Leaves of higher plants are produced in a sequential manner through the differentiation of cells that are derived from the shoot apical meristem. Current evidence suggests that this transition from meristematic to leaf cell fate requires the down-regulation of knotted1-like homeobox (knox) gene expression. If knox gene expression is not repressed, overall leaf shape and cellular differentiation within the leaf are perturbed. In order to identify genes that are required for the aquisition of leaf cell fates, we have genetically screened for recessive mutations that confer phenotypes similar to dominant mutations (e.g. Knotted1 and Rough sheath1) that result in the ectopic expression of class I knox genes. Independently derived mutations at the rough sheath2 (rs2) locus condition a range of pleiotropic leaf, node and internode phenotypes that are sensitive to genetic background and environment. Phenotypes include dwarfism, leaf twisting, disorganized differentiation of the blade-sheath boundary, aberrant vascular patterning and the generation of semi-bladeless leaves. knox genes are initially repressed in rs2 mutants as leaf founder cells are recruited in the meristem. However, this repression is often incomplete and is not maintained as the leaf progresses through developement. Expression studies indicate that three knox genes are ectopically or over-expressed in developing primordia and in mature leaves. We therefore propose that the rs2 gene product acts to repress knox gene expression (either directly or indirectly) and that rs2 gene action is essential for the elaboration of normal leaf morphology.

A unifying feature of higher plant development is the sequential initiation of lateral organs from the shoot apex. Leaf development in particular marks a fundamental change in the behaviour of cells at the apex, however, factors that regulate the timing, position and size of leaf primordia initiated from the apex are largely unknown. Maize initiates leaves in a distichous pattern by the recruitment of cells on the flanks of the shoot apex into a ribbon-like primordium of cells that eventually encircles the apex in an overlapping ring. These cells are termed the ‘disk of insertion’ or founder cells and have been defined by histological, clonal, and molecular criteria (Sharman, 1942; Poethig, 1984; Smith et al., 1992; Scanlon and Freeling, 1997). The first histologically visible signs of leaf development occur when densely cytoplasmic founder cells on the periphery of the meristem begin to divide rapidly in both the L1 and L2 layers. Cell divisions then spread around the circumference of the apex, in both directions from the initiation point, until a young primordium is evident encircling the apex (Sharman, 1942; Sylvester et al., 1990). Fate mapping studies demonstrate that the founder cell population is composed of approximately 200 cells in a 2-3 cell high tier that is about 30 cells in circumference and occupies at least two cell layers (Poethig and Szymkowiak, 1995). In addition to forming the main parts of the leaf (blade, sheath, auricle and ligule), the founder cells also give rise to both the subtending node and internode of the culm (Sharman, 1942; Johri and Coe, 1983; Poethig and Szymkowiak, 1995).

Although meristems from different species vary widely in form, they share a number of common properties with respect to the leaf initiation process. For example, a redirection of cellular growth on the periphery of the shoot apex in the founder cell ribbon leads to the formation of a leaf buttress or a protrusion of cells outward from the apex (Sharman, 1942; Steeves and Sussex, 1989). Other features common to leaf initiation in many flowering plants are an increase in cell division indices and the down-regulation of knotted1-like homeobox (knox) gene expression (Sharman, 1942; Smith et al., 1992; Jackson et al., 1994; Lincoln et al., 1994; Ma et al., 1994; Kerstetter et al., 1995; Schneeberger et al., 1995; Long et al., 1996; Hareven et al., 1996). In maize, the knotted1 (kn1) homeobox gene is expressed in the central zone of the shoot apex but not in the peripheral zone where leaf initiation isoccurring (Smith et al., 1992; Jackson et al., 1994). Other knox genes in maize are also repressed in leaf primordia but each adopts a specific expression pattern within the developing phytomer. For example the rough sheath1 (rs1) gene is expressed in a donut shaped ring underneath leaf primordia (Schneeberger et al., 1995).

Studies of both dominant knox mutants and of transgenic lines overexpressing knox genes, indicate that leaf development in both monocots and dicots is very sensitive to the ectopic expression of knox genes (Smith et al., 1992; Sinha et al., 1993; Schneeberger et al., 1995; Lincoln et al., 1994; Muller et al., 1995; Hareven et al., 1996; Fowler et al., 1996; Muehlbauer et al., 1997; Williams-Carrier et al., 1997). In general, the ectopic expression of knox genes in leaf primordia causes a retardation of leaf development such that cells in distal regions adopt a more basal cell fate (Freeling, 1992; Sylvester et al., 1996; Muehlbauer et al., 1997). This transformation to a more basal cell fate can be extreme in dicots, resulting in the production of leaf-born shoots (Chuck et al., 1996). In tomato, leaf shape is also perturbed when knox genes are ectopically expressed in leaves. In these overexpressing lines, leaves are ‘super compound’ and essentially lack blade lamina (Hareven et al., 1996; Chen et al., 1997). These results strongly suggest that knox genes are important factors in determining the size and shape of leaves in higher plants.

Recent studies with other genes that are involved in leaf formation and meristem function also suggest that the down-regulation of homeobox genes is necessary to signal the switch between a relatively indeterminate non-leaf state and commitment to a leaf development program. For example, mutations at the duplicate factor narrow sheath (ns) loci cause the formation of exceptionally narrow leaves. The mutant phenotype is associated with the failure to down regulate knox gene expression in a small group of approximately 18 cells on the premargin side of the phytomer (Scanlon et al., 1996). Significantly, clonal analysis of leaf development in ns mutants demonstrates that these cells are not incorporated into the leaf primordium and thus do not contribute to blade and internode formation (Scanlon and Freeling, 1997).

All of the studies described above suggest that the downregulation of knox gene expression is essential for normal leaf initiation and development. In order to further examine the role of homeobox genes in leaf development and to identify factors involved in their regulation, we have genetically screened for recessive mutations that mimic the phenotype produced by the ectopic expression of knox genes in leaf primordia (Freeling and Hake, 1985; Becraft and Freeling, 1994; Fowler and Freeling, 1996). This report describes our analysis of maize rough sheath2 (rs2) mutants. Our results show that rs2 plays an important regulatory role in leaf initiation through its activity on knox gene expression.

