08) Vernoud - Maydica

4 downloads 0 Views 3MB Size Report
from the mother plant during seed dispersal. During germination the maize embryo becomes the seedling and consumes its own reserve substances as well as ...
Maydica 50 (2005): 469-483

MAIZE EMBRYOGENESIS V. Vernoud, M. Hajduch, A.-S. Khaled, N. Depège, P.M. Rogowsky* RDP, UMR 5667 CNRS-INRA-ENSL-UCBL, IFR128 BioSciences Lyon-Gerland, ENS-Lyon, 46 Allée d’Italie, F-69364 Lyon Cedex 07, France

Received January 31, 2005

ABSTRACT - After a century of morphological descriptions and classical genetics maize embryogenesis has been approached over the past decade mainly by molecular genetics. Using forward genetics the cloning of a dozen mutations causing aberrant embryo development has been accomplished leading to the conclusion that mutants with developmental blocks before the coleoptilar stage are more likely to be affected in basic cellular functions than mutants with later blocks or mutants with viable but altered seedlings, which are more likely to be impaired in regulatory genes. By reverse genetics numerous genes with well defined, temporally and/or spatially restricted expression patterns in the maize embryo have been isolated and functions inferred based on sequence analysis and/or expression patterns. In parallel the phenotypic analysis of wildtype and mutant embryo morphology and cytology has made a step forward by the integration of novel methods such as confocal laser scanning microscopy, in situ hybridisation with marker genes or TUNEL assays for the detection of PCD. KEY WORDS: Embryogenesis; Morphology; Mutant; Marker gene; Zea mays.

INTRODUCTION Embryogenesis is defined as the sum of all growth and differentiation processes during the development of a single-celled zygote into a multi-cellular, highly organised mature embryo. The formation of the zygote by the fertilisation of the egg cell by one of the sperm cells marks the transition from the gametophytic to the sporophytic generation. It is part of the double fertilisation event typical of flowering plants that also gives rise to the en-

* For correspondence (fax +33 4 72 72 86 00; Peter. [email protected]).

dosperm by the fertilisation of the central cell by the second sperm cell (Fig. 1). As in other plants maize embryogenesis can be divided into three major phases. The first one is devoted to developmental events such as pattern formation responsible for the polarity and position of organs, morphogenesis giving their shape to organs and differentiation sensu strictu setting apart epidermis, ground tissue and vascular tissue. The second or maturation phase is marked by the growth of the embryo and by the accumulation of reserve substances, even though some additional developmental events take place. During the third phase the embryo dehydrates and enters into dormancy to prepare for its separation from the mother plant during seed dispersal. During germination the maize embryo becomes the seedling and consumes its own reserve substances as well as the ones stored in the endosperm that ceases to exist. While the concepts and the vocabulary of plant embryology are strongly influenced by that of animal systems such as Caenorhabditis, Drosophila or mammals, fundamental differences exist between the two eukaryotic kingdoms. Firstly, plant embryogenesis is not a distinct process leading to the formation of a miniature version of the adult organism containing at least primordia of all its future organs. It is rather the beginning of a continuous developmental process interrupted temporarily by dormancy. While the primary root and some but not all leaf primordia are present in the maize embryo, it does not contain primordia of lateral roots, additional leaves or floral organs. However, the stem cells present in the shoot apical meristem and root meristem have the capacity to form secondary meristems which in turn give rise to these additional organs (KAPLAN and COOKE, 1997). Secondly, it is generally recognised that the elaboration of the body plan is not based on cell lineage but on the position of in-

470

V. VERNOUD, M. HAJDUCH, A.S. KHALED, N. DEPÈGE, P.M. ROGOWSKY

FIGURE 1 - Embryogenesis in the maize life cycle. Schematic drawing of the life cycle of a maize plant indicating the different phases of embryo and endosperm development and their duration. DAP, days after pollination.

dividual cells within the embryo (JÜRGENS et al., 1994). Instead gradients of hormones or other signalling molecules seem to determine the fate of individual cells (VOGLER and KUHLEMEIER, 2003). Thirdly, the timing and orientation of cell divisions, which are of uttermost importance for embryo shape, are likely governed differently in animals and plants. Plants have a unique mode of cytokinesis involving plant-specific structures such as the phragmoplast, a highly dynamic cytoskeletal array, or the preprophase band, a transient structure predetermining the division plane (JÜRGENS, 2003). Finally the often cited difference between a coenocytic development in animals and a cellular development at the very beginning of embryo plant development applies to the Drosophila embryo but not to other animal embryos such as the mouse embryo.

RESULTS AND DISCUSSION Morphology of the maize embryo Historically the first morphological descriptions of the maize embryo were side products of work on the double fertilisation (GUIGNARD, 1901; MILLER, 1919; WEATHERWAX, 1919) or on the transformation of the ovary wall into the pericarp of the mature kernel (TRUE, 1893; GUERIN, 1899). Later on more specialised light microscopical studies were undertaken (AVERY, 1930) and the first comprehensive description and illustration of all developmental stages of the maize embryo can be found in the timeless work of Randolph (RANDOLPH, 1936) that remains a reference until our days. Another landmark was the study by Abbe and Stein (ABBE and STEIN, 1954) who focussed on quantitative aspects including statistically valid cell size measurements or cell num-

MAIZE EMBRYOGENESIS

ber counts throughout embryo development. With the advent of electron microscopy in particular the early stages of maize embryogenesis were revisited by DIBOLL (1968) and VAN LAMMEREN (1986b), who provided additional insight in the spatial and temporal aspects of meristem formation. A rather simplified schematic summary of this and other work is presented in Fig. 2. Briefly, the zygote undergoes an asymmetric division into a small apical and a large basal cell giving rise to the embryo proper and the suspensor, respectively. The radial symmetry of the proembryo shifts to a bilateral symmetry at the transition stage, which is also characterised by the formation of a distinct external cell layer, the protoderm. At the onset of the coleoptilar stage the shoot apical meristem (SAM) and the root apical meristem (RAM) can be distinguished and soon thereafter a small protuberance marks the position of the future coleoptile. The subsequent stages are numbered from 1 to 6 according to the number of leaf primordia present in the embryo. While the suspensor degenerates, the other parts of the embryo keep growing and reserve substances are accumulated in the scutellum. Maize has been a model species for the study of monocot embryo development and even today there are no equally detailed descriptions covering the entire embryo development of other grass species such as wheat (SMART and O’BRIEN, 1983) or barley (NORSTOG, 1972; ENGELL, 1989) and in particular rice (MOLDENHAUER and GIBBONS, 2003). In dicots the model species for embryo development has been for many years Capsella bursapastoris (HANSTEIN, 1870; SCHULZ and JENSEN, 1968a,b). In more recent years it has been marginalised by Arabidopsis thaliana for which a reference developmental chart had to be established prior to the study of numerous embryo mutants (MANSFIELD and BRIARTY, 1991, 1992; MANSFIELD et al., 1991). Similarly to the maize zygote the Arabidopsis zygote undergoes an asymmetric division giving rise to the embryo proper and the suspensor. In the next rounds of cell division the suspensor divides more rapidly than the embryo proper but ceases to grow soon thereafter. In the embryo proper a protoderm is set aside at the dermatogen stage. After the globular stage SAM and RAM formation is initiated at the beginning of the heart stage. A vascular system is distinguishable at the torpedo stage. Embryogenesis is complete after accumulation of reserve substances in the expanding cotyledons (Fig. 2). The maize and other monocot embryos share

