Regulation of Vegetative Phase Change by SWI2/SNF2 Chromatin Remodeling ATPase BRAHMA

Abstract

Plants progress from a juvenile vegetative phase of development to an adult vegetative phase of development before they enter the reproductive phase. miR156 has been shown to be the master regulator of the juvenile-to-adult transition in plants. However, the mechanism of how miR156 is transcriptionally regulated still remains elusive. In a forward genetic screen, we identified that a mutation in the SWI2/SNF2 chromatin remodeling ATPase BRAHMA (BRM) exhibited an accelerated vegetative phase change phenotype by reducing the expression of miR156, which in turn caused a corresponding increase in the levels of SQUAMOSA PROMOTER BINDING PROTEIN LIKE genes. BRM regulates miR156 expression by directly binding to the MIR156A promoter. Mutations in BRM not only increased occupancy of the -2 and +1 nucleosomes proximal to the transcription start site at the MIR156A locus but also the levels of trimethylated histone H3 at Lys 27. The precocious phenotype of brm mutant was partially suppressed by a second mutation in SWINGER (SWN), but not by a mutation in CURLEY LEAF, both of which are key components of the Polycomb Group Repressive Complex 2 in plants. Our results indicate that BRM and SWN act antagonistically at the nucleosome level to fine-tune the temporal expression of miR156 to regulate vegetative phase change in Arabidopsis.

Figure 1.

Identification of the brm-7 mutant with a precocious vegetative phase change phenotype. The 18-d-old wild-type, pSPL9::eGFP-rSPL9, pSPL9::eGFP-rSPL9, brm-7 pSPL9::eGFP-SPL9, and brm-7 plants were grown in short days. The first leaf with abaxial trichomes was scored. Numbers indicate the first leaf with abaxial trichomes (n = 27, ±SD). Juvenile leaves are shown in gray and adult leaves in black in the heteroblastic analyses. Different letters indicate significant difference between genotypes using one way ANOVA at P < 0.01. Scale bar = 1 cm.

Figure 2.

BRM activates miR156 expression transcriptionally during vegetative phase change. A, The level of mature miR156 in 14-d-old wild-type and brm-7 seedlings in short days. B, Expression of SPL3, SPL9, and SPL13 in 14-d-old wild-type and brm-7 seedlings in short days. C, The temporal expression pattern of pri-MIR156A in wild type and brm-7 in short days. D, The temporal expression pattern of pri-MIR156C in wild type and brm-7 in short days. E, GUS staining of 11-d-old pMIR156A::GUS and pMIR156A::GUS brm-7 plants in long days. Scale bar = 2 mm. F, Relative expression of GUS transcript in 11-d-old pMIR156A::GUS and pMIR156A::GUS brm-7 plants as shown in E. DAP, Days after planting. Asterisk denotes statistical significance from wild type at P < 0.01 using Student’s t test.

Figure 3.

BRM interacts genetically with the miR156-SPLs pathway. Plants were grown in short days. The first leaf with abaxial trichomes was scored. Numbers indicate the first leaf with abaxial trichomes (n = 27, ±SD). Juvenile leaves are shown in gray and adult leaves in black in the heteroblastic analyses. Different letters indicate significant difference between genotypes using one-way ANOVA at P < 0.01. Scale bar = 1 cm.

Figure 4.

BRM binds to the MIR156A promoter region directly. A, The phenotype of wild type, brm-7, and brm-7 transformed with pBRM::3×FLAG-BRM and pBRM::3×HA-BRM in short days. Numbers indicate the first leaf with abaxial trichomes (n = 32, ±sd). Juvenile leaves are shown in gray and adult leaves in black in the heteroblastic analyses. Scale bar = 1 cm. B, Direct binding of BRM to the MIR156A promoter region shown by ChIP analysis. Red rectangle denotes the MIR156A stem loop structure region. TSS, TSS of MIR156A. Arrows denotes the promoter regions detected by ChIP. Asterisk denotes statistical significance at P < 0.01. ChIP results were averaged from three biological replicates with three technical replicates for each sample.

Figure 5.