Maize stocks and growth conditions

The rs2 reference (rs2-R) allele (isolated by M.I. Hajidov in 1937) was obtained from the Maize Cooperative Stock Center (Columbia, Missouri) and introgressed five times into B73, W22, W23 and Mo17 inbred lines. The rs2-twisted dwarf (rs2-twd) allele was isolated in a transposon mutagenesis experiment that used Spm as the mutagen. Complementation tests between rs2-R and rs2-twd were accomplished by intercrossing heterozygotes and screening the F1 progeny for mutant individuals. In all cases, one quarter of the progeny were mutant. Lines carrying an Sn allele that controls anthocyanin pigmentation of the sheath, auricle and ligule were a gift from Dr Maria Moreno (Yale University).

Plants for genetic crosses were grown to maturity in the field in Berkeley, CA or in New Haven, CT. For scanning electron microscopy (SEM), histological analyses and RNA analyses, plants were grown in the greenhouse with an average daytime temperature of 32°C and a diurnal cycle of 16 hours light (average 300 μmol/m2/second1) and 8 hours dark.

Microscopy

Samples for scanning electron microscopy (SEM) were either prepared using dental impression medium or critical point dried. Dental impression medium replicas were prepared without fixation as described by Sylvester et al. (1990). For critical point drying, tissue was dissected and immediately fixed for 48 hours in formalin:acetic acid:alcohol:Triton X-100 (3.7%:5%:50%:0%) and then dehydrated to 100% ethanol. Tissue was critical point dried in a Tousimis samdri-pvt-3B critical point dryer. Both replicas and critical point dried material were coated with 25 nm platinum using a Polaron E5400 high resolution sputter coater. Subsequent analysis was carried out on a ISI DS-130 SEM with a LaB6 filament operating at 10 kV.

For light microscopy, samples were either sectioned freehand without fixation or were fixed in formalin:acetic acid:ethanol (10%:5%:50%) and embedded in wax as described by Langdale (1994). Wax embedded samples were sectioned using a rotary microtome. Sections were examined and photographed using either a Zeiss Axiophot or a Leica DMRB microscope.

Immunohistochemistry and RNA in situ localization

Immunolocalization of KNOX proteins was carried out as described by Donlin et al. (1995) using anti-KNOX and anti-RS1 antibodies (Scanlon et al., 1996) and paraffin embedded tissue. RNA in situ hybridizations were performed as previously described by Langdale (1994) using 35S-labeled riboprobes and paraffin embedded tissue.

knox gene transcript analysis

RNA was isolated from immature leaves according to the method of Kloeckener-Gruissem (1992). cDNA was subsequently prepared using Superscript reverse transcriptase (Gibco/BRL) and was used as a template for Polymerase Chain Reaction (PCR) amplification of knox genes as described by Bauer et al. (1994). Gene-specific primers used for amplification were as follows:

rough sheath1 a – 5′gagaactacaagccatgcatagacgctac3′

b – 5′ ttctgaagatgacatggacccgaatggtc3′

knotted1 a – 5′agtagtagtgtgggtgcgagatg3′

b – 5′gagatcacccaacactttgg3′

liguleless3 a – 5′gtggaacacgcactaccgctg3′

b – 5′agtggtgtatgattcagggtcc3′.

Control reactions were carried out using primers homologous to a maize ubiquitin gene:

ubi 3 – 5′taagctgccgatgtgcctgcgtcg3′

ubi 4 – 5′ctgaaagacagaacataatgagcacaggc3′.

kn1 primers were a generous gift from T. Foster and S. Hake (USDA, Albany, CA).

rs2 mutant phenotype

rs2 mutant plants have reduced stature and exhibit aberrant development of leaves, branches, floral organs and the stem (Hajidov, 1937). In addition to the original rs2 reference (rs2-R) allele, we have isolated an Spm-induced allele [rs2-twisted dwarf (rs2-twd)] from transposon mutagenesis screens. Both alleles confer similar recessive phenotypes. rs2 mutant plants exhibit a striking dwarf phenotype due to reduced and abnormal growth of internodes that is apparent at the seedling stage and becomes very pronounced at maturity (Fig. 1A,B). Dwarfism as well as the leaf and floral phenotypes discussed below are sensitive to genetic background and environmental conditions. The rs2-R allele was introgressed into four different genetic backgrounds (B73, W22, W23 and Mo17) and phenotypic severity was determined in each case. In general, an increase in dwarfism is correlated with increasing severity of leaf and floral phenotypes. However, background-specific differences are apparent. The most notable of these are male sterility and the formation of bladeless leaves (Fig. 1C). rs2-R plants introgressed into the B73 inbred background display the rough sheath, twisting and dwarfing phenotypes in 100% of homozygous mutants. Up to 90% of these mutants will also show evidence of leaves with multiple midribs, narrow/bladeless leaves and fused nodes. 100% of mutant plants in the B73 background also show a reduction in the number of male and female florets formed plus the generation of multiple silks per ovule, however, pollen is fertile (Fig. 1K). In contrast, the rs2-R allele introgressed in the Mo17 background shows the rough sheath and dwarfing phenotype in 100% of mutant plants but only rarely exhibits narrow/bladeless leaves and fused nodes. 90% of rs2-R mutants in the Mo17 background are male and female fertile although the number of florets in both male and female inflorescences is reduced.

Fig. 1.

Phenotype of rs2 mutant plants. (A) rs2-R mutant (left) and wild-type seedlings. (B) Mature rs2-R mutant (left) and wild-type plants. (C) Severe rs2-R mutant exhibiting semi-bladeless leaves. (D) Upper node showing opposite and decussate paired leaves (white arrow). One of the paired leaves lacks a midrib (asterisk). The leaf one node below the paired leaves has an exceptionally large midrib (M). (E) Wild-type (left) and rs2 (right) culms showing the highly compressed and twisted nature of rs2 mutant internodes. (F) Wild-type plant carrying Sn allele showing purple pigmentation of sheath and auricle. (G) Sn;rs2-twd double mutant showing displaced and ill-defined blade sheath boundary. (H) rs2-R mutant leaf exhibiting triple midribs. (I) rs2-R leaf (left) with half a blade, large midrib and foreshortened sheath compared to wild-type (right). (J) Wild-type carpels each with one silk (pistil) denoted by an asterisk. (K) rs2 mutant carpels with multiple silks.

Fig. 1.

