The maize leaf consists of four distinct tissues along its proximodistal axis: sheath, ligule, auricle and blade. liguleless1 (lg1)functions cell autonomously to specify ligule and auricle, and may propagate a signal that correctly positions the blade-sheath boundary. The dominant Wavy auricle in blade (Wab1) mutation disrupts both the mediolateral and proximodistal axes of the maize leaf. Wab1 leaf blades are narrow and ectopic auricle and sheath extend into the blade. The recessive lg1-R mutation exacerbates the Wab1 phenotype; in the double mutants, most of the proximal blade is deleted and sheath tissue extends along the residual blade. We show that lg1 is misexpressed in Wab1 leaves. Our results suggest that the Wab1 defect is partially compensated for by lg1 expression. A mosaic analysis of Wab1 was conducted in Lg1+ and lg1-R backgrounds to determine if Wab1 affects leaf development in a cell-autonomous manner. Normal tissue identity was restored in all wab1+/–sectors in a lg1-R mutant background, and in three quarters of sectors in a Lg1+ background. These results suggest that lg1can influence the autonomy of Wab1. In both genotypes, leaf-halves with wab1+/– sectors were significantly wider than non-sectored leaf-halves, suggesting that Wab1 acts cell-autonomously to affect lateral growth. The mosaic analysis, lg1 expression data and comparison of mutant leaf shapes reveal previously unreported functions of lg1 in both normal leaf development and in the dominant Wab1mutant.

The diversity of leaf shape in the plant kingdom reflects subtle differences in morphogenetic events that are similar for all plants. Leaves are initiated from groups of cells on the flanks of the shoot apical meristem(SAM) and acquire their characteristic form shortly after emergence. At maturity, leaves are asymmetric about one or more axes: proximodistal (base to tip), adaxial-abaxial (upper to lower surface) and mediolateral (midvein to margins). The asymmetric distribution of cell types, vasculature, air spaces,epidermal hairs and cuticular waxes allow each part of the leaf to perform a specialised function. Relatively little is known about the mechanisms that pattern each axis of the leaf.

Clonal analysis of leaf development indicates that the number of leaf founder cells ranges from about 30 in Arabidopsis to approximately 250 in tobacco and maize (Poethig,1984; Furner and Pumfrey,1992; Irish and Sussex,1992; Poethig and Szymkowiak,1995). In most dicots, the leaf founder cells occupy only a portion of the radial dimension of the SAM, whereas in maize, a complete ring of leaf founder cells surrounds the SAM(Steffensen, 1968). Most dicot leaves first appear as peg-like outgrowths that subsequently grow in the lateral dimension to form a flattened blade. Substantial evidence suggests that the juxtaposition of adaxial and abaxial cell types stimulates lamina outgrowth in dicot leaves (Waites and Hudson, 1995; McConnell and Barton, 1998; Bowman,2000; Kerstetter et al.,2001). In contrast, the maize leaf has a lamina from inception,and furthermore, a number of mutations affect or even delete specific lateral domains without affecting abaxial-adaxial asymmetry(Scanlon et al., 1996; Fowler and Freeling, 1996; Foster et al., 1999). These findings suggest that fundamentally different sequences of pattern formation occur in monocot and dicot leaves.

As the emerging leaf primordium grows away from the SAM, a new proximodistal axis of growth is established, and cells differentiate according to positional cues within the developing organ(McConnell et al., 2001; Matsumoto and Okada, 2003). In maize leaves, four distinct tissues develop along the proximodistal axis. The proximal sheath and distal blade are separated by a fringe of ligule tissue and two wedges of auricle tissue (Fig. 1E). The recessive liguleless1 (lg1) and lg2 mutations remove ligule and auricle, but do not affect the specification of blade and sheath (Emerson,1912; Brink, 1933). lg1 is expressed in leaf primordia in the zone of the developing ligule and encodes a SQUAMOSA PROMOTER-BINDING protein(Moreno et al., 1997). Mosaic analysis has shown that lg1 functions cell autonomously to specify ligule and auricle (Becraft et al.,1990). lg1 has also been implicated in the propagation of a signal to initiate ligule and auricle(Becraft and Freeling, 1991). lg2, which encodes a bZIP transcription factor, is expressed ubiquitously throughout the leaf primordia and functions non-cell autonomously as shown by mosaic analysis (Harper and Freeling, 1996; Walsh et al.,1998). Based on double mutant analysis, it has been suggested that lg1 and lg2 function in the same pathway(Harper and Freeling,1996).

Fig. 1.

Leaf and whole plant phenotypes 8 weeks after planting. (A) Wild-type, (B) Wab1, (C) lg1-R;Wab1 and (D) lg1-R plants. (E-H)Adaxial view of blade-sheath boundary of (E) wild-type, (F) Wab1, (G) lg1-R;Wab1, and (H) lg1-R leaves. a, auricle; b, blade; ae,auricle extension; ea, ectopic auricle; es, ectopic sheath; lg, ligule; s,sheath. Arrowhead in G indicate presumptive blade-sheath boundary in lg1-R;Wab1 leaf.

Fig. 1.

Leaf and whole plant phenotypes 8 weeks after planting. (A) Wild-type, (B) Wab1, (C) lg1-R;Wab1 and (D) lg1-R plants. (E-H)Adaxial view of blade-sheath boundary of (E) wild-type, (F) Wab1, (G) lg1-R;Wab1, and (H) lg1-R leaves. a, auricle; b, blade; ae,auricle extension; ea, ectopic auricle; es, ectopic sheath; lg, ligule; s,sheath. Arrowhead in G indicate presumptive blade-sheath boundary in lg1-R;Wab1 leaf.

Wab1 is a recently described dominant mutation that disrupts both mediolateral and proximodistal patterning of the maize leaf resulting in narrow leaves and inappropriate cell differentiation(Hay and Hake, 2004). We constructed double mutants between Wab1 and lg1-R and lg2-R to analyse the effect that loss of auricle tissue would have on the Wab1 phenotype. Double mutant leaves are very narrow and sheath tissue extends along the residual blade. Genetic mosaics of wab1+/– were created in Wab1 and lg1-R;Wab1mutants to determine if Wab1 disrupts leaf patterning in a cell-autonomous manner. The mosaic analysis indicates that Wab1generally acts autonomously but may be influenced by lg1. Our data reveal previously unreported roles for lg1 in leaf development and support a model in which signalling via lg1 is required to correctly position the blade-sheath boundary (Becraft and Freeling, 1991).

