotime). The primers and probe sequences in Supplementary Data 1.

Microscale thermophoresis assay (MST)

The coding sequence of qRBG1 was amplified and cloned into the pET32a expression vector, and then the fusion vector was expressed in the DE3 strain. His-qRBG1 recombinant protein was purified according to the manufacturer’s protocol (New England Biolabs, Ipswich, MA, USA). The probe was labeled with biotin at the 5’ end, and then the probe was used as the ligand. MST assay was carried out as previously described method⁶². Sixteen samples with constant concentration of fluorescently labeled qRBG1 protein and 2-fold increased concentrations of biotin-labeled probe including the three putative E-Box motifs of D2 promoter region were mixed and incubated for 30 min at room temperature. Then, the specimens were loaded and measured using a Monolith NT.115 instrument (Nano Temper Technologies GmbH, Munich, Germany). Dissociation constant (K_d) was calculated and fitted by the Nano Temper Analysis software. Three independent measurements were performed using the signal from thermophoresis plus T-Jump.

Yeast one-hybrid assay

The full-length coding sequence of qRBG1 was cloned and fused into the pB42AD vector, forming pB42AD-qRBG1. To generate pLacZi-OFP1 reporter vector, 2005 bp promoter region of OFP1 upstream of the ATG starting codon was cloned into the pLacZi vector. The plasmids were co-transformed into yeast strain EGY48. Transformants were grown on SD-Ura/-Trp plates for 3 days at 30 °C and then transferred onto X-gal (Coolarber, CX11921) plates for blue color. The primers used in this assay are listed in Supplementary Data 1.

Haplotypes analysis of qRBG1 variation promoter

The natural variations of qRBG1 from 3,000 rice accessions were downloaded from Rice Functional Genomics Breeding v2.0 (RFGB) database (https://www.rmbreeding.cn/)⁶³. Sequence variations positions (vg0103963675-vg0103963678, vg0103964019, vg0103964148-vg0103964151, vg0103964459, vg0103964461, vg0103964495, vg0103964702, vg0103964745, vg0103965192, vg0103965311) in the 2-kb regions upstream of the start codon were used to execute haplotype analysis, and then nine haplotypes (>50 accessions) were identified based on the diversity. To determine the phenotypic differences among the nine haplotypes associated with grain size traits, including grain length, grain width and 1000-grain weight, we performed one-way ANOVA and Duncan’s multiple comparisons among the five haplotypes in GraphPad Prism 8.0.2 software.

Statistics and reproducibility

Data were presented as means ± standard deviation, shown by error bars. Comparisons were made by two-tailed Student’s t test or one-way ANOVA and Duncan multiple comparisons test (P < 0.05) in statistical software GraphPad Prism 8.0.2, and yield statistical analysis of Nipponbare and Z499 in all plots of three years was conducted by Two-way ANOVA and Duncan multiple comparisons using IBM SPSS Statistics 27.0.1. No statistical method was used to predetermine sample size, no data were excluded from the analyses, the experiments were not randomized, and we were not blinded to allocation during experiments and outcome assessment.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Data supporting the findings of the study are available within this paper and its Supplementary Information files. A reporting summary for the article is available as a Supplementary Information file. The data of grain length, grain width and 1000-grain weight are available on the RFGB v2.0 database (https://www.rmbreeding.cn/Index/) and RiceVarMap (http://ricevarmap.ncpgr.cn/). The gene sequence information of this study can be downloaded from Gramene (https://www.gramene.org/). All primer data generated in this study are provided in the Supplementary Data 1. Source data are provided with this paper.