Phenotype of rs2 mutant plants. (A) rs2-R mutant (left) and wild-type seedlings. (B) Mature rs2-R mutant (left) and wild-type plants. (C) Severe rs2-R mutant exhibiting semi-bladeless leaves. (D) Upper node showing opposite and decussate paired leaves (white arrow). One of the paired leaves lacks a midrib (asterisk). The leaf one node below the paired leaves has an exceptionally large midrib (M). (E) Wild-type (left) and rs2 (right) culms showing the highly compressed and twisted nature of rs2 mutant internodes. (F) Wild-type plant carrying Sn allele showing purple pigmentation of sheath and auricle. (G) Sn;rs2-twd double mutant showing displaced and ill-defined blade sheath boundary. (H) rs2-R mutant leaf exhibiting triple midribs. (I) rs2-R leaf (left) with half a blade, large midrib and foreshortened sheath compared to wild-type (right). (J) Wild-type carpels each with one silk (pistil) denoted by an asterisk. (K) rs2 mutant carpels with multiple silks.

The most penetrant and pronounced leaf phenotype, regardless of the genetic background or environment, is a disruption of the blade-sheath boundary due to disorganized cell growth and acropetal ligule displacement. This can be most easily visualized in plants that are doubly mutant for rs2 and an Sn allele that pigments the sheath and auricle (Fig. 1F,G). In weakly expressing mutants, the resultant ‘rough sheath’ phenotype is localized to the blade-sheath boundary, however, in strongly expressing mutants the entire sheath can appear ‘rough’ (Fig. 1D,I). Although the sheath can be drastically reduced in length, the blade region generally attains normal length (Fig. 1I).

In addition to showing dwarf and rough sheath phenotypes, the most severe mutant plants also manifest a number of other leaf phenotypes. These include leaves with multiple or excessively large midribs, wider leaves with an increased number of veins compared to wild-type and leaves with reduced amounts of blade lamina (Fig. 1C,D,H,I). Leaves with reduced blade lamina range from those with half to three quarters of the normal blade surface area to those that are bladeless and appear to be composed solely of midrib-like tissue (Fig. 1C,I). Such leaves tend to be restricted to upper nodes on the plant and are seldom observed in the first five nodes. An examination of the site of insertion into the stem at the node reveals that reduced lamina leaves are only partially inserted around the circumference of the stem, suggesting that the reduction in blade area is related to the ontogeny of the leaf from the apex. Bladeless leaves often expose the internodes which are normally masked by the encircling sheath in wild-type plants (Fig. 1C). Often semi-bladeless leaves are the result of folding of the leaf margins in toward the point of leaf origin.

The node and internode associated with reduced lamina leaves are also abnormal and display a highly compressed and curved phenotype (Fig. 1E). Occasionally, two nodes will appear to be joined at one side suggesting a fusion of successive leaf nodes. These fused nodes can also take the appearance of opposite and decussate phyllotaxy (Fig. 1D). Leaves initiated at the same node on the same side of the apex are also occasionally observed. In this case, the two leaves show opposite orientations of adaxial and abaxial surfaces similar to the ectopic leaves observed in Lax midrib mutants and in knotted1 loss of function mutants (Schichnes et al., 1997; Kerstetter et al., 1997).

In addition to affecting vegetative development, the rs2 mutation also perturbs floral development. One of the most notable floral phenotypes in the mutant is the production of multiple pistils from the carpel resulting in mature ears that appear exceptionally ‘silky’ (Fig. 1J,K). Significantly, extra silk production is also a feature of the dominant mutant Gnarley1 (Gn1) in which the gn1 homeobox gene is ectopically expressed (T. Foster and S. Hake, personal communication). Staminate flower development in rs2 mutants is also perturbed in that glumes and lemmas display similar leaf folding phenotypes to those observed during vegetative leaf development (data not shown).

Cellular differentiation in rs2 mutant leaves

The phenotype of rs2-twd;Sn mutants indicated that the rough sheath nature of the mutant phenotype was due to a transformation of leaf blade tissue into sheath tissue. In order to determine the extent to which cell fate transformations occur, we examined cellular differentiation patterns in both L1- and L2-derived tissue of wild-type and mutant leaves. An SEM examination of wild-type leaf surfaces revealed distinct differences in epidermal cell shape in the blade (Fig. 2A), auricle (Fig. 2B) and sheath (Fig. 2C). In leaves of rs2-R plants, however, blade-sheath boundaries are ill-defined in that epidermal cells in the blade adopt a mixture of sheath- and auricle-like cell fates (Fig. 2D). The transformed regions frequently appear as sectors running up into the blade with the borders of these sectors often composed of long narrow cells reminiscent of ligule cells (Fig. 2D-F).

Fig. 2.

Cell fate alterations in rs2 leaves. SEM analysis of wild-type and rs2 epidermal cells. Leaves 6-8 were harvested from 6-week old plants. (A) Wild-type blade. (B) Wild-type auricle. (C) Wild-type sheath. (D) rs2 leaf showing a sector of auricle and sheath-like cells adjacent to blade-like cells (black arrow marks boundary of blade and auricle phenotypes). (E) rs2 leaf showing a ridge of ligule cells (marked by the white asterisk) juxtaposed to sheath and auricle cells. Note the cells above the ectopic ligule (L) show a blade cell fate with a sharp boundary (arrows) to the auricle like cells right and sheath like cells left. (F) rs2 leaf replica taken just proximal to that shown in E. Black asterisk, continuation of ligule ridge shown in E; black lines connecting panel E and F indicate continuation of ligule ridges in adjacent replicas; s, sheath; a, auricle; b, blade. The upper side of each panel corresponds to the distal end of the leaf with respect to the point of origin. Size bar, 200 μm.

Fig. 2.

Cell fate alterations in rs2 leaves. SEM analysis of wild-type and rs2 epidermal cells. Leaves 6-8 were harvested from 6-week old plants. (A) Wild-type blade. (B) Wild-type auricle. (C) Wild-type sheath. (D) rs2 leaf showing a sector of auricle and sheath-like cells adjacent to blade-like cells (black arrow marks boundary of blade and auricle phenotypes). (E) rs2 leaf showing a ridge of ligule cells (marked by the white asterisk) juxtaposed to sheath and auricle cells. Note the cells above the ectopic ligule (L) show a blade cell fate with a sharp boundary (arrows) to the auricle like cells right and sheath like cells left. (F) rs2 leaf replica taken just proximal to that shown in E. Black asterisk, continuation of ligule ridge shown in E; black lines connecting panel E and F indicate continuation of ligule ridges in adjacent replicas; s, sheath; a, auricle; b, blade. The upper side of each panel corresponds to the distal end of the leaf with respect to the point of origin. Size bar, 200 μm.

