Plant Cell, Tissue and Organ Culture 74: 201–228, 2003. 201 2003 Kluwer Academic Publishers. Printed in the Netherlands.

Review of Plant Biotechnology and Applied Genetics

Transition of somatic plant cells to an embryogenic state

*´ ´Attila Feher , Taras P. Pasternak & Denes DuditsLaboratory of Functional Cell Biology, Institute of Plant Biology, Biological Research Center, Hungarian

*´Academy of Sciences, Temesvari krt. 62, H-6726 Szeged, Hungary ( requests for offprints; Fax: �36-62-433-434; E-mail: fehera@nucleus.szbk.u-szeged.hu)

Received 16 January 2002; accepted in revised form 17 February 2003

2�Key words: arabinogalactan proteins, Ca , cell division, cell polarity, chromatin, differential screening, pH,protein turnover, somatic embryogenesis, stress

Abstract

Under appropriate in vivo or in vitro conditions, certain somatic plant cells have the capability to initiateembryogenic development (somatic embryogenesis). Somatic embryogenesis provides an unique experimentalmodel to understand the molecular and cellular bases of developmental plasticity in plants. In the last few years,the application of modern experimental techniques, as well as the characterization of Arabidopsis embryogenesismutants, have resulted in the accumulation of novel data about the acquisition of embryogenic capabilities bysomatic plant cells. In this review, we summarize relevant experimental observations that can contribute to thedescription and definition of a transitional state of somatic cells induced to form totipotent, embryogenic cells.During this somatic-to-embryogenic transition, cells have to dedifferentiate, activate their cell division cycle andreorganize their physiology, metabolism and gene expression patterns. The roles of stress, endogenous growthregulators and chromatin remodelling in the coordinated reorganization of the cellular state are especiallyemphasized.

Introduction flexible and this is linked to the reversibility of thedifferentiation state of somatic plant cells. Under

The life strategies of multicellular plant and animal extreme conditions, these cells have to change theirspecies are strikingly different. While active move- fate: either they have to die (‘apoptosis’) or dedif-ment is characteristic of animal organisms, plants are ferentiate and divide, depending on the needs of thesessile. This basic difference is reflected in the evolu- organism. In general, it can be stated that the de-tion of different approaches to short-term environ- velopmental program of plants is much more open tomental adaptation by individuals. Animals respond to alternative pathways compared to that of animals andenvironmental signals or conditions by changes in this is manifested at the level of cellular differentia-their behaviour, while plants accommodate environ- tion (for reviews, Walbot, 1996; Sultan, 2000).mental effects by altering metabolism and/or de- One of the most extreme examples of flexibility invelopment. In the absence of the possibility for de- plant development is the capability of several cellvelopmental cell migration and due to the need for types, in addition to the zygote, to initiate em-continuous organogenesis during their postem- bryogenic development. In flowering plants, sexualbryogenic development, plants maintain organ-form- reproduction involves double fertilization that gener-ing cell files, the meristems. The formation and activi- ates the embryo and the endosperm simultaneously.ty of meristems are highly dependent on environmen- Meiosis precedes the formation of gametes and fertili-tal, as well as developmental (hormonal), factors. zation restores the somatic chromosome number.Thus, the ontogenic program of plants is highly However, in many vascular plants, sexual reproduc-

202

tion is regularly combined with or replaced by differ- pathway of in vivo events most closely (Kranz, 1999).ent kinds of asexual reproduction (reviewed, e.g., by Despite some technical limitations, this experimentalSharma and Thorpe, 1995). In addition to vivipary or approach has provided basic, novel information aboutvegetative reproduction, in seed plants, embryos and early molecular events following fertilization (Kranz,seeds can be generated without fertilization through 1999; Faure and Dumas, 2001). A mass of cells withvarious pathways, which are collectively referred to acquired embryogenic competence can be studied inas apomixis (for reviews, Koltunow, 1993; Koltunow specific cultures of microspores or somatic cellset al., 1995). During gametophytic apomixis, the (Dudits et al., 1991; Reynolds, 1997; Osuga et al.,female gametophyte is formed from an unreduced 1999). There is debate, however, as to the suitabilityembryo-sac initial (diplospory) or from somatic cells of these embryogenic systems as models for zygoticin the nucellus or chalaza (apospory). In the latter embryogenesis (Zimmerman, 1993).case, which is also called adventitious embryony, one Events of zygotic embryogenesis, mainly based onor more embryos are formed directly from the somatic studies in Arabidopsis, can be subdivided into differ-cells in the ovule as nucellar or integumental out- ent stages (Jurgens et al., 1991; Jurgens, 1992; Mayergrowths. In most cases, pollination and fertilization and Jurgens, 1998) such as the first asymmetricare prerequisites for the initiation of adventitious division, and formation of the globular, oblong, heart,embryos, although in some situations pollination is torpedo, cotyledonary and finally the mature dehy-sufficient. Adventitious and zygotic embryos in the drated embryo. Although somatic (and microspore)embryo sac compete with each other until one (or a embryogenesis may serve as a model for most of thesefew) of them dominates and gives rise to the mature stages, there are several obvious differences:embryo(s). How and why somatic cells of the ovule – the lack of endosperm differentiation,change their developmental fate and gain em- – the missing or retarded suspensor development,bryogenic potency is not known. In adventitious – the lack of embryo desiccation and dormancyembryony, although embryogenesis starts from during somatic embryogenesis, andsomatic (in most cases, nucellar) cells, and thus falls – the acquisition phase of embryogenic compe-into the category of somatic embryogenesis, the de- tence by a somatic cell prior to the initiation ofveloping embryos occupy the embryo sac, use the embryo development (for review, Dodeman etendosperm and ‘normal’ seeds are the end product. In al., 1997).other examples, as is the case of Bryophyllum (Yar- While the zygote, formed as a consequence of egg cellbrough, 1932) and Malaxis (Taylor, 1967) species, fertilization, is clearly determined to follow the em-somatic embryos can arise as vegetative propagules bryogenic cell fate, in other forms of plant embryo-from cells at leaf margins. genesis, including apomixis and somatic embryogene-

In addition to these natural, in vivo, forms of sis, there is a transition phase during which competentembryogenesis, there exist at least three ways to and embryogenic cell types are formed. This transi-induce embryo development from in vitro cultured tion phase is very difficult to define, but an under-plant cells (for a comprehensive review see Mor- standing of the underlying mechanisms can providedhorst et al., 1997): in vitro fertilization (for review, insight into the developmental strategy of plants.Kranz, 1999), from microspores (for review, In this review, we aim to summarize selected,Reynolds, 1997) and in vitro somatic embryogenesis recent information on the induction phase of somatic(for review, Dudits et al., 1995). In vitro embryo- embryogenesis up to the formation of globular em-genesis, in addition to its usefulness for cloning and bryos. Early cellular and molecular events of somaticvegetative propagation of a given plant individual, embryogenesis will be described in relation to charac-can serve as a model system to study the molecular, teristic, although overlapping, phases of dedifferentia-cytological, physiological and developmental events tion, cell division, acquisition of competence, induc-underlying plant embryogenesis (Dodeman et al., tion and determination of the embryogenic cell fate.1997). The main advantage of these in vitro ex-perimental systems is that embryogenic cells areaccessible for manipulation by most cellular and Induction of somatic embryogenesismolecular techniques, in contrast to gametic cells andzygotes, which develop embedded in maternal tissues. Stress: the force to switch

In vitro zygote development, initiated by the artifi-cial fusion of isolated gametes, probably follows the As shown by several experimental observations, the

203

differentiated fate of plant cells, dependent on posi- concentration, heavy metal ions or osmotic stresstional information and developmental signals, can be positively influenced somatic embryo induction ineasily altered under in vitro conditions. Drastic diverse plant species (reviewed by Dudits et al.,changes in the cellular environment, such as exposing 1995). These procedures were accompanied by in-wounded cells or tissues to sub-optimal nutrient or creased expression of diverse stress-related genes,hormone supply (e.g., in vitro culture conditions), evoking the hypothesis that somatic embryogenesis isgenerate significant stress effects. The response to an adaptation process of in vitro cultured plant cellsstress conditions depends on two main parameters: the (Dudits et al., 1995).level of stress and the physiological state of the cells.If the stress level exceeds cellular tolerance, the cells Growth regulators: outside and insidedie. In contrast, lower levels of stress enhance metab-olism and induce adaptation mechanisms (Lichten- Hormones are the most likely candidates in the regu-thaler, 1998). Adaptations include the reprogramming lation of developmental switches. Auxins and cyto-of gene expression, as well as changes in the physi- kinins are the main growth regulators in plants in-ology and metabolism of the cells. Stress alters volved in the regulation of cell division and differen-source /sink regulation by switching on genes of sink- tiation. The influences of exogenously applied auxins,specific enzymes in parallel with stress defence genes preferentially 2,4-dichlorophenoxyacetic acid (2,4-D),(Roitsch, 1999). This transient cell state induced by on the induction of somatic embryogenesis are wellstress conditions can be characterized by extensive documented (for reviews, e.g., Dudits et al., 1991;cellular reorganization and allows for a developmen- Yeung, 1995). However, embryo development intal switch, if appropriate signals are perceived. somatic tissues has been reported in the absence of

In vitro tissue culture conditions expose the ex- growth regulators (e.g., Choi et al., 1998), as well asplants to significant stresses, as they are removed in the presence of other growth regulators such asfrom their original tissue environment and placed on cytokinin (e.g., Sagare et al., 2000) or abscisic acidsynthetic media containing non-physiological concen- (e.g., Nishiwaki et al., 2000) (for reviews, Nomuratrations of growth regulators, salts and organic com- and Komamine, 1995; Yeung, 1995, in Leguminosaeponents. Wounding itself is a significant signal in the Lakshmanan; Taji, 2000). Non-hormonal inducersinduction of dedifferentiation: most of the genes can also be used to promote the somatic /embryogenicexpressed in freshly isolated leaf protoplasts, not only transition. Such inducers include high sucrose con-those involved in the stress response, are already centration or osmotic stress (Kamada et al., 1993),induced as a result of wounding (Grosset et al., 1990). heavy metal ions (Kiyosue et al., 1990; Pasternak etIn alfalfa leaf protoplasts, oxidative stress-inducing al., 2002) and high temperature (Kamada et al.,compounds increased the endogenous auxin level of 1989). In most of the cases where no growth reg-the cells and promoted dedifferentiation, as indicated ulators were included for embryogenic induction,by faster acidification of the medium and earlier cell somatic embryos formed directly on the surfaces ofdivision at a smaller size (Pasternak et al., 2002). explants without recognisable callus formation.Dedifferentiation, in many cases, can be clearly corre- Based on the wide variation of inducer types,lated with stress and/or auxin responses of cells. somatic embryogenesis cannot be defined as a specific

Stresses not only promote dedifferentiation, but response to one or more exogenously applied growthalso can be used to induce somatic embryo formation. regulators. Rather, these observations indicate a criti-For example, embryogenic cells could be formed cal role of stress as an embryogenic stimulus (see alsofrom alfalfa leaf protoplasts in response to different Stress: the force to switch section). Endogenous hor-oxidative stress-inducing compounds in the presence mone levels, however, can be considered as major

´of auxin and cytokinin (Feher et al., 2001, 2002; factors in determining the specificity of cellular re-Pasternak et al., 2002). Mitogen-activated protein sponses to these rather general stress stimuli. Duringkinase (MAPK) phosphorylation cascades may link the last few years, a large body of experimentaloxidative stress responses to auxin signalling and cell observations has accumulated on the central roles ofcycle regulation (as reviewed by Hirt, 2000). For endogenous indoleacetic acid (IAA) and abscisic acidexample, the tobacco MAPK Kinase Kinase (ABA) levels during the early phases of embryo-(MAPKKK), NPK1, was shown to be involved in genesis.oxidative stress response, auxin signalling and cell In addition to the absolute requirement of exogen-cycle regulation (Hirt, 2000). Wounding, high salt ous auxins for sustained growth in in vitro cultures,

204

plant cells may produce substantial amounts of the important factor in somatic embryo formation onnative auxin, IAA. Higher endogenous IAA concen- cotyledon explants of ginseng, which did not requiretration has been shown to be associated with increased exogenous growth regulator application (Choi et al.,embryogenic response in various species /explants 1997). These observations altogether suggest that(Rajasekaran et al., 1987; Ivanova et al., 1994; temporal and spatial changes in endogenous auxinMichalczuk and Druart, 1999; Jimenez and Bangerth, levels are important factors controlling the em-2001a, b, c). In carrot cells, exogenous 2,4-D stimu- bryogenic cell fate.lated the accumulation of large amounts of endogen- Among the different auxin analogues used to in-ous IAA (Michalczuk et al., 1992a, b). These authors duce somatic embryogenesis, 2,4-D is by far the mosthypothesized that embryogenic competence of carrot efficient and, therefore, this synthetic growth reg-cells is closely associated with the several-fold in- ulator is used in the majority of embryogenic cell andcrease in endogenous IAA levels due to the presence tissue culture systems. It can be suggested that 2,4-D,of 2,4-D. It was suggested that this synthetic com- above a certain concentration, has a dual effect inpound acts indirectly by disturbing endogenous auxin these cultures, as an auxin (directly or through endog-metabolism and the direct auxin effect of 2,4-D is less enous IAA metabolism; see above) and as a stressor

´significant. (Feher et al., 2001, 2002). 2,4-D is an auxinic her-In alfalfa leaf protoplasts cultured in the presence bicide with diverse effects associated with its

of 2,4-D, the endogenous IAA levels increased con- phytotoxic activity, which cannot be ascribed simplysiderably during the first 2–3 days of culture (Paster- to an auxin overdose. It has been suggested that 2,4-D

nak et al., 2002). This increase was transient and (or other auxin-type herbicides) affect electrical pat-comparable under both embryogenic and non-em- terns (Goldsworthy and Mina, 1991), membrane per-bryogenic conditions; however, the timing of the meability (Schauf et al., 1987), IAA binding to thesynthesis peak showed approximately a 1-day delay auxin-binding protein ABP1 (Deshpande and Hall,under non-embryogenic conditions. This peak shift 2000) and photosynthesis of algae (Matsue et al.,was strongly correlated with the fate of the cells that 1993; Fargasova, 1994). Recently, auxinic herbicideswere manipulated by altering either 2,4-D, iron con- have been shown to interact with ethylene and ABAcentration (the level of stress) or medium pH (Paster- synthesis, increasing the cellular levels of these so-nak et al., 2002). called ‘stress’ hormones (Grossmann, 2000; Wei et

A similar peak of endogenous IAA level has been al., 2000).observed in immature zygotic sunflower embryos In carrot, simple application of abscisic acid to

´induced to form somatic embryos (Charriere et al., seedlings efficiently induced somatic embryo forma-1999). In the sunflower system, cells can be induced tion (Nishiwaki et al., 2000). However, this responseto form either adventitious shoots or somatic embryos was dependent on the presence of the shoot apices, theby simply modifying the sucrose content in the culture main source of auxin in the seedlings. This findingmedium. The tissues grown under embryogenic con- suggested that, as a stress signal, exogenous ABA wasditions showed a 4-fold increase in their IAA content effective only in the presence of an endogenous auxinas compared to those tissues that followed the supply. The level of endogenous IAA was influencedcaulogenic pathway. The timing of the peak (at ap- (increased) by the application of ABA to immatureprox. 24 h of culture) correlated well with the time of zygotic sunflower embryos and resulted in the induc-the irreversible determination of the morphogenetic tion of somatic embryogenesis under sucrose con-response. Immuno-cytochemical localization of IAA ditions which otherwise allow only caulogenesis to

´in the immature zygotic embryos before, during and occur (Charriere et al., 1999). Amounts of endogen-after the induction of somatic embryo development ous ABA were inversely correlated with the em-provided direct evidence that an endogenous auxin bryogenic capacity of alfalfa cultivars (Ivanova et al.,pulse may be one of the first signals leading to 1994). This observation seems to be contradictory tosomatic embryogenesis (Thomas et al., 2002). the positive role of exogenous ABA on somatic

An auxin surge has been reported to take place after embryo induction (e.g., Nishiwaki et al., 2000), but itfertilization in carrot zygotic embryos as well (Rib- can be explained by the differences in the effects ofnicky et al., 2001), further emphasizing the signifi- exogenously applied and endogenous hormones: acance of temporal changes in endogenous auxin levels relatively higher endogenous ABA level may result inin the expression of cellular totipotency. The polar a decreased response to exogenous stress or ABAtransport of endogenous auxin was found to be an signals.