471

with the Arabidopsis embryo the major functional processes established for the embryogenesis of dicots (KAPLAN and COOKE, 1997): formation of a zygote, establishment of an apical-basal polarity leading to a linear proembryo divided into suspensor and embryo proper, initial histogenesis resulting in the formation of a protoderm and organisation of the two apical meristems at the shoot and root end. On the other hand the morphology of the maize and Arabidopsis embryo is quite different. A first difference concerns the first divisions of the embryo that are rather synchronised and equal in Arabidopsis leading to easily recognisable geometric figures and more erratic in maize reflecting possibly the absence of any particular organisation (SHERIDAN, 1995). However, the statement needs to be somehow qualified because some maize embryos show rather geometric figures at the 4-cell stage (RANDOLPH, 1936) and in Arabidopsis cell numbers different from 2n can be observed, in particular in the suspensor (BOWMAN, 1993). Secondly, the formation of leaf primordia occurs after the entrance into dormancy and seed dispersal in Arabidopsis, while 5 to 6 leaf primordia are elaborated in the maize embryo. This difference reinforces the view that there is no real end to embryogenesis in plants and that dormancy is a rather arbitrary interruption of a continuous process covering the entire life span of a plant. Thirdly, the axis between SAM and RAM coincides with apical-basal axis defined by the suspensor and the embryo proper in Arabidopsis but is oblique in maize. This may be the consequence of the absence of a second cotyledon. Finally, the relationship between the cotyledons of Arabidopsis and the scutellum of maize needs to be clarified. They are clearly functionally equivalent due to the fact that the vast majority of reserve lipids and proteins are deposited in these organs. Developmentally both arise without involvement of the SAM (JÜRGENS, 2001), although this is a matter of debate in Arabidopsis (KAPLAN and COOKE, 1997). In maize the debate centers around the question whether the scutellum presents all of the cotyledon, part of it or a different structure (KIESSELBACH, 1949). This question is closely linked to the status of the coleoptile which has been interpreted as a new acquisition (BROWN, 1960), the first leaf (GUIGNARD, 1975) or part of the scutellum (VAN LAMMEREN, 1986b). Beyond these developmental and phylogenetic considerations there are obviously important quantitative differences in the size or cell number of maize and Arabidopsis embryos, the maize one be-

472

V. VERNOUD, M. HAJDUCH, A.S. KHALED, N. DEPÈGE, P.M. ROGOWSKY

FIGURE 2 - Embryo development in Arabidopsis and maize. Schematic drawing of key developmental stages of Arabidopsis (top) or maize embryos (bottom). From left to right the stages depicted for Arabidopsis thaliana are zygote, 2-celled, quadrant, dermatogen, globular, heart, torpedo and mature and for Zea mays zygote, 2-celled, early proembryo, late proembryo, transition, early coleoptilar, late coleoptilar, stage 1 and mature. The time scale in days after pollination (DAP) is merely indicative as embryo development is strongly dependent on the genetic background and environmental conditions. No common scale was used for the drawings to allow detailed views of each species and stage. Only the sizes of the mature embryos are shown.

ing roughly 10 times larger at maturity. The maize embryo also takes a 6 times longer period to reach maturity, the additional time being devoted to the storage of larger quantities of reserve substances and to a further progress in the life cycle of the plant with the elaboration of leaf primordia. A last major difference between maize and Arabidopsis concerns the environment of the embryo in the seed (Fig. 3). In Arabidopsis the endosperm is nonpersistant and the mature seed is essentially composed of the embryo. In maize the endosperm persists and accounts for roughly 2/3 of the mature kernel. Consequently the maize embryo has not only the lipid and protein reserves stored in the scutellum but also the carbohydrate reserves stored in the endosperm available during germination.

Cytology of the maize embryo The light and electron microscopic analysis of embryo tissue sections was not only a means to reconstruct the three dimensional morphology of the maize embryo, it also allowed the observation of cytological details. For example the cells of the embryo proper are small and rich in cytoplasm, while the cells of the suspensor are large and highly vacuolated (SCHEL et al., 1984). Epidermal cells are characterised by a large nucleus in a central position, a high frequency of polysomes, a low number of vacuoles and the absence of starch granules. The formation of the SAM and RAM involves a dedifferentiation of cells present in the respective positions that manifests itself by a decrease in size a loss of vacuoles and an increase of cytoplasm (VAN LAMMEREN, 1986b).

MAIZE EMBRYOGENESIS

Cytological studies also give some insight into the establishment of embryo polarity. A rather dramatic shift in cell polarity takes place in the egg cell upon fertilisation. While the nucleus and most of the cytoplasm are located at the micropylar half of the egg cell prior to fertilisation they are found in the antipodal half afterwards (VAN LAMMEREN, 1986a). More detailed kinetic studies based on the clearing technique documented an intermediate position during karyogamy that occurs 14 to 18 hours after pollination. The shift of polarity is completed during a resting period of 13 to 16 hours before the first division (MOL et al., 1994). After the shift the polarity is the same as in the Arabidopsis or Capsella egg

473

cell and coincides with the orientation of the future embryo: embryo proper cells rich in cytoplasm at the antipodal end and highly vacuolated suspensor cells at the micropylar end. This active reorientation of the cell content is an indication that the polarity of the embryo is not predetermined by the polarity of the egg cell but actively established in the zygote as a consequence of fertilisation. Programmed cell death (PCD) is a genetically and cytologically well defined auto-destructive mechanism triggered by developmental or environmental signals. During development it is a means to eliminate cells that have fulfilled their function and can be seen as the counterpart of cell division in

FIGURE 3 - Seed development in Arabidopsis and maize. Schematic drawing of key developmental stages of Arabidopsis (top) or maize seeds (bottom). The time scale in days after pollination (DAP) is merely indicative as seed development is strongly dependent on the genetic background and environmental conditions. No common scale was used for the drawings to allow detailed views of each species and stage.

474

V. VERNOUD, M. HAJDUCH, A.S. KHALED, N. DEPÈGE, P.M. ROGOWSKY

the morphogenesis of organs. The underlying mechanisms of PCD in plants are less well known than those of PCD in animals, but several morphological and biochemical similarities exist such as DNA laddering, caspase-like proteolytic activity or cytochrome C release from mitochondria (HOEBERICHTS and WOLTERING, 2003). Based on the TUNEL method allowing the visualisation of DNA fragmentation on cytological sections, PCD was detected in the suspensor, scutellum, coleoptile and root cap of the developing maize embryo (GIULIANI et al., 2002). At stage 1 the main activity concerns the suspensor where the PCD progresses from the top to the bottom during further developmental stages. Considerable PCD is also observed in the scutellum with a gradient from the adaxial to the abaxial side, while the TUNEL staining in the coleoptile and root cap is less intense (Fig. 4). These experiments suggest that the “degeneration” of the suspensor that had been described for decades is not a simple necrosis but a developmentally regulated process. Similarly the deposition of reserve substances in the scutellum is followed by a coordinated disassembly of the corresponding cells. None of the TUNEL positive parts of the embryo contribute to the adult plant body leading to the hypothesis that PCD may be the means to eliminate purely embryonic structures as soon as they are no longer needed. In vitro culture of the maize embryo Since the young maize embryo is deeply buried in maternal tissues (Fig. 3), it is not readily accessible for experimentation. Consequently a lot of effort has been invested in the in vitro culture of maize embryos that allows the application of hormones or other signal molecules and facilitates the microscopical observation. In addition to androgenesis (PETOLINO and JONES, 1986), gynogenesis (BECKERT, 1994) and somatic embryogenesis (BAUDINO et al., 2001) the culture of zygotic embryos of various stages has been achieved (MATTHYS-ROCHON et al., 1998). Embryos excised before the transition stage depend for their development on the presence of cytokinins in culture medium, while older embryos develop into mature and fertile plants without any external hormone supply. Embryo culture has been used for the characterisation of mutant embryos allowing conclusions as to the type of the lesion depending on the outcome of the rescue experiment (SHERIDAN and NEUFFER, 1980; CONSONNI et al., 2003). In wheat the application of auxin polar transport inhibitors demonstrated that auxin has a determining influence

FIGURE 4 - Programmed cell death in the maize embryo. In a schematic drawing of a longitudinal (left) or transverse section (right) of a stage 1 maize embryo zones of intense TUNEL staining are indicated in dark red, zones of weak staining in pink and TUNEL negative zones in white. The dashed line in the longitudinal section indicates the plane of the transverse section.