In addition to abnormalities in epidermal cell characteristics, rs2 mutants display abnormal development of subepidermal cells. Transverse sections of wild-type and mutant leaves show that rs2 leaves are thicker in the basal half due to an excess of mesophyll cells. Furthermore, increased numbers of mesophyll cells in the transverse dimension produce a vein spacing pattern similar to that seen in the sheath (i.e. more than four cells separate two veins). It has previously been shown that mesophyll cells in the wild-type leaf sheath adopt a C3 differentiated state in that they accumulate the photosynthetic enzyme ribulose bisphosphate carboxylase (RuBPCase) (Langdale et al., 1988). In contrast, mesophyll cells of the wild-type leaf blade accumulate C4-specific photosynthetic enzymes while RuBPCase accumulates specifically in bundle sheath cells (Fig. 3A). In situ localization of transcripts encoding the large subunit of RuBPCase (rbcL) demonstrate that in regions of rs2 mutant leaf blades where veins are spaced more than four cells apart, rbcL transcripts accumulate in mesophyll cells (Fig. 3B). Similar patterns of accumulation are seen in regions of the leaf where ectopic midribs form i.e. rbcL transcripts accumulate in mesophyll cells (data not shown). Thus, in regions of rs2 mutant leaf blades where vein spacing patterns are perturbed, mesophyll cells adopt a sheath fate.

Fig. 3.

In situ hybridization of rbcL to rs2-twd mutant leaves. Sections through a wild-type leaf blade (A) and a rs2-twd mutant leaf blade with aberrant vascular spacing (B). Second leaves of 3-week old plants were examined. In regions that display normal vein spacing patterns (A; to the left of the arrow in B), rbcL is localized in the bundle sheath cells. However, in regions of the leaf where veins are more than four cells apart (to the right of the arrow in B) rbcL accumulates in mesophyll cells. Thus, ‘patches’ of C3-like tissue can be seen in the leaf. Scale bar, 70 μm.

Fig. 3.

In situ hybridization of rbcL to rs2-twd mutant leaves. Sections through a wild-type leaf blade (A) and a rs2-twd mutant leaf blade with aberrant vascular spacing (B). Second leaves of 3-week old plants were examined. In regions that display normal vein spacing patterns (A; to the left of the arrow in B), rbcL is localized in the bundle sheath cells. However, in regions of the leaf where veins are more than four cells apart (to the right of the arrow in B) rbcL accumulates in mesophyll cells. Thus, ‘patches’ of C3-like tissue can be seen in the leaf. Scale bar, 70 μm.

Development of rs2 leaves

In order to understand the origin of the phenotypes described above, a histological and SEM analysis of early leaf development was undertaken using 2-week old wild-type and rs2 mutant plants. Wild-type maize leaf development is first characterized by the outgrowth of a leaf primordium on the periphery of the meristem flank and is referred to as the plastochron 1 stage (P1). This first point of outgrowth reflects the position of the future midvein and establishes the lateral axis of the leaf (Fig. 4A). Growth of the primordium out from the side of the meristem proceeds around the circumference of the shoot apex while the region corresponding to the future midvein lengthens rapidly to form a hood-like shape which covers the apex of the meristem at the P2 stage (Fig. 4A). The margins of the primordium meet and begin to overlap as rapid cell divisions throughout the primordium result in complete shrouding of the meristem and younger leaves (P3 stage) (Fig. 4B). At the P4-P5 stage, the leaf primordium expands in length and the margins continue to expand and wrap around each other forming a cone shape (Fig. 4C,D).

Fig. 4.

SEM anlysis of leaf development in rs2-R mutants. rs2-R mutant leaf development is disturbed by the P1/P2 stage and is associated with incomplete insertion around the circumference of the shoot. (A-D) Wild-type apices; (A) apex showing meristem (M) and plastochron 1 (P1) and P2 leaf primordia. (B) P3 stage leaf primordium. (C) P4 stage leaf. (D) P5 stage leaf. (E-H) rs2-R mutant apices; (E) P1/P2 apex. (F) P3 apex showing underlying P2. (G) P3/P4 stage apex showing exposed meristem (black arrow) and abnormal leaf growth where the P3 leaf is growing over the margin of the P4 leaf (white arrow). (H) P4/ P5 stage apex. (I-K) Three views of a rs2-R apex showing P3, P4 and P5 leaves; Asterisk indicates reference leaf. Note that the margin of this leaf is severely folded in over its adaxial side. White arrow indicates meristem. (L) Example of a bladeless leaf inserted narrowly at the node; A, axillary bud. Size bars, 50 μm.

Fig. 4.

SEM anlysis of leaf development in rs2-R mutants. rs2-R mutant leaf development is disturbed by the P1/P2 stage and is associated with incomplete insertion around the circumference of the shoot. (A-D) Wild-type apices; (A) apex showing meristem (M) and plastochron 1 (P1) and P2 leaf primordia. (B) P3 stage leaf primordium. (C) P4 stage leaf. (D) P5 stage leaf. (E-H) rs2-R mutant apices; (E) P1/P2 apex. (F) P3 apex showing underlying P2. (G) P3/P4 stage apex showing exposed meristem (black arrow) and abnormal leaf growth where the P3 leaf is growing over the margin of the P4 leaf (white arrow). (H) P4/ P5 stage apex. (I-K) Three views of a rs2-R apex showing P3, P4 and P5 leaves; Asterisk indicates reference leaf. Note that the margin of this leaf is severely folded in over its adaxial side. White arrow indicates meristem. (L) Example of a bladeless leaf inserted narrowly at the node; A, axillary bud. Size bars, 50 μm.

In rs2 mutants, defects in leaf development are apparent immediately after initiation. Leaf primordia often have multiple lateral axes (typically 3; i.e. 3 midveins) and do not always encircle the circumference of the apex (Fig. 4E). At more advanced stages of development younger leaves can be observed growing over the edges of older primordia (Fig. 4F,G). This phenomenon is due in part to folding of the leaf margins toward the midrib. However, the primary cause may be that leaves are initiated too close to one another resulting in competition for founder cells and growth space such that a younger leaf grows over the marginal side of the next oldest leaf. For example, the P3 leaf shown in Fig. 4G forms a distinct boundary (marked by white arrow) where it is growing over the margin of the P4 leaf. This crowded form of leaf development results in a restriction of leaf growth that becomes more pronounced with increasing leaf number (Fig. 4H,I,J,K). In some genetic backgrounds, leaf growth becomes so restricted that leaves resemble radially symmetric structures (Fig. 4L). However, transverse sections of fully developed leaves reveal that abaxial-adaxial symmetry is maintained in all cases (Fig. 5). rs2 mutants also frequently develop wide leaves compared to wild type. The increase in width is accompanied both by the development of extra veins and by extra space between lateral veins (Fig. 3B and J. A. L., unpublished observations). Increased intervein space and widened leaves are also a regular feature of dominant Kn1 mutants (Freeling and Hake, 1985).