205

Direct experimental evidence of the contribution of potential key genes with altered transcriptional profileendogenous ABA to the induction phase of somatic during this process. Most available data comes fromembryogenesis was provided by Senger et al. (2001). genes switched on in freshly isolated protoplasts. LeafThese authors used different approaches to reduce protoplasts are ideal experimental objects of studiescellular ABA levels in Nicotiana plumbaginifolia on cellular dedifferentiation: they are a population ofplants: a homozygous transgenic line was produced single cells derived from fully differentiated, special-which over-expressed an anti-abscisic acid single ized tissues and they respond to in vitro culturechain variable fragment (scFv) antibody, wild-type conditions uniformly and with reliable synchrony.cultures were treated with the ABA synthesis inhibitor The synthesis of proteins involved in photosynthesisflouridone and abscisic acid synthesis mutants (aba1 is completely inhibited, while other proteins, absentand aba2 ) were also used. In all cases of ABA from differentiated mesophyll cells, are actively syn-deficiency, disturbed morphogenesis could be ob- thesized in protoplasts as soon as they are isolatedserved at pre-globular embryoid formation, which from leaves (Fleck and Durr, 1980; Vernet et al.,could be reverted by exogenous ABA application. 1982). Among those genes responsible for the syn-

All of the above-described experimental results and thesis of these proteins, many are associated withobservations (Stress: the force to switch and Growth stress response, primary metabolism, cell wall syn-regulators: outside and inside sections) highlight the thesis or cell division (Grosset et al., 1990; Marty etimportance of the interaction between auxin and al., 1993; Yu et al., 1999). During dedifferentiation,stress /ABA signalling. One can hypothesize that the all these functions participate in the reorganization ofparallel induction of these pathways can lead to the cellular metabolism and the developmental switch.morphological and developmental changes observed Most of these genes, not only those related to stressduring the transition from somatic to embryogenic responses, have been induced by signals resultingcell types due to rapid de- and redifferentiation. The from the mechanical wounding of tissues duringsimultaneous activation of auxin and stress responses protoplast isolation (Grosset et al., 1990; Marty et al.,may be a key event in cellular adaptation, causing 1993). These data also emphasize the role of stress ingenetic, metabolic and physiological reprogramming, cellular reprogramming.which results in the embryogenic competence Nagata et al. (1994) reported the isolation of cDNA(totipotency) of somatic plant cells. clones for three auxin-regulated genes, parA, parB

and parC, from tobacco protoplasts that had regainedtheir division activity. The protein products of these

Changing fate: dedifferentiation and the auxin-induced genes belong to the plant glutathioneacquisition of cellular totipotency S-transferase (GST) family (for review, Marrs, 1996).

Their expression is induced in leaf protoplasts withinAcquisition of embryogenic competence largely relies 20 min of auxin application and fully repressed whenon dedifferentiation, a process whereby existing tran- the protoplasts start to divide (by 48 h of culture)scriptional and translational profiles are erased or (Takahashi et al., 1994). GSTs can have diverse roles,altered in order to allow cells to set a new develop- including the detoxification of xenobiotics, targetingmental program. The activation of cell division is their substrates for transmembrane transport andrequired to maintain the dedifferentiated cell fate, as protection from oxidative stress. It has also beenwell as for embryo differentiation. This is not only shown that some plant GSTs can bind, and probablytrue for those embryogenic systems where em- carry and store, natural auxin (IAA) (for review,bryogenic callus formation precedes somatic embryo Marrs, 1996). There is accumulating evidence that thedevelopment (‘indirect somatic embryogenesis’), but redox status and the glutathione content of cells havealso for those where somatic embryos develop on important roles in developmental processes and espe-primary explants without an intervening callus forma- cially in the reactivation of cell division (for review,tion (‘direct embryogenesis’). den Boer and Murray, 2000).

A carrot cDNA (Dcarg1 ), homologous to the parADedifferentiation-specific genes: are there any? sequence, was isolated as a differentially expressed

gene in hypocotyl cells, induced to form somaticAs dedifferentiation of plant cells is a basic phenom- embryos by a 2-h treatment with high concentrationsenon underlying in vitro cell and tissue culture re- of 2,4-D (Kitamiya et al., 2000). The expression of thesponses, there have been many attempts to isolate Dcarg1 gene was not responsive to stress treatments,

206

but a close relationship between somatic embryo 2001). Many of the early plant responses to auxin areinduction and Dcarg1 expression could be observed apparently mediated by members of a family of auxinunder different hormonal conditions. The expression responsive AUX/IAA proteins that dimerize with,of this gene was repressed during embryo develop- and thus inhibit, members of the auxin response factorment (Kitamiya et al., 2000). These observations may (ARF) family of transcription factors (as review see,indicate that par-related gene products (GSTs) are not e.g., Walker and Estelle, 1998). AUX/IAA proteinsonly involved in the process(es) of dedifferentiation are unstable and their degradation is triggered by anand reactivation of cell division, but may also have ubiquitin–protein ligase complex (SCF type) that isroles in developmental reprogramming. regulated by modification with an ubiquitin-related

In Nicotiana sylvestris, genes expressed in meso- protein (Gray et al., 1999; Gray and Estelle, 2000).phyll protoplasts, before the reinitiation of the DNA Recent genetic and biochemical data indicate thatreplicational activity, were identified through the auxin accelerates the degradation of the already short-differential screening of a cDNA library made from lived AUX/IAA proteins to derepress transcription by6-h-old protoplasts (Criqui et al., 1992). Three ARF proteins (Gray et al., 2001).mRNAs were characterized to represent a novel type The ubiquitin–proteasome pathway of protein deg-of trypsin inhibitor, a moderately anionic peroxidase radation is responsible for the degradation of most celland a protein of unknown function with a nuclear proteins; not only regulators and short-lived proteins,localization signal. The expression of these genes was but also those that serve structural roles and are longindependent of HgCl treatment, heat shock, pathogen lived (Rock et al., 1994). Ubiquitin, a small peptide of2

infection and tumour formation, and declined in 76 amino acids, can be covalently linked to otherprotoplast-derived cells entering cell division. Thus, proteins and serves as a biochemical tag that marksthey were hypothesized to have specific roles during proteins for degradation (Ingvardsen and Veierskov,cellular dedifferentiation. 2001). Dedifferentiation of mature tobacco cells was

As it can be seen, none of the above approaches shown to be accompanied by a sharp increase inresulted in the identification of master genes clearly ubiquitin gene expression (Jamet et al., 1990). Thisresponsible for the dedifferentiation processes. One elevation in ubiquitin gene activity may be related tocan hypothesize, however, that such master genes, if cellular reorganization during dedifferentiation andthey exist, should be involved in the general reprog- might be required for the selective destruction oframming of gene expression (chromatin remodelling, proteins associated with the previous, differentiatedtranscription machinery) and/or protein turnover. cell state. In line with this observation, it has recently

been reported that, in the case of reactivated tobaccoCellular reorganization: proteins are turned over protoplasts, entry into the DNA replication phase

could be prevented by MG132, a specific inhibitor ofEstablishment of a new cellular state is not only the ubiquitin–proteasome pathway, but not by leupep-governed at the level of gene expression, but requires tin, a protease inhibitor (Zhao et al., 2001).modification and/or removal of unnecessary poly- The overall changes in cell metabolism and thepeptides, as well as the proper folding of newly initiation of a new developmental program generatesynthesized proteins and protein complexes. This was the need for new proteins that must be folded anddemonstrated using two-dimensional protein electro- assembled properly to ensure their functions. Proteinphoresis, which showed that dedifferentiation and folding is controlled by special proteins exhibitingsubsequent somatic embryogenesis are associated chaperone activities, including so-called heat shockwith complex changes in the protein pattern (for proteins or HSPs (Morimoto, 1998). In addition torecent publications, e.g., Gianazza et al., 1992; Sallan- their expression during the heat shock response,drouze et al., 1999). Regulation of protein degra- which can also be considered as a transient state whendation is attracting increasing attention as an impor- cellular reprogramming is required, many HSP genestant regulatory mechanism in cell cycle progression have been reported to be developmentally regulatedand development in eukaryotic cells, including those (for reviews, Dudits et al., 1995; Schoffl et al., 1998).of plants (Udvardy, 1996; Genschik et al., 1998; For example, the expression of heat-shock genesCriqui et al., 2000). Auxin signalling has been shown occurred during embryogenesis from somatic cells into be at least partially based on the degradation of alfalfa and carrot (Dudits et al., 1991; Kitamiya et al.,specific proteins (Leyser, 2001; Rogg and Bartel, 2000). Although the induction of somatic and mi-

207

crospore embryogenesis involves the application of form a lateral root primordium (Beeckman et al.,some kind of stress, the developmental phase-depen- 2001). Tobacco microspores can be accumulated indent expression of HSPs, during zygotic embryogene- the G2 phase as a result of sucrose starvation stresssis and seed maturation, indicates that these proteins that induces their embryogenic development (Touraevmay have additional functions apart from those neces- et al., 1996). Alfalfa leaf protoplast populationssary to cope with environmental stresses (for review, treated with high concentrations of 2,4-D are alsoSchoffl et al., 1998). For example, in sunflower, enriched in cells with nuclear DNA content charac-expression of class II small HSPs is roughly parallel teristic of G2 cells (T. Pasternak, P. Miskolczi and A.

´with storage protein and lipid accumulation, whereas Feher, unpublished results). These observations indi-expression of class I small HSPs coincides with seed cate that developmental changes in cells, induced bydesiccation (Coca et al., 1994). In general, it can be environmental or developmental signals, are depen-hypothesized that the developmentally regulated ap- dent on the original DNA content and differentiationpearance of HSPs is associated with their chaperone state of the cells.function, promoting the assembly of newly synthes- The progression of cells through the differentized proteins during developmental transitions (for phases of the cell cycle is regulated primarily by thereviews, Dudits et al., 1995; Schoffl et al., 1998). activity of different cyclin-dependent kinase (CDK)

complexes in all eukaryotes, including plants (for´ ´Cell division and differentiation: being under reviews, Mironov et al., 1999; Meszaros et al., 2000).

control The expression of the gene coding for the kinasecomponent of the complex (Cdc2-related kinase) is

In attempts to understand the molecular levels of the induced by auxin (Hirt et al., 1991) and present nottransition from the somatic to embryogenic cell type, only in dividing plant cells, but also in division-the regulators of the cell cycle can be considered as competent cells (Hemerly et al., 1993), thus repre-key determinants during both dedifferentiation and senting a marker for the degree of differentiation. It isembryo formation. Recent developments in the field well known that in vitro grown plant cells, with fewof plant cell cycle research provide some clues to their exceptions, require exogenous growth regulators,roles. The core of the cell cycle machinery is well namely auxin and cytokinin, for sustained cell divi-conserved among eukaryotes, but obvious differences sion. In embryogenic alfalfa leaf protoplasts, thealso exist (for review, John, 1996). For example, in Cdc2MsA protein re-appeared in response to auxinplants, the synthesis of the new cell wall following treatment; however, the histone H1 phosphorylatingcytokinesis has to be integrated into the events of the activity of this protein was dependent on post-transla-cell cycle, which increases the complexity of regula- tional modifications that required the presence oftion (for review, Sylvester, 2000). The formation of cytokinin (Pasternak et al., 2000). These modifica-special cytoskeletal structures, such as the preproph- tions may include the de-phosphorylation of the CDKase band (PPB) and the phragmoplast, defines the protein on Thre14/Tyr15 regulatory residues (thisplane of cell division. In the absence of cell migration, mainly operates at G2-M transition, but a similar rolethese cellular components strongly influence mor- in G1-S transition cannot be excluded), or by thephogenesis in plants (reviewed by Hemerly et al., induction of the expression of the regulatory cyclin1999; Sylvester, 2000; Meijer and Murray, 2001). It is subunit cyclin D3 (at G1-S transition) (for review,also well accepted that the regulation of plant cell John, 1996).division is highly dependent on external signals af- Although the activation of cell division is a pre-fecting plant development and morphogenesis (see as requisite for embryogenic induction, the activities ofreviews, e.g., Dudits et al., 1998; den Boer and the cell division-regulating cdc2-like kinase complex-Murray, 2000). es are similar in alfalfa protoplasts grown under either

In plants, both the G1-S and G2-M phase transi- embryogenic or non-embryogenic conditions, in spitetions can be controlled by changes in environmental of the fact that earlier activation of the division of

´factors. Rhizobium-induced formation of root nodule embryogenic cells has been observed (Feher et al.,¨primordia from alfalfa cortical cells is initiated by the 2002; Pasternak et al., 2002). Bogre et al. (1990) also

division of cells in G0/G1 phase (Yang et al., 1994), reported that leaf protoplast-derived cells of an em-while in Arabidopsis lateral root development, G2- bryogenic versus a non-embryogenic alfalfa line werearrested pericycle cells respond to auxin treatment and activated earlier. Decreased cell doubling time has

208

been associated with the start of somatic embryo- to differentiated cells (see later). A correlation be-genesis in carrot (Warren, 1980). In the alfalfa leaf tween cell division competence and developmentalprotoplast system, high 2,4-D concentration or stress- determination was indicated by studies on trichomeinduced embryogenic competence were associated initiation in Arabidopsis (Larkin et al., 1997). Tri-with small cell size and earlier cell cycle activation chomes are normally initiated in a field of dividing(Pasternak et al., 2002). In carrot cell cultures, a 2,4-D cells and the ectopic overexpression of the regulatorsconcentration-dependent switch between elongation of trichome initiation can also result in trichomeand division has also been observed and it was formation, exclusively in proliferating tissues (Lloyddemonstrated that 2,4-D inhibited the elongation of et al., 1994; Hulskamp and Schnittger, 1998).cells, not directly, but as a consequence of promoting In vitro cell cultures can be initiated from explantstheir division (Lloyd et al., 1980). In the apical of either meristematic origin or from differentiatedmeristem, the files of smaller cells divided faster as organs. In both cases, the artificial initiation andcompared to their larger neighbours, although cell maintenance of cell division is a prerequisite for thesize and cell cycle time are not always correlated in generation and the establishment of the dedifferen-the same manner (reviewed by Dudits et al., 1998). tiated, meristematic cell fate. In cell cultures, theCell size, cell morphology and the timing of cell dividing cells can follow alternative developmentalactivation have been reported to be correlated with pathways such as unorganized callus growth, root andearlier endogenous auxin synthesis in alfalfa leaf shoot initiation or somatic embryo formation. In theprotoplast-derived cells (Pasternak et al., 2002). It can last case, division of somatic or dedifferentiated cellsbe speculated that this increase in endogenous IAA generates a cellular state similar to that of the zygote,level can be responsible for accelerating the dedif- formed after the fertilization of the egg cell.ferentiation process of an earlier activation of the Therefore, although cell division and differentia-division cycle, preventing cell elongation. tion are frequently considered to be mutually exclu-

Cell division and differentiation are developmental- sive processes (in differentiated somatic cells, divi-ly related processes. The complex relation between sion is arrested while the dedifferentiated cell fate isdifferentiation and cell division, however, might dif- linked with the capacity for sustained division both infer in the various stages of plant development, both in vitro and in planta), their relation is much morevivo and in vitro. In planta, the fusion of differen- complex. Cell cycle progression and differentiationtiated gametes activates a series of cell divisions in the are related but divergent processes in animal cells aszygote, resulting in embryo formation through struc- well, and their regulation is linked by componentstural and functional differentiation into suspensor having important regulatory roles near the G1/S cellcells and embryo proper cells, followed by the estab- cycle transition point (Gao and Zelenka, 1997; Stud-lishment of tissue layers (protoderm, basal tissue and zinski and Harrison, 1999). Some of these compo-vascular primordium), as well as root and shoot nents, such as the retinoblastoma (Rb) repressor pro-meristems (Jurgens, 1992). The sustained cell divi- teins, the E2F-type transcription factors, as well assion in the meristems provides a continuous supply of D-type cyclins and CDK inhibitors (CKIs), exist incells undergoing differentiation processes for or- plants as well (for reviews, e.g., Gutierrez, 1998;

´ ´ganogenesis, and the establishment of the plant body Mironov et al., 1999; Meszaros et al., 2000; Stals andduring the whole ontogenic development of a plant. Inze, 2001), indicating the existence of similar mech-

In vivo, the reactivation of the cell cycle can be a anisms. This is further supported by some experimen-starting point for new organ initiation caused by tal observations. In Arabidopsis, leaf aging and dif-changes in hormonal homeostasis (e.g., lateral root ferentiation are associated with increased expressionformation) or by external signals (e.g., root nodule of the ICK1 CDK inhibitor protein (Wang et al.,formation in response to Nod-factor). Asymmetric 1998). In maize leaf, the regions of older differen-distribution of cell contents among daughter cells, as a tiated cells contain more Rb protein than youngerresult of asymmetric cell division, can be decisive for tissues (Huntley et al., 1998). Ectopic expression oftheir further differentiated cell fate (for review, the cyclinD3 gene caused abnormal meristem and leafScheres and Benfey, 1999). An essential switch in development in transgenic plants (Riou-Khamlichi etdevelopmental pathways may be facilitated in divid- al., 1999). Overexpression of the same protein in calliing cells by a more open chromatin structure, altered induced differentiation and greening and preventedmetabolism and physiological homeostasis, compared shoot regeneration (Riou-Khamlichi et al., 1999).

