The miR167-OsARF12 module regulates rice grain filling and grain size downstream of miR159

Abstract

Grain weight and quality are always determined by grain filling. Plant microRNAs have drawn attention as key targets for regulation of grain size and yield. However, the mechanisms that underlie grain size regulation remain largely unclear because of the complex networks that control this trait. Our earlier studies demonstrated that suppressed expression of miR167 (STTM/MIM167) substantially increased grain weight. In a field test, the yield increased up to 12.90%-21.94% because of a significantly enhanced grain filling rate. Here, biochemical and genetic analyses revealed the regulatory effects of miR159 on miR167 expression. Further analysis indicated that OsARF12 is the major mediator by which miR167 regulates rice grain filling. Overexpression of OsARF12 produced grain weight and grain filling phenotypes resembling those of STTM/MIM167 plants. Upon in-depth analysis, we found that OsARF12 activates OsCDKF;2 expression by directly binding to the TGTCGG motif in its promoter region. Flow cytometry analysis of young panicles from OsARF12-overexpressing plants and examination of cell number in cdkf;2 mutants verified that OsARF12 positively regulates grain filling and grain size by targeting OsCDKF;2. Moreover, RNA sequencing results suggested that the miR167-OsARF12 module is involved in the cell development process and hormone pathways. OsARF12-overexpressing plants and cdkf;2 mutants exhibited enhanced and reduced sensitivity to exogenous auxin and brassinosteroid (BR) treatment, confirming that targeting of OsCDKF;2 by OsARF12 mediates auxin and BR signaling. Our results reveal that the miR167-OsARF12 module works downstream of miR159 to regulate rice grain filling and grain size via OsCDKF;2 by controlling cell division and mediating auxin and BR signals.

Keywords: 2; OsARF12; OsCDKF; cell cycle; grain filling; miR167; rice.

Figure 1

Suppressed expression of miR167 increases grain weight by accelerating grain filling rate. (A and B) Phenotypic observation of grain size of NIP (WT) and STTM167 and MIM167 transgenic plants. Scale bars, 1 cm. (C) Validation of decreased expression of rice miR167 in STTM167 and MIM167 transgenic plants by stem–loop qRT–PCR. (D) Measurement of 1000-hulled grain weight of NIP (WT) and STTM167 and MIM167 transgenic plants. (E and F) Detailed analysis of grain length (E) and width (F) of NIP (WT) and miR167 transgenic plants. (G) Yield per plot of NIP (WT) and STTM167 and MIM167 transgenic plants in a field test. (H) 1000-hulled grain weight of NIP (WT) and STTM167 and MIM167 transgenic plants at different stages of grain filling. (I) Fitted grain filling rate of NIP (WT) and STTM167 and MIM167 transgenic plants at different stages of grain filling. (J) Seed morphologies of NIP (WT) and STTM167 and MIM167 transgenic plants at different stages of grain filling. Scale bars, 1 mm. Experiments were repeated three times, and data are presented as mean ± SD (n = 1000 grains). Statistical analysis was performed by Student’s t-test (∗∗P < 0.01, ∗P < 0.05).

Figure 2

Morphological analysis of the cross between STTM159 and STTM167 and biochemical analysis of OsGAMYBL2 binding to the OsMIR167h promoter. (A) Confirmation of STTM159, crossed plants, STTM167, and the WT by 1% agarose gel electrophoresis. (B) Grain morphology of NIP (WT), STTM167, STTM159, and crossed plants. Scale bars, 1 cm. (C) Measurement of 1000-grain weight of NIP (WT), STTM159, STTM167, and crossed plants (n = 3, n = 1000 grains). (D) Grain length and width of NIP (WT), STTM167, STTM159, and crossed plants (n = 3, n = 1000 grains). (E and F) Expression level of miR167 in STTM159 plant roots (E) and shoots (F) measured by small RNA-seq. (G) Relative expression level of miR167 in NIP (WT), STTM159, and STTM167 plants measured by qRT–PCR. (H) Y1H assay of OsGAMYBL2 and the OsMIR167h promoter in the presence of 50 mM 3-AT. (I–K) Luciferase reporter assays of OsGAMYBL2 and the promoter of MIR167h in rice protoplasts (I) and N. benthamiana(J and K). (L) ChIP–qPCR analysis of binding to the promoter of the miR167h2 region using a FLAG antibody to enriched DNA from seedlings of OsGAMYBL2 transgenic plants. Experiments were repeated three times, and data are presented as mean ± SD. Statistical analysis was performed by Student’s t-test (∗∗P < 0.01, ∗P < 0.05).

Figure 3

Validation of OsARF12 as the main target of miR167 in rice grain filling. (A) Relative expression of OsARF6, OsARF12, OsARF17, OsARF25, and miR167 at 5 DAF (days after fertilization), 10 DAF, 15 DAF, 21 DAF, 27 DAF, and 35 DAF measured by qRT–PCR and stem–loop qRT–PCR. (B) Phenotypic observation of grain size in NIP (WT), ARF6 OE, ARF12 OE, ARF17 OE, and ARF25 OE transgenic plants. Scale bar, 1 cm. (C) 1000-grain weight of NIP (WT) and ARF6 OE, ARF12 OE, ARF17 OE, and ARF25 OE transgenic plants. (D) Correlation analysis of OsARF12 expression level and grain weight in WT and different ARF12 OE transgenic lines. Experiments were repeated three times, and data are shown as means ± SD. Statistical analysis was performed by Student’s t-test (∗∗P < 0.01, ∗P < 0.05).