on the differentiation of the embryonic axis and the scutellum in grass embryos (FISCHER et al., 1997). The culture of very young embryos before the proembryo stage is hampered by the difficulties of proper micro-dissection without surrounding tissues. Two experimental solutions exist: either the zygote or 2-celled embryo is excised with the surrounding endosperm and some nucellar tissue (MOL et al., 1993) or clean zygotes are obtained by treatment with a mix of cell wall degrading enzymes (LEDUC et al., 1996). In both cases only a certain percentage of the cultured zygotes follows a normal development and yields fertile plants. Contrary to the zygotes cultured in the natural environment of the embryo sac and some nucellar tissue, the clean zygotes require the co-culture with androgenetic microspores for their development. In addition their development is not direct but occurs via a callus phase and secondary embryogenesis after reaching the transition stage. This data suggest that the zygote and the very young embryo need external signals for proper development. In vitro culture has also been used to study the dynamics of the fertilisation process. Isolated egg cells and sperm cells can be fused in vitro and fertile plants be generated via secondary embryogenesis (KRANZ and LÖRZ, 1993; FAURE et al., 1994). After an adhesion of several minutes the gametes fuse in less than 10 s. This process is rather specific to male-female gamete pairs and establishes a barrier to polyspermy. Further experiments with fluores-

MAIZE EMBRYOGENESIS

cent Ca2+ indicator dyes showed that the cytoplasmic fusion triggers a transient increase in cytosolic Ca2+ in the fertilised egg cell that lasts several minutes (DIGONNET et al., 1997). The Ca2+ is of extracellular origin and enters the cell at the vicinity of the sperm entry site approximately 2 s after fusion. Subsequently the Ca2+ entry is gradually generalised over the entire zygote membrane (ANTOINE et al., 2000). One may speculate that the initial calcium gradient or the subsequent elevated cytosolic calcium level is a coordinator in space and time of developmental events in the zygote such as the inversion of polarity or the activation of transcription. Mutants of the maize embryo Embryo mutants have been isolated and studied since the beginning of the last century (DEMERC, 1923). As other maize mutants they were obtained either by EMS mutagenesis or from stocks with a high frequency of transposition. These stocks contain active transposable elements, generally of the Mutator (Mu), Activator (Ac) or Enhancer (En) class, that transpose either themselves or mediate the transposition of truncated elements of the same class. From the very beginning the nomenclature of mutants with smaller, incompletely developed or aberrant embryos has caused some problems due to overlaps in the definition of mutant classes, for example between defective seed (de) (JONES, 1920; MANGELSDORF, 1923) and germless (gm) mutants (DEMERC, 1923). Both are part of the large class of defective kernel (dek) mutants that include in the original definition by NEUFFER and SHERIDAN (1980) mutations affecting either the endosperm or the embryo or both of them and where in the last two cases the mutant embryo is either not viable or develops into a seedling with a distinct mutant phenotype (NEUFFER and SHERIDAN, 1980). Work with this class of kernel mutants is hampered by the fact that the vast majority are single gene recessive mutants that are not viable in the homozygous form and need to be propagated as heterozygotes. Viable kernels are always a mixture of wildtype and heterozygote kernels and, in the absence of molecular markers, the resulting plants can only be distinguished a posteriori by scoring self-pollinated ears. Similarly, at early developmental stages homozygous mutant kernels are often not distinguishable from wildtype or heterozygous kernels by simple visual inspection and only a more detailed analysis such as the production of cytological sections of randomly sampled kernels allows the identification and observation of

475

mutant embryos. As a consequence there was a hiatus of 50 years in the work on dek mutants and very often the analysis of the mutant phenotype was restricted to observations at kernel maturity. The modern era of dek mutant characterisation starts with the EMS induced collection of NEUFFER and SHERIDAN (1980) consisting of 855 recessive kernel mutants. Genetic, lethality, morphological, and embryo rescue studies on nearly 200 dek mutants showed (1) that dek mutations can occur on at least 17 of the 20 chromosome arms, (2) that the presence of a normal endosperm can have a positive, neutral or negative effect on a mutant embryo, (3) that the embryo is generally more severely defective than the endosperm, (4) that the majority of embryos is blocked at or after stage 1, and (5) that more than one half of the mutant embryos can be rescued by tissue culture (NEUFFER and SHERIDAN, 1980; SHERIDAN and NEUFFER, 1980). A more detailed study of 14 mutants with defects in both embryo and endosperm at early developmental stages revealed that the embryo phenotype manifests itself even in the presence of wildtype endosperm and that in some instances mutant endosperm can impair the germination of wildtype embryos. These results obtained by pollination of heterozygotes with B-A translocation stock suggest that the embryo phenotype is not simply a consequence of insufficient nutrient supply by mutant endosperm and that mutant endosperm may contain toxic substances (SHERIDAN and NEUFFER, 1982). For 7 of these mutants detailed developmental profiles of the mutant embryo were established by series of cytological sections throughout the development (CLARK and SHERIDAN, 1986; SHERIDAN and THORSTENSON, 1986; CLARK and SHERIDAN, 1988). A great variability of the phenotype was observed because mutant embryos can be arrested at the proembryo, transition or coleoptilar stage, can be necrotic or not, and can simply stop growth or proliferate without ever forming leaf primordia. Since positional cloning of EMS mutants was considered as unfeasible in maize, interest shifted to transposon-tagged embryo mutants. A collection of 63 dek mutants obtained after a cross with active Mutator stock was well characterised genetically by chromosome arm locations for 53 mutants and by linkage mapping and allelism tests for 21 mutants (SCANLON et al., 1994). While the initial phenotypic characterisation concerned only the gross morphology of the kernel at maturity, the detailed phenotypes of individual mutants were the subject of sub-

476

V. VERNOUD, M. HAJDUCH, A.S. KHALED, N. DEPÈGE, P.M. ROGOWSKY

sequent work. The viable embryos of the pleiotropic semaphore1 (sem1) mutant are smaller than wildtype embryos and have fewer if any leaf primordia. Ectopic Knox gene expression in leaves and endosperm and reduced polar auxin transport suggest that Sem1 may be an intermediate in a signalling cascade leading from auxin to Knox genes that in turn are necessary for the correct initiation of leaf primordia (SCANLON et al., 2002). The discoloured-1 (dsc1) mutation results in an arrest of embryo growth at stage 1 followed by tissue degradation. Part of the mutated gene has been cloned but no function has been deduced from the partial sequence (SCANLON and MYERS, 1998). Embryos of the empty pericarp2 (emp2) mutant reach an abnormal coleoptilar stage characterised by a SAM lacking the typical tunica-corpus shape and incapable of forming leaf primordia (SCANLON et al., 1997). The underlying gene shows significant homology to animal heat shock binding proteins and encodes a negative regulator of heat shock response in maize (FU et al., 2002). The embryo lethal phenotype suggests that, in addition to their protective role during heat stress, heat shock proteins (HSP) may have a developmental role in plants as it has been documented in animals (CHRISTIANS et al., 2003). The molecular cloning of the latter two dek mutations was achieved via Mutator tags and confirmed the initial hypothesis that at least some of the newly arising mutants after a cross with active Mutator stock were caused by the insertion of one of the 9 Mutator elements (SCANLON et al., 1994). Several other dek mutants have been characterised, although the phenotypic descriptions frequently focused on the endosperm rather than the embryo. For two of them the underlying genes have been identified via a Mutator and an Activator tag, respectively. Dek1 encodes a membrane protein with a cytoplasmic cysteine proteinase domain at its C-terminus. The exact role of this protein in the development beyond the transition stage and the establishment of the embryo axis remains to be determined (BECRAFT et al., 2002; LID et al., 2002). Lachrima (DekB) encodes a transmembrane protein expressed very early in kernel development with a preference for proliferating tissues of the embryo. Mutant embryos are blocked at the transition stage and undergo little change until kernel maturity. Based on structural rather than sequence similarities with Arabidopsis proteins Lachrima may be an auxin transporter, which would readily explain the embryo phenotype (STIEFEL et al., 1999).