References

Zhang, Q. Strategies for developing green super rice. Proc. Natl. Acad. Sci. USA 104, 16402–16409 (2007).
Article ADS CAS PubMed PubMed Central Google Scholar
Jiang, H. Z., Zhang, A. P., Liu, X. T. & Chen, J. G. Grain size associated genes and the molecular regulatory mechanism in rice. Int. J. Mol. Sci. 23, 3169 (2022).
Article CAS PubMed PubMed Central Google Scholar
Zuo, J. R. & Li, J. Y. Molecular genetic dissection of quantitative trait loci regulating rice grain size. Annu. Rev. Genet. 48, 99–118 (2014).
Article CAS PubMed Google Scholar
Ren, D., Ding, C. & Qian, Q. Molecular bases of rice grain size and quality for optimized productivity. Sci. Bull. 68, 314–350 (2023).
Article CAS Google Scholar
Wang, S. K. et al. The OsSPL16-GW7 regulatory module determines grain shape and simultaneously improves rice yield and grain quality. Nat. Genet. 47, 949–954 (2015).
Article CAS PubMed Google Scholar
Zhao, D. S. et al. GS9 acts as a transcriptional activator to regulate rice grain shape and appearance quality. Nat. Commun. 9, 1240 (2018).
Article ADS PubMed PubMed Central Google Scholar
Si, L. Z. et al. OsSPL13 controls grain size in cultivated rice. Nat. Genet. 48, 447–456 (2016).
Article CAS PubMed Google Scholar
Guo, T. et al. ERECTA1 acts upstream of the OsMKKK10-OsMKK4-OsMPK6 cascade to control spikelet number by regulating cytokinin metabolism in rice. Plant Cell 32, 2763–2779 (2020).
Article CAS PubMed PubMed Central Google Scholar
Tian, X. J. et al. WRKY53 integrates classic brassinosteroid signaling and the mitogen-activated protein kinase pathway to regulate rice architecture and seed size. Plant Cell 33, 2753–2775 (2021).
Article PubMed PubMed Central Google Scholar
Duan, P. G. & Li, Y. H. Size matters: G protein signaling is crucial for grain size control in rice. Mol. Plant 14, 1618–1620 (2021).
Article MathSciNet CAS PubMed Google Scholar
Sun, S. Y. et al. A G-protein pathway determines grain size in rice. Nat. Commun. 9, 851 (2018).
Article ADS PubMed PubMed Central Google Scholar
Shi, C. L. et al. Ubiquitin specific protease 15 has an important role in regulating grain width and size in rice. Plant Physiol. 180, 381–391 (2019).
Article CAS PubMed PubMed Central Google Scholar
Hao, J. Q. et al. The GW2-WG1-OsbZIP47 pathway controls grain size and weight in rice. Mol. Plant 14, 1266–1280 (2021).
Article CAS PubMed Google Scholar
Hirano, K. et al. SMALL ORGAN SIZE 1 and SMALL ORGAN SIZE 2/DWARF AND LOW-TILLERING form a complex to integrate auxin and brassinosteroid signaling in rice. Mol. Plant 10, 590–604 (2017).
Article CAS PubMed Google Scholar
Xiao, Y. H. et al. Big Grain3, encoding a purine permease, regulates grain size via modulating cytokinin transport in rice. J. Integr. Plant Biol. 61, 581–597 (2019).
Article CAS PubMed Google Scholar
Xiong, M. et al. Brassinosteroid regulation in rice seed biology. Seed Biol. 1, 3–11 (2022).
Article Google Scholar
Hong, Z. et al. 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 (2003).
Article CAS PubMed PubMed Central Google Scholar
Sakamoto, T. et al. Erect leaves caused by brassinosteroid deficiency increase biomass production and grain yield in rice. Nat. Biotechnol. 24, 105–109 (2006).
Article CAS PubMed Google Scholar
Wu, Y. Z. et al. CLUSTERED PRIMARY BRANCH 1, a new allele of DWARF11, controls panicle architecture and seed size in rice. Plant Biotechnol. J. 14, 377–386 (2016).
Article CAS PubMed Google Scholar
Mori, M. et al. Isolation and characterization of a rice dwarf mutant with a defect in brassinosteroid biosynthesis. Plant Physiol. 130, 1152–1161 (2002).
Article CAS PubMed PubMed Central Google Scholar
Qin, R. et al. LTBSG1, a new allele of BRD2, regulates panicle and grain development in rice by brassinosteroid biosynthetic pathway. Genes 9, 292 (2018).
Article PubMed PubMed Central Google Scholar
Clouse, S. D. Brassinosteroid signal transduction: from receptor kinase activation to transcriptional networks regulating plant development. Plant Cell 23, 1219–1230 (2011).
Article CAS PubMed PubMed Central Google Scholar