209

Although additional evidence is required, it can be embryogenic development. Below we summarizehypothesized that several key regulators of cell cycle / some of the recent experimental observations support-differentiation control play key roles in somatic ing this hypothesis.embryogenesis through coordinated interactions with In tobacco protoplast-derived cells, probing of thehormonal, environmental and developmental signal- chromatin structure by nuclease digestion, as well asling pathways. the accessibility of DNA by the interchelating fluores-

cent dye, propidium iodide, revealed two separatedChromatin remodelling: open for changes periods of chromatin decondensation. The first took

place during protoplast isolation and was linked toChromatin structure, the DNA organization around dedifferentiation, while the second one was inducedbasic nuclear proteins (histones), is intrinsically in- by plant hormones, auxin and cytokinin, and linked tovolved in the regulation of nuclear processes, such as the reactivation of cell division (Zhao et al., 2001). InDNA repair, replication and especially transcription this model, dedifferentiation (first phase of chromatin(for review, Varga-Weisz and Becker, 1998). Chro- decondensation) represents a transitory stage whenmatin structure changes in a dynamic way and is cells become competent to switch cell fates. However,continuously remodelled during development. Chro- their further fates depend on hormonal signals: in thematin-dependent gene silencing is a common mecha- absence of auxin, cells went through apoptosis, auxinnism for maintaining the differentiated state of cells. and cytokinin were both required for cell division,Thus chromatin remodelling is necessarily linked with while auxin treatment alone resulted in differentiationcellular dedifferentiation and the switching of cell (Zhao et al., 2001).fate. Similar observations were made with leaf proto-

It can be hypothesized that chromatin remodelling plast-derived cells of alfalfa. The structure and size ofplays two major roles during the early stages of the nucleus and especially the ratio of the volumes ofsomatic embryogenesis. Dedifferentiation requires the nucleus and nucleolus, as well as the stainabilityunfolding of the supercoiled chromatin structure, in of the chromatin with fluorescent dyes, could beorder to allow the expression of genes inactivated by correlated with auxin /cytokinin-dependent phases ofheterochromatinization during differentiation, and cell reactivation and division (Figure 1) (Pasternak etsubsequent chromatin remodelling can result in the al., 2000). Both studies indicated that the first phase ofspecific activation of a set of genes required for chromosome decondensation was partial.

Figure 1. Cellular markers of dedifferentiation and totipotency in alfalfa leaf protoplast-derived cells. (A–C) Nuclear structure as a marker ofactivation of leaf protoplast-derived alfalfa cells. 4�,6-diamidino-2-phenylindole (DAPI)-stained isolated nuclei of: (A) Differentiated leaf cell;(B) 1-day-old protoplast-derived cell; (C) 3-day-old dividing protoplast-derived cell. Dedifferentiation is associated with an increase in the sizeof the nucleus and nucleolus that is the most pronounced in dividing cells (see Pasternak et al., 2000). Bar represents 1 �m. (D–G) Cellulardistribution of the pH-dependent fluorescent dye fluorescein diacetate (FDA) is dependent on the cell type developed from alfalfa leafprotoplasts. In elongated, non-embryogenic cells (D) FDA accumulates in the functioning chloroplasts (E), while in the compact, cytoplasmrich embryogenic cells (F) it remains in the cytoplasm (G). Bar represents 10 �m.

210

In Drosophila, a large number of genes, known seed development, and whether the down-regulationcollectively as genes of the Polycomb (Pc) group, of their function is sufficient or only a necessaryregulate the accessibility / inaccessibility of particular condition in the induction of somatic embryogenesis,chromosomal regions. In plants, Pc-like genes play an remain interesting questions to be answered.important role in suppressing the development ofunfertilized seeds (Chaudhury et al., 2001). In medea(mea) and fertilization independent endosperm ( fie) Embryogenic cells: what do they look like?mutants, endosperm, and sometimes embryo, de-velopment is initiated in the absence of fertilization In order to be able to study the earliest phases of(Ohad et al., 1996; Grossniklaus et al., 1998). The somatic-to- embryogenesis, good markers are re-mutated proteins contain domains similar to Drosoph- quired that indicate the acquisition of embryogenicila polycomb proteins: MEA has a domain similar to competence in the induced cells. These markers mustthe SET domain of enhancer of zeste and FIE shares be universal to allow the comparison of differenthomology with the WD domain of extra sex combs embryogenic systems. In addition to serving as(Ohad et al., 1999). On the basis of homology with markers, they may also be useful in revealing specifictheir animal counterparts, plant Pc-like proteins may cellular processes related to the somatic-to-em-be involved in chromatin remodelling. This sugges- bryogenic transition and the determination /descrip-tion is supported by experiments based on the over- tion of cellular states.expression of the chromodomain of a Drosophila Pc-protein in tobacco. The fused GFP protein could Cell morphology: an early marker of embryogenictarget specific chromatin regions and caused a variety competenceof morphological phenotypes in the transgenic plants,indicating interference with endogenous chromatin- Formation of embryogenic cells can be correlatedregulating mechanisms (Ingram et al., 1999). with characteristic morphological changes. The fate

The mutation named fertilization independent seed of embryogenic carrot cells was followed by video( fis2 ) caused the same phenotype as mea or fie, but cell tracking by Toonen et al. (1994). The single cellfis2 encodes for a zinc-finger transcription factor fraction (�22 �m) of the established embryogenic(Chaudhury et al., 1997; Luo et al., 1999). It is cell culture contained cells that could be classifiedhypothesized that FIS2 forms a complex with MEA into five morphological groups. Although all celland FIE and recruits them to particular promoter types were capable of developing into somatic em-regions which regulate the transcriptional activation bryos with varying efficiency, the highest frequencyof downstream genes (Schnittger et al., 1999; Gros- was observed in the case of small, spherical, cyto-sniklaus et al., 2001). plasm-rich cells. In another approach, Nomura and

Additional experimental evidence highlights the Komamine (1985) fractionated single carrot cells byimportance of regulating chromatin structure in de- density gradient centrifugation. They obtained a frac-velopmental transitions. The Arabidopsis mutant, tion of small, isodiametric, cytoplasm-rich cells thatpickle ( pkl), has a phenotype characterized by the could initiate embryo formation with 90% frequency.postembryonic expression of embryo specific markers This cell type was designated as State 0, or em-and the spontaneous regeneration of somatic embryos bryogenic competent cells which, in the presence ofin its roots (Ogas et al., 1997). The product of the pkl auxin (2,4-D), formed the State 1 embryogenic cellgene was characterized as a chromatin-remodelling clusters, consisting of less than 10 cytoplasm-richfactor that represses gene expression and regulates the cells (Nomura and Komamine, 1995). These clustersdevelopmental transition from embryogenic to veg- differentiated into globular embryos on auxin-freeetative state (Ogas et al., 1999). medium. In contrast, in cell cultures initiated from

The above experimental findings strongly support petioles of Medicago sativa, single cells were unablethe view that the initiation of embryogenic develop- to develop into somatic embryos and the fractionment is based on a release from transcriptional re- trapped between 224- and 500-�m meshes repre-pression by the signal of fertilization. Whether MEA- sented embryogenic cell clusters consisting of small,and FIE- or PKL-like proteins and other regulators of rapidly dividing cells (Xu and Bewley, 1992). Similarchromatin remodelling play a similar role during observations have been made in the case of Piceasomatic embryo development as they do during early abies (Filonova et al., 2000).

211

In another approach, the expression of a gene ‘cross talk’ between different signalling cascadescoding for the ‘somatic embryogenesis receptor ki- related to development, metabolism, environmentalnase’ (SERK) was used as a marker to define em- sensing, etc., results in very complex responses tobryogenic cells in the activated hypocotyl explants of diverse and multiple stimuli, even if a relatively smallcarrot (Schmidt et al., 1997). In this case, a class of number of messengers are involved. In plant cells, theelongated cells on the explant surface was reported to list of messengers used by signalling pathways in-

2�be competent for embryogenesis. In Dactylis cludes, among others, pH and Ca . Here we discussglomerata leaf explants, using the expression of the their significance during the induction of somaticDactylis SERK-homologue gene as a marker of embryogenesis.competence, exclusively small, isodiametric cellswith rich cytoplasm proved to be competent and could pHdevelop into somatic embryos. In established em- Numerous studies have demonstrated that changes inbryogenic cell cultures, however, a few percent of cytoplasmic pH (pHc) occurred during metabolic andelongated, vacuolated cells also expressed the marker developmental transitions in a large variety of cells. In(Somleva et al., 2000). Dictiostelium discoideum, cytoplasmic pH deter-

Pennell et al. (1992) using the JIM8 cell wall mined different cell differentiation pathways (Grossepitope as another molecular marker for embryogenic et al., 1983). pHc increase is necessary for transitioncompetence of carrot cells, hypothesized that small, of cells from G0 to G1 and into S-phases in yeasts,cytoplasm-rich cells expressing this epitope represent echinoderm eggs, protozoa, slime moulds and mam-a transitional cell state: they are formed from non- malian cells (for review, Frelin et al., 1988).embryogenic cells and they either elongate and die, or Tumorigenic Chinese hamster cells had a pHc of 0.2divide and form the initial cell of the somatic embryo. pH units higher than that of normal cells (Ober and

In alfalfa, leaf protoplast-derived cells cultured at Pardee, 1987). Preventing the increase in cellular pHdifferent 2,4-D concentrations can develop into either in sea-urchin eggs, by different methods, blockedembryogenic or non-embryogenic cell types. Similar pronuclear movements, diminished protein synthesismorphological traits could be recognized when proto- and prevented cleavage (Swann and Whitaker, 1990).plasts of embryogenic and non-embryogenic geno- In contrast, alkalinization of the cytoplasm with am-types were cultured at the same 2,4-D concentration. monia could reactivate the egg, and induced cyclin

cdc2In both cases, embryogenic cells were small, spherical synthesis, p34 phosphorylation and finally DNAand densely cytoplasmic, while non-embryogenic replication (Whitaker, 1990; Schomer and Epel,

¨ones were elongated and highly vacuolated (Bogre et 1999).´al., 1990; Dudits et al., 1991; Feher et al., 2002; In plants, there is only a limited number of exam-

Pasternak et al., 2002) (Figure 1D,F). Embryogenesis ples of the physiological role of long-term changes incompetent cells formed in chicory leaf explants were cellular pH (as a review, Kurkdjian and Guern, 1989;also characterized by dense cytoplasms (Blervacq et Pichon and Desbiez, 1994; Bibikova et al., 1998). Inal., 1995). Bidens pilosa, Pichon and Desbiez (1994) found that

It can be suggested therefore that in most em- cytoplasmic pH was correlated with cell division.bryogenic systems, including gymnosperms Alkalinization promoted the cell cycle in the meri-(Egertsdotter and von Arnold, 1998), this morphology stematic region of the hypocotyl, while acidificationis associated with embryogenic competence (for re- inhibited it. Initiation of root hair cells in Arabidopsisview, Yeung, 1995). The observed differences in the could also be characterized by the acidification of thecorrelation of cell morphology and embryogenic apoplast and the alkalinization of the cytoplasmcompetence in certain experimental systems can be (Bibikova et al., 1998).ascribed to the different explants and experimental The transition from the somatic to the embryogenicapproaches used. cell state is a complex process that includes dedif-

ferentiation, cellular reactivation, division and meta-2�Physiological competence: pH and Ca bolic, as well as developmental, reprogramming.

Characteristic changes in intracellular pH are hypoth-All living cells use a network of signal transduction esized to be associated with this transition. A relation-pathways to harmonize environmental stimuli with ship between medium (and cellular) pH and develop-the completion of developmental programs. The mental state has been suggested by experiments with

212

wounded carrot zygotic embryos cultured in the pres- 2002). We can assume that FDA accumulation in theence of 1 �M NH Cl (Smith and Krikorian, chloroplasts is related to functional electron transport4

1990a, b). NH Cl-induced cellular alkalinization al- (photosynthesis). The lack of establishment of this4

lowed the establishment of a continuous culture of transthylakoid pH gradient in the embryogenicpreglobular stage proembryos. Medium pH was de- competent cells may indicate the rapid dedifferentia-creased in these cultures (down to pH 4), which was tion of chloroplasts and the loss of photosyntheticlikely correlated with increased pHc. The develop- functions in these cells. Metabolic properties of plas-ment of embryos could only be advanced if the tids have been shown to differ among leaf and embryomedium pH was raised to approximately pH 5.7. It cells of Brassica (Eastmond and Rawsthorne, 1998).had previously been shown that cells of an alfalfa

2�embryogenic-type callus had higher average pHc Ca2�values compared to cells from a non-embryogenic Ca functions as a key regulator of many cellular

type (Schaefer, 1985). and physiological events in plants (for review, San-2�In alfalfa leaf protoplast-derived cells, cytoplasmic ders et al., 1999). Current models of Ca -mediated

and vacuolar alkalinization and medium acidification signalling emphasize the significance of a transientwere shown to be correlated with the activation of cell change, usually an increase, in cytoplasmic calciumdivision (Pasternak et al., 2002). Although a clear concentration, followed by the perception of suchrelationship between embryogenic competence and changes by calcium-binding proteins. The in-

2�the degree of cytoplasmic alkalinization could not be volvement of Ca in a wide variety of stimulus-established in this direct somatic embryogenesis sys- response pathways in plant cells raises several ques-tem, small, cytoplasm-rich embryogenic cells had a tions concerning how the same messenger can reg-tendency to exhibit higher vacuolar pH values in ulate the different responses. The amplitude, duration,

2�comparison to the non-embryogenic vacuolated cells frequency and location of the Ca signal can be(Pasternak et al., 2002). It is supposed that the large considered as key features in the determination ofdifference in the vacuolar pH of the embryogenic and different messages.

2�non-embryogenic cell types is related to the difference The increase in intracellular Ca concentrationin the vacuolar functions (for reviews, Wink, 1993; after fertilization of egg cells has been demonstratedMarty, 1999; Ratajczak, 2000), linked to the fate of in both animal and plant cells (Stricker, 1999; An-these cells: elongated, differentiated cells have large, toine et al., 2000). In brown algae (Robinson et al.,central, lytic vacuoles with more acidic pH, while the 1999) as well as in flowering plants (Antoine et al.,small, dedifferentiated cells have several small 2000), it was hypothesized that gamete fusion-in-storage-type vacuoles. duced calcium influx plays a direct role in egg cell

In the same leaf protoplast system, buffering of the activation.medium pH with 10 mM MES prevented em- Although similar studies at the level of individualbryogenic cell formation under inductive conditions cells have not been performed during the induction of(Pasternak et al., 2002). Increased transport across somatic embryo development, the dependence ofcell membranes, inhibited by extracellular MES, somatic embryogenesis on external and internal cal-might be an important process in the metabolic re- cium concentrations has been demonstrated in differ-programming of embryogenic cells. Embryogenic ent systems. In an embryogenic carrot cell suspension,cells accumulate starch (Yeung, 1995) and have vac- it was shown that an upward shift in the exogenous

�3 �2uoles characterized by low transparency and strong calcium concentration (e.g., from 10 to 10 M) atstaining by toulidene blue, indicating high protein the time of transfer to auxin-free embryo differentia-content, likely to be storage proteins (Pasternak et al., tion medium increased the number of somatic em-2002). bryos approximately 2-fold (Jansen et al., 1990). It

Another marker of dedifferentiated, embryogenic was also observed that the elevated calcium con-alfalfa cells is the intracellular distribution of fluores- centration counteracted the inhibitory effect of 2,4-D

cein diacetate (FDA). This pH-indicator fluorescent on embryo development (Jansen et al., 1990). Over-dye is detectable exclusively in the cytoplasm of voorde and Grimes (1994) reported that exogenous

2�embryogenic competent cells, in contrast to its ac- Ca concentration higher than 200 �M is requiredcumulation in the chloroplasts of the highly vacuol- for optimal embryo formation. The application of

2� 2�ated, elongated cells (Figure 1) (Pasternak et al., either Ca -channel blockers or the Ca ionophore

213

A23187 inhibited embryo initiation. These data indi- 1999; Scheres and Benfey, 1999). Polarized develop-2�cate that exogenous Ca and the maintenance of ment of the zygote and the embryo includes the

2�cellular Ca gradients are required for proper em- formation of shoot and root meristems, which main-bryo development in vitro. tain the apical–basal polarity throughout plant de-

2�Endogenous Ca levels were determined during velopment (Jurgens, 2001). These meristems aresandalwood somatic embryogenesis by Anil and Rao groups of undifferentiated cells and serve as cell(2000). Transfer of cells of the proembryogenic cell sources for tissue and organ differentiation. In themass (PEM) onto growth regulator-free medium to absence of cell migration, the morphogenesis ofpromote embryo development resulted in a several- plants is based on co-ordinated cell division and

2�fold increase in the uptake of exogenous Ca , as well elongation followed by differentiation. Studies on2�as the Ca level in the symplast. This could be root development, with the help of laser ablation of

blocked by chelating exogenous calcium ions, which individual cells, have revealed that position, ratherarrested embryo development, but still allowed cell than lineage, determined patterning and cell fate (vanproliferation to continue. Blocking calcium-mediated den Berg et al., 1997, 1998). Position of a plant cell issignalling by W7 resulted in an 85% decrease in largely determined by its contacts with neighbouringembryo formation frequency. This response indicated cells as a result of the orientation (plane) of cell