Genetic stocks and irradiation conditions

F3 families segregating 1:1 for lg1-R: lg1-R;Wab1/wab1+and 1:1 for lg2-R: lg2-R; Wab1/wab1+ were used for phenotypic analyses. lg1-R introgressed at least four times into the W23 genetic background was used for lg1-R leaf measurements. See supplemental datafor more details. For the mosaic analysis of Wab1, heterozygous Wab1-R plants were crossed to Maize Genetic Coop stocks heterozygous for white seedling3 (w3). For the mosaic analysis of lg1-R;Wab1, lg1-R/lg1-R;Wab1/+ plants were crossed to lg1-R/lg1-R; v4/v4, w3/+ stocks from the Maize Genetic Coop. In accordance with maize nomenclature, we refer to the wild-type alleles as Lg1+ and wab1+, the mutant alleles as lg1-R(reference allele) and Wab1 (Wab1-R), and the wild-type genes as lg1 and wab1. See Fig. S1for a schematic of the genetic material used in the mosaic analysis.

Seeds were allowed to imbibe for 48 hours at 25°C then irradiated with approximately 1,500 rads. The radiation was from a 6 MV photon (X-ray) beam generated by a linear accelerator at the Palmerston North Hospital Radiotherapy Unit, Palmerston North, New Zealand. Irradiated seeds were hand planted into soil at the Institute of Developmental Phenomenology, Raumai, New Zealand.

Sector analysis

Plants were grown to maturity and screened for albino(w3/–), and yellow (v4/–) sectors throughout development. Out of 1681 irradiated seeds, 93 w3 wab1+/–sectored leaves were identified in 42 Wab1/wab1+ plants and 51 sectored leaves were identified in 32 wab1+/wab1+ control plants. In the second experiment, 4,608 seeds were irradiated; 115 w3 wab1+/– sectors and 65 v4 wab1+/– sectors were identified in 81 lg1-R/lg1-R; Wab1/wab1+plants. In 46 lg1-R/lg1-R; wab1+/wab1+ control plants, 90 w3 wab1+/– and 50 v4 wab1+/– sectors were analysed. All sectored leaves were harvested at maturity and photographed and/or photocopied. Leaf number, sector width and the lateral position of the sector within the blade and sheath were recorded. See supplemental materialfor details of data analysis.

Transverse hand sections of freshly harvested sectored leaves were examined by epifluorescence microscopy using a Leica (MZFLIII) stereomicroscope equipped with a 395-440 nm excitation filter and a 470 nm observation filter. All sections were photographed using a DC200 digital camera (Wetzlar,Germany). Under these conditions, normal chloroplasts fluoresce bright red and cell walls appear blue-green. No chlorophyll autofluorescence is detected in w3/– cells. The presence or absence of chlorophyll in epidermal layers was scored by inspecting guard cells, the only chloroplast-containing cell-type in the epidermis. Sectors of v4/– appear yellow because of a delay in the accumulation of chlorophyll, but eventually become green. Boundaries of v4 sectors were marked with a pen.

Scanning electron and light microscopy

Mature leaf tissue was fixed in 3% glutaraldehyde and 2% paraformaldehyde in 0.1 M phosphate buffer. Prior to fixation, a small notch was made at the sector boundary (SEM samples only). Samples for SEM were dehydrated in acetone, critical-point dried in liquid CO2 and sputter coated with 25 nm gold using a Polaron E 5400 sputter coater (SCD-050; Bal-Tec, Balzers,Liechtenstein). Specimens were examined on a Cambridge 250 Mark III scanning electron microscope (Cambridge Instruments, Cambridge, UK) operated at 20 kV,and images were captured on 35 mm film. Samples for light microscopy were infiltrated and embedded in Procure 812 (ProSciTech, Kelso, Australia). Sections (1 μm) were cut, heat mounted, stained with 0.05 (w/v) Toluidine Blue and photographed with an Axioplan microscope equipped with an Axiophot camera (Zeiss, Jena, Germany).

lg1 and lg2 gene expression

Leaf primordia (P9-10) were removed from the shoot, the ligule region dissected and immediately placed in liquid nitrogen. Leaf primordia (P6-8)were removed, dissected 5 mm above their base for a pre-ligule tissue sample and a further 5 mm above this for a blade tissue sample and immediately placed in liquid nitrogen. The remaining five plastochrons and SAM were used as a SAM sample. Each RNA sample consisted of a pool of ten seedlings. Tissue dissection, RNA isolation and cDNA synthesis were each performed twice independently, in each case giving identical results. Gene-specific PCR primers were as described for lg1(Moreno et al., 1997) and lg2 (Walsh et al.,1998). 20 PCR cycles were performed and amplified products were detected by Southern hybridization with gene-specific probes.

lg1-R and lg2-R enhance the Wab1 mutant phenotype

Four distinct tissue types demarcate the proximodistal axis of a wild-type maize leaf, the proximal sheath and distal blade are separated by ligule and auricle tissues (Fig. 1A,E). The ligule is an epidermally derived fringe, and the auricles are thickened wedges of tissue that act as a hinge between blade and sheath(Sharman, 1941; Becraft et al., 1990). Each of these tissue types has characteristic epidermal features and histological organisation, which have been well characterised by scanning electron and light microscopy (Fig. 2A-G)(Sharman, 1942; Esau, 1977; Russell and Evert, 1985; Langdale et al., 1989; Sylvester et al., 1990).

Fig. 2.

Scanning electron and light micrographs illustrating epidermal and histological features of wild-type and mutant leaves. (A-C) SEM of adaxial surface of wild-type (A) blade, (B) auricle and (C) sheath. (D-G) Transverse section through wild-type (D) blade, (E) auricle, (F) internal sheath and (G)marginal sheath tissue. (H) Cartoon depicting regions where tissue was sampled. (I-K) Transverse sections through (I) ectopic auricle in Wab1 blade, (J) ectopic sheath in Wab1 blade and (K) ectopic sheath in lg1-R;Wab1 blade. (L-M) SEM of adaxial surface of (L) Wab1 ectopic auricle and (M) lg1-R;Wab1 ectopic sheath. All sections are oriented with the adaxial surface upwards. Arrows in A and D indicate multicellular base of macrohair. Normal bundle-sheath anatomy indicated by arrowhead in D, abnormal bundle-sheath anatomy indicated by arrowheads in I and J. Scale bars: 100 μm.