In an attempt to focus on developmental rather than metabolic defects in embryo development, a Mutator derived collection was screened for embryo-specific (emb) mutants (CLARK and SHERIDAN, 1991). Considered by some but not all authors as a sub-class of dek mutants, emb mutants are defined by defects concerning only the embryo but not the endosperm. Also called germless or lethal embryo, these mutants are more likely to be affected in developmental processes because metabolic defects should also manifest themselves in the endosperm that shares with the embryo a heterotrophic growth. The embryo morphology of 51 independent mutants was documented at kernel maturity and showed that frequently the embryos not simply stopped growth but were morphologically abnormal, that in about 2/3 of the mutants, the embryo was arrested prior to stage 1 and that embryo necrosis was found only in very few mutants (CLARK and SHERIDAN, 1991; SHERIDAN and CLARK, 1993). A cytological analysis of 5 non-allelic mutants blocked before the transition stage in the course of embryo development revealed three phenotypic groups: morphologically normal embryos that ceased growth at the transition stage, tube-shaped embryos lacking apical-basal differentiation and embryos with a proliferation of the suspensor tissue (HECKEL et al., 1999). The gene responsible for the latter phenotype in mutant emb*-8516 was identified as ZmPRPL35-1 encoding a plastid ribosomal protein (MAGNARD et al., 2004). Interestingly, a lesion in another nuclear-encoded plastid ribosomal protein, ZmPRPS9, also affects only the embryo and not the endosperm in the lem1 mutant (MA and DOONER, 2004). It is not clear at this point in time whether the embryo is more dependent on functional plastids than the endosperm or whether the two proteins have extra-ribosomal functions required for embryogenesis. From the same collection a second group of four emb mutants with slightly later developmental blocks was characterised by confocal laser scanning microscopy. Two of them ceased growth at the proembryo stage. Based on protoderm marker gene expression the other two underwent an incomplete radial organisation that compromised without completely abolishing the formation of a SAM or leaf primordia (ELSTER et al., 2000). The major conclusion from these data is that protoderm and meristem formation are not independent events but that protoderm formation is a prerequisite for proper meristem formation.

MAIZE EMBRYOGENESIS

The characterisation of another group of three independently isolated non-allelic emb mutants with a developmental block prior to the transition stage showed abnormal proliferation of the suspensor and absence of PCD in both the suspensor and the scutellum (CONSONNI et al., 2003). Surprisingly embryo rescue on hormone-free medium with sucrose gives rise to small but otherwise normal seedlings suggesting that the culture conditions trigger the formation of functional meristems. These data further support the observation that the suppression of morphogenesis is frequently accompanied by an uncontrolled pattern of cell division. Other sub-classes of dek mutants can provide valuable information on embryo development, and in particular defective seedling (des) mutants. Embryos of this sub-class germinate but develop into aberrant seedlings that do not yield mature plants. The seedling defects can generally be traced back to aberrations already detectable before germination. In Arabidopsis the screen for seedling rather than embryo defects has been established early on as the method of choice for the isolation of pattern mutants (MAYER et al., 1991). The most prominent example in maize is the shootless phenotype, in which the embryo possesses a scutellum and the root half of the embryo axis but completely lacks the shoot half of the embryo axis. This intriguing phenotype manifests itself only in double mutants at the shootmeristemless (sml) and distorted growth (dgl) loci. Mutant seedlings form an apparently normal primary root but never develop aboveground organs. The in vitro culture of mutants on a medium with high levels of cytokinin leads to the conclusion that the double mutant may be affected in the perception of cytokinin signals (PILU et al., 2002). The class of viviparous (vp) mutants is defined by a precocious germination of the embryo on the ear. These mutants are a good illustration of the concept that desiccation and dormancy are a rather arbitrary endpoint of embryogenesis and that there is a continuity between embryonic and post-embryonic development. While some of the resulting seedlings are lethal, others produce normal fertile plants. Biochemical analyses showed that several vp mutations are affected in carotenoid and ABA synthesis (NEILL et al., 1986). This is the case for vp5 which has been cloned and encodes a phytoene desaturase (LI et al., 1996) but not for the well known vp1 encoding a novel type of plant-specific transcription factors (MCCARTY et al., 1991). A role of the

477

plant hormone ABA in the desiccation process of the kernel and in particular of its balance with GA in the control of germination is well documented (WHITE et al., 2000). Since maize embryos contain up to 6 leaf primordia, many mutants disturbed in the organisation of the SAM and/or leaf morphology after germination are also relevant for embryogenesis. Most of the phenotypic analysis of these mutants has focused on post-embyronic stages and will not be discussed here. Interestingly the genes responsible of the phenotypes in Knotted1 (Kn1) (VOLLBRECHT et al., 1991), Rough sheath1 (Rs1) (SCHNEEBERGER et al., 1995), rough sheath2 (rs2) (TIMMERMANS et al., 1999), Gnarley1 (Gn1) (FOSTER et al., 1999), narrow sheath1 and 2 (ns1, ns2) (NARDMANN et al., 2004) or terminal ear1 (te1) (VEIT et al., 1998) all code for putative transcription factors. Transcriptional regulation plays a major role in animal embryogenesis and consequently the regulatory gene functions altered in meristem mutants are more readily reconcilable with a developmental role than the cellular functions impaired in dek or emb mutants. Genes of the maize embryo Maize is an ancient allo-tetraploid and duplications of large parts of the genome are well documented (GAUT and DOEBLEY, 1997). Consequently the analysis of single gene recessive mutants is likely to uncover only part of the genes involved in a given developmental process. Gene expression and in particular differential gene expression is an alternative criterion to identify additional candidate genes even though their implication in the process needs to be confirmed by the characterisation of mutants or transgenic plants. The most comprehensive analysis to date was based on a thematic array containing 900 selected genes and 600 random cDNA clones from a 20 DAP embryo library. Its hybridisation with kernel or embryo probes obtained between 5 and 45 DAP established several characteristic temporal expression profiles in the embryo that can be correlated with gene function and demonstrated that co-expression was stronger for genes involved in the TCA cycle rather than in glycolysis (LEE et al., 2002). On the protein level recent data with spot identification by mass spectrometry are available for the endosperm (MECHIN et al., 2004), while only profiles of non-identified spots exist for the embryo (SANCHEZ-MARTINEZ et al., 1986). Since it is impossible to review all genes for which expression in the embryo has been demon-

478

V. VERNOUD, M. HAJDUCH, A.S. KHALED, N. DEPÈGE, P.M. ROGOWSKY

strated, we will focus on genes that present either a specific or preferential expression in the embryo as compared to other parts of the maize plants or genes whose expression is restricted to particular parts or developmental stages of the embryo. Gene activation after fertilisation was addressed by a differential screen between an egg cell and a zygote cDNA library. Only 2% of the 4000 surveyed clones exhibited significant changes in transcript levels. Among them was a calreticulin that could possibly play a role in the calcium wave observed in the zygote after fertilisation (DRESSELHAUS et al., 1996). In addition a global silencing of the paternal genome during early embryogenesis can be excluded because a Gfp transgene provided by the male parent is transcribed as early as 4 h after fertilisation (SCHOLTEN et al., 2002). Global paternal silencing has been reported in Arabidopsis but remains a controversial issue (VIELLE-CALZADA et al., 2000; WEIJERS et al., 2001). Genes belonging to the Outer cell layer (OCL) family may play a role in protoderm formation and maintenance because their expression is restricted to the outermost cell layer of the embryo prior to the cytologically visible differentiation of the protoderm at the transition stage (INGRAM et al., 1999). Genes OCL1 to 5 encode putative transcription factors of the HD-ZIP IV family and their overlapping but distinct expression patterns in different parts of the protoderm suggest that the protoderm is not a uniform entity but divided in developmental territories (INGRAM et al., 2000). Similarly the Lipid transfer protein2 (Ltp2) gene is expressed only in the abaxial protoderm of the scutellum and coleoptile (SOSSOUNTZOV et al., 1991). Meristem formation in the embryo is accompanied by the onset of Kn1 expression in the corresponding regions. Restricted to the SAM region at transition stage expression spreads all the way to the RAM at stage 1 but remains excluded from the incipient leaf, leaf primordia and the scutellum (SMITH et al., 1995). Loss-of-function mutants of Kn1 have severe inflorescence and floral phenotypes but still seem to have a functional SAM (KERSTETTER et al., 1997). On the other hand the complete loss of the putative orthologue Shootmeristemless in Arabidopsis causes failure to develop a SAM during embryogenesis (LONG et al., 1996). These results suggest a redundant function in maize and are reminiscent of the uncloned sml mutant in which two loci need to be mutated to uncover the phenotype (PILU et al., 2002). While detailed analyses document the

