The OsSPL16-GW7 regulatory module determines grain shape and simultaneously improves rice yield and grain quality

Journal name:
Nature Genetics
Volume:
47,
Pages:
949–954
Year published:
DOI:
doi:10.1038/ng.3352
Received
Accepted
Published online

The deployment of heterosis in the form of hybrid rice varieties has boosted grain yield, but grain quality improvement still remains a challenge. Here we show that a quantitative trait locus for rice grain quality, qGW7, reflects allelic variation of GW7, a gene encoding a TONNEAU1-recruiting motif protein with similarity to C-terminal motifs of the human centrosomal protein CAP350. Upregulation of GW7 expression was correlated with the production of more slender grains, as a result of increased cell division in the longitudinal direction and decreased cell division in the transverse direction. OsSPL16 (GW8), an SBP-domain transcription factor that regulates grain width, bound directly to the GW7 promoter and repressed its expression. The presence of a semidominant GW7TFA allele from tropical japonica rice was associated with higher grain quality without the yield penalty imposed by the Basmati gw8 allele. Manipulation of the OsSPL16-GW7 module thus represents a new strategy to simultaneously improve rice yield and grain quality.

At a glance

Figures

  1. Positional cloning of qGW7.
    Figure 1: Positional cloning of qGW7.

    (a) Grains from the parents of indica hybrid rice lines. Scale bar, 2 mm. (b) Grain chalkiness of the two hybrid combinations (ZS97A × MH63 and TFA × MH63). (c) QTL locations for grain width and grain length. LOD, logarithm of odds. (d) qGW7 was mapped to a ~20-kb genomic DNA region between markers M1 and M10 using 4,500 BC3F2 plants. The numbers below the bar indicate the number of recombinants between qGW7 and the molecular markers shown. (e) Genotyping of progeny homozygous for qGW7 delimited the locus to a ~2.6-kb stretch flanked by markers S5 and S6. Grain width is shown for recombinant BC4F3 plants (L1–L4) and the parental plant. Filled and open bars represent chromosomal segments homozygous for, respectively, the TFA and HJX74 alleles. Data are shown as means ± s.e.m. (n = 60). (f) Allelic variation in the candidate gene LOC_Os07g41200 (GW7) between TFA and HJX74.

  2. GW7 regulates grain shape by changing cell division patterns.
    Figure 2: GW7 regulates grain shape by changing cell division patterns.

    (a) The transcript abundance of GW7 in developing panicles. BM, branch meristem; YP0.2, YP1, YP3, YP9, YP12 and YP18 represent young panicles with average lengths of 0.2 cm, 1 cm, 3 cm, 9 cm, 12 cm and 18 cm, respectively. Transcript levels are expressed as the number of copies of GS7 per 1,000 copies of rice Actin3. Data are shown as means ± s.e.m. (n = 3). (b) GUS staining. Scale bar, 2 mm. (c) Grain shape. Scale bar, 2 mm. (d) Spikelet hulls. The dashed line indicates the sites of the cross-sections. Scale bar, 2 mm. (e) Cross-sections of spikelet hulls. Scale bars, 0.2 mm and 50 μm (close-up views of the boxed regions). (f) The number of inner parenchyma cells shown in e. The arrows in e indicate the layers corresponding to the numbered cells. Data are shown as means ± s.e.m. (n = 10). (g) Scanning electron microscope analysis. Scale bars, 200 μm. (h) Average numbers of outer epidermal cells shown in g. Data are shown as means ± s.e.m. (n = 30). Student's t tests were used to generate the P values. (i) BiFC analysis. nYFP-tagged GW7 was cotransformed with either cYFP-OsTON1b or cYFP-OsTON2 into Arabidopsis protoplasts. OsSPL16 serves as a negative control. nYFP, N-terminal YFP; cYFP, C-terminal YFP. Scale bars, 10 μm. (j) Overexpression of GW7 produces long and narrow leaf blades in Arabidopsis. Scale bars, 2 cm. (k) Scanning electron microscopy analysis of the leaf blades.