Fig. 2.

Scanning electron and light micrographs illustrating epidermal and histological features of wild-type and mutant leaves. (A-C) SEM of adaxial surface of wild-type (A) blade, (B) auricle and (C) sheath. (D-G) Transverse section through wild-type (D) blade, (E) auricle, (F) internal sheath and (G)marginal sheath tissue. (H) Cartoon depicting regions where tissue was sampled. (I-K) Transverse sections through (I) ectopic auricle in Wab1 blade, (J) ectopic sheath in Wab1 blade and (K) ectopic sheath in lg1-R;Wab1 blade. (L-M) SEM of adaxial surface of (L) Wab1 ectopic auricle and (M) lg1-R;Wab1 ectopic sheath. All sections are oriented with the adaxial surface upwards. Arrows in A and D indicate multicellular base of macrohair. Normal bundle-sheath anatomy indicated by arrowhead in D, abnormal bundle-sheath anatomy indicated by arrowheads in I and J. Scale bars: 100 μm.

The Wab1 mutation disrupts normal patterning of the leaf and results in patches of ectopic auricle, sheath and ligule in the leaf blade(Fig. 1F)(Hay and Hake, 2004). Often a localised increase in blade width occurs immediately distal to patches of ectopic auricle. In addition, the normally placed auricle is more extensive,spreading distally into the leaf blade. Long strips of thickened auricle tissue and the reduced lamina width give Wab1 plants an unusual appearance; narrow, rigid leaves extend from the main axis at a more obtuse angle than wild-type leaves (Fig. 1B). Examination of histological and epidermal features reveals that the Wab1 blade contains cells with auricle and sheath identity(Fig. 2I,J,L). In both ectopic sheath and auricle tissue, intermediate veins fuse into lateral veins, and normal bundle sheath anatomy is absent or incomplete (arrowheads in Fig. 2I,J).

We constructed double mutants between Wab1 and lg1-R and lg2-R to analyse the effect that loss of auricle tissue would have on the Wab1 phenotype. The recessive lg1 and lg2mutations remove ligule and auricle, giving the mutant leaves a more upright appearance (Fig. 1D)(Harper and Freeling, 1996). Despite the lack of ligule and auricle, lg1-R and lg2-Rleaves have a distinct boundary between blade and sheath(Fig. 1H). lg1-Raffects all but the uppermost leaves, whereas lg2-R only affects juvenile leaves (Harper and Freeling,1996). lg1-R;Wab1 double mutants exhibit a striking,narrow leaf phenotype (Fig. 1C,G). Both the normal and ectopic auricle tissue is absent in the double mutant, most of the proximal blade deleted and sheath-like tissue extends along the margins of the residual blade. lg2-R;Wab1 and lg1-R;lg2-R;Wab1 mutants are similarly affected (data not shown).

The ectopic tissue in lg1-R;Wab1 leaf blades has histological and epidermal features of sheath. The adaxial surface is hairless and cells are long with smooth cell wall junctions (Fig. 2M), but the abaxial surface is covered with hairs specific to abaxial sheath tissue (not shown). lg1-R;Wab1 sheath-like tissue is very thin in the transverse dimension (Fig. 2K) and the intervascular spacing and prominent transverse veins resemble those found in marginal sheath tissue(Fig. 2G). In distal positions and near the midrib, lg1-R;Wab1 leaves have normal blade tissue (not shown). Sheath tissue identity is not affected by the lg1, lg2 or Wab1 mutations.

lg1-R leaf shape

Although the lg1-R ligule defect has been well described by others, the altered shape of lg1-R leaves has not been reported. We found that lg1-R leaves are significantly narrower at the blade-sheath boundary than those of non-mutant (Lg1+/lg1-R)siblings (Table S1). The mean width of the ninth leaf counting down from the tassel was 76 mm for lg1-R plants, whereas, the mean width was 102 mm for non-mutant siblings. A similar trend was seen for the tenth and eleventh leaves down from the tassel. We also noted that while the overall lengths of lg1-R and Lg1+/lg1-R leaves are the same, lg1-R blades are shorter and lg1-R sheaths are longer than those of non-mutant siblings (Table S1). This finding indicates that the blade-sheath boundary is established in a more distal position in the lg1-R mutant.

We compared the width of w3-marked clonal sectors in wild-type(Lg1+/Lg1+) and lg1-R plants. Sectors were measured at the base of the blade. Only sectors in adult leaves were included in this analysis to minimise variation in leaf size (see Supplemental data). Sectors near the midrib had similar median widths in wild-type and lg1-R blades. However, sectors in lateral and marginal regions were significantly narrower in lg1-R mutants than in wild-type leaves(Table S2). These data indicate that the lg1-R lateral growth defect is localised to lateral and marginal domains of the blade.

lg1 is misexpressed early in development of Wab1leaves

The dramatic effect of lg mutations on the Wab1 phenotype suggests that the presence of lg1 and lg2 partially compensates for the defects in leaf width and tissue identity in Wab1mutants. We carried out RT-PCR to see if lg1 or lg2 were expressed differently in Wab1 mutants. Previous efforts to detect lg1 or lg2 by in situ hybridisation have not been successful(Moreno et al., 1997) (Walsh and Freeling, personal communication). lg1 transcript is absent from wild-type tissue samples containing the SAM and P1-5 primordia, but expression is detected in equivalent Wab1 tissue at 20 PCR cycles(Fig. 3). This early expression in Wab1 was confirmed using a dilution of cDNA in the PCR reaction. Transcript was never detected in equivalent wild-type tissue at 40 PCR cycles(data not shown). lg1 expression was detected in both wild-type and Wab1 leaf primordia in the preligule band region of P6-8 and the ligule ridge region of P9-10 leaves (Fig. 3). Transcript was absent from the blade region of wild-type P6-8 leaves at 20 PCR cycles. Expression of lg1 was reproducibly detected in the blade region of Wab1 P6-8 primordia at 20 PCR cycles and confirmed using a dilution of cDNA in the PCR reaction. This result indicates that the expression domain of lg1 extends distally in Wab1leaf primordia, consistent with the distal extension of auricle tissue in mature Wab1 leaves.