expression of other members of the Knotted-like homeobox (Knox) family in various part of the vegetative shoot apex, their expression pattern has never been rigorously established in embryos. RNA gel blots prove expression of Knox2, Knox4/Gn1 and Knox6 in 17 DAP embryos but fail to detect the expression of other Knox genes including Kn1, which is known to be expressed by in situ hybridisation (KERSTETTER et al., 1994). An antibody recognising several KNOX proteins marks a territory similar to the one described for Kn1 (SCANLON et al., 2002). Other than Knox genes, Ns1 and Ns2 are involved in the elaboration of lateral organs from the SAM. Their expression is first detected in coleoptilar embryos where they mark the lateral margins of the emerging coleoptile. The post-embryonic expression patterns of these putative transcription factors of the Wuschel family are also highly dynamic and always restricted to small groups of cells at tissue boundaries (NARDMANN et al., 2004). For several other genes detailed developmental expression profiles have been established without shedding much light on their precise function. Among the Fertilisation-independent endosperm (Fie) genes the kernel-specific Fie1 is not expressed in the embryo, while the constitutively expressed Fie2 was found in the embryo proper at the proembryo stage and in leaf primordia at later stages (DANILEVSKAYA et al., 2003). As their Arabidopsis counterparts they are likely involved in chromatin remodelling (LUO et al., 2000). Four ZmHox genes are co-expressed in the embryo proper at the proembryo stage and in the embryo axis at later stages. They encode putative transcription factors of the PHD subclass of homeodomain proteins (KLINGE and WERR, 1995). While ZmHox1 has been isolated as a transcriptional regulator of Shrunken1 encoding sucrose synthase in the endosperm, its role in the embryo is not clear (BELLMANN and WERR, 1992). Hydroxyproline-rich glycoprotein (HPRP) is expressed mainly in the vascular system of the embryo axis and may intervene in early steps of cell wall synthesis (RUIZ-AVILA et al., 1992). Hybrid proline-rich protein (HyPRP) shows a complementary expression pattern with little expression in the embryo axis and a gradient-like expression in the scutellum (JOSE-ESTANYOL et al., 1992). Expression is strongest in cells not yet reached by PCD. During the maturation phase two types of proteins have attracted attention. Oleosins are small hydrophobic proteins localised in the phospholipid mono-layer that constitutes the envelope of oil bod-

MAIZE EMBRYOGENESIS

ies. Oil bodies are organelles filled with fatty acids that present the major reserve substance stored in the embryo. Three Oleosin (Ole) genes have been characterised that are co-ordinately regulated during kernel maturation and that can be found both in the embryo axis and in the scutellum (LEE and HUANG, 1994). Globulins are the major storage proteins of the embryo and the proteins encoded by Globulin1 and 2 (Glb1, Glb2) account for 10 to 20% of the embryo protein (KRIZ, 1989). A group of structurally related proteins collectively called late embryogenesis abundant proteins (LEA) accumulate during late stages of embryogenesis and have been associated with dehydration. A sub-group are the structurally related dehydrins (DHN) that can also be found in vegetative tissues upon water or osmotic stress (CLOSE, 1997) and that possibly act as intracellular sponges (GARAY-ARROYO et al., 2000). Some of the corresponding genes have also been called Responsive to ABA (Rab) due to their induction upon treatment with external ABA in both embryos and vegetative tissues. However, at least in the case of Rab28 different trans-acting factors are involved in gene induction during embryogenesis or by ABA treatment (PLA et al., 1993). Both Rab17 (=Dhn1) and Rab28 are more strongly expressed in the embryo axis as compared to the scutellum (GODAY et al., 1994; NIOGRET et al., 1996), while no expression data are available for Dhn2 (CAMPBELL et al., 1998). On a sub-cellular level nuclear or cytoplasmic localisation of Rab17 is strongly influenced by phosphorylation via protein kinase CK2 (GODAY et al., 1994). This phosphorylation is also essential for the action of Rab17 during seed germination in transgenic Arabidopsis (RIERA et al., 2004). Similar functional approaches are now needed to confirm gene functions of Rab and other expressed genes that have been suggested by sequence and/or expression data.

CONCLUSION AND PERSPECTIVES After a century of morphological descriptions and classical genetics maize embryogenesis has been approached over the past decade mainly by molecular genetics. Over a dozen mutations causing aberrant embryo development have been cloned. Detailed phenotypic characterisations of these and many more mutants integrated novel techniques such as confocal laser scanning microscopy, in situ hybridisation with marker genes or the production

479

of transgenic plants in maize or other species. As a general conclusion mutants with developmental blocks before the coleoptilar stage are more likely to be affected in basic cellular functions, while mutants with later blocks or mutants with viable but altered seedlings are more likely to be impaired in regulatory genes. The success of the transposon tagging approach in forward genetics of embryogenesis is difficult to assess. While successful cloning of dek or emb mutations has been reported for several mutant populations generated from crosses with active Mutator stock, we and others have experienced a lot of frustration with this approach. This may be explained by the technical difficulties to visualise all Mutator elements present in a genome or by an increased “spontaneous” mutation rate in active Mutator stock. One may hypothesise that in active Mutator stock not only the transposition frequency of Mutator elements is increased but also that of numerous other transposable elements (that are not detected by the Mutator specific methods). The creation of more sophisticated populations and/or the conjunction of systematic phenotypic characterisation with systematic sequencing of flanking sequences will certainly improve the situation. While the detailed phenotypic and molecular analysis of dek or emb mutants provided important insights into the mechanisms and genes involved in maize embryogenesis, several limitations also became evident. Firstly, for the majority of the emb and other lethal dek mutants the developmental blocks concern very early stages. Often the morphological analysis only allows to determine the precise stage at which mutant development deviates from wildtype development but not to draw any further conclusions. While viable dek mutants or leaf mutants are generally more informative, so far none of them concern the first two fundamental steps of apical-basal polarisation or protoderm formation. Secondly, none of the underlying genes cloned so far carry annotations suggesting developmental functions. At first sight they are house keeping genes involved in basic cellular functions that are not restricted to embryogenesis. While this was more or less expected for dek mutants in general, this was more surprising in the case of emb mutants. The challenge in the later case is now to find out why the embryo and the endosperm are not affected in the same way, for example by mutations in genes that are part of the translational machinery of plastids. In this particular case one may specu-

480

V. VERNOUD, M. HAJDUCH, A.S. KHALED, N. DEPÈGE, P.M. ROGOWSKY

late that embryo but not endosperm development may be dependent on the action of plant hormones such as GA or ABA that are partly synthesised in plastids. Finally the embryo lethal phenotype makes it difficult to asses the role of mutated genes in post-embryonic development. To determine more precisely the function of these genes it will be necessary to obtain conditional knock-outs that allow normal growth and development during embryogenesis and inhibit expression after germination. This may be accomplished by RNAi constructs under the control of tissue-specific or inducible promoters or by clonal mosaic analysis induced by X-ray treatment of the seed as demonstrated for the emp2 mutant (FU and SCANLON, 2004). Reverse genetics also made valuable contributions to the understanding of embryogenesis. Numerous genes expressed in the embryo have been isolated and characterised and in many cases the sequence and/or expression pattern give more or less precise hints as to their function. Two major pitfalls of this approach are nicely illustrated by the duplicate genes Ns1 and Ns2. Firstly, their expression territory is not only spatially limited to a handful of cells but also very dynamic in time and can easily be missed or missed in part by a low-resolution analysis. Secondly, single mutants do not show a phenotype due to functional redundancy between the two genes and double mutants need to be constructed to draw conclusions as to gene function. The bottle neck of reverse genetics is not the isolation or molecular characterisation of (differentially) expressed genes but the isolation of corresponding mutants or the production of transgenic plants to confirm the suspected function, for example of OCL genes in protoderm formation or of non-cloned Knox genes in meristem formation and maintenance. The establishment of public transformation facilities and the availability of a TILLING population in addition to several transposon-based populations will certainly speed up this process. Several of the genes with well defined, temporally and/or spatially restricted expression patterns in the maize embryo can and have been used as marker genes (BOMMERT and WERR, 2001). Marker genes are a widely used tool in developmental biology that allow to identify morphologically aberrant tissues or to establish hierarchical links between genes. The characterisation of the precise temporal and spatial expression patterns of additional genes and gene families in the maize embryo will not only provide additional candidate genes

for functional validation but also a denser map of marker genes useful for the in depth analysis of existing mutants. ACKNOWLEDGMENTS - MH was supported by Marie Curie Fellowship QLK1-CT-2001-51051 of the European Commission and ASK by a PhD grant of the Egyptian Government.