  3. A field trial of NIL-gw7HJX74 and NIL-GW7TFA plants.
    Figure 3: A field trial of NIL-gw7HJX74 and NIL-GW7TFA plants.

    (a) The morphology of the NIL plants. Scale bar, 12 cm. (b) Plant height. (c) Heading date. (d) Number of tillers per plant. (e) Number of grains per panicle. (f) Grain width. (g) Grain length. (h) Ratio of grain length to width. Data in bh are shown as means ± s.e.m. (n = 60). Student's t tests were used to generate the P values. (i) Time course of the increase in endosperm dry weight. Data are shown as means ± s.e.m. (n = 60). (j) 1,000-grain weight. Data are shown as means ± s.e.m. (n = 180). (k) Overall grain yield per plant. Data are shown as means ± s.e.m. (n = 300). The presence of the same lowercase letter denotes a non-significant difference between the means (P > 0.05; j and k). All phenotypic data were measured from paddy-grown plants under normal cultivation conditions.

  4. OsSPL16 negatively regulates GW7 expression.
    Figure 4: OsSPL16 negatively regulates GW7 expression.

    (a) Transcript levels for GW7 in young panicles. Expression is shown relative to that of NIL-GW8HJX74 plants, which was set to 1. Data are shown as means ± s.e.m. (n = 3). (b) Grain shape. Scale bar, 2 mm. (c) ChIP assays. The diagram depicts the regions used for ChIP-PCR analysis of extracts from young panicles of NIL-gw8Basmati plants carrying the pActin::Myc-OsSPL16 construct. ChIP-PCR results were quantified by normalization of the Myc immunoprecipitation signal by the corresponding input signal. Data are shown as means ± s.e.m. (n = 3). (d) EMSA analysis. Competition for OsSPL16 binding was performed with cold probe containing the GTAC motifs at 10×, 20×, 30× and 50× the amount of labeled probe. (e) OsSPL16 represses transcription of the GW7 gene promoter. Relative luciferase activity was monitored in rice protoplasts cotransfected with different effector and reporter constructs. Mock, cotransfected with reporter construct and an empty effector construct; control, cotransfected with effector construct and an empty reporter construct (set to 1). Data are shown as means ± s.e.m. (n = 3). Student's t tests were used to generate the P values. (f) Variations in the F8 fragment between the HJX74 and TFA alleles. The red box highlights the GTAC motifs. (g) Yeast one-hybrid assays. The 0.5-kb and 1.0-kb DNA fragments upstream of the GW7 transcription start site were used to construct lacZ expression vectors32. Data are shown as means ± s.e.m. (n = 3). The pB42AD and pLacZi2μ empty vectors, pB42AD::OsSPL16 and the pLacZi2μ empty vector, and pgw7HJX74 (1.0 kb)::lacZ and the pB42AD empty vector were used as the negative control (control 1), control 2 and control 3, respectively. The presence of the same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  5. Scanning electron microscopy images of the transverse sections of starch granules from hybrid combinations ZS97A/MH63 and TFA/MH63.
    Supplementary Fig. 1: Scanning electron microscopy images of the transverse sections of starch granules from hybrid combinations ZS97A/MH63 and TFA/MH63.
  6. The contrasting phenotype and grain yield of hybrid combinations ZS97A/MH63 and TFA/MH63.
    Supplementary Fig. 2: The contrasting phenotype and grain yield of hybrid combinations ZS97A/MH63 and TFA/MH63.

    (a) The number of tillers per plant. (b) The number of grains per panicle. (c) 1,000-grain weight. (d) The overall grain yield per plant. All data were measured from plants grown with a spacing of 20 × 20 cm in paddies under normal cultivation conditions. Data are shown as means ± s.e.m. (n = 180). A Student’s t test was used to generate the P values.

  7. The semidominant qGW7 allele from TFA was correlated with the formation of more slender grains.
    Supplementary Fig. 3: The semidominant qGW7 allele from TFA was correlated with the formation of more slender grains.