Morinaka, Y. et al. Morphological alteration caused by brassinosteroid insensitivity increases the biomass and grain production of rice. Plant Physiol. 141, 924–931 (2006).
Article CAS PubMed PubMed Central Google Scholar
Yuan, H. et al. 08SG2/OsBAK1 regulates grain size and number, and functions differently in Indica and Japonica backgrounds in rice. Rice 10, 25 (2017).
Article PubMed PubMed Central Google Scholar
Che, R. H. et al. Control of grain size and rice yield by GL2-mediated brassinosteroid responses. Nat. Plants 2, 15195 (2015).
Article PubMed Google Scholar
Xiao, Y. H., Liu, D. P., Zhang, G. X., Tong, H. N. & Chu, C. C. Brassinosteroids regulate OFP1, a DLT Interacting protein, to modulate plant architecture and grain morphology in rice. Front. Plant Sci. 8, 1698 (2017).
Article PubMed PubMed Central Google Scholar
Tong, H. N. et al. 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 (2012).
Article CAS PubMed PubMed Central Google Scholar
Xiao, Y. H. et al. GSK2 stabilizes OFP3 to suppress brassinosteroid responses in rice. Plant J. 102, 1187–1201 (2020).
Article CAS PubMed Google Scholar
Yang, C., Shen, W. J., He, Y., Tian, Z. H. & Li, J. X. OVATE family protein 8 positively mediates brassinosteroid signaling through Interacting with the GSK3-like Kinase in Rice. PLoS Genet. 12, e1006118 (2016).
Article PubMed PubMed Central Google Scholar
Huang, Y. S. et al. Ribonuclease H-like gene SMALL GRAIN2 regulates grain size in rice through brassinosteroid signaling pathway. J. Integr. Plant Biol. 64, 1883–1990 (2022).
Article CAS PubMed Google Scholar
Lyu, J. et al. Control of grain size and weight by the GSK2-LARGE1/OML4 pathway in rice. Plant Cell 32, 1905–1918 (2020).
Article PubMed PubMed Central Google Scholar
He, J. X. et al. Erratum: BZR1 is a transcriptional repressor with dual roles in brassinosteroid homeostasis and growth responses. Science 307, 1634–1638 (2005).
Article ADS CAS PubMed PubMed Central Google Scholar
Shi, H. Y., Li, X. P., Lv, M. H. & Li, J. BES1/BZR1 family transcription factors regulate plant development via brassinosteroid-dependent and independent pathways. Int. J. Mol. Sci. 23, 10149 (2022).
Article CAS PubMed PubMed Central Google Scholar
Zhu, X. L. et al. Brassinosteroids promote development of rice pollen grains and seeds by triggering expression of Carbon Starved Anther, a MYB domain protein. Plant J. 82, 570–581 (2015).
Article CAS PubMed Google Scholar
Liu, D. P. et al. Diversification of plant agronomic traits by genome editing of brassinosteroid signaling family genes in rice. Plant Physiol. 187, 2563–2576 (2021).
Article CAS PubMed PubMed Central Google Scholar
Ma, F. Y. et al. Identification of QTL for kernel number-related traits in a rice chromosome segment substitution line and fine mapping of qSP1. Crop J. 7, 494–503 (2019).
Article Google Scholar
Hu, J. et al. A rare allele of GS2 enhances grain size and grain yield in rice. Mol. Plant 8, 1455–1465 (2015).
Article CAS PubMed Google Scholar
Nosaki, S. et al. Structural basis for brassinosteroid response by BIL1/BZR1. Nat. Plants 4, 771–776 (2018).
Article CAS PubMed Google Scholar
Kim, T. W. & Wang, Z. Y. Brassinosteroid signal transduction from receptor kinases to transcription factors. Annu. Rev. Plant Biol. 61, 681–704 (2010).
Article CAS PubMed Google Scholar
Bai, M. Y. et al. Functions of OsBZR1 and 14-3-3 proteins in brassinosteroid signaling in rice. Proc. Natl. Acad. Sci. USA 104, 13839–13844 (2007).
Article ADS CAS PubMed PubMed Central Google Scholar
Oh, E., Zhu, J. Y., Ryu, H. J., Hwang, I. & Wang, Z. Y. TOPLESS mediates brassinosteroid-induced transcriptional repression through interaction with BZR1. Nat. Commun. 5, 4140 (2014).
Article ADS CAS PubMed Google Scholar
Youn, J. H. & Kim, T. W. Functional insights of plant GSK3-like kinases: multi-taskers in diverse cellular signal transduction pathways. Mol. Plant 8, 552–565 (2015).
Article CAS PubMed Google Scholar