2�that calmodulin or Ca -dependent protein kinase division and cell elongation during tissue formationcould be involved in this process (Anil and Rao, (van den Berg et al., 1997, 1998).2000). Although there are data showing that cal- The initial zygotic division in higher plants ismodulin levels are increased in dividing cells (e.g., asymmetric (Scheres and Benfey, 1999). This eventPerera and Zielinski, 1992), there were no significant establishes the basic polarity of the plant whichchanges detected in calmodulin levels (Overvoorde probably determines subsequent pattern formationand Grimes, 1994; Dudits et al., 1995) or calmodulin (Jurgens et al., 1997). In Arabidopsis, where em-methylation and calmodulin-binding protein levels bryonic cell divisions are very regular, the gnom(Oh et al., 1992) during carrot, sandalwood and mutation, which affects the first asymmetric division,alfalfa somatic embryo induction. However, an un- causing abnormal pattern formation (Busch et al.,

2�known Ca -binding protein was reported to be tran- 1996). However, in the fass mutant, in spite ofsiently induced in the early phase of somatic embryo disorganized cell divisions, all pattern elements wereinduction in an alfalfa cell culture by Dudits et al. formed in the embryo. Nonetheless subsequent de-(1995). velopment was distorted and resulted in dwarves and

2�Additional data support that Ca -dependent pro- malformed plants (Torres-Ruiz and Jurgens, 1994).tein kinases (CDPKs) are involved in the signalling Thus, although embryogenic patterning may not in-pathways during the formation of somatic embryos. In volve specific cell divisions except the initialsandalwood, two CDPKs could be detected in protein asymmetric division, final form does. Not only theextracts of embryogenic cell cultures (Steenhoudt and pattern, but also the rate of cell division can be an

2�Vanderleyden, 2000). Their strong Ca -dependent important factor in developmental pattern formation.activities were detected in PEMs as well as in somatic Expression of the dominant negative form of the cellembryos of different stages, but not in regenerated cycle regulatory CDK (cdc2aDN) of Arabidopsis inplantlets (Jackson and Casanova, 2000). The expres- embryos resulted in a range of altered phenotypession of the MsCDPK3-encoding gene has been shown (Hemerly et al., 2000). The most severe phenotypeto increase during the early phase of 2,4-D-induced was characterized by unrecognizable tissue organiza-embryogenesis from cultured alfalfa cells (Davletova tion, while in other cases some tissues were missing

2�et al., 2001). The role of Ca in the establishment of and the apical /basal (but not the radial) symmetrycellular polarity during embryogenesis in plants is was distorted.discussed below (Pattern formation: the establishment Much less is known about the establishment ofof polarity section). polarity during somatic embryogenesis and there is

controversy surrounding the role of the initialPattern formation: the establishment of polarity asymmetric cell division. In alfalfa, where leaf proto-

plasts can be used to induce direct development ofIn seed plants, the egg cell and zygote exhibit apical– embryogenic cells (Dijak and Simmonds, 1988; Songbasal polarity (Russell, 1993; Vroemen et al., 1996, et al., 1990; Dudits et al., 1991), 2,4-D stimulates the

214

formation of asymmetrically dividing cells in a con- servation that the application of an electric field tocentration- (Pasternak et al., 2002) and genotype- alfalfa protoplasts enhanced direct embryogenesis

¨(Bogre et al., 1990) dependent manner, and the type (Dijak and Simmonds, 1988), raised the question as toof division determines further development. In the whether there is need for a directional signal toCichorium hybrid 474, leaf cells can be induced to establish polarity in somatic cells that will becomeform somatic embryos directly by stress treatments embryogenic. It is interesting to note here that 2,4-D,(Blervacq et al., 1995). In this system, morphological one of the most efficient inducers of somatic embryo-asymmetry of the dividing embryogenic cells was not genesis in many plant cultures, has been shown toobserved (Blervacq et al., 1995). In carrot, somatic prevent electrical polarity in cultured tobacco cellsembryos were induced in a heterogeneous cell popula- (Goldsworthy and Mina, 1991). However, when 2,4-D

tion and proembryogenic cell mass (PEM) was was removed and replaced by indoleacetic acidformed from various initial cell types, as observed by (IAA), the percentage of electrically polar cells sig-the use of video cell tracking (Toonen et al., 1994). nificantly increased. This may explain why 2,4-D

However, Nomura and Komamine (1985) showed blocks the development of somatic embryos at thethat isolated small, cytoplasm-rich carrot cells, after preglobular stage in carrot cultures, and why itsan unequal first division and polarized synthesis of removal or decreased concentration is required for themacromolecules, have the ability to develop into development of embryogenic polarity.somatic embryos. In certain Arabidopsis ecotypes, Although there are some systems where a polarleaf protoplast-derived embryogenic cells may exhibit input (directional signal) is not needed to achieve cella cell division pattern reminiscent of the first division polarity, external signals are usually required to estab-of the zygote (Luo and Koop, 1997). Further evidence lish the polar axis, not only in plant cells, but in otherhas been provided in experiments where subpopula- eukaryotes as well (Cove, 2000). The best-studiedtions of carrot cells were labelled by the monoclonal plant systems, in this respect, are the brown algae,JIM8 antibody, which recognizes a cell surface epi- Pelvetia and Fucus (Cove, 2000; Souter and Lindsey,tope (McCabe et al., 1997). The JIM8-positive cells 2000). The egg cells of these species are spherical andwent through an asymmetric cell division that resulted fertilization induces the polar outgrowth of a rhizoid.in daughter cells with different antibody staining The first cell division generates a cell plate that ischaracteristics dependent on their further fate. The perpendicular to the long axis of the polar zygote andJIM8� daughter cell died while the cell devoid of the separates the rhizoid and thallus cells. The spermsurface epitope was embryogenic and formed a entry generates a signal which determines the polarsomatic embryo. axis, however the alignment of this axis can be

It is very likely that the establishment of polarity is changed by directional light until the polar axis isan important event, not only for zygotic, but for fixed (Kropf et al., 1999). In Pelvetia, it is suggestedsomatic embryogenesis as well. Even if morphologi- that a rhodopsin-like protein is the photoreceptorcal asymmetry is not obvious, the unequal distribution sensing the light gradient and controlling cellularof cellular determinants can be decisive in the de- cGMP levels (Robinson and Miller, 1997; Robinson

2�termination of cell developmental pathways following et al., 1998). Unequal distribution of Ca channels2�division. The polarity of the egg cell is evident from and the established Ca gradient are also important

the position of the nucleus at the cytoplasm-rich in early determination of the axis (Pu and Robinson,chalazal pole, while the micropylar pole is highly 1998). A calcium influx was triggered in the vicinityvacuolated (Russell, 1993). The microtubular cyto- of the sperm entry site and subsequently spread to theskeleton is particularly dense near the nucleus and has whole egg cell during the in vitro fertilization ofno specific orientation. The actin microfilaments have maize (Antoine et al., 2000).a similar conformation. In alfalfa, protoplast-derived In sunflower protoplast-derived cells, as indicatedembryogenic cells were also characterized by dis- by the fluorescent probe DMBodipy-PAA, there was a

2�ordered microtubules when compared to non-em- translocation of Ca channels, which depended onbryogenic control cells (Dijak and Simmonds, 1988). the division type (Vallee, 1997; Xu et al., 1999). InThe positioning of the nucleus was also peripheral in freshly isolated protoplasts, the probe was strictlythese cells, indicating the early establishment of cel- localized to distinct points on the plasma membrane.lular polarity in response to the electric stimulation This compound also labelled cytoplasmic strands andused to induce embryogenic development. The ob- a region around the nucleus in cultured protoplast-

215

derived cells. When cells divided, the staining was experiments where BrefeldinA was used to treatlocalized homogenously in the plane of cell division embryogenic carrot cells, the polar development ofin symmetrically dividing cells, but marked only a somatic embryos was blocked, but could be restoredperipherical ring around the site of cell division by the application of conditioned medium containingduring asymmetric division (Vallee, 1997; Xu et al., secreted proteins (Capitanio et al., 1997). During the1999). Later, when embryoids developed from the last few years, a large body of experimental data has

2�asymmetrically divided cells, the Ca channel probe accumulated on the role of extracellular proteins onwas clearly localized on basal part of the embryoid, as somatic embryo induction and development.was the case with zygotic embryos (Vallee, 1997; Xuet al., 1999). Top secret(ed): secreted proteins and the

There is evidence that the polar localization of embryogenic cellF-actin filaments and the polarized secretion of cellwall material play roles in axis fixation (Fowler and Callose deposition in walls of embryogenic cells andQuatrano, 1997; Belanger and Quatrano, 2000; Vro- the formation of a glycoproteic extracellular matrixemen et al., 1999). The role of polarized secretion in around superficial cells of globular embryos can servethe establishment of cellular polarity was verified by as cytological markers of cell reactivation and em-the disruption of this process by Brefeldin A, an bryogenic differentiation (Dubois et al., 1990, 1991;inhibitor of Golgi function (Capitanio et al., 1997). Pedroso and Pais, 1995). In chicory, embryogenicThe gnom mutant of Arabidopsis provided some competent leaf cells had thicker walls, exhibiting aexperimental evidence for the existence of similar brighter autofluorescence of cellulose under UV light,mechanisms of axis fixation in higher plants. As was compared to non-reactivated mesophyll cells (Bler-already mentioned, this mutant does not establish the vacq et al., 1995). In the case of direct embryoproper plane of cell division required for the oblique formation from leaf explants of Camellia japonica, asasymmetric division of the zygote. The gene affected an early marker of the morphogenetic response, cellby the gnom mutation was cloned and characterized walls underwent characteristic changes, including theas coding for a guanidine nucleotide exchange factor deposition of callose on the surfaces of induced cellsof a small GTP-ase belonging to the ARF family (Pedroso and Pais, 1995). This was followed by the(Busch et al., 1996; Steinmann et al., 1999). These deposition of other materials (cutin) characteristic forproteins are known to be involved in vesicle move- somatic embryogenesis, where the involvement ofment and secretion (Jackson and Casanova, 2000). lipid transfer proteins is presumed. Expression of lipid

Auxin is transported in a polar manner along the transfer proteins is a well-known early marker ofshoot–root axis, which requires efflux carriers such as somatic embryo induction in different systems (Sterk

¨PIN1 (for review, Palme and Galweiler, 1999). et al., 1991; Schmidt et al., 1997; Sabala et al., 2000).Asymmetric localization of PIN1 develops from a It is also a broadly used marker of embryo differentia-random distribution in Arabidopsis during early tion, as it is also linked to the formation of theembryogenesis, as a result of actin-dependent cycling protoderm layer in developing somatic and zygoticbetween the plasma membrane and endosomal com- embryos (Thoma et al., 1994; Vroemen et al., 1996;partments (Geldner et al., 2001). In gnom mutant Toonen et al., 1997a).Arabidopsis embryos, the localization of the putative Arabinogalactan proteins (AGPs) are a class ofauxin efflux carrier protein, PIN1, is disturbed (Stein- proteoglycans, widely distributed throughout the plantmann et al., 1999). BrefeldinA, a vesicle trafficking kingdom, that have been implicated in diverse pro-inhibitor, was also shown to abolish the polar locali- cesses of plant growth and development (for review,zation of PIN1 and, consequently, the polar transport Showalter, 2001), including somatic embryogenesisof auxin (Geldner et al., 2001). The role of polarized (for review, Vroemen et al., 1999). Although theirauxin transport was implicated in the establishment of exact functions are not clear, AGPs are presumablyapical–basal patterning in Fucus embryos, as well involved in molecular interactions and cellular sig-(Basu et al., 2002). nalling at the cell surface. Their role during somatic

The importance of polarized secretion during embryo formation was demonstrated by the additionsomatic embryogenesis was indicated by the of the �-D-glucosyl Yariv reagent (�GlcY), whichasymmetric distribution of the JIM8 cell wall epitope interacts with AGPs, to the culture medium. Thisin carrot cell cultures (see above). In addition, in compound altered the morphogenetic response (root

216

formation instead of embryogenesis) in the case of lar stage with aberrant protoderm formation at thecarrot (Thompson and Knox, 1998) and blocked restrictive temperature (32 �C) (de Jong et al., 1992).somatic embryo formation in the case of Cichorium Conditioned medium from cultures of wild type cells(Chapman et al., 2000). Promotory and inhibitory (Lo Schiavo et al., 1990) and a 32-kDa acidic endo-effects of certain exogenous arabinogalactan protein chitinase, purified from the same medium (de Jong etfractions in carrot cultures were reported by Toonen et al., 1992), complemented the mutation and allowedal. (1997b). The addition of isolated carrot seed AGPs normal protoderm formation and development atto old, non-embryogenic cell lines re-induced their 32 �C. Cytological investigations highlighted that theembryogenic potential (Kreuger and van Holst, ts11 mutation was associated with defects in the1993). Seed AGPs promoted somatic embryo matura- secretory characteristics of the cells, and not exclu-tion in Norway spruce (Egertsdotter and von Arnold, sively with the reduced level of the endochitinase1998). Several antibodies have been prepared against (Baldan et al., 1997). In embryogenic cultures ofdiverse plant AGPs and used to mark specific cell Cichorium, the major secreted proteins were alsotypes (for reviews, Knox, 1997; Showalter, 2001). In identified as chitinases, glucanases and an osmotin-Cichorium, immunofluorescence and immunogold like protein, all of which accumulated at a signifi-labelling studies localized AGPs to the outer cell cantly higher level in embryogenic compared to non-walls of globular somatic embryos, but they appeared embryogenic cultures (Helleboid et al., 2000b). Simi-internally in later developmental stages. AGPs were larly, in embryogenic alfalfa cultures, alterations inalso abundantly present in the culture medium (Chap- the levels of extracellular proteins homologous to theman et al., 2000). In embryogenic maize cells, the carrot endochitinases have been reported followingextracellular matrix surface network has also been the removal of 2,4-D (Poulsen et al., 1996). Recently,shown to contain AGPs, as well as the JIM4 van Hengel et al. (2001) reported that carrot AGPsarabinogalactan protein epitope, that were not present contained glucosamin and N-acetylglucosamine thaton the surface of non-embryogenic cells (Samaj et al., were sensitive to endochitinase cleavage. It was also1999). demonstrated that the embryogenic competence of

Another AGP epitope recognized by the JIM8 protoplasts could be restored and enhanced by theantibody was originally described as a marker of the application of AGPs, endochitinase-cleaved forms ofvery early transitional stage of cultured carrot cells which were more efficient (van Hengel et al., 2001). Itafter embryogenic induction (Pennell et al., 1992). can be hypothesized that chitinase-modified AGPs areSubsequently, it was shown that most embryos de- extracellular molecules capable of controlling orvelop from cells lacking the JIM8 epitope (Toonen et maintaining the embryogenic competent cell stateal., 1996), and that the AGP fraction containing the (van Hengel et al., 2001).JIM8 epitope actually has an inhibitory effect onsomatic embryogenesis in carrot (Toonen et al.,1997b). Finally it was found that the JIM8 epitope Genes /proteins involved in the formation ofmarks a specific small, spherical cell type which embryogenic cells: who plays the game?asymmetrically transferred the JIM8 epitope, follow-

�ing cell division, to a JIM8 embryogenic and a The processes during which somatic cells acquire�JIM8 apoptotic cell type (McCabe et al., 1997). embryogenic competence obviously involve the re-

Cultured cells, sorted on the basis of JIM8 labelling, programming of gene expression patterns. As waswere attached to secondary antibody-linked paramag- discussed above, the morphology, physiology andnetic beads and were used to demonstrate that the metabolism of the cells are significantly altered due toJIM8 epitope represents a soluble signal produced by dedifferentiation, activation of cell division and a

�JIM8 cells to stimulate embryo development from change in cell fate. All of these changes are dependent�JIM8 cells (Pennell et al., 1995; McCabe et al., on the inactivation of genes functioning in differen-

1997). tiated cells and the activation of those required for theThe first observations, indicating that extracellular, above processes. Obviously, the overall reprogram-

secreted proteins may have important roles during ming of gene expression has to be governed bysomatic embryo development, were derived from regulator genes, including those encoding transcrip-studies on the ts11 temperature sensitive carrot mu- tion factors. Genes regulating changes in the develop-tant. ts11 somatic embryos were arrested at the globu- mental fate of plant cells have attracted the interest of