REFERENCES ABBE E.C., O.L. STEIN, 1954 The growth of the shoot apex in maize: embryogeny. Am. J. Bot. 41: 173-284. ANTOINE A.F., J.E. FAURE, S. CORDEIRO, C. DUMAS, M. ROUGIER, J.A. FEIJO, 2000 A calcium influx is triggered and propagates in the zygote as a wavefront during in vitro fertilization of flowering plants. Proc. Natl. Acad. Sci. USA 97: 10643-10648. AVERY G.S., 1930 Comparative anatomy and morphology of embryos and seedlings of maize, oats and wheat. Bot. Gaz. 89: 1-39. BAUDINO S., S. HANSEN, R. BRETTSCHNEIDER, V.F.G. HECHT, T. DRESSELHAUS, H. LÖRZ, C. DUMAS, P.M. ROGOWSKY, 2001 Molecular characterisation of two novel maize LRR receptor-like kinases, which belong to the SERK gene family. Planta 213: 1-10. BECKERT M., 1994 Advantages and disadvantages of in vitro/in situ produced DH maize plants. Biotechnol. Agricult. For. 25: 201-213. BECRAFT P.W., K.J. LI, N. DEY, Y. ASUNCION CRABB, 2002 The maize dek1 gene functions in embryonic pattern formation and cell fate specification. Development 129: 5217-5225. BELLMANN R., W. WERR, 1992 Zmhox1a, the product of a novel maize homeobox gene, interacts with the Shrunken 26 bp feedback control element. EMBO J. 11: 3367-3374. BOMMERT P., W. WERR, 2001 Gene expression patterns in the maize caryopsis: clues to decisions in embryo and endosperm development. Gene 271: 131-142. BOWMAN J., 1993 Embryogenesis. pp. 351-401. In: J. Bowman (Ed.), Arabidopsis: An atlas of morphology and development. Springer Verlag, New York Berlin Heidelberg. BROWN W.V., 1960 The morphology of the grass embryo. Phytomorphology 10: 215-223. CAMPBELL S.A., D.E. CRONE, T.L. CECCARDI, T.J. CLOSE, 1998 A ca. 40 kDa maize (Zea mays L.) embryo dehydrin is encoded by the dhn2 locus on chromosome 9. Plant Mol. Biol. 38: 417423. CHRISTIANS E.S., Q. ZHOU, J. RENARD, I.J. BENJAMIN, 2003 Heat shock proteins in mammalian development. Semin. Cell. Dev. Biol. 14: 283-290. CLARK J.K., W.F. SHERIDAN, 1986 Developmental profiles of the maize embryo-lethal mutants dek22 and dek23. J. Hered. 77: 83-92. CLARK J.K., W.F. SHERIDAN, 1988 Characterization of the two embryo-lethal defective kernel mutants rgh*-1210 and fl*1253B: Effects on embryo and gametophyte development. Genetics 120: 279-290.

MAIZE EMBRYOGENESIS

CLARK J.K., W.F. SHERIDAN, 1991 Isolation and characterisation of 51 embryo-specific mutations of maize. Plant Cell 3: 935-951. CLOSE T.J., 1997 Dehydrins: a commonalty in the response of plants in dehydration and low temperature. Physiol. Plant. 100: 291-296. CONSONNI G., C. ASPESI, A. BARBANTE, S. DOLFINI, C. GIULIANI, A. GIULINI, S. HANSEN, R. BRETTSCHNEIDER, R. PILU, G. GAVAZZI, 2003 Analysis of four maize mutants arrested in early embryogenesis reveals an irregular pattern of cell division. Sex Plant Reprod. 15: 281-290. DANILEVSKAYA O.N., P. HERMON, S. HANTKE, M.G. MUSZYNSKI, K. KOLLIPARA, E.V. ANANIEV, 2003 Duplicated fie genes in maize: Expression pattern and imprinting suggest distinct functions. Plant Cell 15: 425-438. DEMERC M., 1923 Heritable characters of maize. XV Germless seed. J. Hered. 114: 297-300. DIBOLL A.G., 1968 Fine structural development of the megagametophyte of Zea mays following fertilization. Am. J. Bot. 55: 787-806. DIGONNET C., D. ALDON, N. LEDUC, C. DUMAS, M. ROUGIER, 1997 First evidence of a calcium transient in flowering plants at fertilization. Development 124: 2867-2874. DRESSELHAUS T., C. HAGEL, H. LORZ, E. KRANZ, 1996 Isolation of a full-length cDNA encoding calreticulin from a PCR library of in vitro zygotes of maize. Plant Mol. Biol. 31: 23-34. ELSTER R., P. BOMMERT, W.F. SHERIDAN, W. WERR, 2000 Analysis of four embryo-specific mutants in Zea mays reveals that incomplete radial organisation of the proembryo interfers with subsequent development. Dev. Genes Evol. 210: 300-310.

481

GIULIANI C., G. CONSONNI, G. GAVAZZI, M. COLOMBO, S. DOLFINI, 2002 Programmed cell death during embryogenesis in maize. Ann. Bot. 90: 287-292. GODAY A., A.B. JENSEN, F.A. CULIANEZMACIA, M.M. ALBA, M. FIGUERAS, J. SERRATOSA, M. TORRENT, M. PAGES, 1994 The maize abscisic acid-responsive protein Rab17 is located in the nucleus and interacts with nuclear-localization signals. Plant Cell 6: 351-360. GUERIN P., 1899 Recherches sur le développement du tégument séminal et du péricarpe des graminées. Ann. Sci. Nat. Bot. 9: 1-59. GUIGNARD J.L., 1975 Du cotyledon des monocotyledons. Phytomorphology 25: 193-200. GUIGNARD L., 1901 15: 37-50.

La double fécondation dans le maïs. J. Bot.

HANSTEIN J., 1870 Die Entwicklung des Keimes der Monokotylen und Dikotylen. Bot. Abh. 1: 1-112. HECKEL T., K. WERNER, W.F. SHERIDAN, C. DUMAS, P.M. ROGOWSKY, 1999 Novel phenotypes and developmental arrest in early embryo specific (emb) mutations of maize. Planta 210: 1-8. HOEBERICHTS F.A., E.J. WOLTERING, 2003 Multiple mediators of plant programmed cell death: interplay of conserved cell death mechanisms and plant-specific regulators. Bioessays 25: 47-57. INGRAM G.C., J.L. MAGNARD, P. VERGNE, C. DUMAS, P.M. ROGOWSKY, 1999 ZmOCL1, an HDGL2 family homeobox gene, is expressed in the outer cell layer throughout maize development. Plant Mol. Biol. 40: 343-354.

ENGELL K., 1989 Embryology of barley: Time course and analysis of controlled fertilization and early embryo formation based on serial sections. Nord. J. Bot. 9: 265-280.

INGRAM G.C., C. BOISNARD-LORIG, F. DEGUERRY, C. DUMAS, P.M. ROGOWSKY, 2000 Expression patterns of genes encoding HDZipIV homeodomain proteins define specific domains in the maize embryos and meristems. Plant J. 22: 401-414.

FAURE J.E., C. DIGONNET, C. DUMAS, 1994 An in vitro system for adhesion and fusion of maize gametes. Science 263: 15981600.

JONES D.F., 1920 Heritable characters of maize. IV A lethal factor-defective seeds. J. Hered. 11: 161-167.

FISCHER C., V. SPETH, S. FLEIG-EBERENZ, G. NEUHAUS, 1997 Induction of Zygotic Polyembryos in Wheat: Influence of Auxin Polar Transport. Plant Cell 9: 1767-1780.

JOSE-ESTANYOL M., L. RUIZ-AVILA, P. PUIGDOMENECH, 1992 A maize embryo-specific gene encodes a proline-rich and hydrophobic protein. Plant Cell 4: 413-423. JÜRGENS G., 2001 Apical-basal pattern formation in Arabidopsis embryogenesis. EMBO J. 20: 3609-3616.

FOSTER T., J. YAMAGUCHI, B.C. WONG, B. VEIT, S. HAKE, 1999 Gnarley1 is a dominant mutation in the knox4 homeobox gene affecting cell shape and identity. Plant Cell 11: 12391252.