Fig. 5.

Cellular arrangement in rs2 bladeless leaves showing that dorsoventrality is maintained. Transverse section of the edge of mature wild-type (A) and bladeless rs2-twd mutant (B) leaves stained with Fast Green FCF. Ad, adaxial surface; ab, abaxial surface; m, margin. Size bar, 100 μm.

Fig. 5.

Cellular arrangement in rs2 bladeless leaves showing that dorsoventrality is maintained. Transverse section of the edge of mature wild-type (A) and bladeless rs2-twd mutant (B) leaves stained with Fast Green FCF. Ad, adaxial surface; ab, abaxial surface; m, margin. Size bar, 100 μm.

Leaf initiation from axillary meristems is also irregular in rs2 mutants. Fig. 6 shows an SEM comparison of prophyll and husk leaf initiation from wild-type and rs2-R axillary meristems. Whereas husk leaf initiation from the wild-type axillary meristem is distichous (Fig. 6A), a clear example of altered primordium initiation can be seen in the mutant apex (Fig. 6B). Here, two primordia of similar developmental stage are positioned on the same side of the meristem in an unusual phyllotactic arrangement. In addition, evidence of leaf margin folding similar to that seen in the primary apical meristem is visible (Fig. 6B).

Fig. 6.

Axillary bud development in normal and rs2-R mutant plants. Plants were harvested at 3-weeks old and meristems in the axils of leaf 3 examined. (A) Wild-type axillary meristem (M) exhibiting prophyll primordia (P) and a first husk leaf primordium (H1). (B) rs2-R axillary shoot showing the same organs as in A plus two small equally staged primordia in an abnormal phyllotactic arrangement; b, air bubble. Size bars, 50 μm.

Fig. 6.

Axillary bud development in normal and rs2-R mutant plants. Plants were harvested at 3-weeks old and meristems in the axils of leaf 3 examined. (A) Wild-type axillary meristem (M) exhibiting prophyll primordia (P) and a first husk leaf primordium (H1). (B) rs2-R axillary shoot showing the same organs as in A plus two small equally staged primordia in an abnormal phyllotactic arrangement; b, air bubble. Size bars, 50 μm.

Ectopic accumulation of KNOX proteins in rs2 leaves

Since leaf development in maize is associated with a down-regulation of genes encoding KN1-like homeobox proteins, shoot apices of wild-type and mutant siblings were examined 2 weeks after germination using a KNOX antiserum that recognizes both KN1 and RS1 proteins and a RS1-specific antiserum (Scanlon et al., 1996). Fig. 7A and C show that in wild-type apices, KNOX proteins are present in most cells of the meristem. However, in the group of cells that constitute the incipient leaf primordium (P0), no protein is detected (Fig. 7A). Significantly, KNOX proteins are also absent from P0 cells in rs2 mutant meristems (Fig. 7B) suggesting that the rs2 gene may not play a role in the initial down-regulation of knox genes in the meristem.

Fig. 7.

KNOX proteins accumulate ectopically in rs2 mutants. Immunolocalization of KNOX proteins in wild-type (A,C) and rs2-R (B,D,G) shoot apices. (E) Immunolocalization of RS1 protein in rs2-R vegetative shoot apex. (F) Ectopic accumulation of RS1 proteins in a P7 leaf at the site of ligule (L) development (white arrow). A,B,C and F are longitudinal sections whereas C, D and G are transverse sections. G shows a rs2 apex with demarcation between P0 and P1 primordia. P0-P2 leaf primordia are indicated. Black arrows indicate ectopic accumulation of homeodomain proteins. White arrows in A,B,D and E indicate absence of KNOX proteins in P0 and P(-1) stage leaves. Size bar, 200 μm (A-E); 100 μm (F,G).

Fig. 7.

KNOX proteins accumulate ectopically in rs2 mutants. Immunolocalization of KNOX proteins in wild-type (A,C) and rs2-R (B,D,G) shoot apices. (E) Immunolocalization of RS1 protein in rs2-R vegetative shoot apex. (F) Ectopic accumulation of RS1 proteins in a P7 leaf at the site of ligule (L) development (white arrow). A,B,C and F are longitudinal sections whereas C, D and G are transverse sections. G shows a rs2 apex with demarcation between P0 and P1 primordia. P0-P2 leaf primordia are indicated. Black arrows indicate ectopic accumulation of homeodomain proteins. White arrows in A,B,D and E indicate absence of KNOX proteins in P0 and P(-1) stage leaves. Size bar, 200 μm (A-E); 100 μm (F,G).

As development proceeds in wild-type plants, KNOX proteins are down-regulated in cells as they become recruited into the primordium (Fig. 7C). In mutant meristems, which are shorter in height and more oval in morphology than wild-type, this recruitment process is less ordered. For example, KNOXcells may extend only partially around one side of the apex creating an uneven ring with respect to normal phyllotaxy (Fig. 7D,G). The most striking phenotype observed in rs2 mutant apices, however, is the ectopic accumulation of KNOX proteins in leaf primordia (Fig. 7B,D). As early as P1, KNOX proteins accumulate in the proximal regions of the primordium (Fig. 7B). Later in development, although a few leaves accumulate KNOX proteins throughout, ectopic accumulation is more frequently seen in ‘patches’ (Fig. 7D). This is observed both with the general KNOX antiserum and with the RS1-specific antibody (Fig. 7E). Particularly noticeable is the accumulation of RS1 protein in cells of the ligular region (Fig. 7F). These data are consistent with the idea that RS2 maintains wild-type cells of the leaf primordium in a KNOX-off state.