217

plant biologists. It is not surprising, therefore, that globulin-like protein, histones, ribosomal proteins,since the beginning of plant molecular biological elongation factor 1� and ubiquitin fusion protein. Theresearch, there have been attempts in many laborator- expression patterns of these clones were highly vari-ies to identify genes which may render somatic cells able and could serve as markers of the developmentalembryogenic. Interestingly, although it was generally stages of the embryos. However, most could also bebelieved that somatic embryogenesis would serve as a detected in undifferentiated callus cells. This limita-model to identify key regulators in zygotic embryo tion showed that the strategy used for the screeningdevelopment, recent studies of seed and embryo de- was not appropriate to identify genes characteristic forvelopment mutants seem to provide the clues to an the very early stages. This was considered by Sato etimproved understanding of the basic regulation of al. (1995), who used cDNA prepared from undif-somatic embryogenesis. ferentiated callus cells to subtract cDNA from pro-

globular cell clusters developed 3 days after theApproaches to identify genes activated during the transfer to hormone-free medium. However, only oneinduction phase of somatic embryogenesis of the nine identified clones proved to be specifically

characteristic for somatic embryo formation. ThisThe most common approach to identify somatic clone, named as CEM6, was identified as a glycine-embryogenesis-related genes is to compare gene ex- rich protein with a possible function in cell wallpression in somatic embryos to that of callus cells. synthesis.This approach resulted in the identification of a few The determination of the transcripts of the earliest-abundant transcripts linked to specific stages of activated genes during carrot somatic embryo induc-embryogenesis, rather than to the induction period of tion was the goal of the experiments carried out byembryogenic development (for review, Zimmerman, Yasuda et al. (2001). They used the differential dis-1993). More recently, similar experiments have re- play reverse transcription polimerase chain reactionsulted in the identification of a set of genes showing (DDRT PCR) technique to compare mRNA popula-differential transcript accumulation during somatic tions in embryogenic and non-embryogenic cultures.embryo development in conifers (Dong and Dunstan, Three cDNA clones were isolated. Clone �43 was1996, 1999). preferentially expressed very early in the em-

As this review focuses on the very early stage of bryogenic cell clusters and its expression declined insomatic embryogenesis, when the developmental globular embryos. It was also expressed in hypo-transition actually takes place, only those experiments cotyls, the only tissue from which somatic embryosthat aimed to isolate genes activated during the induc- could be initiated, and not in other organs or tissues.tion period will be described. During the last years, This cDNA coded for a thaumatin-like protein and itmore progress has been achieved in this field due to was suggested that its expression was related to theimprovements in experimental techniques, including stress response of the plant cells. Clone �93 repre-tissue culture systems, as well as mRNA isolation and sented a cDNA coding for a homologue of the DCcDNA synthesis. 2.15 proline-rich protein (Holk et al., 1996), but

Most of the results are derived from the three best- without the characteristic Pro-X motifs. The expres-studied systems: carrot, alfalfa and chicory. In carrot, sion of this clone, unlike the heart-shaped embryo-the phase of embryogenic induction can not be clearly enhanced Dc 2.15, was characteristic of early em-defined. Lin et al. (1996) used an approach whereby bryogenic cell clusters, but could also be detected atthe mRNA population of globular embryos was com- later stages and in suspension cultured cells. It waspared to that of seedlings. The sensitivity of the suggested that clone �93 represents a transductioncDNA library screening was improved by using a factor during somatic embryogenesis. A similar ex-subtracted probe enriched for globular embryo en- pression pattern could be ascribed to clone �87,hanced transcripts. Thirty-eight cDNA clones, repre- which had no homologous sequence in the databases.senting genes with altered expression during somatic Schmidt et al. (1997) followed a more extensiveembryogenesis, were identified. The majority of pro- approach. Three different screening techniques wereteins encoded by these genes could be classified into used to search for genes specifically active in em-the categories of cell wall proteins, enzymes, patho- bryogenic versus non-embryogenic carrot cultures:genesis-related (PR) proteins, heat-shock proteins, differential cDNA library screening, cold plaquelate-embryogenic abundant (Lea) proteins, oleosins, a screening and differential display. All putative, dif-

218

ferentially expressed clones obtained from the three Medicago sativa L. genotypes using a conventionaltypes of experiments were subjected to a second differential cDNA library screening approach. Prior toscreening based on a dot-blot reverse Northern hy- mRNA isolation, cells were fractionated and the 230-bridization. Differential screening resulted in 30 puta- �m fraction enriched in embryogenic cells was usedtive clones, but only one of them could be verified as a to prepare the cDNA library and the probes. Thegene exclusively expressed in the embryogenic cul- screening resulted in three cDNA clones (ASET1 –3 )tures, after the second screening. Six clones derived specifically present in cells of the embryogenic geno-from differential display RT PCR were sequenced and type. Two of the cDNA clones could be sequenced,two of the fragments represented the lipid transfer but clear functions could not be ascribed to either ofprotein (LTP) gene previously identified as a marker them based on the sequence information. The cloneof early embryogenesis and protoderm formation ASET1 was partial and coded for a hydrophobic(Sterk et al., 1991; Toonen et al., 1997a). More protein. The ASET2 cDNA was full length and codedinterestingly, one of the 26 clones obtained after cold for a protein with structural similarities to integralplaque screening proved to be the cDNA of a gene membrane proteins. It was suggested that these pro-expressed at a very low level, but in a very specific teins could be important in membrane related sig-way, in embryogenic cultures. The sequence of this nalling events.cDNA clone showed homology to plant and animal In Medicago falcata, somatic embryo developmentreceptor kinases and was named as somatic embryo- can be induced in leaf explants by parallel woundinggenesis receptor kinase (SERK). Characteristics of and 2,4-D application (Denchev et al., 1991). mRNAthis putative regulator of the somatic-to-embryogenic samples were isolated at different time points, follow-transition will be discussed later (Genes in transition. ing the induction of direct somatic embryogenesis,Genes /proteins involved in embryogenic cell forma- and converted to cDNA prior to RNA arbitrarilytion section). primed PCR (RAP-PCR) by Fowler et al. (1998).

Other tissue culture systems, where embryos can Two different primer combinations were used and thedevelop directly from somatic cells, were considered cDNA fragments with differential accumulation dur-to be more appropriate for the isolation of genes ing embryo induction were identified and sequenced.regulating early events of somatic embryogenesis. Most of the clones had no significant homology toOne of these systems is based on a chicory genotype database sequences and the expression patterns of(Cichorium hybrid ‘474’), where somatic embryos only two of them have been verified. The clone A1.4develop directly and synchronously on leaf explants was characterized as a calnexin homologue with aunder specific conditions (Blervacq et al., 1995). The potential chaperone function (Huang et al., 1993). Itsapproaches to identify genes activated during the expression might be regulated in a manner similar toearly phases of chicory embryogenesis resulted in the that of heat shock protein genes, also characteristic ofidentification of cDNAs of a �-1,3-glucanase (Hel- somatic embryogenesis (see before). Chaperone ac-leboid et al., 2000a) and other pathogenesis-related tivities might be important due to the intensive syn-proteins and a non-symbiotic haemoglobin (Hendriks thesis of many new proteins during the reactivationet al., 1998). Although the activities of these genes and developmental transition of somatic cells. Thewere characteristic for the transition of somatic cells clone A2.5 exhibited homology with a family ofto embryogenic cells as being related to stress, metab- multidrug resistance genes, like the yeast SNQ2 geneolism and cytoplasmic reorganization (membrane (Servos et al., 1993). It was expressed only at the timetrafficking), it is very likely that none of them is a key of globular embryo formation. The role of thesedeterminant of embryogenic cell fate. proteins during somatic embryogenesis needs to be

Direct somatic embryogenesis is well established clarified.for various alfalfa species and genotypes. More ad- Differential screening of a cDNA library madevantageously, there are also many closely related from embryogenic Medicago sativa cell cultures,genotypes that are non-embryogenic under the same treated with a high 2,4-D concentration to induceconditions. It is not surprising, therefore, that many somatic embryogenesis, and screened with cDNAattempts have been made to identify early embryo- from un-induced cells maintained in the presence ofgenesis-related genes in these species. Giroux and NAA, resulted in the isolation of cDNAs coding forPauls (1997) compared mRNA populations of estab- small heat shock proteins (Dudits et al., 1991) a

¨lished cultures of embryogenic and non-embryogenic proline-rich protein (Gyorgyey et al., 1997) and a

219

calcium-binding protein with unknown function (Schmidt et al., 1997). The Arabidosis homologue of(Dudits et al., 1991). the carrot SERK cDNA has also been cloned as one of

In another approach, PCR-based cDNA subtraction five members of a small gene family (Hecht et al.,was used to identify differentially expressed genes in 2001). It was shown that in planta, AtSERK1 expres-induced and non-induced Medicago falcata leaves sion was first expressed during megasporogenesis and(Russinova et al., 1998). Due to the sensitivity of the then in the functional megaspore, in all cells of theapproach, a very high number of clones were iden- embryo sac until fertilization and in the embryo up totified. They were classified into ‘early’ (from the the heart stage. After this stage, expression wasinduction onward), ‘medium’ (from 3 days after the undetectable in any part of the developing seed. Lowinduction) and ‘late’ (expression after 5–10 days) expression was, however, detected in adult vascularexpression categories. The sequences obtained re- tissues. AtSERK1 gene expression was also observedvealed the presence of many regulatory genes, such as in the shoot apical meristem and cotyledon of auxin-genes of transcription factors, kinases, the phospha- grown Arabidopsis seedlings used to initiate em-tase PP2C, and auxin-induced genes. Several bryogenic callus cultures. These observations indi-ribosomal proteins and others related to translation cated that AtSERK1 expression was not restricted toand post-translational protein processing were also embryogenic cells, but characteristic of those cellsrepresented, as well as proteins linked to stress re- capable of rapid response to hormonal signals andsponse (pathogenesis related and ABA-responsive competent to form somatic embryos or embryogenicproteins, etc.). In this system, somatic embryos were cell cultures (Hecht et al., 2001).formed on leaf pieces as a result of the combined Ectopic expression of the AtSERK1 cDNA underaction of wounding and 2,4-D. However, many of the the control of 35S promoter did not cause any specificidentified genes could be induced by wounding or phenotype, however, the efficiency of the initiation of2,4-D alone. That is why further analysis is required to somatic embryos was increased by approximately 4-elucidate the potential significance of the protein fold in the transgenic seedlings (Hecht et al., 2001).products of these genes during somatic embryogene- AtSERK1 gene expression was higher in cell culturessis. derived from the altered meristem program 1 (amp1 )

In Lycium barbarum (Kairong et al., 1999) three mutant of Arabidopsis, which has enhanced em-transcripts were identified by differential display that bryogenic capabilities (Mordhorst et al., 1998; Hechtwere expressed in embryogenic calli and in early et al., 2001). These mutants also had an increasedsomatic embryos, but not in non-embryogenic callus number of undifferentiated cells in their shoot meri-culture. The partial sequences could not verify any stem, and it was hypothesized that AMP1 activity issignificant homology to known sequences in the required to suppress embryogenic development (e.g.,databases. by suppressing SERK activity) after germination

(Hecht et al., 2001).Genes in transition: genes /proteins involved in The expression of a SERK homologue gene wasembryogenic cell formation also tested in Dactylis glomerata, where somatic

embryogenesis could be initiated directly from leafOf the above-mentioned cDNA sequences, there is explants, as well as indirectly from leaf-derived callusonly one gene known to play a role in the acquisition tissues (Somleva et al., 2000). In this system, SERKof embryogenic competence in plant cells. This is the expression was present in small, cytoplasm-rich iso-somatic embryogenesis receptor kinase (SERK) gene diametric cells that were shown to form somaticisolated by Schmidt et al. (1997). In carrot, SERK embryos by cell tracking. In contrast to carrot andexpression was shown to be characteristic of em- Arabidopsis, SERK expression was maintainedbryogenic cell cultures and somatic embryos, but its beyond the globular stages of embryogenesis in meri-expression ceased after the globular stage (Schmidt et stematic regions (Somleva et al., 2000). Based on allal., 1997). It could also be detected in zygotic em- of the above experiments, the expression of the SERKbryos up to the early globular stage, but not in gene may be used as a marker of embryogenic compe-unpollinated flowers or in any other tissue. Using the tence. And the SERK may represent a possible candi-SERK promoter fused to the luciferase gene and video date for mediating signals which are required tocell tracking, it was also shown that SERK-expressing initiate the embryogenic development. Other leucine-single cells could develop into somatic embryos rich repeat-containing receptor-like kinases are

220

known to be involved in different developmental the formation of somatic embryos and other organ-processes in plants, such as the ARABIDOPSIS like structures and conferred embryonic characters toCLAVATA and ERECTA proteins (for review, Flet- vegetative tissues (Stone et al., 2001). Thus, both lec1cher and Meyerowitz, 2000) and the Petunia PRK1 and lec2 can be considered to be transcriptionalprotein, which is involved in signalling during pollen regulators that can establish a cellular environmentdevelopment and pollination (Mu et al., 1994). For sufficient to initiate embryo development. WhetherSERK, the possible ligands and the downstream similar genes play key roles in the initiation of em-targets have to be elucidated in order to define its bryo formation during zygotic and in vitro inducedexact role in embryogenesis. somatic embryogenesis remains to be determined.

While ectopic expression of the SERK gene was Another family of transcription factors, those con-shown to enhance the embryogenic response of cul- taining the so-called MADS-box domain, are alsotured cells induced to form somatic embryos, the important regulators of many plant developmentalectopic overexpression of the lec1 gene of Arabidop- processes (Jack, 2001a). One of the best-describedsis was shown to induce embryo development directly examples of their importance in plant development ison vegetative tissues in the absence of any exogenous the regulation of flower development and floral organtreatments (Lotan et al., 1998). The lec1 gene was identity (Jack, 2001b). It was shown by Perry et al.identified by studies of a mutation causing defects in (1999) that the MADS-box containing AGAMOUS-embryo maturation and desiccation (Meinke, 1992; like 15 (AGL-15) transcription factor accumulated inMeinke et al., 1994). Cloning and sequencing of this embryos of diverse origin, including zygotic, apomic-mutated gene revealed that it coded for a protein tic, microspore-derived and somatic embryos. In addi-homologous to a subunit of a CCAAT box-binding tion to its high level in embryonic tissues, the subcel-transcription factor (Lotan et al., 1998). Further lular localization of the protein was also changedstudies also revealed that the protein might function when embryonic development began: its originallyduring early stages of embryogenesis in the suppres- cytoplasmic localization became nuclear. Using pro-sion of suspensor proliferation, among other roles moter-GUS gene fusions, it was shown that its expres-(Lotan et al., 1998). The lec1 gene is expressed from sion is not restricted to embryogenesis and it is morethe octan stage up to the late torpedo stage of embryo- likely linked to a juvenile tissue state. This wasgenesis. When the lec1 cDNA was expressed in confirmed by the ectopic expression of the AGL-15transgenic plants under the control of the 35S promot- cDNA, which caused a delay in many aging-relateder, several abnormalities were observed, which indi- processes (Fernandez et al., 2000). These resultscated that embryo-specific programs were not com- indicated that although this protein is associated withpletely shut off or had been reinitiated in the seed- embryogenesis, it has a more general function in plantlings. This was confirmed by the ectopic expression of development. Recently, a cDNA of another MADS-embryo-specific markers in these plants. More inter- box containing transcription factor was identified byestingly, two plants among the T2 progenies of the investigating differentially expressed genes duringfew fertile transgenic lines exhibited a phenotype in Brassica microspore embryogenesis (Boutilier et al.,which embryo-like structures developed on their 2000). When this cDNA was overexpressed under theleaves. These structures expressed embryogenic control of the 35S promoter in transgenic plants,markers, thus confirming the activation or mainte- ectopic formation of embryos and cotyledons onnance of a gene expression program characteristic of leaves was observed, due to which the gene wasembryogenesis (Lotan et al., 1998). named babyboom (bbm). There are no data about its

More recently, another gene, lec2, has been iden- role during zygotic or somatic embryogenesis, but ittified in Arabidopsis, having similar characteristics can be supposed to have a general role during the(Stone et al., 2001). The function of lec2 was also different forms of plant embryogenesis.shown to be required during both early and late Gene activation tagging was used by Zuo et al.zygotic embryogenesis. It also codes for a transcrip- (2002) to identify genes whose over-expression in-tion factor, but has a plant specific B3 protein domain duces the formation of somatic embryos on Arabidop-and as such, is similar to the transcription factors sis tissues without the need for external hormonalviviparous1/ABA insensitive3 and fusca3, acting pri- treatments. The identified allele, PGA6, was found tomarily in developing seeds. Ectopic, postembryogenic be identical to WUSCHEL (WUS), a homeodomainexpression of lec2 in transgenic plants also resulted in protein previously shown to be involved in specifying

221

Figure 2. A schematic and generalized model of somatic embryogenic cell formation. The data are from different experimental systems andhave not been experimentally proved to act together on the hypothetical way shown in the figure. The proper order and timing of the differentevents is not indicated to maintain simplicity. Differentiated cells, if subjected to appropriate signals and hormones (wounding, stress, auxin,cytokinin, ABA) dedifferentiate and divide in parallel with chromatin remodelling. General chromatin decondensation characterizesdedifferentiation while remodelling of chromatin structure might also be required for the initiation of the embryogenic path of development.Dedifferentiation and cell division are delayed by high medium pH and embryogenic cell formation is also dependent on cellular pH changes.Embryogenic competent cell is formed if the cell division-promoting auxin signal coincides with a stress signal. Effectiveness of inducing

