Epidermal cells in the shoot meristematic regions of many plants undergo predominantly anticlinal divisions to give more epidermal cells. However, clonal analysis has led to periclinal divisions being observed in the epidermal lineage of organ primordia, and sometimes also in shoot apical meristems. The degree to which L1 cells divide periclinally appears to be dependent upon both species and developmental stage. For example, incursions between the tunica cell-layers of Arabidopsis and Antirrhinum meristems appear to be relatively rare although incursions in later organ development of many dicots are observed especially at organ margins. In maize it is clear that the L1 can divide periclinally during both meristem and organ development (Sharman 1942), and in later nodes of the maize plant, most tissues can be derived from cells situated in the L1 layer of the embryonic SAM.
In a seductive series of mathematical models Jens Wegner proposes that the propensity of the L1 layer of meristems to divide periclinally is a func tion of the tangential strain placed on the outer cell layers of the tissue in question. This in turn is dependent on both cell size and the radius of the meristem or organ margin in question (Wegner 2000a,b). Thus, a corpus containing many cells and forming a wide meristematic dome, would generate a relatively large tangential strain within the outermost cell layers. This would have two predicted consequences; firstly it would encourage the occurrence of anticlinal divisions at the expense of periclinal divisions in the outermost (L1) cell layers and secondly, it would cause the formation of more tunica layers (i.e. layers of cells undergoing predominantly anticlinal divisions). In support of this theory maize vegetative meristems, which have only one tunica layer showing periclinal division, have a radius of only a few cell lengths. In comparison, dicot meristems, such as those of Antirrhinum are broader with a radius of many more cell widths and in agreement with Wegner's model contain two anticlinally dividing and clonally distinct tunica layers.
Wagner's models are based on the assumption that all meristematic cells divide at the same rate and are the same size at division, and that there is no genetic control over cell division planes. The first of these assumptions is not met, as several studies have demonstrated that the cells at the centre/apex of the shoot meristem, even in angiosperms, divide more slowly than those on the flanks (Grandjean et al. 2004) (reviewed in Carraro et al. 2006; Fleming 2006), although this discrepancy appears small enough in angiosperms that it does not completely destroy the predictions of the model. In contrast, the apical meristems of many gymnosperms (for example Cupressus species) do not possess a discrete tunica layer. Instead they have a population of one to three enlarged and histologically distinct apical initial cells which undergo both anticlinal and periclinal divisions to give rise to both epidermal and underlying cells (Pillai 1963). However, there is a tendency for periclinal divisions in superficial cell-layers to be lost in some gymnosperms [for example Monkey Puzzle (Araucaria)] in favour of the production of a more uniform anticli-nally dividing tunica layer. In a study of the seasonal variation in structure of the shoot apices of Araucaria columnaris, Pillai (1964) notes that in the summer, when the meristem is relatively narrow, it contains only one tunica layer, whereas winter meristems, which are much broader, contain two distinct tunica layers, as predicted by Wegner's model (Pillai 1964). The second assumption underlying these models (that there is no genetic control over cell division planes) is controversial, and will be dealt with in more detail later.
Wegner also notes a correlation between the frequency of L1 incursion into underlying cell-layers at leaf borders and the form of these borders in transverse sections. In general, species or cultivars with "narrow" (in terms of cell radii) leaf margins are more likely to develop L1 derived mesophyll at leaf margin (Wegner 2003a,b). Although the assumptions underlying Wagner's models are open to discussion, the fact remains that L1 cells can and do undergo periclinal divisions and contribute to underlying tissues in some angiosperm species, an observation that complicates analysis of both how epidermal growth is controlled, and to what extent the epidermis is responsible for controlling the growth of underlying tissue.
The control of cell division planes of cells which are not at the organ margin is obviously also an important issue for plant growth and development. In particular maintenance of predominantly anticlinal division planes in the developing organ surface is predicted to be crucial in allowing laminar outgrowth due to the apparent role of epidermal cells in blade expansion. Mutants where epidermal cell specification is compromised (see later sections) tend to loose their ability to regulate epidermal cell division planes, suggesting that restriction of epidermal cell divisions to the anticlinal plane is an intrinsic consequence of the acquisition of epidermal identity. However, interpreting the phenotypes of such mutants in terms of growth regulation can be difficult. One mutant in which cell division planes appear to be uncoupled without noticeable loss of epidermal identity is the Extra Cell Layer1 (Xcll) mutant of maize, in which late oblique divisions in the developing epidermis lead to the production of two epidermal cell-layers. In support of the proposed role of the epidermis in lamina expansion, the blade width of Xcll mutant leaves is considerably reduced (Kessler et al. 2002). The division defect in this mutant is proposed to occur late enough in development that cells are irreversibly committed to an epidermal fate. However, the mutation is semi-dominant and the gene responsible has not yet been cloned making it difficult to draw mechanistic conclusions from this particular mutant.
Was this article helpful?