table of contents
The authors reviewed the history, milestones and perspectives in plant developmental biology...
Biology Articles » Developmental Biology » Plant Development » Historical perspectives on plant developmental biology » Milestones in plant developmental biology
Milestones in plant developmental biology
- Historical perspectives on plant developmental biology
Milestones in plant developmental biology
Milestones during the last 25 years were the cloning of genes corresponding to mutations affecting key steps in developmental processes (forward genetics), their molecular analysis and the study of their genetic interactions in order to build genetic models for a given process (reviewed for Arabidopsis by Somerville and Koornneef, 2002; Van Lijsebettens et al., 2002). This moleculargenetic approach is now replaced by a large-scale functional genomics approach in which the function of all members of a gene family is analyzed by reverse genetics. In the following sections, we will exemplify a number of specific cases, with milestones listed in Table 1.
Embryogenesis in plants starts with the asymmetric division of the zygote resulting in an upper cell that will develop into an embryo and in a lower cell that generates the suspensor, which is the connection to the maternal tissue. For ease of conceptualization approximately 20 stages have been distinguished in embryo formation of which the early ones are important for axis formation and patterning and the later ones for growth and maturation. Mature plant embryos have a very simple body plan, in which the apical-basal and radial axes are specified. At the end of the apicalbasal axis the root and shoot apical meristems (SAM) are situated that become active upon germination and generate the primary root and shoot, respectively. Along the radial axis, patterning in progenitor tissue layers occurs during early embryogenesis. One of the key questions has been the identification of regulators that control the switches from globular to heart stage or from heart to torpedo stage. However, such master switches have not been detected despite extensive mutagenization programs for embryolethal (emb) mutants in Arabidopsis (Meinke and Sussex, 1979a, 1979b; Franzmann et al., 1995; McElver et al., 2001) and in maize and extensive studies of the Daucus carota embryogenic cell suspension (Giuliano et al., 1984; De Jong et al., 1992). Currently these structures are assumed to arise progressively. Approximately 750 EMB loci have been described to date that are essential for embryogenesis (Franzmann et al., 1995; McElver et al., 2001). Some of these genes are important in the communication between suspensor and embryo, in the control of cell number in the embryos and in the control of embryo maturation. A number of them will reveal essential enzymes for primary metabolism and numerous loci correspond to unknown proteins (Berg et al., 2004). A comprehensive database has been developed containing information on genes that give a seed phenotype upon mutation in Arabidopsis (Tzafrir et al., 2003). The conclusion is that the development of the zygote into the embryo is a progressive process in which the action of many genes together is required.
Key regulatory genes that control axis formation have been identified: phenotypes predicted from defects along the apicalbasal or the radial axis were obtained upon mutagenization and screening for seedling lethals (Mayer et al., 1991); their corresponding genes are involved in cytokinesis and auxin transport (Shevell et al., 1994; Lukowitz et al., 1996; Hardtke and Berleth, 1998; Assaad et al., 2001). These studies confirmed the model for embryo formation obtained through cell biology and clonal analyses describing the different domains and boundaries within a developing embryo and the embryonal origin of the different parts of the germinating seedling (Dolan et al., 1993; Scheres et al., 1994).