    (a) Segregation of the BC1F2 population derived from TFA and HJX74 (recurrent parent). (bd) Comparisons of grain width (b), grain length (c) and ratio of grain length to width (d) among homozygotes for the TFA qGW7 allele, heterozygotes for the TFA qGW7 allele and the HJX74 qgw7 allele, and homozygotes for the HJX74 qgw7 allele. Data are shown as means ± s.e.m. (n = 120). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  8. Phylogenetic tree of TRM family proteins based on the protein sequences.
    Supplementary Fig. 4: Phylogenetic tree of TRM family proteins based on the protein sequences.

    The names of Arabidopsis and rice TRM genes were downloaded from the Arabidopsis Functional Genomics Network (http://www.dbg-afgn.de) and the Rice Annotation Project Database (http://rapdb.dna.affrc.go.jp).

  9. C-terminal motifs of the GW7 protein are present in CAP350.
    Supplementary Fig. 5: C-terminal motifs of the GW7 protein are present in CAP350.

    A significant match was obtained with a human centrosomal protein CAP350, which contains the three conserved motifs M3-M4-M2 in the same configuration as in both rice GW7 and Arabidopsis TRM1. The numbers on the right indicate the position of the residues in the full-length protein. Identical residues are indicated by dark boxes, conserved residues are indicated by gray boxes and variant residues are indicated by light boxes. The C-terminal motifs of human centrosomal protein CAP350 (NP_055625.4) are from NCBI (http://www.ncbi.nlm.nih.gov).

  10. Comparison of GW7 transcription between NIL-GW7TFA and NIL-gw7HJX74 plants in vegetable growth stage.
    Supplementary Fig. 6: Comparison of GW7 transcription between NIL-GW7TFA and NIL-gw7HJX74 plants in vegetable growth stage.

    SAM, shoot apical meristem. Expression levels were expressed as the relative copies of rice actin3. Data are shown as means ± s.e.m. (n = 3). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  11. Comparison of GW7 transcription between NIL-GW7TFA and NIL-gw7HJX74 plants in developing rice endosperms.
    Supplementary Fig. 7: Comparison of GW7 transcription between NIL-GW7TFA and NIL-gw7HJX74 plants in developing rice endosperms.

    Total RNA was extracted from the developing rice endosperms. DAP, days after pollination. Expression levels were expressed as the relative copies of rice actin3. Data are shown as means ± s.e.m. (n = 3). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  12. The effect of GW7 on the average length and width of the outer epidermal cells of spikelet hulls.
    Supplementary Fig. 8: The effect of GW7 on the average length and width of the outer epidermal cells of spikelet hulls.

    Data are shown as means ± s.e.m. (n = 30). A Student’s t test was used to generate the P values.

  13. The GW7 protein interacts with both OsTON1b and OsTON2.
    Supplementary Fig. 9: The GW7 protein interacts with both OsTON1b and OsTON2.

    (a) The colored boxes correspond to the position of the motif present in the GW7 and TRM1/LNG2 proteins. (b) The M2 motif of GW7 was involved in the interaction between GW7 and OsTON1b, and the M3 motif of GW7 was involved in the interaction between GW7 and OsTON2. The schematic shows the GW7 fragments tested for interaction using yeast two-hybrid assays.

  14. Different transcriptional levels of starch biosynthesis genes in the developing rice endosperm between NIL-GW7TFA and NIL-gw7HJX74 plants.
    Supplementary Fig. 10: Different transcriptional levels of starch biosynthesis genes in the developing rice endosperm between NIL-GW7TFA and NIL-gw7HJX74 plants.

    Total RNA was extracted from the developing rice endosperms on day 9 after pollination. BEI, branching enzyme I; BEII, branching enzyme II; AGPL1, AGP large subunit; AGPL2, AGP large subunit; AGPS1, AGP small subunit; SSI, soluble starch synthase; SSIIa, soluble starch synthase; SSIIIa, soluble starch synthase; GBSS, granule-bound starch synthase; ISA1, isoamylase; PUI, pullulanase; SuSy2, sucrose synthase; UGP1, UDP-glucose pyrophosphorylase1. Expression levels are expressed as the relative copies of rice actin3, and the values are expressed relative to the level of transcript in NIL-gw7HJX74, which was set to 1. Data are shown as means ± s.e.m. (n = 3). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  15. The effect of GW7 transcript abundance on scanning electron microscopy images of the transverse sections of starch granules.
    Supplementary Fig. 11: The effect of GW7 transcript abundance on scanning electron microscopy images of the transverse sections of starch granules.

