Short grain1 decreases organ elongation and brassinosteroid response in rice

Abstract

We identified a short-grain mutant (Short grain1 (Sg1) Dominant) via phenotypic screening of 13,000 rice (Oryza sativa) activation-tagged lines. The causative gene, SG1, encodes a protein with unknown function that is preferentially expressed in roots and developing panicles. Overexpression of SG1 in rice produced a phenotype with short grains and dwarfing reminiscent of brassinosteroid (BR)-deficient mutants, with wide, dark-green, and erect leaves. However, the endogenous BR level in the SG1 overexpressor (SG1:OX) plants was comparable to the wild type. SG1:OX plants were insensitive to brassinolide in the lamina inclination assay. Therefore, SG1 appears to decrease responses to BRs. Despite shorter organs in the SG1:OX plants, their cell size was not decreased in the SG1:OX plants. Therefore, SG1 decreases organ elongation by decreasing cell proliferation. In contrast to the SG1:OX plants, RNA interference knockdown plants that down-regulated SG1 and a related gene, SG1-LIKE PROTEIN1, had longer grains and internodes in rachis branches than in the wild type. Taken together, these results suggest that SG1 decreases responses to BRs and elongation of organs such as seeds and the internodes of rachis branches through decreased cellular proliferation.

Figure 1.

Identification of the rice SG1 gene by activation tagging. A to C, Phenotypes of the Sg1-D rice mutant. A, Heading stage: In comparison with the wild-type plants (WT), the Sg1-D homozygote showed a semidwarf phenotype. B, The short-grain phenotype cosegregated with the T-DNA insertion in a dose-dependent manner. Seeds of WT (+/+), Sg1-D heterozygotes (Sg1-d/+), and Sg1-D homozygotes (Sg1-D homo) are shown. Means and sd from at least 30 measurements of lengths (L, mm), widths (W, mm), and the ratio of length to width (L/W) are shown below the seeds. C, Leaf blades of the Sg1-D homozygote were dark green and were shorter and wider than those of wild type. Means and sd from at least nine measurements of lengths (L, cm) and widths (W, cm) of second uppermost leaves are shown. D, Schematic representation of the T-DNA flanking region of the Sg1-D mutant. A candidate for the ORF responsible for the phenotype (Os09g0459200, which encodes a novel protein with unknown function, colored blue), referred to as SG1, is located 1.4-kb downstream of the T-DNA insertion. The cauliflower mosaic virus 35S minimal promoter (P35S) and a tetramer of the 35S enhancer (4×EN) are shown by the black triangles and the green box, respectively. LB and RB are left and right borders of the T-DNA. Pnos, Nopaline synthase promoter sequence; PAT, ORF-containing region of the phosphinothricin acetyltransferase gene. E, The levels of SG1 mRNA in the WT, Sg1-D heterozygote, and Sg1-D homozygote plants. For each genotype, RNA from two independent plants was used for qPCR, and values were normalized using RUBQ2 as a standard. F, The Sg1-D mutant shows reduced panicle internode elongation in a dose-dependent manner. Bases of the terminal spikelets and nodes of the primary rachis branches are connected with dotted lines. Asterisks show that significantly different from WT. **, P < 0.01; ***, P < 0.001 (Student’s t test).

Figure 2.

Amino acid sequences and phylogenetic analysis of the SG1 and SGL1 proteins. A, Sequence alignment of the SG1, SGL1, and SG1-like proteins in rice and Arabidopsis. Amino acid residues identical and similar to those of SG1 are shaded in black and gray, respectively. The multiple sequence alignment was performed using the CLUSTAL W analysis tool provided by the DNA Data Bank of Japan. B, Phylogenetic relationships among the SG1, SGL1, and SG1-like proteins in plants. The corresponding sequence alignment is shown in Supplemental Figure S2. Bootstrap values from 1,000 replicates are indicated at each node. At, Arabidopsis; Os, O. sativa; Pt, Populus trichocarpa; Rc, Ricinus communis; Sb, S. bicolor; Vv, Vitis vinifera; Zm, Z. mays. The bar corresponds to 0.1 amino acid substitutions per site.

Figure 3.