The self-regulatory shoot apical meristem
The SAM has an embryonic origin based on the expression of the SHOOTMERISTEMLESS (STM) SAM marker gene and the stm knockout phenotype that produces seedlings without SAM (Barton and Poethig, 1993; Long et al., 1996; Long and Barton, 1998). The STM gene is the orthologue of the maize KNOTTED1 (KN1) homeobox gene that upon ectopic expression in maize leaves reverts the determinate to the indeterminate state, resulting in the production of knots (Vollbrecht et al., 1991). KN1 was the first plant protein for which plasmodesmal trafficking has been shown to occur (Lucas et al., 1995); this report was one of the first to emphasize the importance of plasmodesmal cell-to-cell communication in developmental processes. After germination, the SAM starts to produce the lateral organs and stem tissue that are organized in the so-called phytomers. The SAM consists of zones that are distinct with respect to their cell division activity and developmental destination and was subject of early developmental research in plants and is still today (Vaughan, 1952; Steeves and Sussex, 1989; Potten and Loeffler, 1990; Laufs et al., 1998) (Figure 1A). In the central zone, stem cells stay in an indeterminate state. Upon division, stem cells replenish themselves but also produce daughter cells that are displaced into the peripheral zone where they are recruited to initiate leaf primordia or into the rib zone where they contribute to the formation of stem tissue. The SAM is also layered: L1, L2 and L3 layers that are the progenitor of epidermal tissue, of palisade and spongy parenchyma (and the sporogenic cells) and of vascular tissue, respectively. A genetic model has been proposed for the self-regulation of the SAM. The CLAVATA (CLV ) genes are responsible for the repression of the growth in the central zone (Clark et al., 1993, 1997; Fletcher et al., 1999; Brand et al., 2000) and encode components of a signaling cascade that regulates WUSCHEL (WUS ) activity (Laux et al., 1996; Trotochaud et al., 1999; Schoof et al., 2000). WUS is a homeodomain protein that keeps stem cells in their indeterminate state through a negative feedback loop with CLV3 (Mayer et al., 1998). The site of expression of the WUS domain, just beneath the stem cell zone is called the «organizing center» and is comparable to the quiescent center in the root apical meristem (van den Berg et al., 1997; for review, see Weigel and Jürgens, 2002).
Leaf phyllotaxis, initiation and polarity
Leaf primordia initiate at the SAM peripheral zone and have a multicellular origin because cells are recruited from the different SAM layers. The leaf initiation site or phyllotaxis is delineated by molecular markers, such as ASYMMETRIC1, whose position at the periphery of the SAM depends on the position of previously formed primordia. Phyllotaxis is species specific, can be opposite, decussate, or spiral according to the mathematical Fibonacci series (Mitchison, 1977). «Biophysical forces» regulating local epidermal cell wall extensibility was one of the mechanisms proposed to explain phyllotaxis (Green, 1996; Fleming et al., 1997). Recently mutational analysis and pharmacological tests combined with micro-manipulation have demonstrated that the hormone auxin is crucial in determining the leaf initiation site (Reinhardt et al., 2000, 2003).
At the leaf inception site, no STM expression fits the exit from the proliferative state into a differentiation state of the primordium founder cells with a de-repressed AS1 gene activity. AS1 in Arabidopsis (Byrne et al., 2000), its orthologue PHANTASTICA (PHAN) in Antirrhinum (Waites and Hudson 1995; Waites et al., 1998) and ROUGH SHEATH2 (RS2) in maize (Timmermans et al., 1999; Tsiantis et al., 1999) are important for promoting adaxial fate in leaf primordia, their function being conserved in monocots and dicots. Upon recessive mutation of the AS1/PHAN/RS2 genes, some of the KNOX genes are ectopically expressed in the leaves where they are normally inactive (Schneeberger et al., 1998; Byrne et al., 2000; Ori et al., 2000). Microsurgical experiments on the potato shoot apex have shown that the SAM communicates with leaf primordia and that a signal is required to induce polarity in the leaf primordium. Incisions between the SAM and the primordium resulted in radial symmetrical rather than dorsoventral asymmetrical leaves (Figure 1B) (Sussex, 1951, 1955). Although the signal is still not known today, the genetic factors for polarity have been identified: these are AS1 and transcription factors of the HD-ZIPIII and GARP class (Sawa et al., 1999; Siegfried et al., 1999; Kerstetter et al., 2001; McConnell et al., 2001). Analysis of their genetic interactions resulted in a model for dorsoventrality in leaves (for review, see Bowman, 2004).
Organ size and shape
A very intriguing question in organ formation is how size and shape are determined. Cell expansion and its direction have been considered for a long time to be the major determinants. However, recent work has demonstrated that cell division activity, rate of cell division and termination of division activity are also important determinants for organ morphology as shown by mutational analysis and manipulation of the cell cycle. Two theories have been postulated: the Cell Theory which states organ size and shape are merely determined by their building blocks, the cells; in the Organismal Theory, cells just fill up the organ form that is determined by higher order control (for review, see Tsukaya, 2002; Beemster et al., 2003).
