OsMAPKKK5 affects brassinosteroid signal transduction via phosphorylating OsBSK1-1 and regulates rice plant architecture and yield

Abstract

Improving plant architecture and increasing yields are the main goals of rice breeders. However, yield is a complex trait influenced by many yield-related traits. Identifying and characterizing important genes in the coordinated network regulating complex rice traits and their interactions is conducive to cultivating high-yielding rice varieties. In this study, we determined that the interaction between mitogen-activated protein kinase kinase kinase5 (OsMAPKKK5) and brassinosteroid-signalling kinase1-1 (OsBSK1-1) regulates yield-related traits in rice. Specifically, OsMAPKKK5 phosphorylates OsBSK1-1, which enhances the interaction between these two proteins, but adversely affects the OsBSK1-1-OsBRI1 (BR insensitive1) and OsBSK1-1-OsPPKL1 (protein phosphatase with two Kelch-like domains) interactions. Additionally, OsMAPKKK5 disrupts brassinosteroid signal transduction, which prevents OsBZR1 (brassinazole-resistant1) from efficiently entering the nucleus, thereby negatively modulating its function as a transcription factor regulating downstream effector genes, ultimately adversely affecting plant architecture and yield. This study revealed the relationship between the MAPK cascade and the regulatory effects of brassinosteroid on the rice grain yield involves OsMAPKKK5 and OsBSK1-1. The study data may be important for future investigations on the rice yield-regulating molecular network.

Keywords: brassinosteroids; crosstalk; mitogen‐activated protein kinase kinase kinase; plant architecture; rice yield.

Figure 1

Phenotypic effects of OsMAPKKK5 on plant and grain yield traits in transgenic rice lines. (a) The knockout target and mutation sequence of the OsMAPKKK5 knockout lines in the wild‐type Bing1B background. (b–e) Phenotypic comparison between the wild‐type Bing1B and OsMAPKKK5 knockout lines. (f–k) Phenotypic differences in plant and grain yield traits between wild‐type Bing1B and OsMAPKKK5 knockout line (n = 10). (l–o) Phenotypic comparison between the wild‐type Bing1B and OsMAPKKK5 overexpressing lines. (p–u) Phenotypic differences in plant and grain yield traits between wild‐type Bing1B and OsMAPKKK5 overexpressing lines (n = 10). (v) The relative expression level of OsMAPKKK5 in wild‐type Bing1B and overexpressing lines (n = 3). (w) Stem cell sections. The measured cells are marked with red box, Scale bar = 100 μm; (x) Statistical analysis of cell length in (W). The data are shown as mean ± SEM (n = 80). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2

Experimental evidence for the interaction between OsMAPKKK5 and OsBSK1–1. (a) Yeast two‐hybrid assays. OsMAPKKK5 was fused with the DNA‐binding domain (BD) of GAL4, and full‐length of OsBSK1–1 were fused with the activation domain (AD) of GAL4. (b) Pull‐down assay confirmation of the in vitro interaction of OsMAPKKK5 and OsBSK1–1. (c) BiFC assay demonstration of the interaction of OsMAPKKK5 and OsBSK1–1 in rice protoplasts, Scale bar = 5 μm. cYFP‐OstMAPKKK5 and nYFP‐OsBSK1–1 (NT) served as negative controls. (d) Verification of the in vivo interaction between OsMAPKKK5 and OsBSK1–1 by the split‐luciferase assay. OsMAPKKK5 and OsBSK1–1 were fused with either the N‐terminal (nLUC) or C‐terminal (cLUC) portion of firefly luciferase (LUC). OstMAPKKK5‐nLUC and cLUC‐OsBSK1–1 (NT) served as negative controls. (e) Co‐immunoprecipitation assays. Pro35S:flag‐OsMAPKKK5 and Pro35S:OsBSK1–1‐GFP fusions were co‐expressed in N. benthamiana leaves. Proteins were extracted (Input) and immunoprecipitated (IP) with flag beads. The immunoblot assays were performed using anti‐flag and anti‐GFP antibodies. (f) Subcellular localization of OsMAPKKK5 and OsBSK1–1 in rice protoplasts, Scale bar = 5 μm. (g) The relative spatial and temporal expression levels of OsMAPKKK5 and OsBSK1–1. The values on the y‐axis represent the ratio of the gene's expression level to the expression level of OsUBQ5. The data are shown as mean ± SEM (n = 3).

Figure 3

Phenotypic effects of OsBSK1–1 on plant and grain yield traits in transgenic rice lines. (a) Phenotypic comparison between the wild‐type Bing1B and OsBSK1–1overexpressing lines. (b–d) Phenotypic comparison between the wild‐type Bing1B and OsBSK1–1 overexpressing lines. (e–j) Phenotypic differences in plant and grain yield traits between wild‐type Bing1B and OsBSK1–1 overexpressing lines (n = 10). (k) The relative expression level of OsBSK1–1 in wild‐type Bing1B and overexpressing lines (n = 3). (l) Stem cell sections. The measured cells are marked with red box, Scale bar = 100 μm; (m) Statistical analysis of cell length in (W), The data are shown as mean ± SEM (n = 60). **P < 0.01, ***P < 0.001.

