Control of cell proliferation in Arabidopsis thaliana by microRNA miR396

Abstract

Cell proliferation is an important determinant of plant form, but little is known about how developmental programs control cell division. Here, we describe the role of microRNA miR396 in the coordination of cell proliferation in Arabidopsis leaves. In leaf primordia, miR396 is expressed at low levels that steadily increase during organ development. We found that miR396 antagonizes the expression pattern of its targets, the GROWTH-REGULATING FACTOR (GRF) transcription factors. miR396 accumulates preferentially in the distal part of young developing leaves, restricting the expression of GRF2 to the proximal part of the organ. This, in turn, coincides with the activity of the cell proliferation marker CYCLINB1;1. We show that miR396 attenuates cell proliferation in developing leaves, through the repression of GRF activity and a decrease in the expression of cell cycle genes. We observed that the balance between miR396 and the GRFs controls the final number of cells in leaves. Furthermore, overexpression of miR396 in a mutant lacking GRF-INTERACTING FACTOR 1 severely compromises the shoot meristem. We found that miR396 is expressed at low levels throughout the meristem, overlapping with the expression of its target, GRF2. In addition, we show that miR396 can regulate cell proliferation and the size of the meristem. Arabidopsis plants with an increased activity of the transcription factor TCP4, which reduces cell proliferation in leaves, have higher miR396 and lower GRF levels. These results implicate miR396 as a significant module in the regulation of cell proliferation in plants.

Fig. 1.

Accumulation of miR396 during leaf development. (A) Scheme representing the GRF genes. The interaction of the seven miRNA-regulated GRFs from Arabidopsis with miR396 is shown. QLQ and WRC indicate the conserved domains that define the GRF family (Kim et al., 2003). (B) Pattern of miR396 accumulation during development of leaf 5. Samples were collected every 3 days starting 16 DAS. miR396 accumulation was analyzed by small RNA blots. Scale bar: 1 cm. (C) Accumulation of MIR396a precursor and mature miR396 during development of leaf 5. Both precursor and mature miRNA were detected by RT-qPCR. The expression levels were normalized to the earliest time point. The data shown are mean ± s.e.m. of three biological replicates. (D) Expression of miR396-regulated GRFs during leaf development. The GRFs were detected by RT-qPCR. Determination procedures were carried out as in C. (E) Expression of GRF5-6, GIF1, KNOLLE and CYCB1;1 during leaf development. Determination procedures were carried out as in C.

Fig. 2.

Dynamic pattern of miR396 expression. (A) Scheme representing two reporters with different sensitivity to miR396: GRF2:wtGRF2-GUS, which is sensitive to the miRNA, and GRF2:rGRF2-GUS, which is insensitive. (B) GUS staining of GRF2:wtGRF2-GUS plants. Arrowheads point to leaves 1 and 2. Numerals indicate plant age expressed as DAS. (C) GUS staining of GRF2:rGRF2-GUS plants. Arrowheads point to leaves 1 and 2. Numerals indicate plant age expressed as DAS. (D) Accumulation of mature miR396 in different leaves of a rosette. The GRF2:wtGRF2-GUS pattern of the same leaves is indicated at the top of each lane. (E) Expression of miR396 in the proximal (P) and distal (D) parts of leaf 3 (taken from same plants as for panel D). (F-H) Quantification of miR396, cell proliferation markers and GRFs in proliferative (leaf 5 plus apex) and non-proliferative (leaves 1 and 2) organs by RT-qPCR. The expression levels were normalized to the sample: leaf 5 plus apex. The data shown are mean ± s.e.m. of three biological replicates. (F) Accumulation of MIR396a and MIR396b precursors and mature miR396. (G) KNOLLE and CYCB1;1. (H) GRF1-9.

Fig. 3.

Effects of miR396 on GRF expression and leaf development. (A) Expression of miR396 in 10-day-old rosettes of wild-type plants and miR396b overexpressers. Numerals indicate the miR396 content of transgenic plants relative to that of wild-type plants. (B) Effect of miR396 overexpression on wild-type plants and grf5 mutants. (C) Expression levels of GRFs in 10-day-old rosettes of wild-type and 35S:miR396b #28 plants. The GRF levels were determined by RT-qPCR and normalized to wild-type plants. The data shown are mean ± s.e.m. of five biological replicates.