BRM is required to maintain low occupancy of the −2 and +1 nucleosome at the MIR156A locus. qPCR was used to detect the nucleosome positioning and occupancy at the MIR156A locus using MNase-digested DNA from 10-d-old wild-type and brm-7 plants in short days. The fraction of digested mononucleosome DNA amplified for each amplicon was normalized to that of the −73 position of the negative control locus of gypsy-like retrotransposon. Values represent mean ± sd from three technical replicates in a representative experiment. The number on the x axis denotes distance (bp) from the TSS (0 bp) of MIR156A. A diagram of the positioned nucleosomes is shown below the x axis. Blue ovals, nucleosomes; black arrow, TSS; black lines, genomic DNA; red rectangle, MIR156A stem loop structure region. Two figure insets show the detailed increase in occupancy at the −2 and +1 nucleosomes in brm-7 compared with wild type. Asterisk denotes statistical significance from wild type at P < 0.01.

Figure 6.

SWN, but not CLF, antagonizes BRM to promote vegetative phase change. Plants were grown in short days. The first leaf with abaxial trichomes was scored. Numbers indicate the first leaf with abaxial trichomes (n = 18, ±sd). Juvenile leaves are shown in gray and adult leaves in black in the heteroblastic analyses. Different letters indicate significant difference between genotypes using one-way ANOVA at P < 0.01. Scale bar = 1 cm.

Figure 7.

swn-3 delays vegetative phase change by modulating the expression of genes in the miR156-SPLs pathway. A, The level of mature miR156 in 14-d-old wild-type, brm-7, swn-3, and swn-3 brm-7 seedlings in short days. B, The level of pri-MIR156A in 14-d-old swn-3 seedlings in short days. C, The level of pri-MIR156C in 14-d-old swn-3 seedlings in short days. D, The level of SPL3, SPL9, and SPL13 in 14-d-old swn-3 seedlings in short days. Asterisk denotes statistical significance from wild type at P < 0.01 using Student’s t test.

Figure 8.

The level of H3K27me3 at MIR156A and temporal expression of miR156 in wild type and brm-7. A, The level of H3K27me3 in the promoter and coding region of MIR156A in 14-d-old wild-type, brm-7, swn-3, and swn-3 brm-7 seedlings in short days. Values represent mean ± sd from two technical replicates in a representative experiment. B, The level of H3K27me3 in 10-d-old wild-type juvenile plants and 21-d-old wild adult plants in short days. Values represent mean ± sd from two technical replicates in a representative experiment. C, Temporal expression pattern of miR156 in wild type and brm-7 in short days. Wild-type and brm-7 plants were harvested for qRT-PCR analysis of mature miR156 levels at different time points. Asterisk denotes statistical significance from wild type at P < 0.01 using Student’s t test.

Figure 9.

Nucleosome occupancy at the MIR156A locus in different mutants. A, Nucleosomes occupancy at the MIR156A locus in 10-d-old wild-type, brm-7, swn-3, and swn-3 brm-7 plants in short days. B, Nucleosome occupancy at the MIR156A locus in juvenile and adult wild-type plants in short days. qPCR was used to detect the nucleosome positioning and occupancy at the MIR156A locus using MNase-digested DNA from wild-type, brm-7, swn-3, and swn-3 brm-7 plants (A) or from 10-d-old juvenile wild-type plants and 21-d-old adult wild-type plants (B). The fraction of digested mononucleosome DNA amplified for each amplicon was normalized to that of the −73 position of the negative control locus of gypsy-like retrotransposon. Values represent mean ± sd from three technical replicates in a representative experiment. Asterisk denotes the statistical significance from wild type at P < 0.01 using Student’s t test.

Figure 10.

A model for the function of BRM and SWN in the regulation of MIR156A expression during vegetative phase change in Arabidopsis. A, Regulation of MIR156A by BRM and SWN during vegetative phase change in wild-type plants. At the juvenile phase, BRM binds to the MIR156A promoter to antagonize SWN function to reduce the level of H3K27me3 at the MIR156A locus as well as to reduce nucleosome occupancy to activate MIR156A expression. When plants enter the adult phase, BRM still functions normally, but SWN overtakes the function of BRM to increase the level of H3K27me3 at the MIR156A locus to repress MIR156A expression. Red star: H3K27me3 modification. Number of red stars denotes the degree of H3K27me3. Temporal changes in the level of miR156 and H3K27me3 were illustrated by the shaded bars during the juvenile-to-adult development. B, Regulation of MIR156A by BRM and SWN during vegetative phase change in the brm mutant plants. When BRM is mutated, SWN takes over BRM function to increase the level of H3K27me3, and the brm mutation increases nucleosome occupancy at the MIR156A locus to repress MIR156A expression.