Figure 4

Responses of wild type and transgenic plants to 24‐epibrassinolide (24‐eBL). (a) Rice young root morphology under1 μM 24‐eBL treatment in seedlings. (b) Coleoptile length in the presence or absence of 1 μM 24‐eBL; (c) Statistical analysis of coleoptile length in (b), n = 18. (d) Leaf angle in response to gradient concentration of 24‐eBL; (e) Statistical analysis of bending angle in (d), n = 8. The data are shown as mean ± SEM, *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 5

Experimental evidence for the phosphorylation OsBSK1–1 by OsMAPKKK5. (a) Protein function domains of OsMAPKKK5. (b) Yeast two‐hybrid assays. OsBSK1–1 was fused with the activation domain (AD) of GAL4, and full‐length and various truncated forms of OsMAPKKK5 were fused with the DNA‐binding domain (BD) of GAL4. (c) In vitro experimental verification of phosphorylation of OsBSK1–1 by OsMAPKKK5 and itself. His‐OsBSK1–1 migration rate slows down after the addition of GST‐OsMAPKKK5^K416M when phos‐tag and ATP are added to the protein gel. His‐OsBSK1–1 slows down more with longer incubation time, resulting in the shift band. His‐OsBSK1–1 can be autophosphorylated in the presence of ATP. *1, 55 kDa, *2, 72 kDa. (d) The mass spectrometry analysis results showing the 2 sites in OsBSK1–1 phosphorylated by OsMAPKKK5 (marked in red, stated as P1, P2) and by itself (marked in orange). (e) Interaction intensity between OsMAPKKK5 and OsBSK1–1 (intact and mutated variants: OsBSK1–1^WT, OsBSK1–1^2A and OsBSK1–1^2D) (n = 5). (f) FRET images. Scale bars = 5 μm. (g) FRET efficiency (n = 10). (h) Activity of firefly luciferase (LUC) after adding flag‐OsMAPKKK5. (i) Activity of LUC. OsBSK1–1 (intact and mutated variants: OsBSK1–1^WT, OsBSK1–1^2A and OsBSK1–1^2D) and OsBRI1 or OsPPKL1 were fused with either the N‐terminal (N‐LUC) or C‐terminal (C‐LUC) portion of LUC. The data are shown as mean ± SEM, ns represents not significant, *P < 0.05, **P < 0.01, ns represents no significance.

Figure 6

Identification of the pathways that synergistically affected by OsMAPKKK5 and OsBSK1–1. (a) Venn of the differentially expressed genes (DEGs) of overexpression (O.E.)‐OsMAPKKK5 lines and bsk1–1 lines. (b) Heatmap of the common DEGs. (c) Expression levels of OsBZR1 downstream target genes in wild‐type Bing1B (B1B) and OsMAPKKK5‐overexpressing lines. The genes that are negatively regulated by OsBZR1 were shown on the left of the dotted line, while genes that are positively regulated by OsBZR1 were shown on the right of the dotted line (n = 3). (d) Subcellular localization of OsBZR1 in the presence or absence of OsMAPKKK5 with eBL. OsPAY1‐mCHERRY served as the mark of nucleus. Scale bars = 5 μm. (e) Quantitative analysis of nucleus‐cytoplasm ratio of OsBZR1 with or without OsMAPKKK5 (n = 15). (f) The content of OsBZR1‐EGFP in cytoplasm of tobacco leaves. The number below the lane represents the relative expression level of OsBZR1‐EGFP and is equal to the grayscale value of α‐GFP divided by α‐ACTIN. α‐H3 has no signal, indicating there are no nuclear protein contamination. (g) The relative expression level of OsBZR1‐EGFP in cytoplasm (n = 3). The data are shown as mean ± SEM, ns represents no significance, *P < 0.05.

Figure 7

Proposed model of the mechanism underlying the functions for OsMAPKKK5 and OsBSK1–1 in modulating rice plant height and yield traits. OsBSK1–1 is a brassinosteroids (BR) signalling receptor that transmits BR signals from OsBRI1 to OsPPKL1. (a) The OsMAPKKK5 phosphorylates OsBSK1–1, which strengthens the interaction between OsMAPKKK5 and OsBSK1–1, thereby preventing OsBSK1–1 from interacting with OsBRI1 and OsPPKL1 and resulting the weak BR signals. Accordingly, OsBZR1 is prevented from entering the nucleus and regulating the expression of the downstream genes. (b) In the OsMAPKKK5 knockout lines, the strong BR signals are transmitted by OsBSK1–1 from OsBRI1 to OsPPKL1, OsBZR1 enters the nucleus and regulates the expression of the downstream genes, ultimately leading to the improvement in the plant height and yield traits.