Fig. 3.

lg1 and lg2 expression in Wab1 leaves. RT-PCR gel blot analysis of lg1 and lg2 expression in 4-week-old seedlings of wild-type and Wab1/+ at the following stages of leaf development: SAM including P1-5 (S), P6-8 pre-ligule region (PL), P6-8 blade(B), P9-10 ligule region (L). Lanes 9-16 are as for lanes 1-8 using a four-fold dilution of cDNA. Control with no cDNA included in PCR (C). Control amplification of ubiquitin (ub) indicates equal amounts of cDNA present in each sample.

Fig. 3.

lg1 and lg2 expression in Wab1 leaves. RT-PCR gel blot analysis of lg1 and lg2 expression in 4-week-old seedlings of wild-type and Wab1/+ at the following stages of leaf development: SAM including P1-5 (S), P6-8 pre-ligule region (PL), P6-8 blade(B), P9-10 ligule region (L). Lanes 9-16 are as for lanes 1-8 using a four-fold dilution of cDNA. Control with no cDNA included in PCR (C). Control amplification of ubiquitin (ub) indicates equal amounts of cDNA present in each sample.

No significant difference in lg2 mRNA expression could be detected in Wab1 compared with wild-type plants in any of the tissue samples analysed (Fig. 3).

Mosaic analysis of Wab1

Sectors of tissue lacking the dominant Wab1 allele(wab1+/–) were created in both Wab1 and lg1-R;Wab1 mutants to determine if Wab1 disrupts leaf patterning in a cell-autonomous manner (Fig. S1). Stocks carrying Wab1 in repulsion to white seedling3(w3) were X-irradiated to induce random chromosome breaks. Radiation-induced breaks proximal to W3 resulted in albino,non-Wab1 (w3 wab1+) sectors in otherwise green, Wab1 or lg1-R;Wab1 plants. In lg1-R;Wab1 plants,chromosome breaks proximal to Virescent4 (V4) created yellow, lg1-R (v4 wab1+;lg1-R) sectors. The loss of W3 in normal plants, and either W3 or V4 in lg1-R plants, provided control sectors that were hemizygous for chromosome 2L.

To ensure that the chromosome arm carrying Wab1 was lost early in leaf development, only sectors that extended through both the sheath and blade were analysed. Given the variability of the Wab1 phenotype, only sectors adjacent to tissue displaying a mutant phenotype could be scored. Thus, of 273 total sectors, only 77 were scorable for tissue identity. Sectors adjacent to ectopic auricle and sheath tissue were analysed for phenotypic expression (mutant or wild type) and cell layer composition (green or albino)(Table 1). Because v4sectors eventually accumulate normal amounts of chlorophyll, it was difficult to determine the internal layer composition of yellow, v4 sectors in mature leaves. Thus, only w3 sectors were scored for albino versus green mesophyll and epidermal layers. We predicted that white or yellow sectors would have normal blade tissue if Wab1 functions cell autonomously, while the white or yellow sectors would have the same mutant phenotype as the adjacent green tissue if Wab1 acts in a non-autonomous manner.

Table 1.

Summary of wab1+/– sector phenotypes

Summary of wab1+/– sector phenotypes
Summary of wab1+/– sector phenotypes

Ectopic auricle and auricle extension phenotypes in Wab1mutants

Sectors were examined using SEM and hand-cut sections to determine the phenotype of wab1+/– tissue. In 77% (24/31) of scorable sectors, wab1+/– cells exhibited normal blade characteristics whereas adjacent Wab1 tissue displayed either ectopic auricle or extension of auricle phenotypes (Fig. 4A-E; Table 1). These results indicate that Wab1 generally acts in a cell-autonomous manner in the lateral dimension to condition ectopic auricle and auricle extension phenotypes.

Fig. 4.

Sector phenotypes in Wab1 mutants. (A-E) Abaxial view of Wab1 leaves with auricle extension and/or ectopic auricle phenotypes with albino sectors exhibiting normal blade characteristics. (F) Scanning electron micrograph of adaxial surface of boxed region in C, illustrating unexpanded, dovetailed auricle-like cells in Wab1/wab1+ ectopic auricle tissue (right of arrowhead), and normal blade epidermal characteristics such as macrohairs in the wab1+/– sector (left of arrowhead). (G) Fluorescence micrograph of a transverse section through a sector adjacent to auricle extension, illustrating a sharp boundary between green, Wab1/wab1+ auricle-like tissue (which fluoresces red), and albino, wab1+/– tissue (which appears blue-green). (H) SEM of adaxial surface of sector boundary near that shown in G, with hairy, fully expanded auricle-like cells in the Wab1/wab1+ tissue (right of arrowhead) and normal blade cells including prickle hairs and macrohairs in the wab1+/– tissue (left of arrowhead). (I) Fluorescence micrograph of a transverse section through the sector shown in J. Green, Wab1/wab1+ tissue is very thin with widely spaced veins, a hairless adaxial surface and abaxial hairs characteristic of marginal sheath tissue. Albino wab1+/– tissue has the histological organization of normal blade tissue. (J) Abaxial view of sector adjacent to ectopic sheath tissue, arrowhead marks sector boundary. (K) Cartoon depicting an albino non-mutant (w3 wab1+/–) sector in a green Wab1/wab1+leaf. Scale bar: 500 μm (F) and 100 μm (G-I).

Fig. 4.

Sector phenotypes in Wab1 mutants. (A-E) Abaxial view of Wab1 leaves with auricle extension and/or ectopic auricle phenotypes with albino sectors exhibiting normal blade characteristics. (F) Scanning electron micrograph of adaxial surface of boxed region in C, illustrating unexpanded, dovetailed auricle-like cells in Wab1/wab1+ ectopic auricle tissue (right of arrowhead), and normal blade epidermal characteristics such as macrohairs in the wab1+/– sector (left of arrowhead). (G) Fluorescence micrograph of a transverse section through a sector adjacent to auricle extension, illustrating a sharp boundary between green, Wab1/wab1+ auricle-like tissue (which fluoresces red), and albino, wab1+/– tissue (which appears blue-green). (H) SEM of adaxial surface of sector boundary near that shown in G, with hairy, fully expanded auricle-like cells in the Wab1/wab1+ tissue (right of arrowhead) and normal blade cells including prickle hairs and macrohairs in the wab1+/– tissue (left of arrowhead). (I) Fluorescence micrograph of a transverse section through the sector shown in J. Green, Wab1/wab1+ tissue is very thin with widely spaced veins, a hairless adaxial surface and abaxial hairs characteristic of marginal sheath tissue. Albino wab1+/– tissue has the histological organization of normal blade tissue. (J) Abaxial view of sector adjacent to ectopic sheath tissue, arrowhead marks sector boundary. (K) Cartoon depicting an albino non-mutant (w3 wab1+/–) sector in a green Wab1/wab1+leaf. Scale bar: 500 μm (F) and 100 μm (G-I).