He, J. X., Gendron, J. M., Yang, Y. L., Li, J. M. & Wang, Z. Z. The GSK3-like kinase BIN2 phosphorylates and destabilizes BZR1, a positive regulator of the brassinosteroid signaling pathway in Arabidopsis. Proc. Natl. Acad. Sci. USA 99, 10185–10190 (2002).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhang, C. et al. Dynamics of brassinosteroid response modulated by negative regulator LIC in rice. PLoS Genet 8, e1002686 (2012).
Article CAS PubMed PubMed Central Google Scholar
Qiao, S. L. et al. The RLA1/SMOS1 transcription factor functions with OsBZR1 to regulate brassinosteroid signaling and rice architecture. Plant Cell 29, 292–309 (2017).
Article CAS PubMed PubMed Central Google Scholar
Wu, Q. et al. OsNAC016 regulates plant architecture and drought tolerance by interacting with the kinases GSK2 and SAPK8. Plant Physiol. 189, 1296–1313 (2022).
Article CAS PubMed PubMed Central Google Scholar
Zhang, G. Target chromosome-segment substitution: a way to breeding by design in rice. Crop J. 9, 658–668 (2021).
Article Google Scholar
Xue, W. Y. et al. Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice. Nat. Genet. 40, 761–767 (2009).
Article Google Scholar
Wang, D. C. et al. Identification, pyramid and candidate genes of QTLs for associated traits based on a dense erect panicle rice CSSL-Z749 and five SSSLs, three DSSLs and one TSSL. Rice 14, 1–55 (2021).
Article CAS PubMed PubMed Central Google Scholar
Akagi, H. et al. Positional cloning of the rice Rf-1 gene, a restorer of BT-type cytoplasmic male sterility that encodes a mitochondria-targeting PPR protein. Theor. Appl. Genet. 108, 1449–1457 (2004).
Article CAS PubMed Google Scholar
Etsuko, I. et al. The fertility restorer gene, Rf2, for lead rice-type cytoplasmic male sterility of rice encodes a mitochondrial glycine-rich protein. Plant J. 65, 359–367 (2011).
Article Google Scholar
Cai, J., Fan, H. Y., Liao, Q. P. & Zhang, G. Q. Allelic differentiations and effects of the Rf3 and Rf4 genes on fertility restoration in rice with wild abortive cytoplasmic male sterility. Biol. Plant 57, 274–280 (2013).
Article CAS Google Scholar
Tang, H. W. et al. The rice restorer Rf4 for wild-abortive cytoplasmic male sterility encodes a mitochondrial-localized PPR protein that functions in reduction of WA352 transcripts. Mol. Plant 7, 1497–1500 (2014).
Article CAS PubMed Google Scholar
Chen, C. J. et al. TBtools: an integrative toolkit developed for interactive analyses of big biological data. Mol. Plant 13, 1194–1202 (2020).
Article CAS PubMed Google Scholar
Lescot, M. et al. PlantCARE, a database of plant cis-acting regulatory elements and a portal to tools for in silico analysis of promoter sequences. Nucleic Acids Res. 30, 325–327 (2002).
Article CAS PubMed PubMed Central Google Scholar
Tamura, K. et al. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739 (2011).
Article CAS PubMed PubMed Central Google Scholar
Zhang, T. et al. LATERAL FLORET 1 induced the three-florets spikelet in rice. Proc. Natl. Acad. Sci. USA 114, 9984–9989 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Du, D. et al. The CC-NB-LRR OsRLR1 mediates rice disease resistance through interaction with OsWRKY19. Plant Biotechnol. J. 19, 1052–1064 (2021).
Article CAS PubMed PubMed Central Google Scholar
You, J. et al. The APC/CTAD1-WIDE LEAF 1-NARROW LEAF 1 pathway controls leaf width in rice. Plant Cell 34, 4313–4328 (2022).
Article PubMed PubMed Central Google Scholar
Fujioka, S., Noguchi, T., Takatsuto, S. & Yoshida, S. Activity of brassinosteroids in the dwarf rice lamina inclination bioassay. Phytochemistry 49, 1841–1848 (1998).
Article CAS Google Scholar
Ma, L. et al. ABNORMAL VASCULAR BUNDLES regulates cell proliferation and procambium cell establishment during aerial organ development in rice. New. Phytol. 213, 275–286 (2017).
Article CAS PubMed Google Scholar
Jerabek-Willemsen, M., Wienken, C. J., Braun, D., Baaske, P. & Duhr, S. Molecular interaction studies using microscale thermophoresis. Assay. Drug Dev. Technol. 9, 342–353 (2011).
Article CAS PubMed PubMed Central Google Scholar
Wang, C. C. et al. Towards a deeper haplotype mining of complex traits in rice with RFGB v2.0. Plant Biotechnol. J. 18, 14–16 (2020).
Article PubMed Google Scholar