To investigate whether rs2 regulates the expression of one or multiple homeodomain-encoding genes, we examined the expression of kn1 and two other knox genes in wild-type and rs2-R immature sheath tissue. cDNA prepared from immature sheath tissue of leaf 7-8 (3 weeks after germination) was PCR amplified with primers specific for each of three knox genes. The predicted amplification products were 615 bp (rs1), 248 bp (liguleless3; lg3) and 342 bp (kn1). The results shown in Fig. 8 demonstrate that rs1 and lg3 transcripts are not present in wild-type leaves whereas kn1 transcripts are detected at very low levels. In contrast, all three transcripts are present in rs2 mutant leaves. Ectopic expression of rs1 is particularly obvious. Ectopic expression was also observed when cDNA was prepared from mature leaf sheath tissue (data not shown). These results suggest that the rs2 gene product acts as a general repressor of class I knox gene expression during leaf development.

Fig. 8.

rs2-R mutants ectopically express multiple knox genes. RT-PCR analysis of homeobox gene expression in wild-type (+) and rs2-R (rs2) immature leaves using gene-specific primers for ubiquitin (ub), rough sheath1 (rs1), liguleless3 (lg3) and knotted1 (kn1).Asterisk indicates expected size of PCR products. The smear in the lg3 lane represents non-specific amplification as determined by DNA gel blot analysis (data not shown).

Fig. 8.

rs2-R mutants ectopically express multiple knox genes. RT-PCR analysis of homeobox gene expression in wild-type (+) and rs2-R (rs2) immature leaves using gene-specific primers for ubiquitin (ub), rough sheath1 (rs1), liguleless3 (lg3) and knotted1 (kn1).Asterisk indicates expected size of PCR products. The smear in the lg3 lane represents non-specific amplification as determined by DNA gel blot analysis (data not shown).

DISCUSSION

Changes in regulatory gene expression patterns often mark a fundamental shift in the developmental potential of a cell. Although many studies have shown that leaf development is associated with down-regulation of knox homeobox genes, the mechanism by which this regulation is achieved is not understood. We have examined the role of the rough sheath2 gene in leaf development and find that it defines a new class of genes that function to either directly or indirectly regulate plant knotted1-like homeobox genes. rs2 mutants display a wide range of leaf phenotypes that resemble those caused by dominant mutations in several maize homeobox genes, such as Kn1 and Rs1 (Freeling and Hake 1985; Becraft and Freeling 1992). The most penetrant leaf phenotype is inappropriate cell fate acquisition whereby cells in the leaf blade adopt a sheath-like identity (Figs 1, 2, 3). Consistent with the phenotypic similarity to dominant homeobox mutants, rs2 mutants show ectopic accumulation of KNOX homeodomain proteins in leaf primordia (Fig. 7). Transcript analysis further indicates that multiple knox genes including kn1, rs1 and lg3 are ectopically expressed (Fig. 8). We suggest that the rs2 phenotype results from the elimination of a negative regulator, the rs2 gene product, that normally represses knox gene expression in developing leaf primordia. Consistent with this model is the finding that the maize rs2 gene, which has been cloned using the rs2-twd allele and Spm as a molecular tag, encodes a myb-like transcription factor (M. T., R. S., M. F. and J. A. L., unpublished observations).

In addition to the rough sheath phenotype, rs2 mutant plants also show abnormalities in leaf shape, leaf insertion points and internode development. Thus it is probable that the rs2 gene product also influences recruitment of cells into the leaf primordium. However, immunolocalization studies of even the most severe mutant plants indicate that KNOX proteins are down-regulated appropriately at P0 (Fig. 7). Thus, a mechanism that does not involve RS2 may be responsible for the initial down-regulation of KNOX proteins in the meristem. Due to the duplicated nature of the maize genome (Helentjaris et al., 1988), a second rs2-like gene may act to initially suppress KNOX accumulation at P0 while RS2 itself maintains cells in the leaf primordium in a KNOX-off state. However, transgenic plants expressing knox genes under the control of a constitutive promoter, also show appropriate repression of knox gene expression in PO (Chuck et al., 1996; Williams-Carrier et al., 1997). This would suggest that the initial down-regulation at P0 is facilitated by post-transcriptional mechanisms.

Two further aspects of the rs2 phenotype, bladeless leaves and aberrant vascular patterning, provide an insight into mechanisms that may be operating during mutant leaf initiation and development. Bladeless and semi-bladeless leaves fail to encircle the stem, thus exposing the internode. Immunolocalizations of KNOX proteins during leaf initiation in rs2 mutants suggest that this phenotype is correlated with a reduction in the number of cells recruited into the primordium at P1 (Fig. 7D,G). A similar observation in maize ns mutants led to the proposal that if founder cells fail to fully encircle the shoot apex then the leaf primodium lacks information for leaf margin development (Scanlon et al., 1996; Scanlon and Freeling, 1997). In support of this idea is the observation that, as seen in ns mutants, narrow leaves of rs2 mutants have blunt edges and lack a tapering margin (Fig. 5B). rs2;ns1;ns2 multiple mutant homozygotes appear additive in phenotype and mainly exhibit an increase in the narrowness of semi-bladeless leaves (R. S., unpublished observations). This suggests that NS and RS2 function in separate pathways and that NS function preceeds that of RS2. Failure to maintain down-regulation of knox genes in the marginal domain defined by the ns genes (Scanlon et al., 1996) could thus explain the elaboration of half leaf and narrow leaf phenotypes in rs2 mutants (Fig. 9).

Fig. 9.

Model for generating semi-bladeless and narrow leaves during leaf initiation in rs2 shoot apices. The lower half of the figure represents transverse sections through the meristem as a P0-P1 stage leaf is formed. The upper half of the figure represents mature leaf phenotypes resulting from the meristematic configurations depicted. (A) Wild-type. Black and white diamonds indicate region of knox gene expression. White and grey regions peripheral to this knox region denote leaf founder cells. The grey region denotes the marginal domain defined by ns genes. (B,C) Model depicting how loss of knox gene down-regulation in a portion of the meristem leads to semi-bladeless and bladeless leaf phenotypes.

Fig. 9.