Fig. 4F is a SEM of the adaxial epidermis of the boxed region shown in Fig. 4C. There is a clear transition from albino wab1+/– tissue, which has blade characteristics such as macrohairs (left of arrowhead), to green, hairless Wab1/wab1+ tissue with unexpanded dovetailed cells typical of immature auricle (right of arrowhead, Fig. 4F). Green, Wab1/wab1+ tissue fluoresces red under UV illumination while albino wab1+/– tissue appears blue-green. When sectored tissue is viewed in transverse section, abrupt changes in histology are apparent at the sector boundaries. For example, in Fig. 4G, the albino wab1+/– tissue has characteristics of blade tissue, whereas the adjacent Wab1/wab1+ tissue is thicker and auricle-like. The SEM in Fig. 4H shows the same sector boundary, with albino tissue to the left of the arrowhead and green tissue to the right. The green Wab1/wab1+ tissue has larger mesophyll cells and is densely covered by long hairs without multicellular bases, these cells are characteristics of mature auricle tissue. The albino wab1+/–tissue has prickle hairs, macrohairs and cell types typical of blade tissue.

Of the seven sectors that displayed auricle characteristics through all or part of the sector, six had one or more inner layers of green, Wab1cells (Table 1). These results indicate that Wab1 generally acts cell autonomously in the lateral dimension, but may act non-cell autonomously between cell layers.

An interesting pattern was observed in Wab1 plants with mild auricle extension phenotypes. In most cases, sectors in these plants had an auricle extension to the midrib side of the sector, but recovered normal tissue identity both within the sector and on the marginal side of the sector(e.g. Fig. 4A,B). In plants exhibiting more severe phenotypes such as ectopic auricle and sheath, sectors with normal tissue identity were flanked by mutant tissue on both sides. These results suggest that normal (wab1+) cells may have a directional effect on adjacent Wab1 cells. There was no obvious relationship between recovery of tissue identity and sector size.

Ectopic sheath in Wab1 and lg1-R;Wab1 mutants

In Wab1 mutants, 75% (15/20) of the sectors adjacent to ectopic sheath exhibited normal blade characteristics. These results also indicate that Wab1 generally disrupts tissue patterning in a cell-autonomous manner (Table 1). Fig. 4J shows an albino sector adjacent to a region of ectopic marginal sheath tissue, and Fig. 4I is a transverse section through this sector boundary. The green, Wab1/wab1+ mutant tissue has characteristics of marginal sheath tissue; it is thin, has widely spaced veins, the adaxial surface is hairless, and the abaxial surface has long hairs without multicellular bases (Fig. 4I). In contrast, the adjacent albino wab1+/–tissue exhibits histological organisation and epidermal features specific to normal blade tissue (Fig. 2D).

In lg1-R;Wab1 double mutants, all (26/26) scorable wab1+/– sectors exhibited normal blade characteristics,indicating that Wab1 acts completely autonomously in the absence of Lg1+ (Table 1). The widest sectors were located at the margin, and restored the leaf half to a more normal shape and width (Fig. 5A,B). Fig. 5Ashows a yellow wab1+ v4/– sector that occurred at the margin. The yellow blade tissue has almost doubled the width of the leaf base. The sector shown in Fig. 5C was sectioned and examined by SEM (Fig. 5E,D). In transverse section, there is a sharp boundary between albino wab1+/– blade tissue and green Wab1/wab1+tissue with veins appressed against the abaxial surface, typical of sheath(Fig. 5E). The SEM shows crenulated blade cells to the left of the sector boundary (arrowhead), and smooth-walled, elongated sheath-like cells to the right(Fig. 5D). Fig. 5F is a transverse section through another sector boundary, illustrating the abrupt transition between albino blade tissue and green tissue with long abaxial hairs and other characteristics typical of marginal sheath.

Fig. 5.

Sector phenotypes in lg1-R;Wab1 mutants. (A) Abaxial view of yellow v4, wab1+/– sector, and (B) adaxial and (C) abaxial view of white, w3 wab1+/– sectors adjacent to ectopic sheath tissue. Arrowheads in A and B mark sector borders. (E) Fluorescence micrograph of a transverse section through the inner sector boundary of leaf shown in C. Green lg1-R;Wab1 tissue has the histological organisation of sheath, and albino wab1+/– tissue that of the blade. (D) SEM of adaxial surface of sector boundary shown in C and D, illustrating the sharp boundary between epidermal cell types in lg1-R;Wab1/wab1+ tissue (right of arrowhead) and lg1-R;wab1+/– tissue (left of arrowhead). (F)Fluorescence micrograph of a transverse section through another sector adjacent to marginal sheath-like tissue. The adaxial surface of green sheath-like tissue in D and F is hairless, whereas the albino tissue has hairs specific to blade. Scale bar: 100 μm (D-F).

Fig. 5.

Sector phenotypes in lg1-R;Wab1 mutants. (A) Abaxial view of yellow v4, wab1+/– sector, and (B) adaxial and (C) abaxial view of white, w3 wab1+/– sectors adjacent to ectopic sheath tissue. Arrowheads in A and B mark sector borders. (E) Fluorescence micrograph of a transverse section through the inner sector boundary of leaf shown in C. Green lg1-R;Wab1 tissue has the histological organisation of sheath, and albino wab1+/– tissue that of the blade. (D) SEM of adaxial surface of sector boundary shown in C and D, illustrating the sharp boundary between epidermal cell types in lg1-R;Wab1/wab1+ tissue (right of arrowhead) and lg1-R;wab1+/– tissue (left of arrowhead). (F)Fluorescence micrograph of a transverse section through another sector adjacent to marginal sheath-like tissue. The adaxial surface of green sheath-like tissue in D and F is hairless, whereas the albino tissue has hairs specific to blade. Scale bar: 100 μm (D-F).