    (a) NIL-gw7HJX74 plants. (b) NIL-GW7TFA endosperm comprised largely sharp-edged, compactly arranged polygonal starch granules when compared to NIL-gw7HJX74 ones. (c) Transgenic NIL-gw7HJX74 plants overexpressing the TFA GW7 cDNA driven by its native promoter.

  16. EMSA assays.
    Supplementary Fig. 12: EMSA assays.

    (a,b) The DNA fragment (F8) containing two GATC motifs (a) or two mutated ATAC motifs (b) was incubated with GST-OsSPL16 as indicated. Competition for OsSPL16 binding was performed with 10×, 20×, 30× and 50× cold probes containing the GTAC motif.

  17. Yeast one-hybrid assays.
    Supplementary Fig. 13: Yeast one-hybrid assays.

    The 0.5-kb TFA promoter fragments and mutated promoters, which contained an 18-bp deletion and/or an 11-bp insertion located in the F8 region, were used to analyze the binding activity of OsSPL16 to the GW7 promoter. Data are shown as means ± s.e.m. (n = 3). The presence of the same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  18. Appearance of the NIL grains.
    Supplementary Fig. 14: Appearance of the NIL grains.

    The four contrasting allelic combinations of the qGW7 and qGW8 loci were assembled on a near-isogenic HJX74 background. Scale bar, 2 mm.

  19. Improving appearance quality of the rice grain by QTL pyramiding.
    Supplementary Fig. 15: Improving appearance quality of the rice grain by QTL pyramiding.

    The four contrasting allelic combinations of the qGW7 and qGS3 loci were assembled on a near-isogenic HJX74 background. Scale bar, 2 mm.

