GEM, a Novel Factor in the Coordination of Cell Division to Cell Fate Decisions in the Arabidopsis Epidermis

Abstract

Cell division and cell fate decisions are highly regulated processes that need to be coordinated both spatially and temporally for correct plant growth and development. Gaining a deeper molecular and cellular understanding of these links is especially relevant for plant biology since, unlike in animals, formation of new organs is a process that takes place after embryogenesis and continues throughout the entire plant lifespan. The recent identification of a novel factor, GEM, has provided a molecular framework that coordinates cell division to cell fate in the Arabidopsis epidermis. GEM is an inhibitor of cell division through interacting with CDT1, a DNA replication protein. It also inhibits the expression of the homeobox GLABRA2 (GL2) gene that determines the hair/non-hair fate and the pavement/trichome fate in the root and leaf epidermis, respectively. GEM seems to be crucial in controlling the balance of activating/repressing histone modifications at its target promoters.

Cell Fate Decisions

The epidermis of Arabidopsis thaliana has proved a very good model to study both cell division and differentiation during organ development. In the Arabidopsis root epidermis, cell fate specification is affected by cell division. Radial symmetry depends on the occurrence of longitudinal divisions that generate a characteristic pattern of tricho- and atrichoblast files in the epidermis.¹ In addition, the expression of the homeobox GLABRA2 (GL2) gene contributes to fix the hair/non-hair epidermal cell fate. This requires, at least, three convergent processes. One is the processing of a positional information from the underlying cortical cell layer, where the SCRAMBLED (SCM) receptor-like kinase is of primary importance.² Another is the activity of a transcriptional complex containing TRANSPARENT TESTA GLABRA1 (TTG1), a WD40-repeat protein, the bHLH proteins GL3 and EGL3, and the Myb transcription factors WEREWOLF (WER) and CAPRICE (CPC).^3,4 Finally, a change in chromatin accessibility favors GL2 expression.⁵ However, whether these pathways act in a coordinated manner has not been defined.

Cell Division

The availability of the Arabidopsis, rice and poplar genomes^6–8 has been instrumental for the identification of most cell cycle genes required for cell cycle progression and regulation in higher plants. Progression of plant cells through the different phases of the cell cycle has revealed as an extremely complex and finely regulated process.⁹ Cell cycle regulators and effectors, in addition to their direct participation in the cell cycle machinery, appear to play functions directly impinging on plant development.^10,11

We had already found that proteins that control initiation of DNA replication, such as CDT1, also play a role in controlling the cell division potential of meristemoid cells that give rise to the stomata.¹² Now we have shown that this effect is not restricted to developing leaves since in root meristems elevated levels of CDT1 also increase the rate of longitudinal anticlinal divisions in the epidermis. Furthermore, we also found that CDT1 can activate GL2 expression and this information was the first hint of a possible way of coupling cell division to cell fate decision mechanisms.¹³

A Framework for Coordinating Cell Fate and Cell Division

To define the molecular basis for such coordination, a yeast two-hybrid screening using CDT1 as a bait, identified a novel protein, GEM (GL2-expression modulator). GEM inhibits the expression not only of GL2 but also of CPC, both responsible for cell fate decisions in the root epidermis. GEM has a role in determining the spatial pattern of GL2 expression, as revealed in plants expressing the GUS reporter gene under the control of the GL2 promoter in the different GEM backgrounds. Furthermore, GEM also acts as a repressor of cell division in the epidermal and cortical layers of the root, inhibiting the occurence of the longitudinal anticlinal divisions that are responsible for the increase in thickness of the root. However, where is GEM placed in the complex network of cellular factors that have been genetically identified in the pathway specifying cell fate in the root epidermis?

Chromatin immunoprecipitation (ChIP) experiments show that GEM is recruited to the GL2 and CPC promoters through its specific interaction with TTG1. We analyzed the genetic interactions of GEM with TTG1 and found that they interact genetically since the gem-1, ttg1-1 double mutant showed the ttg1-1 phenotype, indicating that GEM acts upstream TTG1. Furthermore, our data suggest that CDT1 and TTG1 compete in vivo for their binding to GEM. Therefore, we concluded that GEM is part of the complex that represses GL2 and CPC expression through TTG1. We also analyzed the genetic interactions of GEM with SCRAMBLED (SCM), which encodes a receptor-like kinase required to interpret positional signals during epidermal cell fate specification. The phenotype of scm-2, GEM^OE double mutant indicates that GEM and SCM act, at least in part, in different pathways, that finally converge in controlling the spatial control of GL2 expression.

Chromatin organization in the root epidermis responds to positional signals, as cell fate does, in such a way that accessibility of the GL2 locus is reset every cell cycle.⁵ Based on FISH analysis, it was speculated that a large chromosomal region contains a global “open” chromatin state in atrichoblasts.¹⁴ In fact, epigenetic marks characteristic of active or repressed euchromatic genes (H3K9me3 or H3K9me2, respectively) change in the GL2 and CPC promoters in a GEM-dependent manner.¹³ However, ChIP experiments scanning the GL2 and CPC loci indicate that the occurrrence of such modifications just upstream of the corresponding ORFs, instead of over a large chromosomal region, is sufficient to explain the implication of GL2 and CPC in cell fate specification.⁴ In addition, the expression of both genes in highly-synchronized Arabidopsis cells has also revealed that it is cell cycle-regulated, being low in G₂ and high in early G₁. Furthermore, the same GEM-dependent changes in the pattern of epigenetic modifications found are also cell cycle-regulated.¹³ This is reminiscent of the DNA replication licensing mechanism, which also operates in late mitosis and early G₁.¹⁵ The implications of such parallels await future experimental efforts which are under way.

The identification of the novel regulatory protein GEM provides a molecular framework for the coordination of cell division and cell fate decisions in the root epidermis. GEM, as a regulator of GL2 expression, also participates in trichome specification in the leaves, suggesting that GEM acts as a general regulator of epidermal cell fate decisions. However, the identification of this GEM function might just be one of many. The relevance of GEM in other processes is underscored by the fact that it interacts with TTG1, a protein that not only controls epidermal cell fate.¹⁶ Thus, GEM may reveal as a multifaceted regulator of a variety of developmental processes through its gene-specific targeting to finely regulate the epigenetic modifications required for proper gene expression in a highly temporally and spatially controlled manner.

Addendum to: Caro E, Castellano MM, Gutierrez C. A Chromatin Link That Couples Cell Division to Root Epidermis Patterning in Arabidopsis. Nature. 2007;447:213–217. doi: 10.1038/nature05763.