JÜRGENS G., 2003 Growing up green: cellular basis of plant development. Mech. Dev. 120: 1395-1406.

FU S., R. MEELEY, M.J. SCANLON, 2002 Empty pericarp2 encodes a negative regulator of the heat shock response and is required for maize embryogenesis. Plant Cell 14: 3119-3132.

JÜRGENS G., R.A. TORRES RUIZ, T. BERLETH, 1994 Embryonic pattern formation in flowering plants. Annu. Rev. Genet. 28: 351-371.

FU S., M.J. SCANLON, 2004 Clonal mosaic analysis of EMPTY PERICARP2 reveals nonredundant functions of the duplicated HEAT SHOCK FACTOR BINDING PROTEINs during maize shoot development. Genetics 167: 1381-1394.

KAPLAN D.R., T.J. COOKE, 1997 Fundamental Concepts in the Embryogenesis of Dicotyledons: A Morphological Interpretation of Embryo Mutants. Plant Cell 9: 1903-1919.

GARAY-ARROYO A., J.M. COLMENERO-FLORES, A. GARCIARRUBIO, A.A. COVARRUBIAS, 2000 Highly hydrophilic proteins in prokaryotes and eukaryotes are common during conditions of water deficit. J. Biol. Chem. 275: 5668-5674. GAUT B.S., J.F. DOEBLEY, 1997 DNA sequence evidence for the segmental allotetraploid origin of maize. Proc. Natl. Acad. Sci. USA 94: 6809-6814.

KERSTETTER R., E. VOLLBRECHT, B. LOWE, B. VEIT, J. YAMAGUCHI, S. HAKE, 1994 Sequence analysis and expression patterns divide the maize knotted1-like homeobox genes into two classes. Plant Cell 6: 1877-1887. KERSTETTER R.A., D. LAUDENCIA-CHINGCUANCO, L.G. SMITH, S. HAKE, 1997 Loss-of-function mutations in the maize homeobox gene, knotted1, are defective in shoot meristem maintenance. Development 124: 3045-3054.

482

V. VERNOUD, M. HAJDUCH, A.S. KHALED, N. DEPÈGE, P.M. ROGOWSKY

KIESSELBACH T.A., 1949 The structure and reproduction of corn. Agricultural Experiment Station, University of Nebraska, Lincoln, Nebraska 161.

MANSFIELD S.G., L.G. BRIARTY, 1992 Cotyledon cell development in Arabidopsis thaliana during reserve deposition. Can. J. Bot. 70: 151-164.

KLINGE B., W. WERR, 1995 Transcription of the Zea mays homeobox (ZmHox) genes is activated early in embryogenesis and restricted to meristems of the maize plant. Dev. Genet. 16: 349-357.

MATTHYS-ROCHON E., F. PIOLA, E. LE DEUNFF, R. MOL, C. DUMAS, 1998 In vitro development of maize immature embryos: a tool for embryogenesis analysis. J. Exp. Bot. 49: 839-845.

KRANZ E., H. LÖRZ, 1993 In vitro fertilization with isolated single gametes results in zygotic embryogenesis and fertile maize plants. Plant Cell 5: 739-746.

MAYER U., R.A.T. RUIZ, T. BERLETH, S. MISERA, G. JURGENS, 1991 Mutations affecting body organization in the Arabidopsis embryo. Nature 353: 402-407.

KRIZ A.L., 1989 Characterization of embryo globulins encoded by the maize Glb genes. Biochem. Genet. 27: 239-251.

MCCARTY D.R., T. HATTORI, C.B. CARSON, V. VASIL, M. LAZAR, I.K. VASIL, 1991 The Viviparous-1 developmental gene of maize encodes a novel transcriptional activator. Cell 66: 895-905.

LEDUC N., E. MATTHYS-ROCHON, M. ROUGIER, L. MOGENSEN, P. HOLM, J.L. MAGNARD, C. DUMAS, 1996 Isolated maize zygotes mimic in vivo embryonic development and express microinjected genes when cultured in vitro. Dev. Biol. 177: 190-203.

MECHIN V., T. BALLIAU, S. CHATEAU-JOUBERT, M. DAVANTURE, O. LANGELLA, L. NEGRONI, J.L. PRIOUL, C. THEVENOT, M. ZIVY, C. DAMERVAL, 2004 A two-dimensional proteome map of maize endosperm. Phytochem. 65: 1609-1618.

LEE J.M., M.E. WILLIAMS, S.V. TINGEY, J.A. RAFALSKI, 2002 DNA array profiling of gene expression changes during maize embryo development. Funct. Integr. Genomics 2: 13-27.

MILLER E.C., 1919 Development of the pistillate spikelet and fertilization in Zea mays. J. Agr. Res. 18: 255-293.

LEE K., A.H. HUANG, 1994 Genes encoding oleosins in maize kernel of inbreds Mo17 and B73. Plant Mol. Biol. 26: 19811987. LI Z.H., P.D. MATTHEWS, B. BURR, E.T. WURTZEL, 1996 Cloning and characterization of a maize cDNA encoding phytoene desaturase, an enzyme of the carotenoid biosynthetic pathway. Plant Mol. Biol. 30: 269-279. LID S.E., D. GRUIS, R. JUNG, J.A. LORENTZEN, E. ANANIEV, M. CHAMBERLIN, X. NIU, R. MEELEY, S. NICHOLS, O.A. OLSEN, 2002 The defective kernel 1 (dek1) gene required for aleurone cell development in the endosperm of maize grains encodes a membrane protein of the calpain gene superfamily. Proc. Natl. Acad. Sci. USA 99: 5460-5465. LONG J.A., E.I. MOAN, J.I. MEDFORD, M.K. BARTON, 1996 A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature 379: 66-69. LUO M., P. BILODEAU, E.S. DENNIS, W.J. PEACOCK, A. CHAUDHURY, 2000 Expression and parent-of-origin effects for FIS2, MEA, and FIE in the endosperm and embryo of developing Arabidopsis seeds. Proc. Natl. Acad. Sci. USA 97: 10637-10642. MA Z., H.K. DOONER, 2004 A mutation in the nuclear-encoded plastid ribosomal protein S9 leads to early embryo lethality in maize. Plant J. 37: 92-103. MAGNARD J.L., T. HECKEL, A. MASSONNEAU, J.P. WISNIEWSKI, S. CORDELIER, H. LASSAGNE, P. PEREZ, C. DUMAS, P.M. ROGOWSKY, 2004 Morphogenesis of maize embryos requires ZmPRPL35-1 encoding a plastid ribosomal protein. Plant Physiol. 134: 649663. MANGELSDORF P.C., 1923 The inheritence of defective seeds in maize. J. Hered. 14: 119-125. MANSFIELD S.G., L.G. BRIARTY, 1991 Early embryogenesis in Arabidopsis thaliana. II. The developing embryo. Can. J. Bot. 69: 461-476. MANSFIELD S.G., L.G. BRIARTY, S. ERNI, 1991 Early embryogenesis in Arabidopsis thaliana. I. The mature embyro sac. Can. J. Bot. 69: 447-460.