Accession codes

Referenced accessions

NCBI Reference Sequence

References

  1. Sasaki, A. et al. A mutant gibberellin-synthesis gene in rice. Nature 416, 701702 (2002).
  2. Spielmeyer, W., Ellis, M.H. & Chandler, P.M. Semidwarf (sd-1), “green revolution” rice, contains a defective gibberellin 20-oxidase gene. Proc. Natl. Acad. Sci. USA 99, 90439048 (2002).
  3. Yuan, L. Hybrid rice breeding for super high yield. Hybrid Rice 12, 16 (1997).
  4. Xing, Y. & Zhang, Q. Genetic and molecular bases of rice yield. Annu. Rev. Plant Biol. 61, 421442 (2010).
  5. Tian, Z. et al. Allelic diversities in rice starch biosynthesis lead to a diverse array of rice eating and cooking qualities. Proc. Natl. Acad. Sci. USA 106, 2176021765 (2009).
  6. Ahn, S.N., Bollich, C.N., McClung, A.M. & Tanksley, S.D. RFLP analysis of genomic regions associated with cooked-kernel elongation in rice. Theor. Appl. Genet. 87, 2732 (1993).
  7. Aluko, G. et al. QTL mapping of grain quality traits from the interspecific cross Oryza sativa × O. glaberrima. Theor. Appl. Genet. 109, 630639 (2004).
  8. Wang, L.Q. et al. Genetic basis of 17 traits and viscosity parameters characterizing the eating and cooking quality of rice grain. Theor. Appl. Genet. 115, 463476 (2007).
  9. Zhou, L. et al. Fine mapping of the grain chalkiness QTL qPGWC-7 in rice (Oryza sativa L.). Theor. Appl. Genet. 118, 581590 (2009).
  10. Nelson, J.C. et al. Mapping QTL main and interaction influences on milling quality in elite US rice germplasm. Theor. Appl. Genet. 122, 291309 (2011).
  11. Cai, X.L., Wang, Z.Y., Xing, Y.Y., Zhang, J.L. & Hong, M.M. Aberrant splicing of intron 1 leads to the heterogeneous 5′ UTR and decreased expression of waxy gene in rice cultivars of intermediate amylose content. Plant J. 14, 459465 (1998).
  12. Wang, S. et al. Control of grain size, shape and quality by OsSPL16 in rice. Nat. Genet. 44, 950954 (2012).
  13. Li, Y. et al. Chalk5 encodes a vacuolar H+-translocating pyrophosphatase influencing grain chalkiness in rice. Nat. Genet. 46, 398404 (2014).
  14. Fan, C. et al. 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, 11641171 (2006).
  15. Azimzadeh, J. et al. Arabidopsis TONNEAU1 proteins are essential for preprophase band formation and interact with centrin. Plant Cell 20, 21462159 (2008).
  16. Drevensek, S. et al. The Arabidopsis TRM1-TON1 interaction reveals a recruitment network common to plant cortical microtubule arrays and eukaryotic centrosomes. Plant Cell 24, 178191 (2012).
  17. Patel, H., Truant, R., Rachubinski, R.A. & Capone, J.P. Activity and subcellular compartmentalization of peroxisome proliferator-activated receptor α are altered by the centrosome-associated protein CAP350. J. Cell Sci. 118, 175186 (2005).
  18. Shao, G. et al. Allelic variation for a candidate gene for GS7, responsible for grain shape in rice. Theor. Appl. Genet. 125, 13031312 (2012).
  19. Qiu, X., Gong, R., Tan, Y. & Yu, S. Mapping and characterization of the major quantitative trait locus qSS7 associated with increased length and decreased width of rice seeds. Theor. Appl. Genet. 125, 17171726 (2012).
  20. Kirik, A., Ehrhardt, D.W. & Kirik, V. TONNEAU2/FASS regulates the geometry of microtubule nucleation and cortical array organization in interphase Arabidopsis cells. Plant Cell 24, 11581170 (2012).
  21. Lee, Y.K. et al. LONGIFOLIA1 and LONGIFOLIA2, two homologous genes, regulate longitudinal cell elongation in Arabidopsis. Development 133, 43054314 (2006).
  22. Yan, X., Habedanck, R. & Nigg, E.A. A complex of two centrosomal proteins, CAP350 and FOP, cooperates with EB1 in microtubule anchoring. Mol. Biol. Cell 17, 634644 (2005).
  23. Spinner, L. et al. A protein phosphatase 2A complex spatially controls plant cell division. Nat. Commun. 4, 1863 (2013).
  24. Sun, H. et al. Heterotrimeric G proteins regulate nitrogen-use efficiency in rice. Nat. Genet. 46, 652656 (2014).
  25. Lu, Z. et al. Genome-wide binding analysis of the transcription activator ideal plant architecture1 reveals a complex network regulating rice plant architecture. Plant Cell 25, 37433759 (2013).
  26. Xie, K., Wu, C. & Xiong, L. Genomic organization, differential expression, and interaction of SQUAMOSA promoter-binding-like transcription factors and microRNA156 in rice. Plant Physiol. 142, 280293 (2006).
  27. Huang, X. et al. Natural variation in the DEP1 locus enhances grain yield in rice. Nat. Genet. 41, 494497 (2009).
  28. Jiang, C. et al. Root architecture and anthocyanin accumulation of phosphate starvation responses are modulated by the GA-DELLA signaling pathway in Arabidopsis. Plant Physiol. 145, 14601470 (2007).
  29. Gendrel, A.V. et al. Profiling histone modification patterns in plants using genomic tiling microarrays. Nat. Methods 2, 213218 (2005).
  30. Chern, M. et al. A rice transient assay system identifies a novel domain in NRR required for interaction with NH1/OsNPR1 and inhibition of NH1-mediated transcriptional activation. Plant Methods 8, 6 (2012).
  31. Bracha-Drori, K. et al. Detection of protein-protein interactions in plants using bimolecular fluorescence complementation. Plant J. 40, 419427 (2004).
  32. Li, G. et al. Coordinated transcriptional regulation underlying the circadian clock in Arabidopsis. Nat. Cell Biol. 13, 616622 (2011).