Figure 4

OsARF12 positively regulates grain filling and grain weight. (A) Subcellular localization analysis of OsARF12. Scale bars, 5 μm. (B and C) Phenotypic observation of grain size in NIP (WT) and ARF12 OE and ARF12 RNAi transgenic plants. Scale bars, 1 cm. (D) Validation of decreased expression of OsARF12 in ARF12 OE and ARF12 RNAi transgenic plants by qRT–PCR. (E) Measurement of 1000-grain weight of NIP (WT) and ARF12 OE and ARF12 RNAi transgenic plants. (F and G) Detailed analysis of grain length (F) and width (G) in NIP (WT) and ARF12 OE and ARF12 RNAi transgenic plants. (H) Yield per plot of NIP (WT) and ARF12 OE and ARF12 RNAi transgenic plants in a field test. (I) 1000-hulled grain weight of NIP (WT) and ARF12 OE and ARF12 RNAi transgenic plants at different stages of grain filling. (J) Fitted grain filling rate of NIP (WT) and ARF12 OE and ARF12 RNAi transgenic plants at different stages of grain filling. (K) Seed morphologies of NIP (WT) and ARF12 OE and ARF12 RNAi transgenic plants at different stages of grain filling. Scale bars, 0.5 cm. (L) Number of cells that contained 2C DNA and 4C DNA in young panicles of WT and ARF12 OE transgenic plants. (M) Percentage of cells in different phases of the cell cycle in young panicles of WT and ARF12 OE transgenic plants.Experiments were repeated three times, and data are presented as mean ± SD (n = 1000 grains). Statistical analysis was performed by Student’s t-test (∗∗P < 0.01, ∗P < 0.05).

Figure 5

OsARF12 directly activates expression of OsCDKF;2. (A) Potential binding sequence of OsARF12 calculated by scoring matrix analysis. (B) Increased and decreased expression of OsCDKF;2 in ARF12 OE and ARF12 RNAi transgenic plants, respectively. (C) Location of the putative OsARF12 binding site in the OsCDKF;2 promoter. (D) Y1H assay of OsARF12 and the promoter of OsCDKF;2 in the presence of 80 mM 3-AT. The empty pGADT7-Rec2 vector was used as a negative control. (E) Transactivation assay of OsARF12 with the luciferase reporter system. (F and G) ChIP–qPCR analysis of OsARF12 binding to the promoter of OsCDKF;2 using a FLAG antibody to enriched DNA from spikelets and panicles of ARF12 OE transgenic plants. Experiments were repeated three times, and data are presented as mean ± SD. Statistical analysis was performed by Student’s t-test (∗∗P < 0.01, ∗P < 0.05).

Figure 6

Positive effects of OsCDKF;2 on grain filling and grain size. (A) Comparison of grains from NIP (WT) and cdkf;2 and ARF12 OE-cdkf;2 mutants. Scale bars, 1 cm.(B) Measurement of 1000-grain weight of NIP (WT) and cdkf;2 and ARF12 OE-cdkf;2 mutants (n = 3, n = 1000 grains). (C–E) Grain length (C), width (D), and thickness (E) of NIP (WT) and cdkf;2 and ARF12 OE-cdkf;2 mutants (n = 3, n = 1000 grains). (F) Fitted grain filling rate of NIP (WT) and cdkf;2 mutants at different stages of grain filling.(G) Fitted grain filling rate of ARF12 OE transgenic plants and ARF12 OE-cdkf;2 mutants at different stages of grain filling. (H) Cross-sections of spikelet hulls from NIP (WT), ARF12 OE transgenic plants, and ARF12 OE-cdkf;2 and cdkf;2 mutants (scale bars, 500 μm) and magnified views of spikelet hull cross-sections of NIP (WT), ARF12 OE transgenic plants, and ARF12 OE-cdkf;2 and cdkf;2 mutants (scale bars, 50 μm). (I) Statistical analysis of cell numbers at the outer parenchyma layer of spikelet hulls from NIP (WT), ARF12 OE transgenic plants, and ARF12 OE-cdkf;2 and cdkf;2 mutants (n = 10). Experiments were repeated three times, and data are presented as mean ± SD. Statistical analysis was performed by Student’s t-test (∗∗P < 0.01, ∗P < 0.05).

Figure 7

Proposed model explaining how the miR167-OsARF12 module acts downstream of miR159 to regulate grain filling and grain size via OsCDKF;2 in rice. miR159 is thought to suppress expression of miR167 through OsGAMYBL2 binding to the promoter of miR167h. OsARF12 is identified as the main target of miR167 in rice grain filling regulation. OsARF12 is thought to upregulate expression of OsCDKF;2 by directly binding to the TGTCGG motif in the OsCDKF;2 promoter. OsCDKF;2 targeted by OsARF12 mediates auxin and BR signals and is involved in cell cycle processes, resulting in increased grain filling and grain size. Dashed lines indicate a putative signaling pathway.