Fig. 4.

Control of cell proliferation by miR396. (A) GRF2:wtGRF2-GUS expression 4-7 DAS. (B) CYCB1;1:GUS reporter expression in plants the same age as in A. (C) CYCB1;1:GUS reporter expression in 35S:miR396b #28 transgenic plants. Scale bar: 1 mm.

Fig. 5.

Regulation of leaf development by the balance between miR396 and the GRFs. (A) Diagram of rGRF2 indicating the synonymous mutations that alter the interaction with miR396. The graph below shows expression levels of GRF2 in seedlings with extra copies of a wild-type (GRF2) or miRNA-resistant (rGRF2) form of the transcription factor. The data shown are mean ± s.e.m. of three biological replicates. (B-D) Complementation of the miR396 overexpression phenotype by rGRF2. (B) Rosette phenotype of wild-type, 35S:miR396b and rGRF2 plants overexpressing miR396. (C) Leaf area determined for fully expanded first leaves of wild-type, 35S:miR396b and rGRF2+35S:miR396b plants. Values expressed relative to wild-type plants. (D) miR396 expression in wild-type, 35S:miR396b and rGRF2 plants overexpressing miR396. (E) Rosette phenotype of 18-day-old plants with different GRF levels. 35S:miR396b plants have reduced GRF levels, whereas rGRF2 plants have increased GRF levels. (F) Paradermal view of palisade cells in the subepidermal layer of leaf 1 of 20-day-old plants. (G-I) Leaf area (G), cell size (H) and palisade cell number (I) per leaf in wild-type, 35S:miR396b and rGRF2 transgenic plants. The data shown are the mean ± s.e.m. of eight leaves or 50 cells.

Fig. 6.

Control of meristem function by miR396. (A) Effect of miR396 overexpression on an3-1 mutants. (B) Medial-longitudinal sections of 15-day-old seedlings stained with toluidine blue. Note the lack of a meristem dome in 35S:miR396b an3-1 plants. (C) In situ hybridization of vegetative shoot apices with an miR396 probe in the wild type. The inset shows an in situ hybridization to a section of hyl1-1 mutant, which has reduced miRNA levels. (D) In situ hybridization of vegetative shoot apices with an antisense probe against GRF2 mRNA. An in situ hybridization with a sense probe is shown in the inset. (E-F) Expression pattern in sections of shoot apices of GRF2:wtGRF2-GUS (E) and GRF2:rGRF2-GUS (F) reporters. (G) HISTONE H4 expression in shoot apices of 30-day-old plants grown in short photoperiods, detected by in situ hybridization. (H) Morphology and cell proliferation of the SAM. Width and height were estimated from medial-longitudinal sections of the SAMs of 30-day-old plants grown in short photoperiods. HISTONE H4 positive cells were counted from sections such as those of panel G. Numbers are mean ± s.e.m. (n=9). Asterisks indicate significant differences between transgenic and wild-type plants, as determined by ANOVA (P<0.05).

Fig. 7.

Regulation of miR396 and the GRFs by TCP4. (A) Estimated palisade cell number per leaf in wild-type and soj8 plants. The data shown are the mean ± s.e.m of eight leaves or 50 cells. (B) GRF2:wtGRF2:GUS and CYCB1;1:GUS expression in wild-type and soj8 plants. (C) Accumulation of TCP4 and miR396 in moderate rTCP4 transgenic plants and soj8 mutants. Determinations were carried out by RT-qPCR and normalized to wild-type plants. RNA was prepared from apices of 30-day-old plants grown in short photoperiods, including developing leaves smaller than 3 mm. The data shown are mean ± s.e.m. of three biological replicates. (D) Measurements of GRF1-9, GIF1 and KNOLLE in soj8 and rTCP4 plants. Samples were analyzed as in C.