Download references

Author information

Affiliations

  1. State Key Laboratory of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China.

    • Shaokui Wang,
    • Shan Li,
    • Qian Liu,
    • Kun Wu,
    • Jianqing Zhang,
    • Shuansuo Wang,
    • Yi Wang,
    • Xiangbin Chen,
    • Yi Zhang,
    • Caixia Gao &
    • Xiangdong Fu
  2. State Key Laboratory for Conservation and Utilization of Subtropical Agro-Bioresources, South China Agricultural University, Guangzhou, China.

    • Shaokui Wang
  3. Rice Research Institute of the Guangdong Academy of Agricultural Sciences, Guangzhou, China.

    • Feng Wang
  4. Jiaxing Academy of Agricultural Sciences, Jiaxing, China.

    • Haixiang Huang

Contributions

Shaokui Wang performed most of the experiments. S.L. and F.W. conducted QTL analysis. K.W. and Y.W. developed the NILs. Y.Z. and C.G. performed rice transformation. S.L. and Q.L. analyzed genetic diversity. K.W. and H.H. analyzed grain quality. Shuansuo Wang and J.Z. performed yeast two-hybrid screening. X.C. and Shaokui Wang performed ChIP and EMSA assays. X.F. designed the experiments and wrote the manuscript. All authors have discussed the results and contributed to the drafting of the manuscript.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: Scanning electron microscopy images of the transverse sections of starch granules from hybrid combinations ZS97A/MH63 and TFA/MH63. (161 KB)
  2. Supplementary Figure 2: The contrasting phenotype and grain yield of hybrid combinations ZS97A/MH63 and TFA/MH63. (91 KB)

    (a) The number of tillers per plant. (b) The number of grains per panicle. (c) 1,000-grain weight. (d) The overall grain yield per plant. All data were measured from plants grown with a spacing of 20 × 20 cm in paddies under normal cultivation conditions. Data are shown as means ± s.e.m. (n = 180). A Student’s t test was used to generate the P values.

  3. Supplementary Figure 3: The semidominant qGW7 allele from TFA was correlated with the formation of more slender grains. (116 KB)

    (a) Segregation of the BC1F2 population derived from TFA and HJX74 (recurrent parent). (bd) Comparisons of grain width (b), grain length (c) and ratio of grain length to width (d) among homozygotes for the TFA qGW7 allele, heterozygotes for the TFA qGW7 allele and the HJX74 qgw7 allele, and homozygotes for the HJX74 qgw7 allele. Data are shown as means ± s.e.m. (n = 120). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  4. Supplementary Figure 4: Phylogenetic tree of TRM family proteins based on the protein sequences. (230 KB)

    The names of Arabidopsis and rice TRM genes were downloaded from the Arabidopsis Functional Genomics Network (http://www.dbg-afgn.de) and the Rice Annotation Project Database (http://rapdb.dna.affrc.go.jp).

  5. Supplementary Figure 5: C-terminal motifs of the GW7 protein are present in CAP350. (150 KB)

    A significant match was obtained with a human centrosomal protein CAP350, which contains the three conserved motifs M3-M4-M2 in the same configuration as in both rice GW7 and Arabidopsis TRM1. The numbers on the right indicate the position of the residues in the full-length protein. Identical residues are indicated by dark boxes, conserved residues are indicated by gray boxes and variant residues are indicated by light boxes. The C-terminal motifs of human centrosomal protein CAP350 (NP_055625.4) are from NCBI (http://www.ncbi.nlm.nih.gov).