2�signals is dependent on the endogenous hormone levels. Formation of the embryogenic cell type is promoted by external Ca . In embryogeniccompetent and embryogenic cells, polarity is established as indicated by the asymmetric distribution of the JIM8 arabinogalactan protein

2�(AGP) epitope, the Ca -channels, actin filaments and by polarized secretion of cell wall material. Division of embryogenic competent cells�results in JIM8 nurse cells providing soluble signals (AGPs?) to promote development of embryogenic cells. The AGP-signal is generated by

the contribution of endochitinases. Embryogenic competent cells are marked by the expression of the somatic embryogenesis receptor kinase(SERK). Neither the ligand(s) nor the targets of this putative receptor are known at present. At the end of the signalling cascade transcriptionfactors like bbm, lec1 or lec2 are activated and they switch on the embryogenic pathway of development.

stem cell fate in shoot and floral meristems. WUS/ plants. During the last few years there has been aPGA6 presumably promotes a vegetative-to-em- considerable increase in the amount of informationbryogenic transition and/or maintains embryonic related to somatic embryogenesis gained in differentstem cell identity. culture systems, especially carrot, chicory, alfalfa and

Finally, as we have discussed above (Chromatin conifer tissue cultures. In spite of this accumulation ofremodelling: open for changes section), regulators of experimental data, we are still far from understandingchromatin remodelling likely play a key role in sup- the key events underlying the transition of differen-ressing embryogenic developmental pathways, and tiated somatic cells to the totipotent and embryogenicthus can be considered as main regulators of the cell state. The recent identification of genes that aredevelopmental switch during somatic embryo induc- markers of switch in cell fate (like SERK) or aretion (for review, e.g., Grossniklaus et al., 2001). themselves capable of inducing embryogenic de-

velopment in somatic cells (like bbm, lec1 and lec2 )has opened up new approaches to the question of

Concluding remarks developmental flexibility and determination in plants.Interestingly, although somatic embryogenesis is

Somatic embryogenesis is a unique experimental considered to be a model proposed to understand earlymodel to study developmental flexibility in higher events of zygote development, investigations on

222

Boutilier K, Offringa R, Fukuoka H, Sharma V, Kieft H, vanArabidospsis embryogenesis mutants ( gnom, lec1Lammeren AAM, Ouellett T & van Lookeren C (2000) Ectopicand lec2, pkl, etc.) have provided, and may stillexpression of the Brassica napus baby boom gene triggers aprovide, important clues to somatic embryogenesis.homeotic conversion of vegetative tissues into embryos and

In addition to regulation at the level of gene expres- cotyledons (Abstract). Plant Mol. Biol. Rep. 18: s11–s14´sion, the importance of the physiological state of the ¨ ´Bogre L, Stefanov I, Abraham M, Somogyi I & Dudits D (1990)

Differences in the responses to 2,4-dichlorophenoxyacetic acidcells, the presence of cell wall-derived or secreted(2,4-D) treatment between embryogenic and non-embryogenicsignals, endogenous hormones and the interaction oflines of alfalfa. In: Nijkamp HJJ, van der Plas LHW & Vandifferent signalling cascades is evident from manyAartrijk J (eds) Progress in Plant Cellular and Molecular Biology

studies. Based on these data, derived from different (pp. 427–436). Kluwer Academic Publishers, Dordrechtexperimental systems, including those where somatic Busch M, Mayer U & Jurgens G (1996) Molecular analysis of theembryo development is directly induced on the sur- Arabidopsis pattern formation gene GNOM: gene structure and

intragenic complementation. Mol. Gen. Genet. 250: 681–691face of explants and those where the establishment ofCapitanio G, Baldan B, Filippini F, Terzi M, LoSchiavo F &embryogenic cell cultures is necessary, we propose a

Mariani P (1997) Morphogenetic effects of Brefeldin A onmodel describing the present knowledge of the embryogenic cell cultures of Daucus carota L. Planta 203: 121–characteristics of an embryogenic cell. This schematic 128and generalized model of a ‘virtual’ embryogenic cell Chapman A, Blervacq AS, Vasseur J & Hilbert JL (2000)

Arabinogalactan-proteins in Cichorium somatic embryogenesis:is shown in Figure 2.effect of beta-glucosyl Yariv reagent and epitope localizationduring embryo development. Planta 211: 305–314

´´Charriere F, Sotta B, Miginiac E & Hahne G (1999) Induction ofAcknowledgements adventitious or somatic embryos on in vitro cultured zygotic

embryos of Helianthus annuus: Variation of endogenous hor-mone levels. Plant Physiol. Biochem. 37: 751–757The presented experiments on leaf protoplast-derived

Chaudhury AM, Ming L, Miller C, Craig S, Dennis ES & Peacockembryogenic alfalfa cells were supported by the WJ (1997) Fertilization-independent seed development ingrants IC15-CT96-0906, OTKA T034818 and BIO- Arabidopsis thaliana. Proc. Natl. Acad. Sci. USA 94: 4223–

´00062/2000. AF is also thankful for the ‘Janos 4228Chaudhury AM, Koltunow A, Payne T, Luo M, Tucker MR,´Bolyai’ Research Fellowship for its support.

Dennis ES & Peacock WJ (2001) Control of early seed develop-ment. Annu. Rev. Cell Dev. Biol. 17: 677–699

Choi YE, Kim HS, Soh WY & Yang DC (1997) Developmental andReferences structural aspects of somatic embryos formed on medium con-

taining 2,3,5-triiodobenzoic acid. Plant Cell Rep. 16: 738–744Anil VS & Rao KS (2000) Calcium-mediated signaling during Choi YE, Yang DC, Park JC, Soh WY & Choi KT (1998) Re-

sandalwood somatic embryogenesis. Role for exogenous calcium generative ability of somatic single and multiple embryos fromas second messenger. Plant Physiol. 123: 1301–1311 cotyledons of Korean ginseng on hormone-free medium. Plant

Antoine AF, Faure JE, Cordeiro S, Dumas C, Rougier M & Feijo Cell Rep. 17: 544–551JA (2000) A calcium influx is triggered and propagates in the Coca MA, Almoguera C & Jordano J (1994) Expression of sun-zygote as a wavefront during in vitro fertilization of flowering flower low-molecular-weight heat-shock proteins duringplants. Proc. Natl. Acad. Sci. USA 97: 10643–10648 embryogenesis and persistence after germination: localization

Baldan B, Guzzo F, Filippini F, Gasparian M, LoSchiavo F, Vitale and possible functional implications. Plant Mol. Biol. 25: 479–A, deVries SC, Mariani P & Terzi M (1997) The secretory nature 492of the lesion of carrot cell variant ts11, rescuable by endochitin- Cove DJ (2000) The generation and modification of cell polarity. J.ase. Planta 203: 381–389 Exp. Bot. 51: 831–838

Basu S, Sun H, Brian L, Quatrano RL & Muday GK (2002) Early Criqui MC, Plesse B, Durr A, Marbach J, Parmentier Y, Jamet E &embryo development in Fucus distichus is auxin sensitive. Plant Fleck J (1992) Characterization of genes expressed in mesophyllPhysiol. 130: 292–302 protoplasts of Nicotiana sylvestris before the re-initiation of the

Beeckman T, Burssens S & Inze D (2001) The peri-cell-cycle in DNA replicational activity. Mech. Dev. 38: 121–132Arabidopsis. J. Exp. Bot. Suppl. 52: 403–411 Criqui MC, Parmentier Y, Derevier A, Shen WH, Dong AW &

Belanger KD & Quatrano RS (2000) Polarity: the role of localized Genschik P (2000) Cell cycle-dependent proteolysis and ectopicsecretion. Curr. Opin. Plant Biol. 3: 67–72 overexpression of cyclin B1 in tobacco BY2 cells. Plant J. 24:

Bibikova TN, Jacob T, Dahse I & Gilroy S (1998) Localized 763–773´ ´ ¨ ¨changes in apoplastic and cytoplasmic pH are associated with Davletova S, Meszaros T, Miskolczi P, Oberschall A, Torok K,

´root hair development in Arabidopsis thaliana. Development Magyar Z, Dudits D & Deak M (2001) Auxin and heat shock125: 2925–2934 activation of a novel member of the calmodulin like domain

Blervacq AS, Dubois T, Dubois J & Vasseur J (1995) First protein kinase gene family in cultured alfalfa cells. J. Exp. Bot.divisions of somatic embryogenic cells in Cichorium hybrid 52: 215–221‘474’. Protoplasma 186: 163–168 de Jong AJ, Cordewener J, Lo SF, Terzi M,Vandekerckhove J, van

223

Kammen A & de Vries SC (1992) A carrot somatic embryo postembryonically: Inhibition of perianth senescence and abscis-mutant is rescued by chitinase. Plant Cell 4: 425–433 sion via constitutive expression. Plant Cell 12: 183–197

den Boer BG & Murray JA (2000) Triggering the cell cycle in Filonova LH, Bozhkov PV & von Arnold S (2000) Developmentalplants. Trends Cell Biol. 10: 245–250 pathway of somatic embryogenesis in Picea abies as revealed by

Denchev PD, Velcheva MR & Atanassov AI (1991) A new ap- time-lapse tracking. J. Exp. Bot. 51: 249–264proach to direct somatic embryogenesis in Medicago. Plant Cell Fleck J & Durr A (1980) Comparison of proteins synthetized inRep. 10: 338–341 vivo and in vitro by mRNA of isolated protoplasts. Planta 148:

Deshpande S & Hall JC (2000) Auxinic herbicide resistance may 453–454be modulated at the auxin-binding site in wild mustard (Sinapis Fletcher JC & Meyerowitz EM (2000) Cell signaling within thearvensis L.): A light scattering study. Pestic. Biochem. Physiol. shoot meristem. Curr. Opin. Plant Biol. 3: 23–3066: 41–48 Fowler JE & Quatrano RS (1997) Plant cell morphogenesis: Plasma

Dijak M & Simmonds DH (1988) Microtubule organization during membrane interactions with the cytoskeleton and cell wall.early direct embryogenesis from mesophyll protoplasts of Annu. Rev. Cell Dev. Biol. 13: 697–743Medicago sativa L. Plant Sci. 58: 183–191 Fowler MR, Ong LM, Russinova E, Atanassov AI, Scott NW,

Dodeman VL, Ducreux G & Kreis M (1997) Zygotic embryo- Slater A & Elliott MC (1998) Early changes in gene expressiongenesis versus somatic embryogenesis. J. Exp. Bot. 48: 1493– during direct somatic embryogenesis in alfalfa revealed by RAP-1509 PCR. J. Exp. Bot. 49: 249–253

Dong JZ & Dunstan DI (1996) Expression of abundant mRNAs Frelin C,Vigne P, Ladoux A & Lazdunski M (1988) The regulationduring somatic embryogenesis of white spruce [Picea glauca of the intracellular pH in cells from vertebrates. Eur. J. Biochem.(Moench) Voss]. Planta 199: 459–466 174: 3–14

Dong JZ & Dunstan DI (1999) Cloning and characterization of six Gao CY & Zelenka PS (1997) Cyclins, cyclin-dependent kinasesembryogenesis-associated cDNAs from somatic embryos of and differentiation. Bioessays 19: 307–315Picea glauca and their comparative expression during zygotic Geldner N, Friml J, Stierhof YD, Jurgens G & Palme K (2001)embryogenesis. Plant Mol. Biol. 39: 859–864 Auxin transport inhibitors block PIN1 cycling and vesicle traf-

Dubois T, Guedira M, Dubois J & Vasseur J (1990) Direct somatic ficking. Nature 413: 425–428embryogenesis in roots of Cichorium. Is callose an early marker? Genschik P, Criqui MC, Parmentier Y, Derevier A & Fleck J (1998)Ann. Bot. 65: 539–545 Cell cycle-dependent proteolysis in plants: Identification of the

Dubois T, Guedira M, Dubois J & Vasseur J (1991) Direct somatic destruction box pathway and metaphase arrest produced by theembryogenesis in leaves of Cichorium. A histological and proteasome inhibitor MG132. Plant Cell 10: 2063–2075S.E.M. study of early stages. Protoplasma 162: 120–127 Gianazza E, De Ponti P, Scienza A, Villa P & Martinelli L (1992)

¨ ¨Dudits D, Bogre L & Gyorgyey J (1991) Molecular and cellular Monitoring by two-dimensional electrophoresis somatic embryo-approaches to the analysis of plant embryo development from genesis in leaf and petiole explants from Vitis. Electrophoresissomatic cells in vitro. J. Cell Sci. 99: 475–484 13: 203–209

¨ ¨ ´Dudits D, Gyorgyey J, Bogre L & Bako L (1995) Molecular Giroux RW & Pauls KP (1997) Characterization of somaticbiology of somatic embryogenesis. In: Thorpe TA (ed) In Vitro embryogenesis-related cDNAs from alfalfa (Medicago sativa L.).Embryogenesis in Plants (pp. 267–308). Kluwer Academic Plant Mol. Biol. 33: 393–404Publisher, Dordrecht Goldsworthy A & Mina MG (1991) Electrical patterns of tobacco

´ ´ ´Dudits D, Magyar Z, Deak M, Meszaros T, Miskolczi P, Feher A, cells in media containing indole-3-acetic acid or 2,4-dichloro-Brown S, Kondorosi E, Athanasiadis A, Pongor S, Bako L, phenoxyacetic acid. Planta 183: 386–373

¨Koncz C & Gyorgyey J (1998) Cyclin- and calcium-dependent Gray WM & Estelle M (2000) Function of the ubiquitin–protea-kinase families: response of cell division cycle to hormone and some pathway in auxin response. Trends Biochem. Sci. 25:stress signals. In: Francis D, Dudits D & Inze D (eds) Plant Cell 133–138Division (pp. 21–47). Portland Press, London, UK Gray WM, del Pozo JC, Walker L, Hobbie L, Risseeuw E, Banks T,

Eastmond PJ & Rawsthorne S (1998) Comparison of the metabolic Crosby WL, Yang M, Ma H & Estelle M (1999) Identification ofproperties of plastids isolated from developing leaves or embryos an SCF ubiquitin–ligase complex required for auxin response inof Brassica napus L. J. Exp. Bot. 49: 1105–1111 Arabidopsis thaliana. Genes Dev. 13: 1678–1691

Egertsdotter U & von Arnold S (1998) Development of somatic Gray WM, Kepinski S, Rouse D, Leyser O & Estelle M (2001)embryos in Norway spruce. J. Exp. Bot. 49: 155–162 Auxin regulates SCFTIR1-dependent degradation of AUX/IAA

Fargasova A (1994) Comparative study of plant growth hormone proteins. Nature 414: 271–276(herbicide) toxicity in various biological subjects. Ecotoxicol. Gross JD, Bradbury J, Kay RR & Peacey MJ (1983) IntracellularEnviron. Saf. 29: 359–364 pH and the control of cell differentiation in Dictyostelium

Faure JE & Dumas C (2001) Fertilization in flowering plants. New discoideum. Nature 303: 244–245approaches for an old story. Plant Physiol. 125: 102–104 Grosset J, Marty I, Chartier Y & Meyer Y (1990) mRNAs newly

´Feher A, Pasternak T, Miskolczi P, Ayaydin F & Dudits D (2001) synthetized by tobacco mesophyll protoplasts are wound-induc-Induction of the embryogenic pathway in somatic plant cells. ible. Plant Mol. Biol. 15: 485–496Acta Hort. 560: 293–298 Grossmann K (2000) Mode of action of auxinic herbicides: a new

¨´ ¨Feher A, Pasternak T, Otvos K, Miskolczi P & Dudits D (2002) ending to a long, drawn out story. Trends Plant Sci. 5: 506–508Induction of embryogenic competence in somatic plant cells: a Grossniklaus U, Vielle-Calzada JP, Hoeppner MA & Gagliano WBreview. Biologia 57: 5–12 (1998) Maternal control of embryogenesis by MEDEA, a poly-

Fernandez DE, Heck GR, Perry SE, Patterson SE, Bleecker AB & comb group gene in Arabidopsis. Science 280: 446–450Fang SC (2000) The embryo MADS domain factor AGL15 acts Grossniklaus U, Spillane C, Page DR & Kohler C (2001) Genomic

224

imprinting and seed development: endosperm formation with and chromodomain show developmental alterations: possible role ofwithout sex. Curr. Opin. Plant Biol. 4: 21–27 Pc chromodomain proteins in chromatin-mediated gene regula-

Gutierrez C (1998) The retinoblastoma pathway in plant cell cycle tion in plants. Plant Cell 11: 1047–1060and development. Curr. Opin. Plant Biol. 1: 492–497 Ingvardsen C & Veierskov B (2001) Ubiquitin- and proteasome-¨Gyorgyey J, Nemeth K, Magyar Z, Kelemen Z, Alliotte T, Inze D & dependent proteolysis in plants. Physiol. Plant. 112: 451–459

Dudits D (1997) Expression of a novel-type small proline-rich Ivanova A,Velcheva M, Denchev P, Atanassov A & Van Onckelenprotein gene of alfalfa is induced by 2,4-dichlorophenoxiacetic H (1994) Endogenous hormone levels during direct somaticacid in dedifferentiated callus cells. Plant Mol. Biol. 34: 593– embryogenesis in Medicago falcata. Physiol. Plant. 92: 85–89602 Jack T (2001a) Plant development going MADS. Plant Mol. Biol.