MOL R., E. MATTHYS-ROCHON, C. DUMAS, 1993 In vitro culture of fertilized embryo sacs of maize: Zygotes and two-celled proembryos can develop into plants. Planta 189: 213-217. MOL R., E. MATTHYS-ROCHON, C. DUMAS, 1994 The kinetics of cytological events during double fertilisation in Zea mays L. Plant J. 5: 197-206. MOLDENHAUER K.A.K., J.H. GIBBONS, 2003 Rice morphology and development. pp. 103-127. In: C.W. Smith (Ed.), Rice: Origin, History, Technology and Production. John Wiley & Sons, Inc. NARDMANN J., J. JI, W. WERR, M.J. SCANLON, 2004 The maize duplicate genes narrow sheath1 and narrow sheath2 encode a conserved homeobox gene function in a lateral domain of shoot apical meristems. Development 131: 2827-2839. NEILL S.J., R. HOGAN, A.D. PARRY, 1986 The carotenoid and abscisic acid content of viviparous kernels and seedlings of Zea mays L. Planta 169: 87-96. NEUFFER M.G., W.F. SHERIDAN, 1980 Defective kernel mutants of maize: I. Genetic and lethality studies. Genetics 95: 929-944. NIOGRET M.F., F.A. CULIANEZMACIA, A. GODAY, M.M. ALBA, M. PAGES, 1996 Expression and cellular localization of rab28 mRNA and Rab28 protein during maize embryogenesis. Plant J. 9: 549-557. NORSTOG K., 1972 Early development of the barley embryo: fine structure. Am. J. Bot. 59: 123-132. PETOLINO J.F., A.M. JONES, 1986 Anther culture of elite genotypes of maize. Crop Sci. 26: 1072-1074. PILU R., G. CONSONNI, E. BUSTI, A.P. MACCABE, A. GIULINI, S. DOLFINI, G. GAVAZZI, 2002 Mutations in two independent genes lead to suppression of the shoot apical meristem in maize. Plant Physiol. 128: 502-511. PLA M., J. VILARDELL, M.J. GUILTINAN, W.R. MARCOTTE, M.F. NIOGRET, R.S. QUATRANO, M. PAGÈS, 1993 The cis-regulatory element CCACGTGG is involved in ABA and water-stress responses of the maize gene rab 28. Plant Mol. Biol. 21: 259-266. RANDOLPH L.F., 1936 Developmental morphology of the caryopsis in maize. J. Agric. Res. 53: 882-916.

MAIZE EMBRYOGENESIS

RIERA M., M. FIGUERAS, C. LOPEZ, A. GODAY, M. PAGES, 2004 Protein kinase CK2 modulates developmental functions of the abscisic acid responsive protein Rab17 from maize. Proc. Natl. Acad. Sci. USA 101: 9879-9884. RUIZ-AVILA L., S.R. BURGESS, V. STIEFEL, M.D. LUDEVID, P. PUIGDOMENECH, 1992 Accumulation of cell wall hydroxyprolinerich glycoprotein mRNA is an early event in maize embryo cell differentiation. Proc. Natl. Acad. Sci. USA 89: 2414-2418. SANCHEZ-MARTINEZ D., P. PUIGDOMENECH, M. PAGES, 1986 Regulation of gene-expression in developing zea-mays embryos protein-synthesis during embryogenesis and early germination of maize. Plant Physiol. 82: 543-549. SCANLON M.J., A.M. MYERS, 1998 Phenotypic analysis and molecular cloning of discolored-1 (dsc1), a maize gene required for early kernel development. Plant Mol. Biol. 37: 483-493. SCANLON M.J., P.S. STINARD, M.G. JAMES, A.M. MYERS, D.S. ROBERTSON, 1994 Genetic analysis of 63 mutations affecting maize kernel development isolated from Mutator stocks. Genetics 136: 281-294. SCANLON M.J., A.M. MYERS, R.G. SCHNEEBERGER, M. FREELING, 1997 The maize gene empty pericarp-2 is required for progression beyond early stages of embryogenesis. Plant J. 12: 901-909. SCANLON M.J., D.C. HENDERSON, B. BERNSTEIN, R. JUNG, J.A. LORENTZEN, E. ANANIEV, M. CHAMBERLIN, X. NIU, R. MEELEY, S. NICHOLS, O.A. OLSEN, 2002 SEMAPHORE1 functions during the regulation of ancestrally duplicated knox genes and polar auxin transport in maize. Development 129: 2663-2673. SCHEL J.H.N., H. KIEFT, A.A.M. LAMMEREN, 1984 Interactions between embryo and endosperm during early developmental stages of maize caryopses (Zea mays). Can. J. Bot. 62: 28422853. SCHNEEBERGER R.G., P.W. BECRAFT, S. HAKE, M. FREELING, 1995 Ectopic expression of the knox homeo box gene rough sheath1 alters cell fate in the maize leaf. Genes Dev. 9: 2292-2304. SCHOLTEN S., H. LORZ, E. KRANZ, 2002 Paternal mRNA and protein synthesis coincides with male chromatin decondensation in maize zygotes. Plant J. 32: 221-231. SCHULZ S.R., W.A. JENSEN, 1968a Capsella embryogenesis: the early embryo. J. Ultrastruct. Res. 22: 376-392. SCHULZ S.R., W.A. JENSEN, 1968b Capsella embryogenesis: the egg, zygote, and young embryo. Am. J. Bot. 55: 807-819. SHERIDAN W.F., 1995 Genes and embryo morphogenesis in angiosperms. Dev. Genet. 16: 291-297. SHERIDAN W.F., M.G. NEUFFER, 1980 Defective kernel mutants of maize: II. Morphological and embryo culture studies. Genetics 95: 945-960. SHERIDAN W.F., M.G. NEUFFER, 1982 Maize developmental mutants. Embryos unable to form leaf primordia. J. Hered. 73: 318-329. SHERIDAN W.F., Y.R. THORSTENSON, 1986 Developmental profiles of three embryo-lethal maize mutants lacking leaf primordia: ptd*-1130, cp*-1418 and bno*-747B. Dev. Genet. 7: 35-49.

483

SHERIDAN W.F., J.K. CLARK, 1993 Mutational analysis of morphogenesis of the maize embryo. Plant J. 3: 347-358. SMART M.G., T.P. O’BRIEN, 1983 The development of the wheat embryo in relation to the neighbouring tissues. Protoplasma 114: 1-13. SMITH L.G., D. JACKSON, S. HAKE, 1995 Expression of knotted1 marks shoot meristem formation during maize embryogenesis. Dev. Genet. 16: 344-348. SOSSOUNTZOV L., L. RUIZ-AVILA, F. VIGNOLS, A. JOLLIOT, V. ARONDEL, F. TCHANG, M. GROSBOIS, F. GUERBETTE, E. MIGINIAC, M. DELSENY, P. PUIGDOMENECH, J.-C. KADER, 1991 Spatial and temporal expression of a maize lipid transfer protein gene. Plant Cell 3: 923-933. STIEFEL V., E.L. BECERRA, R. ROCA, M. BASTIDA, T. JAHRMANN, E. GRAZIANO, P. PUIGDOMENECH, 1999 TM20, a gene coding for a new class of transmembrane proteins expressed in the meristematic tissues of maize. J. Biol. Chem. 274: 2773427739. TIMMERMANS M.C., A. HUDSON, P.W. BECRAFT, T. NELSON, 1999 ROUGH SHEATH2: a Myb protein that represses knox homeobox genes in maize lateral organ primordia. Science 284: 151-153. TRUE R.H., 1893 On the development of the caryopsis. Bot. Gaz. 18: 212-226. VAN

LAMMEREN A.A.M., 1986a A comparative ultrastructural study of the megagametophytes in two strains of Zea mays L. before and after fertilization. Agric. Univ. Wageningen Pap. 86: 1-37.

VAN

LAMMEREN A.A.M., 1986b Developmental morphology and cytology of the young maize embryo (Zea mays L.). Acta Bot. Neerl. 35: 169-188.

VEIT B., S.P. BRIGGS, R.J. SCHMIDT, M.F. YANOFSKY, S. HAKE, 1998 Regulation of leaf initiation by the terminal ear 1 gene of maize. Nature 393: 166-168. VIELLE-CALZADA J.P., R. BASKAR, U. GROSSNIKLAUS, 2000 Delayed activation of the paternal genome during seed development. Nature 404: 91-94. VOGLER H., C. KUHLEMEIER, 2003 Simple hormones but complex signalling. Curr. Opin. Plant Biol. 6: 51-56. VOLLBRECHT E., B. VEIT, N. SINHA, S. HAKE, 1991 The developmental gene Knotted-1 is a member of a maize homeobox gene family. Nature 350: 241-243. WEATHERWAX P., 1919 Gametogenesis and fecundation in Zea mays as the basis of xenia and heredity in endosperm. Bull. Torrey Bot. Club 46: 73-90. WEIJERS D., N. GELDNER, R. OFFRINGA, G. JURGENS, 2001 Seed development: Early paternal gene activity in Arabidopsis. Nature 414: 709-710. WHITE C.N., W.M. PROEBSTING, P. HEDDEN, C.J. RIVIN, 2000 Gibberellins and seed development in maize. I. Evidence that gibberellin/abscisic acid balance governs germination versus maturation pathways. Plant Physiol. 122: 1081-1088.