A WUSCHEL-LIKE HOMEOBOX gene represses a YABBY gene expression required for rice leaf development

Abstract

YABBY and WUSCHEL-LIKE HOMEOBOX (WOX) genes have been shown to play important roles in lateral organ formation and meristem function. Here, we report the characterization of functional relationship between rice (Oryza sativa) YAB3 and WOX3 in rice leaf development. Rice YAB3 is closely related to maize (Zea mays) ZmYAB14 and Arabidopsis (Arabidopsis thaliana) FILAMENTOUS FLOWER (FIL), whereas rice WOX3 is highly conserved with maize narrow sheath1 (NS1) and NS2 and Arabidopsis PRESSED FLOWER (PRS). In situ hybridization experiments revealed that the expression of both genes was excluded from the shoot apical meristem, but the transcripts were detected in leaf primordia, young leaves, and reproductive organs without any polar distribution. The function of the two genes was studied by both overexpression and RNA interference (RNAi) in transgenic rice. YAB3 RNAi induced twisted and knotted leaves lacking specialized structures such as ligule and auricles, while no phenotypic change was observed in YAB3 overexpression plants, suggesting that rice YAB3 may be required for leaf cell growth and differentiation. Overexpression of WOX3 repressed YAB3 and showed a YAB3 RNAi phenotype. The expression of class I KNOTTED-LIKE HOMEOBOX (KNOX) genes was ectopically induced in leaves of YAB3 RNAi or WOX3 overexpression plants. Data from inducible WOX3 expression and DNA-protein interaction assays suggested that WOX3 acted as a transcriptional repressor of YAB3. These data reveal a regulatory network involving YAB3, WOX3, and KNOX genes required for rice leaf development.

Figure 1.

Phylogeny analysis of YABBY and WOX families. Neighbor-joining trees of YABBY (A) and WOX (B) proteins from rice (Os), Arabidopsis (At), Antirrhinum majus (Am), maize (Zm), Triticum aestivum (Ta), Solanum tuberosum (St), Ipomoea nil (In), Nymphaea alba (Na), Nymphaea colorata (Nc), and Solanum lycopersicum (Le). GenBank accession numbers are shown in Table I.

Figure 2.

In situ detection of YAB3 and WOX3 transcripts. A to G, Sections hybridized with YAB3 antisense (A–F) or sense (G) probes. H to N, Sections hybridized with WOX3 antisense (H–M) or sense (N) probes. A and H, Longitudinal sections of SAM. B and I, Transverse sections of SAM; insets, enlarged views of the central areas of the apexes. C and J, Developing panicle longitudinal sections. D to F and K to M, Longitudinal sections of florets at different development stages. The midrib regions of leaf primordia (A and H) are designated by plastochron (P) numbers, such that the leaf incipient primordium is labeled P0, the next older leaf is labeled P1, and so on. FAM, Floral apical meristem; st, stamen; ca, carpel; l, lemma; p, palea. Bar = 100 μm.

Figure 3.

Nuclear localization of YAB3 and WOX3 GFP alone was localized in cytosol of onion skin cells (A). YAB3-GFP (B) and WOX3-GFP (C) are nucleus localized. a, GFP images. b, Transmission images. c, DAPI staining. d, Merged images. Bar = 50 μm.

Figure 4.

Shoot phenotypes induced by YAB3 RNAi or by WOX3 overexpression. A, Wild-type leaf at abaxial (a) and adaxial (b) views showing the structures of ligule (l), auricle (au), and lamina joint (lj). bl, Blade; s, sheath. B, Phenotypes in YAB3 RNAi (a–d) and WOX3 (e–h) overexpressing plants. a and e, Profile of the transgenic plants. b and f, Enlarged view of the parts with twisted leaves in transgenic plants. Insets, junctions between leaf blades and sheaths with ligules and auricles missing, indicated by triangles. c and g, Knots in the lower parts of some transgenic leaves; arrows indicate the position of the knots. d and h, Knots in the upper parts of transgenic leaves, arrows indicate the position of the knots. The knotted regions in c, d, g, and h are enlarged (insets).

Figure 5.

Expression analysis of YAB3, WOX3, and KNOX genes in YAB3 RNAi and WOX3 overexpression plants. A, YAB3 transcript levels revealed by RT-PCR in wild-type (WT), a nonphenotypic (Y-L), and three phenotypic (Y1 to Y3) YAB3 RNAi transgenic plants. B, WOX3 transcript levels revealed by northern blots in wild-type (WT), a nonphenotypic (W-L), and three phenotypic (W1–W3) WOX3 overexpression transgenic plants. C, Transcript levels of rice KNOX genes OSH1 and OSH3 revealed by RT-PCR in leaves from wild-type (WT), phenotypic YAB3 RNAi (Y1 and Y2), and WOX3 overexpression plants (W1 and W2). Shoot apex mRNA was used as a positive control (S).

Figure 6.

Analysis of YAB3 overexpression and WOX3 RNAi plants. A, Expression analysis of YAB3 in wild-type (WT) and five independent YAB3 overexpression transgenic lines (OY1–OY6). B, Comparison of wild type (left) to the YAB1 overexpression transgenic line OY6 (right). C, Expression analysis of WOX3 in wild-type (WT) and five independent WOX3 RNAi transgenic plants (RW1–RW5). D, No phenotypic change observed in RW1 (right) compared to wild type (left). [See online article for color version of this figure.]

Figure 7.

Expression analysis of WOX3 and YAB3 in the transgenic and wild-type plants. A, Semiquantitative RT-PCR analysis of four rice YABBY gene expression in WOX3 overexpression line W1 compared to wild type. Rice actin1 transcripts were measured as controls. B, Semiquantitative RT-PCR analysis of WOX3 and the YABBY genes in the YAB3 RNAi line Y1 compared to wild type. Rice actin1 transcripts were used as control. C, In situ hybridization to detect YAB3 transcripts in wild type (a and d) and WOX3 overexpression (b and e) shoot apexes (top) and florets (bottom). A shoot apex and a floret hybridized with a sense probe of YAB3 were used as controls (c and f). D, Expression analysis of YAB3 in wild-type and WOX3 RNAi 1-week-old seedlings, inflorescences, and florets. Rice actin1 transcripts were measured and used as controls.

Figure 8.

Activation of WOX3 directly repressed the expression of YAB3. A, Identification of transgenic lines expressing the WOX3-GR fusion by real-time RT-PCR. B, Relative expression levels of YAB3 in wild-type and three WOX3-GR transgenic plants treated with or without DEX and/or CHX. The YAB3 transcript levels were normalized with the actin1 mRNA levels, and that of the wild-type control was assessed as 1. Bars are sd ± three biological repeats.