Hecht V, Vielle-Calzada JP, Hartog MV, Schmidt ED, Boutilier K, 46: 515–520Grossniklaus U & de Vries SC (2001) The Arabidopsis somatic Jack T (2001b) Relearning our ABCs: new twists on an old model.embryogenesis receptor kinase 1 gene is expressed in developing Trends Plant Sci. 6: 310–316ovules and embryos and enhances embryogenic competence in Jackson CL & Casanova JE (2000) Turning on ARF: the Sec7culture. Plant Physiol. 127: 803–816 family of guanine–nucleotide-exchange factors. Trends Cell

Helleboid S, Chapman A, Hendriks T, Inze D, Vasseur J & Hilbert Biol. 10: 60–67JL (2000a) Cloning of beta-1,3-glucanases expressed during Jamet E, Durr A, Parmentier Y, Criqui MC & Fleck J (1990) IsCichorium somatic embryogenesis. Plant Mol. Biol. 42: 377– ubiquitin involved in the dedifferentiation of higher plant cells?386 Cell. Differ. Dev. 29: 37–46

Helleboid S, Hendriks T, Bauw G, Inze D, Vasseur J & Hilbert JL Jansen MAK, Booij H, Schel JHN & de Vries SC (1990) Calcium(2000b) Three major somatic embryogenesis related proteins in increases the yield of somatic embryos in carrot embryogenicCichorium identified as PR proteins. J. Exp. Bot. 51: 1189–1200 suspension cultures. Plant Cell Rep. 9: 221–223

Hemerly AS, Ferreira P, de Almeida EJ,Van Montagu M, Engler G Jimenez VM & Bangerth F (2001a) Endogenous hormone con-& Inze D (1993) cdc2a expression in Arabidopsis is linked with centrations and embryogenic callus development in wheat. Plantcompetence for cell division. Plant Cell 5: 1711–1723 Cell Tiss. Org. Cult. 67: 37–46

Hemerly AS, Ferreira PCG,Van Montagu M & Inze D (1999) Cell Jimenez VM & Bangerth F (2001b) Endogenous hormone levels incycle control and plant morphogenesis: is there an essential link? explants and in embryogenic and non-embryogenic cultures ofBioessays 21: 29–37 carrot. Physiol. Plant. 111: 389–395

Hemerly AS, Ferreira PC, Van Montagu M, Engler G & Inze D JimenezVM & Bangerth F (2001c) Hormonal status of maize initial(2000) Cell division events are essential for embryo patterning explants and of the embryogenic and non-embryogenic callusand morphogenesis: studies on dominant-negative cdc2aAt mu- cultures derived from them as related to morphogenesis in vitro.tants of arabidopsis. Plant J. 23: 123–130 Plant Sci. 160: 247–257

Hendriks T, Scheer I, Quillet MC, Randoux B, Delbreil B,Vasseur J John PC (1996) The plant cell cycle: conserved and unique features& Hilbert JL (1998) A non-symbiotic hemoglobin gene is in mitotic control. Prog. Cell Cycle Res. 2: 59–72expressed during somatic embryogenesis in Cichorium. Biochim. Jurgens G (1992) Pattern formation in the flowering plant embryo.Biophys. Acta 1443: 193–197 Curr. Opin. Genet. Dev. 2: 567–570

Hirt H (2000) Connecting oxidative stress, auxin, and cell cycle Jurgens G (2001) Apical–basal pattern formation in Arabidopsisregulation through a plant mitogen-activated protein kinase embryogenesis. EMBO J. 20: 3609–3616

´pathway. Proc. Natl. Acad. Sci. USA 97: 2405–2407 Jurgens G, Mayer U, Torres R, Berleth T & Misera S (1991)¨ ¨Hirt H, Pay A, Gyorgyey J, Bako L, Nemeth K, Bogre L, Schweyen Genetic analysis of pattern formation in the Arabidopsis embryo.

RJ, Heberle-Bors E & Dudits D (1991) Complementation of a Development Suppl. 1: 27–38yeast cell cycle mutant by an alfalfa cDNA encoding a protein Jurgens G, Grebe M & Steinmann T (1997) Establishment of cellkinase homologous to p34cdc2. Proc. Natl. Acad. Sci. USA 88: polarity during early plant development. Curr. Opin. Cell Biol. 9:1636–1640 849–852

Holk A, Kaldenhoff R & Richter G (1996) Regulation of an Kairong C, Gengsheng X, Lin Q, Xinmin L & Yafu W (1999) Theembryogenic carrot gene (DC 2.15) and identification of its analysis of differential gene expression in early somatic embryo-active promoter sites. Plant Mol. Biol. 31: 1153–1161 genesis on Lycium barbarum. Plant Cell Tiss. Org. Cult. 59:

Huang L, Franklin AE & Hoffman NE (1993) Primary structure 169–174and characterization of an Arabidopsis thaliana calnexin-like Kamada H, Kobayashi K, Kiyosue T & Harada H (1989) Stressprotein. J. Biol. Chem. 268: 6560–6566 induced somatic embryogenesis in carrot and its application to

Hulskamp M & Schnittger A (1998) Spatial regulation of trichome synthetic seed production. In Vitro Cell Dev. Biol. 25: 1163–116formation in Arabidopsis thaliana. Semin. Cell. Dev. Biol. 9: Kamada H, Ishikawa K, Saga H & Harada H (1993) Induction of213–220 somatic embryogenesis in carrot by osmotic stress. Plant Tiss.

Huntley R, Healy S, Freeman D, Lavender P, de Jager S, Green- Cult. Lett. 10: 38–44wood J, Makker J, Walker E, Jackman M, Xie Q, Bannister AJ, Kitamiya E, Suzuki S, Sano T & Nagata T (2000) Isolation of twoKouzarides T, Gutierrez C, Doonan JH & Murray JA (1998) The genes that were induced upon the initiation of somatic embryo-maize retinoblastoma protein homologue ZmRb-1 is regulated genesis on carrot hypocotyls by high concentrations of 2,4-D.during leaf development and displays conserved interactions Plant Cell Rep. 19: 551–557with G1/S regulators and plant cyclin D (CycD) proteins. Plant Kiyosue T, Takano K, Kamada H & Harada H (1990) Induction ofMol. Biol. 37: 155–169 somatic embryogenesis in carrot by heavy metal ions. Can. J.

Ingram R, Charrier B, Scollan C & Meyer P (1999) Transgenic Bot. 68: 2301tobacco plants expressing the Drosophila Polycomb (Pc) Knox JP (1997) The use of antibodies to study the architecture and

225

developmental regulation of plant cell walls. Int. Rev. Cytol. Mayer U & Jurgens G (1998) Pattern formation in plant embryo-genesis: a reassessment. Semin. Cell. Dev. Biol. 9: 187–193171: 79–120

McCabe PF, Valentine TA, Forsberg LS & Pennell RI (1997)Koltunow A (1993) Apomixis: embryo sacs and embryos formedSoluble signals from cells identified at the cell wall establish awithout meiosis or fertilization in ovules. Plant Cell 5: 1425–developmental pathway in carrot. Plant Cell 9: 2225–22411437

Meijer M & Murray JA (2001) Cell cycle controls and the develop-Koltunow AM, Bicknell RA & Chaudhury AM (1995) Apomixis:ment of plant form. Curr. Opin. Plant Biol. 4: 44–49Molecular strategies for the generation of genetically identical

Meinke DW (1992) A homeotic mutant of Arabidopsis thalianaseeds without fertilization. Plant Physiol. 108: 1345–1352with leafy cotyledons. Science 258: 1647–1650Kranz E (1999) In vitro fertilization with isolated single gametes.

Meinke DW, Franzmann LH, Nickle TC & Yeung EC (1994) LeafyMethods Mol. Biol. 111: 259–267cotyledon mutants of Arabidopsis. Plant Cell 6: 1049–1064

Kreuger M & van Holst GJ (1993) Arabinogalactan proteins are´ ´ ´Meszaros T, Miskolczi P, Ayaydin F, Pettko-Szandtner A, Magyar

essential in somatic embryogenesis of Daucus carota L. Planta´ ´ ´Z, Horvath VG, Bako L, Feher A & Dudits D (2000) Multiple

189: 243–248cyclin-dependent kinase complexes and phosphatases control

Kropf DL, Bisgrove SR & Hable WE (1999) Establishing a growthG2/M progression in alfalfa cells. Plant Mol. Biol. 43: 595–605

axis in fucoid algae. Trends Plant Sci. 4: 490–494 Michalczuk L & Druart P (1999) Indole-3-acetic acid metabolismKurkdjian A & Guern J (1989) Intracellular pH: measurement and in hormone-autotrophic, embryogenic callus of Inmil (R) cherry

importance in cell activity. Annu. Rev. Plant Physiol. Plant Mol. rootstock (Prunus incisaserrula ‘GM 9’) and in hormone-Biol. 40: 277–303 dependent, non-embryogenic calli of Prunus incisaserrula and

Lakshmanan P & Taji A (2000) Somatic embryogenesis in Prunus domestica. Physiol. Plant. 107: 426–432leguminous plants. Plant Biol. 2: 136–148 Michalczuk L, Cooke TJ & Cohen JD (1992a) Auxin levels at

Larkin JC, Marks MD, Nadeau J & Sack F (1997) Epidermal cell different stages of carrot somatic embryogenesis. Phytochemis-fate and patterning in leaves. Plant Cell 9: 1109–1120 try 31: 1097–1103

Leyser O (2001) Auxin signalling: the beginning, the middle and Michalczuk L, Ribnicky DM, Cooke TJ & Cohen JD (1992b)the end. Curr. Opin. Plant Biol. 4: 382–386 Regulation of indole-3-acetic acid biosynthetic pathways in

Lichtenthaler HK (1998) The stress concept in plants: an intro- carrot cell cultures. Plant Physiol. 100: 1346–1353duction. Ann. NY Acad. Sci. 851: 187–198 Mironov V, De Veylder L, Van Montagu M & Inze D (1999)

Lin X, Hwang GJ & Zimmerman JL (1996) Isolation and charac- Cyclin-dependent kinases and cell division in plants – Theterization of a diverse set of genes from carrot somatic embryos. nexus. Plant Cell 11: 509–521Plant Physiol. 112: 1365–1374 Mordhorst AP, Toonen MAJ & de Vries SC (1997) Plant embryo-

Lloyd AM, Schena M, Walbot V & Davis RW (1994) Epidermal genesis. Crit. Rev. Plant Sci. 16: 535–576cell fate determination in Arabidopsis: patterns defined by a Mordhorst AP, Voerman KJ, Hartog MV, Meijer EA, van Went J,steroid-inducible regulator. Science 266: 436–439 Koornneef M & de Vries SC (1998) Somatic embryogenesis in

Lloyd CW, Lowe SB & Peace GW (1980) The mode of action of Arabidopsis thaliana is facilitated by mutations in genes re-2,4-D in counter-acting the elongation of carrot cells grown in pressing meristematic cell divisions. Genetics 149: 549–563culture. J. Cell Sci. 195: 309–312 Morimoto RI (1998) Regulation of the heat shock transcriptional

Lo Schiavo F, Giuliano G, deVries SC, Genga A, Bollini R, Pitto L, response: cross talk between a family of heat shock factors,Cozzani F, Nuti-Ronchi V & Terzi M (1990) A carrot cell molecular chaperones, and negative regulators. Genes Dev. 12:variant temperature sensitive for somatic embryogenesis reveals 3788–3796a defect in the glycosylation of extracellular proteins. Mol. Gen. Mu JH, Lee HS & Kao TH (1994) Characterization of a pollen-Genet. 223: 385–393 expressed receptor-like kinase gene of Petunia inflata and the

Lotan T, Ohto M, Yee KM, West MAL, Lo R, Kwong RW, activity of its encoded kinase. Plant Cell 6: 709–721Yamagishi K, Fischer RL, Goldberg RB & Harada JJ (1998) Nagata T, Ishida S, Hasezawa S & Takahashi Y (1994) GenesArabidopsis LEAFY COTYLEDON1 is sufficient to induce involved in the dedifferentiation of plant cells. Int. J. Dev. Biol.embryo development in vegetative cells. Cell 93: 1195–1205 38: 321–327

Luo M, Bilodeau P, Koltunow A, Dennis ES, Peacock WJ & Nishiwaki M, Fujino K, Koda Y, Masuda K & Kikuta Y (2000)Chaudhury AM (1999) Genes controlling fertilization-indepen- Somatic embryogenesis induced by the simple application ofdent seed development in Arabidopsis thaliana. Proc. Natl. abscisic acid to carrot (Daucus carota L.) seedlings in culture.Acad. Sci. USA 96: 296–301 Planta 211: 756–759

Luo Y & Koop HU (1997) Somatic embryogenesis in cultured Nomura K & Komamine A (1985) Identification and isolation ofimmature zygotic embryos and leaf protoplasts of Arabidopsis single cells that produce somatic embryos at a high frequency inthaliana ecotypes. Planta 202: 387–396 a carrot cell suspension culture. Plant Physiol. 79: 988–991

Marrs KA (1996) The functions and regulation of glutathione Nomura K & Komamine A (1995) Physiological and biologicalS-transferases in plants. Annu. Rev. Plant Physiol. Plant Mol. aspects of somatic embryogenesis. In: Thorpe TA (ed) In VitroBiol. 47: 127–158 Embryogenesis in Plants (pp. 249–266). Kluwer Academic

Marty F (1999) Plant vacuoles. Plant Cell 11: 587–600 Publishers, DordrechtMarty I, Brugidou C, Chartier Y & Meyer Y (1993) Growth-related Ober SS & Pardee AB (1987) Intracellular pH is increased after

gene expression in Nicotiana tabacum mesophyll protoplasts. transformation of Chinese hamster embryo fibroblasts. Proc.Plant J. 4: 265–278 Natl. Acad. Sci. USA 84: 2766–2770

Matsue T, Koike S & Uchida I (1993) Microamperometric estima- Ogas J, Cheng JC, Sung ZR & Somerville C (1997) Cellulartion of photosynthesis inhibition in a single algal protoplast. differentiation regulated by gibberellin in the ArabidopsisBiochem. Biophys. Res. Commun. 197: 1283–1287 thaliana pickle mutant. Science 277: 91–94

226

Ogas J, Kaufmann S, Henderson J & Somerville C (1999) PICKLE Ratajczak R (2000) Structure, function and regulation of the plantis a CHD3 chromatin-remodeling factor that regulates the transi- vacuolar H(�)-translocating ATPase. Biochim. Biophys. Actation from embryonic to vegetative development in Arabidopsis. 1465: 17–36Proc. Natl. Acad. Sci. USA 96: 13839–13844 Reynolds TL (1997) Pollen embryogenesis. Plant Mol. Biol. 33:

Oh SH, Steiner HY, Dougall DK & Roberts DM (1992) Modulation 1–10of calmodulin levels, calmodulin methylation, and calmodulin Ribnicky DM, Cohen JD, Hu WS & Cooke TJ (2001) An auxinbinding proteins during carrot cell growth and embryogenesis. surge following fertilization in carrots: a mechanism for regulat-Arch. Biochem. Biophys. 297: 28–34 ing plant totipotency. Planta DOI 10.1007/s004250100639

Ohad N, Margossian L, Hsu Y, Williams C, Repetti P & Fischer RL Riou-Khamlichi C, Huntley R, Jacqmard A & Murray JA (1999)(1996) A mutation that allows endosperm development without Cytokinin activation of Arabidopsis cell division through a D-fertilization. Proc. Natl. Acad. Sci. USA 93: 5319–5324

type cyclin. Science 283: 1541–1544Ohad N,Yadegari R, Margossian L, Hannon M, Michaeli D, Harada

Robinson KR & Miller BJ (1997) The coupling of cyclic GMP andJJ, Goldberg RB & Fischer RL (1999) Mutations in FIE, a WD

photopolarization of Pelvetia zygotes. Dev. Biol. 187: 125–130polycomb group gene, allow endosperm development without

Robinson KR, Lorenzi R, Ceccarelli N & Gualtieri P (1998) Retinalfertilization. Plant Cell 11: 407–416

identification in Pelvetia fastigiata. Biochem. Biophys. Res.Osuga K, Masuda H & Komamine A (1999) Synchronization ofCommun. 243: 776–778somatic embryogenesis at high frequency using carrot suspen-

Robinson KR, Wozniak M, Pu R & Messerli M (1999) Symmetrysion cultures: model systems and application in plant develop-breaking in the zygotes of the fucoid algae: controversies andment. Methods Cell Sci. 21: 129–140recent progress. Curr. Top. Dev. Biol. 44: 101–125Overvoorde PJ & Grimes HD (1994) The role of calcium and