  6. Supplementary Figure 6: Comparison of GW7 transcription between NIL-GW7TFA and NIL-gw7HJX74 plants in vegetable growth stage. (34 KB)

    SAM, shoot apical meristem. Expression levels were expressed as the relative copies of rice actin3. Data are shown as means ± s.e.m. (n = 3). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  7. Supplementary Figure 7: Comparison of GW7 transcription between NIL-GW7TFA and NIL-gw7HJX74 plants in developing rice endosperms. (44 KB)

    Total RNA was extracted from the developing rice endosperms. DAP, days after pollination. Expression levels were expressed as the relative copies of rice actin3. Data are shown as means ± s.e.m. (n = 3). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  8. Supplementary Figure 8: The effect of GW7 on the average length and width of the outer epidermal cells of spikelet hulls. (39 KB)

    Data are shown as means ± s.e.m. (n = 30). A Student’s t test was used to generate the P values.

  9. Supplementary Figure 9: The GW7 protein interacts with both OsTON1b and OsTON2. (51 KB)

    (a) The colored boxes correspond to the position of the motif present in the GW7 and TRM1/LNG2 proteins. (b) The M2 motif of GW7 was involved in the interaction between GW7 and OsTON1b, and the M3 motif of GW7 was involved in the interaction between GW7 and OsTON2. The schematic shows the GW7 fragments tested for interaction using yeast two-hybrid assays.

  10. Supplementary Figure 10: Different transcriptional levels of starch biosynthesis genes in the developing rice endosperm between NIL-GW7TFA and NIL-gw7HJX74 plants. (45 KB)

    Total RNA was extracted from the developing rice endosperms on day 9 after pollination. BEI, branching enzyme I; BEII, branching enzyme II; AGPL1, AGP large subunit; AGPL2, AGP large subunit; AGPS1, AGP small subunit; SSI, soluble starch synthase; SSIIa, soluble starch synthase; SSIIIa, soluble starch synthase; GBSS, granule-bound starch synthase; ISA1, isoamylase; PUI, pullulanase; SuSy2, sucrose synthase; UGP1, UDP-glucose pyrophosphorylase1. Expression levels are expressed as the relative copies of rice actin3, and the values are expressed relative to the level of transcript in NIL-gw7HJX74, which was set to 1. Data are shown as means ± s.e.m. (n = 3). The same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  11. Supplementary Figure 11: The effect of GW7 transcript abundance on scanning electron microscopy images of the transverse sections of starch granules. (111 KB)

    (a) NIL-gw7HJX74 plants. (b) NIL-GW7TFA endosperm comprised largely sharp-edged, compactly arranged polygonal starch granules when compared to NIL-gw7HJX74 ones. (c) Transgenic NIL-gw7HJX74 plants overexpressing the TFA GW7 cDNA driven by its native promoter.

  12. Supplementary Figure 12: EMSA assays. (38 KB)

    (a,b) The DNA fragment (F8) containing two GATC motifs (a) or two mutated ATAC motifs (b) was incubated with GST-OsSPL16 as indicated. Competition for OsSPL16 binding was performed with 10×, 20×, 30× and 50× cold probes containing the GTAC motif.

  13. Supplementary Figure 13: Yeast one-hybrid assays. (45 KB)

    The 0.5-kb TFA promoter fragments and mutated promoters, which contained an 18-bp deletion and/or an 11-bp insertion located in the F8 region, were used to analyze the binding activity of OsSPL16 to the GW7 promoter. Data are shown as means ± s.e.m. (n = 3). The presence of the same lowercase letter denotes a non-significant difference between the means (P > 0.05).

  14. Supplementary Figure 14: Appearance of the NIL grains. (74 KB)

    The four contrasting allelic combinations of the qGW7 and qGW8 loci were assembled on a near-isogenic HJX74 background. Scale bar, 2 mm.

  15. Supplementary Figure 15: Improving appearance quality of the rice grain by QTL pyramiding. (68 KB)

    The four contrasting allelic combinations of the qGW7 and qGS3 loci were assembled on a near-isogenic HJX74 background. Scale bar, 2 mm.

PDF files

  1. Supplementary Text and Figures (4,650 KB)

    Supplementary Figures 1–15 and Supplementary Tables 1–10.

Additional data