In summary, mixed layer sectors behaved differently in Wab1 and lg1-R;Wab1 plants. In Wab1 plants, some mixed layer sectors exhibited the Wab1 phenotype, whereas none of the mixed layer sectors in lg1-R;Wab1 plants exhibited the Wab1 phenotype. Thus, Wab1 may act non-autonomously between layers or laterally, but only in a Lg1+ background.

Effect of Wab1 sectors on leaf width

The leaf blades of Wab1 and especially lg1-R;Wab1 mutants are significantly narrower than those of non-mutant siblings. To investigate the effect of wab1+/– sectors on Wab1 and lg1-R;Wab1 leaf width, the width of sectored and non-sectored halves of each leaf were measured and compared (see Supplemental data). In both Wab1 and lg1-R;Wab1 plants, there is a small but significant increase in the median width of the sectored half of the blade relative to the non-sectored half (Table 2, Table S3). Measurements made at the sheath midpoint show no significant difference in width between leaf-halves, indicating that the Wab1 mutation specifically disrupts lateral growth of blade tissue(Table S3). No difference between the widths of sectored and non-sectored leaf-halves is found in wild-type and lg1 control plants.

Table 2.

Median differences in sectored and non-sectored leaf-half widths, and effect of sector position

Median differences in sectored and non-sectored leaf-half widths, and effect of sector position
Median differences in sectored and non-sectored leaf-half widths, and effect of sector position

Analysis of the data suggested a relationship between sector position and leaf width. In both Wab1 and lg1-R;Wab1 plants, sectors near the midrib were not associated with a significant difference in leaf-half width (Table 2A), whereas leaf-halves with sectors in lateral and marginal positions were significantly wider than non-sectored leaf halves.

Many of the widest sectors in lg1-R;Wab1 plants were yellow, v4 sectors. To test if there was a difference in behaviour between v4 and w3 sectors, the median difference in leaf-half widths was evaluated separately for yellow and white sectors in lg1-R;Wab1and lg1-R plants. Sectors near the midrib were not included in this analysis as we had previously determined that they have no significant effect on leaf width. Surprisingly, yellow sectors had a significantly greater effect on leaf width than the white sectors (Table 2B). No difference was detected between yellow and white sectors in lg1-R controls, indicating that this effect is not inherent to v4 sectors, but only occurs in a Wab1 background.

In the maize leaf, ligule and auricle form at the boundary between the blade and sheath. lg1 has been implicated in the propagation of a signal that correctly positions this boundary, and is necessary for the development of ligule and auricle tissues. The dominant Wab1 mutation affects mediolateral and proximodistal patterning, resulting in narrow leaves and inappropriate cell differentiation. The recessive lg1 mutation exacerbates the Wab1 phenotype. lg1 expression is activated precociously in Wab1 leaves and may counteract some effects of the Wab1 mutation. We conducted a mosaic analysis of Wab1 in Lg1+ and lg1-R backgrounds to determine if Wab1affects leaf development in a cell-autonomous manner. This analysis shows that Wab1 generally acts cell autonomously to disrupt proximodistal patterning. Examples of Wab1 non autonomy were only observed in a Lg1+ background, supporting a role for lg1 in signal propagation. The mosaic analysis, lg1 expression data and comparison of mutant leaf shapes reveal previously unreported functions of lg1in both normal leaf development and in the dominant Wab1 mutant.

Lg1+ influences cell autonomy of the Wab1phenotype

In the majority of Wab1 and in all lg1-R;Wab1 leaves,scorable wab1+/– sectors exhibited characteristics of normal blade tissue, whereas adjacent tissue differentiated inappropriately as sheath or auricle. The sharp boundaries between tissue types were coincident with sector boundaries, indicating that Wab1 generally acts cell autonomously in the lateral dimension to disrupt normal proximodistal patterning.

Twelve of the 51 scorable sectors showed some aspect of the Wab1mutant phenotype and thus are exceptions to the general rule of cell autonomy. Of these 12, two were completely albino and therefore had no layer with Wab1. Of the 10 remaining sectors, half carried Wab1 only in the epidermis, and half carried Wab1 in the epidermis and/or one mesophyll layer. Thus, Wab1 may act non-autonomously in the lateral dimension and/or between cell layers to influence cell identity in wab1+/– tissue.

In contrast, all sectors in lg1-R;Wab1 plants had normal cell types. Wab1 in either the epidermis or a single mesophyll layer did not condition the mutant phenotype (12 of 26 sectors). These results suggest that normal lg1 function is required for Wab1 to act non-autonomously.

We also observed a few cases of non autonomy in which the normal blade phenotype (wab1+) was seen on the margin side of the sector as well as within the sector (e.g. Fig. 4A,B). The extension of the normal phenotype from the sector into genetically Wab1 cells was only seen in mildly affected plants. This pattern could reflect the fact that the Wab1 phenotype is most severe in the lateral domain. Alternatively, it may be that once correct proximodistal patterning is established within wab1+/– tissue,this information can be propagated towards the margins into Wab1/wab1+ tissue, but only if Wab1 activity is low. Interestingly, non-autonomy was never documented in lg1-R;Wab1plants. Cells in lg1-R;wab+/– sectors always had blade identity and cells outside of these sectors always had sheath identity. Thus, lg1 may affect the autonomy of Wab1 in both lateral and transverse dimensions.

Mosaic analyses of lg1-R have indicated that lg1 is involved in signal propagation, while also acting cell autonomously to induce ligule and auricle (Becraft et al.,1990; Becraft and Freeling,1991). One of the key findings of this work was the observation that ligule and auricle reinitiated immediately within Lg1+/lg1-Rtissue on the margin side of lg1-R/– sectors, but was displaced proximally. The authors interpreted this as evidence that lg1 is involved in the propagation of a `make ligule and auricle' signal, and that this signal moves from the midrib towards the margins. Our observations support this hypothesis.

Our data suggest that Wab1 alters positional information in an autonomous manner, resulting in inappropriate cell differentiation. We speculate that lg1 is responsible for the non-autonomous effects of Wab1 that were observed in our mosaic analysis. According to this model, lg1 may occasionally transmit the signal to initiate ectopic auricle from Wab1 tissue into wab1+/– sectors(Fig. 6A). Similarly, lg1 may relay correct positioning of the auricle and ligule from wab1+/– sectors into adjacent Wab1 tissue. In the absence of lg1, there is no lateral signalling from Wab1tissue into wab1+/– sectors, or from sectors into Wab1tissue (Fig. 6B).

