Genetic and molecular analyses of shoot development in model organisms, maize, and Arabidopsis have shown that the KNOTTED-like homoebox Class I (KNOX I) of genes plays an important role in maintaining indeterminacy at the shoot apical meristem (SAM) and may also be involved in the production of determinate leaf primordia. In order to investigate whether this class of genes is also involved in shoot development in plants with divergent morphology, we analyzed the developmental morphology and gene expression patterns in Welwitschia mirabilis, a gymnosperm with an interesting phylogenetic placement and a peculiar growth habit. It produces only two photosynthetic leaves over its entire life span, and these leaves continue to grow from a basal meristem throughout the life of the plant. Since KNOX I genes have been hypothesized as required for maintenance of indeterminate shoot apices, and their absence as necessary for determinate leaf production, we were able to test robustness of the specificity of KNOX I gene expression in these two organ types with altered fates in W. mirabilis. Our results show that W. mirabilis produces three pairs of leaflike organs with very distinct terminal fates, that the SAM is present and active even after these leaf pairs are initiated, and that the KNOX I class of genes are a good marker for meristem identity and maintenance. KNOX I gene expression is absent from leaf primordia in this species, indicating that these expression patterns are ancient and functionally important. However, as the leaf takes on an indeterminate growth pattern, the KNOX I genes show expression in the leaf base region where cell proliferation is seen to occur.

Recent work on species with simple leaves suggests that the juxtaposition of abaxial (lower) and adaxial (upper) cell fates (dorsiventrality) in leaf primordia is necessary for lamina outgrowth. However, how leaf dorsiventral symmetry affects leaflet formation in species with compound leaves is largely unknown. In four non-allelic dorsiventrality-defective mutants in tomato, wiry, wiry3, wiry4 and wiry6, partial or complete loss of ab-adaxiality was observed in leaves as well as in lateral organs in the flower, and the number of leaflets in leaves was reduced significantly. Morphological analyses and expression patterns of molecular markers for ab-adaxiality [LePHANTASTICA (LePHAN) and LeYABBY B (LeYAB B)] indicated that ab-adaxial cell fates were altered in mutant leaves. Reduction in expression of both LeT6 (a tomato KNOX gene) and LePHAN during post-primordial leaf development was correlated with a reduction in leaflet formation in the wiry mutants. LePHAN expression in LeT6 overexpression mutants suggests that LeT6 is a negative regulator of LePHAN. KNOX expression is known to be correlated with leaflet formation and we show that LeT6 requires LePHAN activity to form leaflets. These phenotypes and gene expression patterns suggest that the abaxial and adaxial domains of leaf primordia are important for leaflet primordia formation, and thus also important for compound leaf development. Furthermore, the regulatory relationship between LePHAN and KNOX genes is different from that proposed for simple-leafed species. We propose that this change in the regulatory relationship between KNOX genes and LePHAN plays a role in compound leaf development and is an important feature that distinguishes simple leaves from compound leaves.

KNOTTEDI-like homeobox (KNOXI) genes regulate development of the leaf from the shoot apical meristem (SAM) and may regulate leaf form. We examined KNOXI expression in SAMs of various vascular plants and found that KNOXI expression correlated with complex leaf primordia. However, complex primordia may mature into simple leaves. Therefore, not all simple leaves develop similarly, and final leaf morphology may not be an adequate predictor of homology.

We examined leaf morphology, anatomy, and margin development in wild-type and mutant leaves in order to define the genetic pathways controlling the different stages of leaf development in tomato. Tomato mutants were placed into four categories based on the degree of leaf complexity and margin development. Mutant phenotypes ranged from little or no blade expansion (Type I mutant wiry [w]), reduced leaf complexity (Type II mutants Lanceolate [La] and entire [e]), and reduced margin complexity (Type III mutants solanifolia [sf], trifoliate2 [tf2], and potato leaf [c]) to excessive leaf complexity (Type IV mutants Mouse ears [Me], Curl [Cu], clausa [clau], and complicata [com]). Based on our analyses, we conclude that the central domain of the terminal leaflet is established first, followed by blade expansion and further partitioning of the blade into a proximal domain that will promote the initiation of lobes and lateral leaflets. The mutant w causes a defect in blade expansion, La inhibits the establishment of the proximal domain, e restricts partitioning of leaf into leaflets, and Me, Cu, clau, and com cause excessive leaf complexity by promoting lateral leaflet initiation. The mutants with defective margin development (sf, tf2, c) also had decreased leaf lobing and complexity, indicating that the margin domain plays a role in the initiation of lobes and lateral leaflets.