Phenotype of the SG1:OX rice plants. A, Mature plants at the heading stage. Wild-type (WT) and SG1:OX plants with the short-grain phenotype (S) and the extremely short-grain phenotype (XS) are shown. B, Patterns of internode elongation of the WT and SG1:OX plants with the short-grain phenotype (S). More than 10 culms of the wild-type and SG1:OX plants were measured. I to V represent the internode number, with I representing the uppermost internode. C, Seeds of wild-type plants, and of SG1:OX plants with moderately short (MS), short (S), and extremely short (XS) length are shown. D, qPCR analysis of SG1 mRNA in the wild-type and SG1:OX plants. Symbols (MS, S, XS) correspond to those in A to C. E, Comparison of seed length in the WT and SG1:OX plants. Values are means ± sd for more than 250 seeds. F, Panicles of WT plants and SG1:OX plants with the short-grain phenotype. G, Close-up view of the primary rachis branches. Arrows and triangles indicate the bases of the terminal spikelets and of the pedicels, respectively. Values are means ± sd for the length of each internode (mm). At least 30 internodes were measured for each line of plants. Asterisks show that significantly different from WT. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Student’s t test).

Figure 4.

Expression analysis of SG1 using qPCR and SG1::GUS transgenic plants. A and B, qPCR analysis of SG1 in various organs of wild-type plants. Expression of SG1 was normalized using 18S rRNA. A, Vegetative organs. Shoots and roots of 10-d-old seedlings and leaf blades (LB) and leaf sheaths (LS) of 2-month-old plants were used. B, Reproductive organs. Young panicles (YP) at 0.3, 1, 5, 15, and 18 cm and mature panicles (MP) at 0 and 3 d after heading (DAH) were used. C, qPCR analysis of SG1 in various parts of the young panicle. Schematic representation of the rice panicle (left). Young panicles were dissected into three parts: spikelets, upper rachis, and rachis branches, and lower rachis and rachis branches. These parts were then used for qPCR analysis (right). qPCR analyses are shown for SG1 in the young panicles at the 4-cm stage and the 10-cm stage. Expression of SG1 was normalized using 18S rRNA. D to K, Histochemical staining of transgenic plants harboring the SG1::GUS construct. D to H, Seven-day-old seedlings: Views are of the whole plant (D), a cross section of the seminal root (E), a close-up of the coleoptile (F), a cross section of the coleoptile (G), and a cross section of the embryo (H). I, Vegetative node of the 2-month-old plant. J and K, Young panicles: at the 2-cm stage (J) and 10-cm stage (K) are shown.

Figure 5.

SG1:OX plants showed a BR-insensitive lamina-joint phenotype. A, Typical response of the lamina joint of the second leaf of wild-type plants (WT) and SG1:OX plants (overexpressor) treated with 0, 1, 10, 100, and 1,000 ng of BL. B, The response of the bending angle to BL as a function of dose in the WT and SG1:OX plants. Data represent means ± sd of the results from at least six plants. Asterisks show that significantly different from WT. **, P < 0.01; ***, P < 0.001 (Student’s t test).

Figure 6.

Phenotypes of RNAi knockdown plants that down-regulated expression of both SG1 and SGL1. A, Expression of SG1 and SGL2 in the wild-type (WT) and SG1SGL1i RNAi knockdown lines. B, Panicles of WT and SG1SGL1-RNAi plants. C, Seeds of WT and SG1SGL1-RNAi lines. D, Comparison of seed lengths of WT and SG1SGL1-RNAi lines. Values represent means ± sd of at least 25 measurements. E, Close-up view of the primary rachis branches of the wild-type and SG1SGL1-RNAi lines. Bases of the terminal spikelets and of the pedicels are indicated by arrows and triangles, respectively. Values are shown for means ± sd based on at least 30 rachis internodes for each plant. Asterisks show that significantly different from WT. *, P < 0.05; ***, P < 0.001 (Student’s t test).

Figure 7.

Cellular sizes of the SG1:OX plants are comparable to those of the wild-type (WT) plants. A to C, Comparisons of epidermal imprint images of the WT and SG1:OX plants. A, Adaxial surface of the lemma. B, An internode of the rachis branch. C, The adaxial surface of the leaf sheath. D to F, Size distribution of epidermal cell lengths. Triangles at the right of each plot show the mean cellular length of the WT and SG1:OX plants. D, Lower epidermis of the lemma. E, Epidermal cells between the stomata in the rachis branch. F, Adaxial surface of the leaf sheath. For both types of plants, 250 cells (D and E) or 120 cells (F) from at least five independent organs were measured. Asterisks show that significantly different from WT. *, P < 0.05; ***, P < 0.001 (Student’s t test). NS, Nonsignificant (P > 0.05).