Model for generating semi-bladeless and narrow leaves during leaf initiation in rs2 shoot apices. The lower half of the figure represents transverse sections through the meristem as a P0-P1 stage leaf is formed. The upper half of the figure represents mature leaf phenotypes resulting from the meristematic configurations depicted. (A) Wild-type. Black and white diamonds indicate region of knox gene expression. White and grey regions peripheral to this knox region denote leaf founder cells. The grey region denotes the marginal domain defined by ns genes. (B,C) Model depicting how loss of knox gene down-regulation in a portion of the meristem leads to semi-bladeless and bladeless leaf phenotypes.

rs2 mutants exhibit significant aberrations in vascular patterning which range from an increased number of intermediate veins to the presence of multiple midribs. Since a large body of data suggest that auxin acts as an inductive signal for the development of vascular elements in the shoot (Sachs, 1991; Shininger, 1979), it seems reasonable to suggest that auxin homeostasis is perturbed in rs2 mutants. Consistent with this idea is the fact that mutants that disrupt basipetal auxin transport, such as lop1 in Arabidopsis, also exhibit oversized midribs and bifurcated midveins (Carland and McHale, 1996). Significantly, preliminary studies indicate that there are polar auxin transport defects in rs2 mutants (M. T. and J. A. L., unpublished observations). Of interest with respect to these data are observations regarding maize homeobox gene expression and vascular differentiation. First, kn1 and rs1 homeobox gene expression is observed in vascular elements in the stem leading up to leaf traces (Smith et al., 1992; Jackson et al., 1994; Schneeberger et al., 1995). Second, ectopically expressed homeodomain proteins cause abnormalities in vein differentiation and may as a consequence disrupt the normal pattern of auxin transport (Becraft and Freeling, 1994; Sinha et al., 1993). Third, lateral vein formation in maize closely follows knox gene repression during leaf initiation (Smith et al., 1992; Schneeberger et al., 1995). Thus, the challenge now remains to determine the interelationship between developmental pathways that regulate knox gene expression and those that coordinate the action of plant growth regulators. However, one model for RS2 function that incorporates all of the above observations, is that the gene functions indirectly in early leaf development i.e. recruitment of cells into the primordium, and directly in late development i.e. repression of KNOX genes. This model would predict that rs2 transcripts accumulate primarily after P1 and that signals from the developing primordia (possibly auxin?) influence the recruitment of cells from the apex into younger primordia.

Since rs2 mutants exhibit floral phenotypes that are consistent with ectopic expression of knox genes in floral primordia (Jackson et al., 1994; Schneeberger et al. 1995; Kerstetter et al. 1997), it is likely that rs2 negatively regulates knox gene expression during both vegetative and reproductive growth. Furthermore, since extra silks are produced from rs2 mutant carpels (Fig. 1I,J,K) it is possible that rs2 also regulates expression of the floral homeotic genes. As such, our results support the idea that genes like rs2 are involved in specifying (or maintaining) a developmental state through the exclusion of genes which promote the elaboration of a different fate. A conceptually similar role has been proposed for the CURLY LEAF gene in Arabidopsis which acts to exclude AGAMOUS gene expression from leaves (Goodrich et al., 1997). It therefore becomes possible to view shoot development as the regulated expression of distinct developmental modules. Evolutionary variations in the temporal and spatial expression patterns of genes like rs2 could thus explain the multitude of distinct shoot morphologies that are apparent in nature.

The authors would like to thank Steve Ruzin, Hans Holtan and Denise Schichnes-Porter from the Department of Plant and Microbial Biology Center for Biological Imaging for assistance with light microscopy and histology. Thanks are also extended to Paula Sicurello and Doug Davis of the U.C. Berkeley Electron Microscope Lab for assistance with SEM operation and sample preparation. Thanks to Petra Bauer and Gary Muelhbauer for helpful discussions and comments on the manuscript, to Randall Tyers for help with the figures and to Sarah Hake for a very constructive review. Work in M. F.’s lab was funded by NIH fellowship GM14578 to RS and NIH grant GM42610 to M. F. Work in J. A. L.’s lab was funded by the BBSRC and The Gatsby Charitable Foundation. M. T. is the recipient of a University of Oxford Glasstone Research Fellowship.

