Induction of Leaf Primordia by the Cell Wall Protein Expansin

See allHide authors and affiliations

Science  30 May 1997:
Vol. 276, Issue 5317, pp. 1415-1418
DOI: 10.1126/science.276.5317.1415


Expansins are extracellular proteins that increase plant cell wall extensibility in vitro. Beads loaded with purified expansin induced bulging on the leaf-generating organ, the apical meristem, of tomato plants. Some of these bulges underwent morphogenesis to produce leaflike structures, resulting in a reversal of the direction of phyllotaxis. Thus, expansin can induce tissue expansion in vivo, and localized control of tissue expansion may be sufficient to induce leaf formation. These results suggest a role for biophysical forces in the regulation of plant development.

Leaves form by reiterative organogenesis from a specialized organ, the shoot apical meristem (1). Although spatial domains of transcription factor activity can dictate where and when a leaf is initiated, the mechanism by which this information is transduced into morphogenesis is unknown (2). One model predicts that the regulation of epidermal cell wall extensibility controls tissue expansion and thus the initial steps of primordium formation (3). Recently, a family of cell wall proteins, expansins, that modulate cell wall extension in vitro has been characterized, although the ability of these proteins to induce cell expansion in vivo was not demonstrated (4). We now show that the localized application of expansin to the apical meristem induces expansion in living tissue and that the resultant bulging is sufficient to induce primordium formation.

In the tomato plant, leaves are initiated in a spiral in which the youngest primordium is designated P1, the next oldest P2, and so on in a developmental gradient (Fig. 1, A and B). The position of the cells from which the next leaf primordia (I1 and I2) arise can be predicted. Biophysical analysis suggests that the outermost cell layers of the apical meristem are under tension, whereas the inner cell mass is subjected to compression (3), in which case a localized increase in cell wall extensibility in the outer cells of the meristem should result in expansion and accompanying bulging out of the tissue (3). To test this hypothesis, we placed beads loaded with purified expansin onto the I2 position (5) and indeed observed the formation of a bulge (I′2) at the position of the bead during the subsequent plastochron (Fig. 1, C and D). At the same time, the formation of a primordium at position I1 was suppressed.

Figure 1

Induction of primordia by expansin. (A) Tomato apex showing phyllotaxis. (B) Diagram of apex in (A). The meristem has generated primordia P6 to P1. The next leaves will arise at position I1 and then I2. (C) Apex of a plant 5 days after the placing of expansin-loaded beads (5) at I2. (D) Diagram of apex in (C). A bulge, I′2, on the apical meristem is present adjacent to an expansin-loaded bead. (E) Confocal laser-scanning microscopy section (6) through an incomplete bulge at the I2 position induced by expansin. Samples (A) and (C) were stained with safranin red to reveal cut surfaces, shown as hatched areas in (B) and (D). Scale bars: 50 μm (A and C) and 20 μm (E).

In a first series of experiments, 37 of the 122 apices analyzed showed some effect 5 days after treatment, with a broad spectrum of changes observed (Table 1). Confocal laser-scanning microscopy (6) indicated that several of the expansin-induced bulges were incomplete, with internal areas devoid of tissue surrounded by an intact surface layer (Fig. 1E). Most of the component cells of these bulges were not substantially enlarged, suggesting that expansin-induced tissue expansion was accompanied by cell division. In seven apices, intact primordia were observed at the I2 position where the expansin-loaded beads had been positioned.

Table 1

Specific induction of leaflike structures on the apical meristem by expansin. Beads loaded with various agents in 50 mM sodium acetate buffer (pH 4.8) were manipulated onto the I2 position of meristems (5). After 5 or 14 days, the apices were analyzed with either a binocular microscope or confocal laser-scanning microscopy (6).

View this table:

In a second series of experiments, in which the apices were analyzed 14 days after expansin treatment, 9 of the 70 apices examined generated leaflike organs at the I2 position (Fig.2A and Table 1). These organs were green, produced trichomes, and had intact internal tissue (Fig. 2B). However, cytological differentiation was aberrant, with no vascular differentiation apparent. Nevertheless, in situ hybridization (7) revealed the expression of an rbcS gene previously shown to be a positive marker for leaf differentiation and a negative marker for the apical meristem (8) (Fig. 2C). Some of the expansin-induced primordia were more elongated (Fig. 2D) and, in addition to lacking appropriate cellular differentiation (Fig. 2E), did not express the rbcS marker gene (Fig. 2F).