Fig. 6.

lg1 affects cell autonomy of the Wab1 phenotype. (A)Cartoon illustrating spread of the Wab1 phenotype (wavy lines) into a wab1+/– sector and spread of normal phenotype (no wavy lines)from a sector into Wab1/wab1+ tissue. We propose that lg1 may transmit both correct and incorrect positional signals towards the margins of Wab1 leaves (red arrows). (B) In the absence of lg1, the Wab1 phenotype is strictly cell-autonomous.

Fig. 6.

lg1 affects cell autonomy of the Wab1 phenotype. (A)Cartoon illustrating spread of the Wab1 phenotype (wavy lines) into a wab1+/– sector and spread of normal phenotype (no wavy lines)from a sector into Wab1/wab1+ tissue. We propose that lg1 may transmit both correct and incorrect positional signals towards the margins of Wab1 leaves (red arrows). (B) In the absence of lg1, the Wab1 phenotype is strictly cell-autonomous.

Loss of Wab1 is associated with an increase in leaf width

We found a significant difference between the widths of sectored (i.e. cells that have lost Wab1) and non-sectored leaf-halves in Wab1 and lg1-R;Wab1 plants, but not in wild-type or lg1-R control plants. No obvious differences in cell width were seen between similar cell types in sectored and non-sectored tissue (data not shown). Therefore, we infer that the increase in lateral growth associated with wab1+/– sectors was the result of increased longitudinal cell divisions. It is difficult to determine if increased cell division was confined to the sectored tissue or occurred throughout the sectored leaf-half. The fact that we recorded a significant difference between sectored and non-sectored leaf-half widths suggests that Wab1 restricts lateral growth with at least some degree of autonomy. If Wab1 were entirely non autonomous, then the presence of a wab1+/– sector would have no effect on leaf-half width. A mosaic analysis of the tangled(tan) mutation in maize found that the cell division defect is autonomous in both the lateral and transverse dimensions(Walker and Smith, 2002). Thus, there is precedence for strict autonomy of cell division patterns in chimeric tissues.

In both lg1-R;Wab1 and Wab1 leaves, sectors positioned near the midrib had no significant effect on leaf-half width, whereas sectors in the outer two thirds of the leaf-half were associated with significant differences between sectored and non-sectored leaf-half widths. This result could reflect the fact that sectors near the midrib tend to be very narrow,and hence have a minimal effect on lateral growth. Alternatively, it may reflect the fact that the Wab1 phenotype primarily affects regions outside the midrib domain (Hay and Hake,2004).

The role of lg1 in leaf morphogenesis

Our study provides evidence of previously unreported functions of lg1 in leaf morphogenesis. We found that lg1-R leaves have longer sheaths and shorter blades than non-mutant(Lg1+/lg1-R) siblings, thus the blade-sheath boundary is shifted distally (this is shown schematically in Fig. 7A,B). This suggests that lg1 is required for correct positioning of the blade-sheath boundary.

Fig. 7.

Model for lg1 function in leaf morphogenesis. (A) Cartoon of wild-type leaf illustrating distal blade (green) and proximal sheath (blue)separated by ligule and auricle (red). (B) In the absence of lg1,ligule and auricle are deleted, the blade-sheath boundary is shifted distally,and the base of the blade is narrower than in wild-type leaves. (C) Wab1 disrupts proximodistal patterning and restricts lateral growth of the blade. Misexpression of lg1 in Wab1 leaves results in ectopic ligule and auricle, and partially compensates for the narrow leaf phenotype at the base of the blade. The lateral signalling function of lg1 permits some recovery of proximodistal patterning at the margins of Wab1 leaves. (D) In the absence of lg1, Wab1 leaves never establish blade at the margins and are extremely narrow.

Fig. 7.

Model for lg1 function in leaf morphogenesis. (A) Cartoon of wild-type leaf illustrating distal blade (green) and proximal sheath (blue)separated by ligule and auricle (red). (B) In the absence of lg1,ligule and auricle are deleted, the blade-sheath boundary is shifted distally,and the base of the blade is narrower than in wild-type leaves. (C) Wab1 disrupts proximodistal patterning and restricts lateral growth of the blade. Misexpression of lg1 in Wab1 leaves results in ectopic ligule and auricle, and partially compensates for the narrow leaf phenotype at the base of the blade. The lateral signalling function of lg1 permits some recovery of proximodistal patterning at the margins of Wab1 leaves. (D) In the absence of lg1, Wab1 leaves never establish blade at the margins and are extremely narrow.

We also determined that lg1-R leaves are narrower at the base of the blade than non-mutant siblings (see Fig. 7A,B). A comparison of clonal sectors in lg1-R and wild-type control plants indicates that margin and lateral sectors were significantly narrower in lg1-Rmutants than in wild-type leaves. These results imply that lg1promotes lateral growth at the base of the blade. Sylvester and co-workers(Sylvester et al., 1990)showed that a localised increase in both longitudinal and transverse anticlinal divisions generates a band of small cells across the base of the leaf primordium. This preligule band is a necessary prerequisite for the formation of ligule and auricle. lg1-R mutants are specifically defective in longitudinal divisions in the preligule region and at the base of the blade. lg1 may promote lateral growth via a direct effect on the rate and orientation of cell divisions. Alternatively, the development of the auricle itself may drive lateral growth of the lower leaf blade, ensuring the coordinated expansion of the leaf.

We suggest that the misexpression of lg1 is responsible for the ectopic auricle and much of the lateral growth at the base of Wab1blades. This is reflected in the shape of Wab1 leaves, which are relatively wide at the base of the blade, but narrow in more distal areas(Fig. 7C). We speculate that the lateral signalling function of lg1 permits some recovery of proximodistal patterning at the margins of Wab1 leaves(Fig. 7C). In the absence of lg1, Wab1 leaves never establish blade in this region, and are extremely narrow (Fig. 7D). A cell lineage analysis in lg1-R, Wab1 and double mutant backgrounds,similar to that carried out for narrow sheath(Scanlon and Freeling, 1997),would help elucidate cell division patterns in these mutants.