Bauer
P.
,
Crespi
,
M. D.
,
Szecsi
,
J.
,
Allison
,
L. A.
,
Schultze
,
M.
,
Ratet
,
P.
,
Kondorosi
,
E.
and
Kondorosi
,
A.
(
1994
).
Alfalfa Enod12 genes are differentially regulated during nodule development by nod factors and Rhizobium invasion
.
Pl. Physiol
.
105
,
585
592
.
Becraft
,
P. W.
and
Freeling
,
M.
(
1994
).
Genetic analysis of rough sheath1 developmental mutants of maize
.
Genetics
136
,
295
311
.
Carland
,
F. M.
and
McHale
,
N. A.
(
1996
).
LOP1: A gene involved in auxin transport andvascular patterning in Arabidopsis
.
Development
122
,
1811
1819
.
Chen
,
J.-J.
,
Janssen
,
B.-J.
,
Williams
,
A.
and
Sinha
,
N.
(
1997
).
A gene fusion at a homeobox locus: alterations in leaf shape and implications for morphologocal evolution
.
The Plant Cell
9
,
1289
1304
.
Chuck
,
G.
,
Lincoln
,
C.
and
Hake
,
S.
(
1996
).
KNAT1 induces lobed leaves withectopic meristems when overexpressed in Arabidopsis
.
The Plant Cell
8
,
1277
1289
.
Donlin
,
M. J.
,
Lisch
,
D.
and
Freeling
,
M.
(
1995
).
Tissue-specific accumulation of MURB, a protein encoded by MuDR, the autonomous regulator of the mutator transposable element family
.
The Plant Cell
7
,
1989
2000
.
Freeling
,
M.
and
Hake
,
S.
(
1985
).
Developmental genetics of mutants that specify knotted leaves in maize
.
Genetics
111
,
617
634
.
Freeling
,
M.
(
1992
).
A conceptual framework for maize leaf development
.
Dev. Biol
.
153
,
44
58
.
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
.
Fowler
,
J. E.
,
Muehlbauer
,
G. J.
and
Freeling
,
M.
(
1996
).
Mosaic analysis of the liguleless3 mutant phenotype in maize by coordinate suppression of mutator-insertion alleles
.
Genetics
143
,
489
503
.
Goodrich
,
J.
,
Puangsomlee
,
P.
,
Martin
,
M.
,
Long
,
D.
,
Meyerowitz
,
E. M.
and
Coupland
,
G.
(
1997
).
A polycomb-group gene regulates homeotic gene expression in Arabidopsis
.
Nature
386
,
44
51
.
Hajidov
,
M. I.
(
1937
).
Genes of ‘Rough Sheath’ in maize: Rough sheath-1 and rough sheath-2
.
Bul. Appl. Bot. Gen. Plant Breed. Ser. II
7
,
247
258
.
Hareven
,
D.
,
Gutfinger
,
T.
,
Parnis
,
A.
,
Eshed
,
Y.
and
Lifschitz
,
E.
(
1996
).
The making of a compound leaf: Genetic manipulation of leaf architecture in tomato
.
Cell
84
,
735
744
.
Helentjaris
,
T.
,
Weber
,
D.
and
Wright
,
S.
(
1988
).
Identification of the genomic locations of duplicate nucleotide sequences in maize by analysis of restriction fragment length polymorphisms
.
Genetics
118
,
363
363
.
Jackson
,
D.
,
Veit
,
B.
and
Hake
,
S.
(
1994
).
Expression of the maize KNOTTED1 related homeobox genes in the shoot apical meristem predicts patterns of morphogenesis in the vegetative shoot
.
Development
120
,
405
413
.
Johri
,
M. M.
and
Coe
,
E. H.
(
1983
).
Clonal analysis of corn plant development I. The development of the tassel and the ear shoot
.
Dev. Biol
.
97
,
154
172
.
Kerstetter
,
R. A.
,
Vollbrecht
,
E.
,
Lowe
,
B.
,
Veit
,
B.
,
Yamaguchi
,
J.
and
Hake
,
S.
(
1995
).
Sequence analysis and expression patterns divide the maize knotted-1-like genes into two classes
.
The Plant Cell
6
,
1877
1887
.
Kerstetter
,
R. A.
,
Laudencia-Chingcuanco
,
D.
,
Smith
,
L. G.
and
Hake
,
S.
(
1997
).
Loss of function mutations in the maize homeobox gene knotted1 are defective in shoot meristem maintenance
.
Development
124
,
3045
3054
.
Kloeckener-Gruissem
,
B.
,
Vogel
,
J. M.
and
Freeling
,
M.
(
1992
).
The TATA box promoter region of maize Adh1 affects its organ-specific expression
.
EMBO J
.
11
,
157
66
.
Langdale
,
J. A.
(
1994
). In situ hybridization. In
The Maize Handbook
(ed.
M.
Freeling
and
V.
Walbot
), pp
165
-
179
.
New York
:
Springer Verlag
.
Langdale
,
J. A.
,
Zelitch
,
I.
,
Miller
,
E.
and
Nelson
,
T.
(
1988
).
Cell position and light influence C4 versus C3 patterns of photosynthetic gene expression in maize
.
EMBO J
.
7
,
3643
3651
.
Lincoln
,
C.
,
Long
,
J.
,
Yamaguchi
,
J.
,
Serikawa
,
K.
and
Hake
,
S.
(
1994
).
A knotted1-like homeobox gene in Arabidopsis is expressed in the vegetative meristem and dramatically alters leaf morphology when overexpressed in transgenic plants
.
The Plant Cell
6
,
1859
1876
.
Long
,
J. A.
,
Moan
,
E. I.
,
Medford
,
J. I.
and
Barton
,
K. A.
(
1996
).
A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis
.
Nature
379
,
66
69
.
Ma
,
H.
,
McMullen
,
M. D.
and
Finer
,
J. J.
(
1994
).
Identification of a homeobox containing gene with enhanced expression during soybean (Glycine max. L.) somatic embryo development
.
Plant Mol. Biol
.
24
,
465
473
.
Muehlbauer
,
G.
,
Fowler
,
J. E.
and
Freeling
,
M.
(
1997
).
Sectors expressing the homeobox gene liguleless 3 implicate a time dependent mechanism for cell fate aquisition along the proximal-distal axis of the maize leaf
.
Development
124
,
5097
5106
.
Muller
,
K. J.
,
Romano
,
N.
,
Gerstner
,
O.
,
Garcia-Maroto
,
F.
,
Pozzi
,
C.
,
Salamini
,
F.
and
Rohde
,
W.
(
1995
).
The barley Hooded mutation is caused by a duplication in a homeobox gene intron
.
Nature
374
,
727
730
.
Poethig
,
R. S.
(
1984
). Cellular parameters of leaf morphogenesis in maize and tobacco. In
Contemporary Problems of Plant Anatomy
(ed.
R. A.
White
and
W. C.
Dickinson
), pp.
235
259
.
New York
:
Academic Press
.
Poethig
,
R. S.
and
Szymkowiak
,
E. J.
(
1995
).
Clonal analysis of leaf development in maize
.
Maydica
40
,
67
76
.
Sachs
,
T.
(
1991
).
Cell polarity and tissue patterning in plants
.
Development
Supplement
1
,
83
93
.
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
.
Schichnes
,
D.
,
Schneeberger
,
R.
and
Freeling
,
M.
(
1997
).
Induction of leaves directly from leaves in the maize mutant Lax midrib1-0
.
Dev. Biol
.
186
,
36
45
.
Schneeberger
,
R. G.
,
Becraft
,
P. W.
,
Hake
,
S.
and
Freeling
,
M.
(
1995
).
Ectopic expression of the knox homeobox gene rough sheath1 alters cell fate in the maize leaf
.
Genes. Dev
.
9
,
2292
2304
.
Sharman
,
B. C.
(
1942
).
Developmental anatomy of the shoot of Zea mays L
.
Ann. Bot
.
6
,
245
284
.
Shinninger
,
T. L.
(
1979
).
The control of vascular development
.
Annu. Rev. Plant Physiol
.
30
,
313
337
.
Sinha
,
N. R.
,
Williams
,
R. E.
and
Hake
,
S.
(
1993
).
Overexpression of the maize homeobox gene, KNOTTED-1, causes a switch from determinate to indeterminate cellfates
.
Genes Dev
.
7
,
787
795
.
Smith
,
L.
,
Greene
,
B.
,
Veit
,
B.
and
Hake
,
S.
(
1992
).
A dominant mutation in the maize homeobox gene, Knotted-1, causes its ectopic expression in leaf cells with altered fates
.
Development
116
,
21
30
.
Steeves
,
T. A.
and
Sussex
,
I. M.
(
1989
).
Patterns in Plant Development
, pp.
1
5
.
Cambridge
:
University Press
.
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
.
Sylvester
,
A. W.
,
Smith
,
L. G.
and
Freeling
,
M.
(
1996
).
Acquisition of identity in the developing leaf
.
Annu. Rev. Cell Dev. Biol
.
12
,
257
304
.
Williams-Carrier
,
R. E.
,
Lie
,
Y. S.
,
Hake
,
S.
,
Lemaux
,
P. G.
(
1997
).
Ectopic expression of the maize kn1 gene phenocopies the Hooded mutant of barley
.
Development
124
,
3737
3741
.