Figure 2

Induction of leaflike structures by expansin. (A) An intact leaflike organ 14 days after positioning of expansin-loaded beads on the I2 position. (B) Section (7) through (A). (C) In situ hybridization (7) of (B) with an antisense probe for the leaf marker gene rbcS. Signals (red dots) are apparent throughout the structure. (D) An elongated leaflike organ 14 days after positioning of expansin-loaded beads on the I2 position. (E) Section through (D). (F) In situ hybridization of part of (E) with an antisense probe for rbcS. No signal is apparent within the structure, although an intense signal is visible in the subepidermal stem cell layers. (G) Longitudinal section through a vegetative apex showing the meristem (M) and primordium (P). (H) In situ hybridization of (G) with an antisense probe for expansin (15). A relatively high signal is apparent in the emerging leaf primordium. Scale bars: 200 μm (A through F) and 25 μm (G and H).

Analysis of apices in which beads loaded with various substances (boiled expansin, buffer, bovine serum albumin, cellulase, and oligogalacturonic acid) were placed at the I2 position did not reveal any induction of primordia, indicating that the effects observed with the expansin-loaded beads were specific (Table 1).

The generation of a primordium at the I2 position after expansin treatment resulted in a reversal of the subsequent phyllotaxis of the plant. The formation of a structure (I′2) at position I2 suppressed the initiation of the primordium at position I1 (Fig. 3, A and B). Thus, the anticlockwise order of leaf initiation from P4 to P1 became reversed to clockwise for P1 to I′2 (Fig. 3, C and D). This reversal of phyllotaxis was maintained during subsequent generation of leaf primordia by the meristem (Fig. 3, E and F). Thus, the formation of leaf primordia around the meristem after the formation of an I′2 structure appeared to follow the classical rules by which the site of leaf primordium initiation is influenced by the adjacent preexisting leaf primordia (9). In this context, the I′2 structures functioned as leaves.

Figure 3

Reversal of the direction of phyllotaxis by expansin-induced primordia. (A) Distal view of an apex that has generated an organ at the I2 position 5 days after expansin treatment. (B) Diagram of (A). Phyllotaxis P4 to P1 is anticlockwise. Primordium I′2 is clockwise from P1. (C) Side view of apex in (A). (D) Diagram of (C). The insertion point of the I′2 primordium lies distal and clockwise to that of P1. (E) View along the stem of a plant that has generated several primordia subsequent to the production of the structure at I′2. (F) Diagram of (E). Phyllotaxis P2 to P1 is clockwise. Twenty-one days after the induction of the I′2 leaflike structure, phyllotaxis of the subsequent leaves (I′3 to I′5) is anticlockwise. In (A), (C), and (E), primordia have been cut to reveal the meristem and the tissue stained with safranin red to reveal cut surfaces [shown as hatched areas in (B), (D), and (F)]. Scale bars: 50 μm (A and C) and 500 μm (E).

We suggest that, in our experiments, expansin induces changes in the cell wall that lead to an altered physical stress pattern in the meristem, so that tissue bulging occurs and, as a result, cells gain “primordium” identity (2, 3). The observed frequency of incomplete leaf structures is consistent both with a requirement for endogenous factors in the correct completion of leaf development (10) and with a spatiotemporal specificity of expansin action. A role for endogenous expansin in leaf initiation is indicated by in situ hybridization analysis showing not only that expansin genes are expressed in the apical meristem, but also that transcript abundance is highest in cells forming a primordium bulge (Fig. 2, G and H).

Our data add to a growing body of evidence that the physical environment influences morphogenesis and differentiation (3,11), that signaling events (of either a chemical or physical nature) occur within the meristem (1, 3), and that cell-wall components are determinantal for development (12). Although, in a few instances, extracellular proteins have been identified that can affect plant morphogenesis, such as arabinogalactan proteins (13) and cell wall enzymes (14), the mechanism by which these effects are transduced is unclear. We have now described specific morphogenetic effects of a defined gene product that are consistent with both the expected biochemical action of the protein (4) and established biophysical theories of leaf initiation (3). Our data suggest that cell wall structure plays a regulatory role in plant morphogenesis.

  • * Present address: Institute of Plant Sciences, Eidgenössische Technische Hochschule–Zürich, CH-8092 Zürich, Switzerland.

  • To whom correspondence should be addressed. E-mail: kuhlemeier{at}


Stay Connected to Science

Navigate This Article