Rock KL, Gramm C, Rothstein L, Clark K, Stein R, Dick L, Hwangcalmodulin in carrot somatic embryogenesis. Plant Cell Physiol.D & Goldberg AL (1994) Inhibitors of the proteasome block the35: 135–144degradation of most cell proteins and the generation of peptides¨Palme K & Galweiler L (1999) PIN-pointing the molecular basis ofpresented on MHC class I molecules. Cell 78: 761–771auxin transport. Curr. Opin. Plant Biol. 2: 375–381

Rogg LE & Bartel B (2001) Auxin Signaling. Derepression through´ ´Pasternak T, Miskolczi P, Ayaydin F, Meszaros T, Dudits D &regulated proteolysis. Dev. Cell. 1: 595–604´Feher A (2000) Exogenous auxin and cytokinin dependent

Roitsch T (1999) Source-sink regulation by sugar and stress. Curr.activation of CDKs and cell division in leaf protoplast-derivedOpin. Plant Biol. 2: 198–206cells of alfalfa. Plant Growth Regul. 32: 129–141

Russell SD (1993) The egg cell: Development and role in fertiliza-Pasternak T, Prinsen E, Ayaydin F, Miskolczi P, Potters G, Asardtion and early embryogenesis. Plant Cell 5: 1349–1359´H, Van Onckelen H, Dudits D & Feher A (2002) The role of

Russinova E, Shao C-Y, Fowler M, Iantcheva A, Atanassov AI,auxin, pH and stress in the activation of embryogenic cellScott NW, Elliott MC & Slater A (1998) The regulation of genedivision in leaf protoplast-derived cells of alfalfa (Medicago

sativa L.). Plant Physiol 129: 1807–1819 expression and cell division during direct somatic embryogenesisPedroso MC & Pais MS (1995) Factors controlling somatic in alfalfa. (Abstract) 5th Plant Embryogenesis Workshop. 19–21

embryogenesis – Cell wall changes as an in vivo marker of November 1998. Barcelona, Spainembryogenic competence. Plant Cell Tiss. Org. Cult. 43: 147– Sabala I, Elfstrand M, Farbos I, Clapham D & von Arnold S (2000)154 Tissue-specific expression of Pa18, a putative lipid transfer

Pennell RI, Janniche L, Scofield GN, Booij H, de Vries SC & protein gene, during embryo development in Norway spruceRoberts K (1992) Identification of a transitional cell state in the (Picea abies). Plant Mol. Biol. 42: 461–478developmental pathway to carrot somatic embryogenesis. J. Cell Sagare AP, Lee YL, Lin TC, Chen CC & Tsay HS (2000) Cyto-Biol. 119: 1371–1380 kinin-induced somatic embryogenesis and plant regeneration in

Pennell RI, Cronk QC, Forsberg LS, Stohr C, Snogerup L, Kjel- Corydalis yanhusuo (Fumariaceae) – a medicinal plant. Plantlbom P & McCabe PF (1995) Cell-context signalling. Phil. Sci. 160: 139–147Trans. R. Soc. Lond. Biol. Sci. 350: 87–93 Sallandrouze A, Faurobert M, El Maataoui M & Espagnac H

Perera IY & Zielinski RE (1992) Structure and expression of the (1999) Two-dimensional electrophoretic analysis of proteinsArabidopsis CaM-3 calmodulin gene. Plant Mol. Biol. 19: 649– associated with somatic embryogenesis development in Cupres-664 sus sempervirens L. Electrophoresis 20: 1109–1119

Perry SE, Lehti MD & Fernandez DE (1999) The MADS-domain Samaj J, Baluska F, Bobak M & Volkmann D (1999) Extracellularprotein AGAMOUS-like 15 accumulates in embryonic tissues matrix surface network of embryogenic units of friable maizewith diverse origins. Plant Physiol. 120: 121–129 callus contains arabinogalactan-proteins recognized by mono-

Pichon O & Desbiez MO (1994) Is cytoplasmic pH involved in the clonal antibody JIM4. Plant Cell Rep. 18: 369–374regulation of cell cycle in plants? Physiol. Plant. 92: 261–265 Sanders D, Brownlee C & Harper JF (1999) Communicating with

Poulsen GB, Frugis G, Albrechtsen M & Mariotti D (1996) calcium. Plant Cell 11: 691–706Synthesis of extracellular proteins in embryogenic and non- Sato S, Toya T, Kawahara R, Whittier RF, Fukuda H & Komamineembryogenic cell cultures of alfalfa. Plant Cell Tiss. Org. Cult. A (1995) Isolation of a carrot gene expressed specifically during44: 257–260 early-stage somatic embryogenesis. Plant Mol. Biol. 28: 39–46

Pu R & Robinson KR (1998) Cytoplasmic calcium gradients and Schaefer J (1985) Regeneration in alfalfa tissue culture. Plantcalmodulin in the early development of the fucoid alga Pelvetia Physiol. 79: 584compressa. J. Cell Sci. 111: 3197–3207 Schauf CL, Bringle B & Stillwell W (1987) Membrane-directed

Rajasekaran K, Hein MB, Davis GC, Carnes MG & Vasil IK (1987) effects of the plant hormones abscisic acid, indole-3-acetic acidExogenous growth regulators in leaves and tissue cultures of and 2,4-dichlorophenoxyacetic acid. Biochem. Biophys. Res.Pennisetum purpureum Schum. J. Plant Physiol. 130: 13–25 Commun. 143: 1085–1091

227

Scheres B & Benfey PN (1999) Asymmetric cell division in plants. Stricker SA (1999) Comparative biology of calcium signalingAnnu. Rev. Plant Physiol. Plant Mol. Biol. 50: 505–537 during fertilization and egg activation in animals. Dev. Biol. 211:

Schmidt ED, Guzzo F, Toonen MA & de Vries SC (1997) A 157–176leucine-rich repeat containing receptor-like kinase marks somatic Studzinski GP & Harrison LE (1999) Differentiation-relatedplant cells competent to form embryos. Development 124: 2049– changes in the cell cycle traverse. Int. Rev. Cytol. 189: 1–582062 Sultan SE (2000) Phenotypic plasticity for plant development,

Schnittger A, Schellmann S & Hulskamp M (1999) Plant cells – function and life history. Trends Plant Sci. 5: 537–542young at heart? Curr. Opin. Plant Biol. 2: 508–512 Swann K & Whitaker MJ (1990) Second messengers at fertilization

Schoffl F, Prandl R & Reindl A (1998) Regulation of the heat- in sea-urchin eggs. J. Reprod. Fertil. Suppl. 42: 141–153shock response. Plant Physiol. 117: 1135–1141 Sylvester AW (2000) Division decisions and the spatial regulation

Schomer M & Epel D (1999) The roles of changes in NADPH and of cytokinesis. Curr. Opin. Plant Biol. 3: 58–66pH during fertilization and artificial activation of the sea urchin Takahashi Y, Ishida S & Nagata T (1994) Function and modulationegg. Dev. Biol. 216: 394–405 of expression of auxin-regulated genes. Int. Rev. Cytol. 152:

Senger S, Mock HP, Conrad U & Manteuffel R (2001) Immuno- 109–144modulation of ABA function affects early events in somatic Taylor RL (1967) The foliar embryos of Malaxis paludosa. Can. J.embryo development. Plant Cell Rep. 20: 112–120 Bot. 45: 1553–1556

Servos J, Haase E & Brendel M (1993) Gene SNQ2 of Sac- Thoma S, Hecht U, Kippers A, Botella J, deVries S & Somerville Ccharomyces cerevisiae, which confers resistance to 4-nitro- (1994) Tissue-specific expression of a gene encoding a cellquinoline-N-oxide and other chemicals, encodes a 169 kDa wall-localized lipid transfer protein from Arabidopsis. Plantprotein homologous to ATP-dependent permeases. Mol. Gen. Physiol. 105: 35–45Genet. 236: 214–218 Thomas C, Bronner R, Molinier J, Prinsen E, Van Onckelen H &

Sharma KK & Thorpe TA (1995) Asexual embryogenesis in Hahne G (2002) Immuno-cytochemical localization of indole-3-vascular plants in nature. In: Thorpe TA (ed) In Vitro Embryo- acetic acid during induction of somatic embryogenesis in cul-genesis in Plants (pp. 17–72). Kluwer Academic Publishers, tured sunflower embryos. Planta 215: 577–583Dordrecht Thompson HJM & Knox JP (1998) Stage-specific responses of

Showalter AM (2001) Arabinogalactan-proteins: structure, expres- embryogenic carrot cell suspension cultures to arabinogalactansion and function. Cell. Mol. Life Sci. 58: 1399–1417 protein-binding beta-glucosyl Yariv reagent. Planta 205: 32–38

Smith DL & Krikorian AD (1990a) pH control of carrot somatic Toonen MAJ, Hendriks T, Schmidt EDL, Verhoeven HA, vanembryogenesis. In: Nijkamp HJJ, Van der Plas LHW & Van Kammen A & deVries SC (1994) Description of somatic-embryoAartrijk J (eds) Progress in Plant Cellular and Molecular Biology forming single cells in carrot suspension cultures employing(pp. 449–453). Kluwer Academic Publishers, Dordrecht video cell tracking. Planta 194: 565–572

Smith DL & Krikorian AD (1990b) Somatic proembryo production Toonen MAJ, Schmidt EDL, Hendriks T, Verhoeven HA, vanfrom excised, wounded zygotic carrot embryos on hormone-free Kammen A & deVries SC (1996) Expression of the JIM8 cellmedium: evaluation of the effects of pH, ethylene and activated wall epitope in carrot somatic embryogenesis. Planta 200: 167–charcoal. Plant Cell Rep. 9: 34 173

Somleva MN, Schmidt EDL & de Vries SC (2000) Embryogenic Toonen MA, Verhees JA, Schmidt ED, van Kammen A & de Vriescells in Dactylis glomerata L. (Poaceae) explants identified by SC (1997a) AtLTP1 luciferase expression during carrot somaticcell tracking and by SERK expression. Plant Cell Rep. 19: embryogenesis. Plant J. 12: 1213–1221718–726 Toonen MAJ, Schmidt EDL, van Kammen A & deVries SC (1997b)

Song J, Sorensen EL & Liang GH (1990) Direct embryogenesis Promotive and inhibitory effects of diverse arabinogalactanfrom single mesophyll protoplasts in alfalfa (Medicago sativa proteins on Daucus carota L. somatic embryogenesis. PlantaL.). Plant Cell Rep. 9: 21–25 203: 188–195

Souter M & Lindsey K (2000) Polarity and signalling in plant Torres-Ruiz RA & Jurgens G (1994) Mutations in the FASS geneembryogenesis. J. Exp. Bot. 51: 971–983 uncouple pattern formation and morphogenesis in Arabidopsis

Stals H & Inze D (2001) When plant cells decide to divide. Trends development. Development 120: 2967–2978Plant Sci. 6: 359–364 Touraev A, Pfosser M,Vicente O & HeberleBors E (1996) Stress as

Steenhoudt O & Vanderleyden J (2000) Azospirillum, a free-living the major signal controlling the developmental fate of tobacconitrogen-fixing bacterium closely associated with grasses: ge- microspores: Towards a unified model of induction of micros-netic, biochemical and ecological aspects. FEMS Microbiol. pore /pollen embryogenesis. Planta 200: 144–152Rev. 24: 487–506 Udvardy A (1996) The role of controlled proteolysis in cell–cycle

Steinmann T, Geldner N, Grebe M, Mangold S, Jackson CL, Paris regulation. Eur. J. Biochem. 240: 307–313S, Galweiler L, Palme K & Jurgens G (1999) Coordinated polar Vallee N (1997) Role Des Protons, Du Calcium Et Des Canauxlocalization of auxin efflux carrier PIN1 by GNOM ARF GEF. Ioniques Dans La Division Des Protoplastes De TournesolScience 286: 316–318 (Helianthus annuus L.)., PhD Thesis, L’Institute National Poly-

Sterk P, Booij H, Schellekens GA, van Kammen A & de Vries SC technique de Toulouse, Toulouse, France(1991) Cell-specific expression of the carrot EP2 lipid transfer van den Berg C,WillemsenV, Hendriks G,Weisbeek P & Scheres Bprotein gene. Plant Cell 3: 907–921 (1997) Short-range control of cell differentiation in the

Stone SL, Kwong LW, Yee KM, Pelletier J, Lepiniec L, Fischer RL, Arabidopsis root meristem. Nature 390: 287–289Goldberg RB & Harada JJ (2001) LEAFY COTYLEDON2 van den Berg C, Weisbeek P & Scheres B (1998) Cell fate and cellencodes a B3 domain transcription factor that induces embryo differentiation status in the Arabidopsis root. Planta 205: 483–development. Proc. Natl. Acad. Sci. USA 98: 11806–11811 491

228

van Hengel AJ, Tadesse Z, Immerzeel P, Schols H, van Kammen A Xu N & Bewley JD (1992) Contrasting pattern of somatic and& de Vries SC (2001) N-Acetylglucosamine and glucosamine- zygotic embryo development in alfalfa (Medicago sativa L.) ascontaining arabinogalactan proteins control somatic embryogen- revealed by scanning electron microscopy. Plant Cell Rep. 11:esis. Plant Physiol. 125: 1880–1890 279–284

Varga-Weisz PD & Becker PB (1998) Chromatin-remodeling fac- Xu XH, Briere C,Vallee N, Borin C, van Lammeren AAM, Aliberttors: machines that regulate? Curr. Opin. Cell Biol. 10: 346–353 G & Souvre A (1999) In vivo labeling of sunflower embryonic

Vernet T, Fleck J, Durr A, Fritsch C, Pinck M & Hirth L (1982) tissues by fluorescently labeled phenylalkylamine. ProtoplasmaExpression of the gene coding for the small subunit for ribulose 210: 52–58biphosphate carboxylase during differentiation of tobacco plant Yang WC, de Blank C, Meskiene I, Hirt H, Bakker J, van Kammenprotoplasts. Eur. J. Biochem. 126: 489–494 A, Franssen H & Bisseling T (1994) Rhizobium nod factors

¨Vroemen CW, Langeveld S, Mayer U, Ripper G, Jurgens G, Van reactivate the cell cycle during infection and nodule primordiumKammen A & de Vries SC (1996) Pattern formation in the formation, but the cycle is only completed in primordium forma-arabidopsis embryo revealed by position-specific lipid transfer tion. Plant Cell 6: 1415–1426protein gene expression. Plant Cell 8: 783–791 Yarbrough JA (1932) Anatomical and developmental studies of the

Vroemen C, de Vries S & Quatrano R (1999) Signalling in plant foliar embryos of Bryophyllum calcynum. Am. J. Bot. 19: 443–embryos during the establishment of the polar axis. Semin. Cell 453Dev. Biol. 10: 157–164 Yasuda H, Nakajima M, Ito T, Ohwada T & Masuda H (2001)

Walbot V (1996) Sources and consequences of phenotypic and Partial characterization of genes whose transcripts accumulategenotypic plasticity in flowering plants. Trends Plant Sci. 1: preferentially in cell clusters at the earliest stage of carrot27–32 somatic embryogenesis. Plant Mol. Biol. 45: 705–712

Walker L & Estelle M (1998) Molecular mechanisms of auxin Yeung EC (1995) Structural and developmental patterns in somaticaction. Curr. Opin. Plant Biol. 1: 434–439 embryogenesis. In: Thorpe TA (ed) In Vitro Embryogenesis in

Wang H, Qi Q, Schorr P, Cutler AJ, Crosby WL & Fowke LC Plants (pp. 205–248). Kluwer Academic Publishers, Dordrecht(1998) ICK1, a cyclin-dependent protein kinase inhibitor from Yu HJ, Moon MS, Lee HS, Mun JH, Kwon YM & Kim SG (1999)Arabidopsis thaliana interacts with both Cdc2a and CycD3, and Analysis of cDNAs expressed during first cell division of petuniaits expression is induced by abscisic acid. Plant J. 15: 501–510 petal protoplast cultures using expressed sequence tags. Mol.

Warren GS (1980) A decrease in intercellular space accompanies Cell. 9: 258–264auxin-induced abortion in somatic carrot embryos. Biochem. Zhao J, Morozova N,Williams L, Libs L, Avivi Y & Grafi G (2001)Soc. Trans. 8: 627–628 Two phases of chromatin decondensation during dedifferentia-

Wei YD, Zheng HG & Hall JC (2000) Role of auxinic herbicide- tion of plant cells: distinction between competence for cell fateinduced ethylene on hypocotyl elongation and root /hypocotyl switch and a commitment for S-phase. J. Biol. Chem. 276:radial expansion. Pest. Manage. Sci. 56: 377–387 22772–22778

Whitaker MJ (1990) Cell cycle control proteins are second mes- Zimmerman JL (1993) Somatic embryogenesis. Plant Cell 5:senger targets at fertilization in sea-urchin eggs. J. Reprod. Fertil. 1411–1423Suppl. 42: 199–204 Zuo J, Niu QW, Frugis G & Chua NH (2002) The WUSCHEL gene

Wink M (1993) The plant vacuole: a multifunctional compartment. promotes vegetative-to-embryonic transition in Arabidopsis.J. Exp. Bot. 44: 231–246 Plant J. 30: 349–359