The leaf has been exploited as a model to study the genetic and environmental factors that control size and shape. Early leaf growth is mainly due to cell division processes that cease gradually from the tip to the base of the organ, from its margin to the midvein and from the ventral to the dorsal side of the lamina (Figure 1C) (Pyke et al., 1991; Van Lijsebettens and Clarke 1998; Donnelly et al., 1999). Interference with early growth by modulation of cell cycle regulatory genes has resulted in changes in leaf size and shape (De Veylder et al., 2001; Fleming, 2002; Wyrzykowska et al., 2002; Dewitte et al., 2003). The AINTEGUMENTA transcription factor controls organ size by regulating the number and the extent of cell divisions during organogenesis (Mizukami and Fischer, 2000). Later growth is assumed to be due to polar and non-polar cell expansion processes. Expansion growth is perturbed by modifying the expression of genes coding for enzymes involved in hormone biosynthesis or cell wall composition and results in altered leaf size and shape (Fleming et al., 1997; Kim et al., 1999; Cho and Cosgrove, 2000; Pien et al., 2001; Fleming, 2002). ANGUSTIFOLIA, a transcriptional co-repressor, is required for polar cell expansion (Folkers et al., 2002; Kim et al., 2002). Over 100 gene loci have been identified to date with a function in the making of the leaf, of which 94 originate from an ethyl methane sulfonate mutagenization program (Berná et al., 1999; Tsukaya, 2002). The systematic cloning and molecular-genetic analyses of these genes will further our knowledge on the molecular mechanisms directing organ formation.
In addition to the above-mentioned internal factors, leaf growth is also modulated by environmental factors, such as water, light and CO2 availability, that affect leaf size and shape. These parameters influence the number of cell cycles during leaf formation In Arabidopsis, mutational analysis showed that leaf growth is controlled at the transcriptional level not only by transcription factors but presumably also by chromatin modification. Functional analysis of structural components and a putative regulator of the Elongator histone acetyltransferase complex, associated with the RNA polymerase II transcription elongation complex resulted in plants with a leaf phenotype (Nelissen et al., 2005), suggesting that the chromatin status is important during organogenesis (Figure 2).
Universal flower model
Flowers consist of four types of organs that are arranged in whorls. The genetic control of flower organ identity was a major discovery of plant developmental biology in the nineties and was based on the study of homeotic flower mutants with normal floral organs at ectopic positions, which replace the flower organs usually present. Such mutants were described in a number of species in ancient literature all over the world. Homeotic mutants and their genetic interactions have been studied extensively in snapdragon (Carpenter and Coen, 1990) and Arabidopsis (Komaki et al., 1988; Meyerowitz et al., 1989). This research resulted in the famous ABC flower model (Bowman et al., 1989; Schwarz-Sommer et al., 1990; Coen and Meyerowitz, 1991). In this model, floral organs are specified by the action of A, B and C genes, the so-called floral organ identity genes, that are active in two subsequent whorls. From 1990 on, the genes corresponding to the flower homeotic genes were cloned in Antirrhinum, Arabidopsis and Petunia by making use of the first mutant collections tagged either with endogenous transposable elements (Carpenter and Coen, 1990), with the Agrobacterium T-DNA (Azpiroz-Leehan and Feldmann, 1997), with the endogenous transposon dTph1 in petunia; or using reverse genetics strategies (Vandenbussche et al., 2003a, 2004). The first homeotic genes cloned were the C function gene AGAMOUS in Arabidopsis (Figure 3B) (Yanofsky et al., 1990) and the B function gene DEFICIENS in snapdragon (Sommer et al., 1990; Schwarz-Sommer et al., 1990). In the following years all the floral homeotic genes were cloned and appeared to be MADSbox transcription factors, except for APETALA2 that identified a plant-specific transcription factor class (Jofuku et al., 1994). The ABC model has been verified by double and triple mutants, overexpression constructs and gene expression analyses; it still stands today even though it has been extended with D and E function genes (Colombo et al., 1995; Pelaz et al., 2000; Honma and Goto, 2001). The identity of the floral organs has been postulated to be specified by tetrameric complexes of floral organ identity gene products that bind to promoters of downstream targets thereby activating or repressing their activity and resulting in specific floral organ identities (Honma and Goto, 2001; Theiβen and Saedler, 2001). ABCDE genes were identified in a lot of species of flowering plants, monocots and dicots, confirming the conservation of the overall molecular mechanism of flower organ formation in evolution (for recent review, see Ferrario et al., 2004). All genes, except for A-type function, are present in gymnosperms indicating ancient mechanisms for reproductive organ formation (Tandre et al., 1995; Theissen et al., 2000). In a-b-c- triple mutants, every flower organ is converted into a leaf and overexpression of B and D function genes is sufficient to transform rosette leaves into petals, which is genetic proof of leaves being the ground state (Figure 3C) (Weigel and Meyerowitz, 1994; Honma and Goto, 2001). von Goethe (1790) had already pointed out that different types of organs, such as leaves, petals and stamens, were variations on a common underlying theme. His theory was also based on the study of abnormal flower morphologies in which one organ type was replaced by another one.