Download references

Acknowledgements

We thank professor Chengcai Chu (South China Agricultural University) and professor Hongning Tong (Institute of Crop Sciences, Chinese Academy of Agricultural Sciences) for providing the OsGSK2 relative mutants, Xuelu Wang (Henan University) for providing Osbzr1-D mutant and professor Ming Luo (Biotechnology Research Center of S.W.U.) for his assistance in reviewing and modifying the manuscript. We thank professor Shaokui Wang (South China Agricultural University) and professor Hai Du (College of Agronomy and Biotechnology of S.W.U.) for their valuable suggestions on haplotype analysis. This work was partially supported by funds from the National Natural Science Foundation of China (32072039 to F.Z.), NSFC-Key projects of the Joint Fund (U23A20184 to G.H.), Science and Technology Innovation 2030-Major Project (2023ZD0406804), Chongqing Natural Science Foundation Innovation Group (cstc2021jcyj-cxttX0004), and Natural Science Foundation of Chongqing (CSTB2023NSCQ-MSX0124 to F.Z.).

Author information

These authors contributed equally: Qiuli Zhang, Renhong Wu.

Authors and Affiliations

Rice Research Institute, Key Laboratory of Crop Molecular Improvement, Academy of Agricultural Sciences, Ministry of Education, Southwest University, Chongqing, 400715, China
Qiuli Zhang, Renhong Wu, Tao Hong, Dachuan Wang, Qiaolong Li, Jiayi Wu, Han Zhang, Kai Zhou, Hongxia Yang, Ting Zhang, JinXiang Liu, Nan Wang, Yinghua Ling, Zhenglin Yang, Guanghua He & Fangming Zhao

Authors

Qiuli Zhang
View author publications
Search author on:PubMed Google Scholar
Renhong Wu
View author publications
Search author on:PubMed Google Scholar
Tao Hong
View author publications
Search author on:PubMed Google Scholar
Dachuan Wang
View author publications
Search author on:PubMed Google Scholar
Qiaolong Li
View author publications
Search author on:PubMed Google Scholar
Jiayi Wu
View author publications
Search author on:PubMed Google Scholar
Han Zhang
View author publications
Search author on:PubMed Google Scholar
Kai Zhou
View author publications
Search author on:PubMed Google Scholar
Hongxia Yang
View author publications
Search author on:PubMed Google Scholar
Ting Zhang
View author publications
Search author on:PubMed Google Scholar
JinXiang Liu
View author publications
Search author on:PubMed Google Scholar
Nan Wang
View author publications
Search author on:PubMed Google Scholar
Yinghua Ling
View author publications
Search author on:PubMed Google Scholar
Zhenglin Yang
View author publications
Search author on:PubMed Google Scholar
Guanghua He
View author publications
Search author on:PubMed Google Scholar
Fangming Zhao
View author publications
Search author on:PubMed Google Scholar

Contributions

F.Z. directed and supervised the research and reviewed and modified the manuscript. Q.Z. and R.W. conceived the research and designed the experiments. Q.Z. performed the experiments and wrote the manuscript. T.H., D.W., Q.L., J.W., H.Z., K.Z., and H.Y. assisted in carrying out the experiments. T.Z., J.L., N.W., Y.L., Z.Y., and G.H. contributed valuable suggestions for the manuscript. All the authors read and approved the manuscript.

Corresponding authors

Correspondence to Guanghua He or Fangming Zhao.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Zhang, Q., Wu, R., Hong, T. et al. Natural variation in the promoter of qRBG1/OsBZR5 underlies enhanced rice yield. Nat Commun 15, 8565 (2024). https://doi.org/10.1038/s41467-024-52928-9

Download citation

Received: 19 December 2023
Accepted: 24 September 2024
Published: 03 October 2024
Version of record: 03 October 2024
DOI: https://doi.org/10.1038/s41467-024-52928-9

This article is cited by

Positional variations of rice protein compositions accumulation within a panicle during the grain filling
- Min Xi
- Zhong Li
- Linsheng Yang
BMC Plant Biology (2025)
Genome-wide identification and functional characterization of CLG family genes reveal likely roles in epidermal development in Arabidopsis
- Qiuli Zhang
- Tao Hong
- Renhong Wu
Plant Cell Reports (2025)