Supplemental data available online

We thank David Hantz for exporting the seed to New Zealand, Ben Parkinson at the Palmerston North Hospital for irradiation of the seeds, Doug Hopcroft and Raymond Bennett for assistance with tissue fixation and plant photography,Alla Seleznyova for help with data analysis, past and present members of the Freeling lab for lg1-R material, ideas and information, Jane Langdale, Jennifer Fletcher, Joe Colasanti and our reviewers for helpful comments on the manuscript, and Bruce Veit for tractor driving and stimulating conversations. This work was supported by a Marsden grant from the Royal Society of New Zealand (T.F. and R.J.), a Fulbright scholarship (A.H.) and the United States Department of Agriculture-Agricultural Research Service(S.H.).

Becraft, P. and Freeling, M. (
1991
). Sectors of liguleless-1 tissue interrupt an inductive signal during maize leaf development.
Plant Cell
3
,
801
-808.
Becraft, P. W., Bongard Pierce, D. K., Sylvester, A. W.,Poethig, R. S. and Freeling, M. (
1990
). The liguleless-1 gene acts tissue specifically in maize leaf development.
Dev. Biol.
141
,
220
-232.
Bowman, J. L. (
2000
). The YABBY gene family and abaxial cell fate.
Curr. Opin. Plant Biol.
3
,
17
-22.
Brink, R. A. (
1933
). Heritable characters in maize XLVI – liguleless2.
J. Hered.
24
,
325
-326.
Emerson, R. A. (
1912
). The inheritance of the ligule and auricles of corn leaves.
Nebraska Agricultural Experimental Station Annual Report
25
,
81
-88.
Esau, K. (
1977
).
Anatomy of Seed Plants
. New York: John Wiley & Sons.
Foster, T., Yamaguchi, J., Wong, B. C., Veit, B. and Hake,S. (
1999
). Gnarley1 is a dominant mutation in the knox4 homeobox gene affecting cell shape and identity.
Plant Cell
11
,
1239
-1252.
Fowler, J. E. and Freeling, M. (
1996
). Genetic analysis of mutations that alter cell fates in maize leaves: dominant Liguleless mutations.
Dev. Genet.
18
,
198
-222.
Furner, I. J. and Pumfrey, J. E. (
1992
). Cell fate in the shoot apical meristem of Arabidopsis thaliana.
Development
115
,
755
-764.
Harper, L. and Freeling, M. (
1996
). Interactions of liguleless1 and liguleless2 function during ligule induction in maize.
Genetics
144
,
1871
-1882.
Hay, A. and Hake, S. (
2004
). The dominant mutant Wavy auricle in blade1 (Wab1) disrupts patterning in a lateral domain of the maize leaf.
Plant Phys.
(in press).
Irish, V. F. and Sussex, I. M. (
1992
). A Fate map of the Arabidopsis embryonic shoot apical meristem.
Development
115
,
745
-753.
Kerstetter, R., Bollman, K., Taylor, R. A., Bomblies, K, and Poethig, R. S. (
2001
). KANADI regulates organ polarity in Arabidopsis.
Nature
411
,
706
-709.
Langdale, J. A., Lane, B., Freeling, M. and Nelson, T.(
1989
). Cell lineage analysis of maize bundle sheath and mesophyll cells.
Dev. Biol.
133
,
128
-139.
Matsumoto, N. and Okada, K. (
2003
). A homeobox gene, PRESSED FLOWER, regulates lateral axis-dependent development of Arabidopsis flowers.
Genes Dev.
15
,
3355
-3364.
McConnell, J. R. and Barton, M. K. (
1998
). Leaf polarity and meristem formation in Arabidopsis.
Development
125
,
2935
-2942.
McConnell, J. R., Emery, J, Eshed, Y., Bao, N., Bowman, J. and Barton, K. M. (
2001
). Role of PHABULOSA and PHAVOLUTA in determining radial patterning in shoots.
Nature
411
,
709
-712.
Moreno, M. A., Harper, L. C., Krueger, R. W., Dellaporta, S. L. and Freeling, M. (
1997
). liguleless1 encodes a nuclear-localized protein required for induction of ligules and auricles during maize leaf organogenesis.
Genes Dev.
11
,
616
-628.
Poethig, R. S. (
1984
). Cellular parameters of leaf morphogenesis in maize and tobacco. In
Contemporary Problems in Plant Anatomy
(ed. R. A. White and W. C. Dickison). Orlando:Academic Press.
Poethig, R. S. and Szymkowiak, E. J. (
1995
). Clonal analysis of leaf development in maize.
Maydica
40
,
67
-76.
Russell, S. H. and Evert, R. F. (
1985
). Leaf vasculature in Zea Mays. L.
Planta
164
,
448
-458.
Scanlon, M. J. and Freeling, M. (
1997
). Clonal sectors reveal that a specific meristematic domain is not utilized in the maize mutant narrow sheath.
Dev. Biol.
182
,
52
-66.
Scanlon, M. J., Schneeberger, R. G. and Freeling, M.(
1996
). The maize mutant narrow sheath fails to establish leaf margin identity in a meristematic domain.
Development
122
,
1683
-1691.
Sharman, B. C. (
1941
). Development of the ligule in Zea Mays L.
Nature
147
,
641
.
Sharman, B. C. (
1942
). Developmental anatomy of the shoot of Zea mays L.
Ann. Bot.
6
,
245
-284.
Steffensen, D. M. (
1968
). A reconstruction of cell development in the shoot apex of maize.
Am. J. Bot.
55
,
354
-369.
Sylvester, A. W., Cande, W. Z. and Freeling, M.(
1990
). Division and differentiation during normal and liguleless-1 maize leaf development.
Development
110
,
985
-1000.
Waites, R. and Hudson, A. (
1995
). Phantastica: a gene required for dorsiventrality of leaves in Antirrhinum majus.
Development
121
,
2143
-2154.
Walker, K. L. and Smith, L. G. (
2002
). Investigation of the role of cell-cell interactions in division plane determination during maize leaf development through mosaic analysis of the tangled mutation.
Development
129
,
3219
-3226.
Walsh, J., Waters, C. A. and Freeling, M.(
1998
). The maize gene liguleless2 encodes a basic leucine zipper protein involved in the establishment of the leaf blade-sheath boundary.
Genes Dev.
12
,
208
-218.