The famous botanist C. Linnaeus (1749) noticed that occasionally individuals of a plant species had altered flower morphologies. He described a naturally occurring mutant of Linaria vulgaris (toadflax) with radial flower symmetry instead of dorsoventral asymmetry; the mutant flower was called peloric (Greek for monstruous). Similar mutations have been obtained in Antirrhinum from the large transposon-mutagenized population generated at Norwich (Carpenter and Coen, 1990; Coen, 1996). In the mutant flowers, the bilateral symmetry was converted into a radial symmetry with ventral-type of petals and stamens, indicating that a dorsalizing factor had been affected. The dorsalizing factor represented two closely related genes, CYCLOIDEA (CYC) and DICHOTOMA (DICH) (Luo et al., 1996; Luo et al., 1999), both members of the so-called TCP class of plant-specific DNAbinding proteins (Cubas et al., 1999a). The peloric mutant described by Linnaeus had an epigenetic mutation in a CYC orthologue (Cubas et al., 1999b). The CYC and DICH genes are expressed very early in flower meristems before flower organ initiation. Superimposition of a dorsal domain onto the radial symmetry of the flower meristem is necessary to create dorsoventral asymmetry. Differences in flower morphology between the closely related species Antirrhinum and Mohavea have been explained by ectopic expression patterns of the CYC and DICH genes (Hileman et al., 2003), which is one of the first examples that explains evolutionary morphological divergence in terms of variations in gene expression.
The cellular organization in the primary root
The development of the primary root has been neglected for a long time and it was only until Arabidopsis started to be a model that people got interested in root biology. Until then, the root was mainly studied for its role in gravitropism. The root is a good model for cell biology because it is transparent (no chlorophyll) and, hence amenable to confocal microscopy on living explants by using fluorescent dyes or reporter genes. Tissue patterning and cellular communication have been studied extensively and with great success in the root. Clonal analyses showed the embryonic origin of the root meristem initials and its radial organization in a constant cell number with root cell initials giving rise to one or two cell layers (Dolan et al., 1993; Scheres et al., 1994). The root meristem consists of quiescent center cells that keep the surrounding initials in an indeterminate state (van den Berg et al., 1997). The daughter cells of the initials differentiate into specific tissue, the identity of which is reinforced by signals from more mature cells (van den Berg et al., 1995). Patterning in the root epidermis was subject to cell biology and genetic analyses (Dolan et al., 1994) and a number of regulatory genes have been identified (Larkin et al., 2003). Mutagenization programs were initiated to look for regulatory genes of root cell specification (Benfey et al., 1993; Scheres et al., 1995). SCARECROW and SHORT ROOT are essential for the asymmetric cell division in the generation of the cortex/endodermal tissue layers (Di Laurenzio et al., 1996). A huge number of marker lines exist with cell typespecific expression in the root, which have been obtained by promoter trapping with a modified GFP reporter gene (Haseloff et al., 1997). A major breakthrough technology was the use of these marker lines in cell sorting to purify specific cell types and study their transcript profiles (Birnbaum et al., 2003). The technology will become applicable to a large number of cell types or cell domains in planta that can be distinguished by marker genes. Cell differentiation studies were restricted in plant research because in vitro culture of a specific cell type has not been achieved so far. This restriction has been alleviated by the above-mentioned approach.
rating: 1.40 from 5 votes | updated on: 28 Oct 2006 | views: 11259 |