Cell division events are essential for embryo patterning and morphogenesis: studies on dominant-negative cdc2aat mutants of Arabidopsis

Transcription

1 The Plant Journal (2000) 23(1), 123±130 SHORT COMMUNICATION Cell division events are essential for embryo patterning and morphogenesis: studies on dominant-negative cdc2aat mutants of Arabidopsis Adriana S. Hemerly 1, Paulo C.G. Ferreira 1, Marc Van Montagu 2, Gilbert Engler 3 and Dirk Inze 2, * 1 Departamento do BioquõÂmica MeÂdica, ICB, Universidade Federal do Rio de Janeiro, 21941±590, Rio de Janeiro, RJ, Brazil, 2 Vakgroep Moleculaire Genetica en Departement Plantengenetica, Vlaams Interuniversitair Instituut voor Biotechnologie, Universiteit Gent, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium, and 3 Laboratoire Associe de l'institut National de la Recherche Agronomique (France), Universiteit Gent, B-9000 Gent, Belgium Received 20 December 1999; revised 2 May 2000; accepted 11 May *For correspondence (fax ; diinz@gengenp.rug.ac.be). Summary During plant development, cell division events are coordinately regulated, leading to speci c growth patterns. Experimental evidence indicates that the morphogenetic controls that act at the vegetative plant growth stage are exible and tolerate distortions in patterns and frequencies of cell division. To address questions concerning the relationship between cell division and embryo formation, a novel experimental approach was used. The frequencies of cell division were reduced exclusively during embryo development of Arabidopsis by the expression of a dominant cdc2a mutant. The ve independent transgenic lines with the highest levels of the mutant cdc2a affected embryo formation. In the C13 line, seeds failed to germinate. The C1, C5 and C12 lines displayed a range of distortions on the apical±basal embryo pattern. In the C3 line, the shoot apical meristem of the seedlings produced leaves defective in growth and with an incorrect phyllotactic pattern. The results demonstrate that rates of cell division do not dictate cellular differentiation of embryos. Nevertheless, whereas cell divisions are uncoupled from vegetative development, they are instrumental in elaborating embryo structures and modulating embryo and seedling morphogenesis. Keywords: apical±basal pattern, cell division, dominant negative cdc2a mutant, embryogenesis, morphogenesis. Introduction Cell division plays a key role in plant development by generating the cells that will expand and form the plant body. At the vegetative meristems, most cell divisions increase the cell number and, consequently, the growth potential of an organ. During the rst stages of Arabidopsis thaliana embryo development, cell division does not represent a growth mechanism, thus resulting in smaller cells without a signi cant increase in embryo volume (JuÈ rgens and Mayer, 1994; Mans eld and Briarty, 1991). At later stages, intense cell proliferation modulates embryo growth. A fundamental question in plant development is how cell division events are coordinately regulated and integrated with cell expansion, leading to the normal growth patterns. The basic cell cycle machinery in plants, which includes conserved families of plant cyclin-dependent kinases (CDKs) and cyclins, is controlled by a variety of internal and external signals. Upstream controls can act at distinct levels, modulating CDK level and activation, cell cycle timing and cell division pattern and frequency to generate ã 2000 Blackwell Science Ltd 123

2 124 Adriana S. Hemerly et al. the nal plant architecture (for a review, see Hemerly et al., 1999; Mironov et al., 1999). Characterization of mutant phenotypes has revealed that the mechanisms dictating plant morphogenesis during vegetative growth are not always dependent on rates and precise planes of cell division (Reynolds et al., 1998; Smith et al., 1996). Furthermore, plant morphogenesis was shown to tolerate alterations of cell numbers at meristems during iterative development (Doerner et al., 1996; Hemerly et al., 1995; van den Berg et al., 1995). These experimental data suggest that when the basic machinery that controls cell division is somewhat perturbed, it can be locally sensed via cell-tocell communication and nally balanced with cell expansion to maintain normal growth patterns (Hemerly et al., 1999; Meyerowitz, 1996; Scheres and Heidstra, 1999). An important issue to be considered is whether the morphogenetic controls governing cell division during vegetative growth act similarly at embryogenesis. The Arabidopsis embryo is formed by a sequential number of highly reproducible steps that imply stringent control of rate and orientation of cell division, shape and differentiation. Embryo mutants with irregular cell division patterns and defects in cellular shapes, together with perturbations in development, have been described previously (for a review see Berleth, 1998; Laux and JuÈ rgens, 1997). Whereas research on embryo mutants demonstrates that the nal shape and position of the cells are often important for the correct apical±basal embryo pattern and normal morphogenesis, no evidence was found that regulated cell division is instrumental in pattern formation and morphogenesis during embryo formation. The embryo mutant genes isolated so far do not code for regulators of the cell cycle machinery (Hardtke and Berleth, 1998; Long et al., 1996; Lynn et al., 1999; Mayer et al., 1998; Shevell et al., 1994). In order to directly assess the role of cell division in embryo formation, we experimentally modi ed the frequency of cell divisions during embryo development. We used an approach similar to that described previously (Hemerly et al., 1995), in which a dominant-negative mutant of the Arabidopsis cdc2a cdna (cdc2a.n147) was used to de-regulate the normal function of the wildtype gene. The mutant cdc2a.n147 was expressed in Arabidopsis plants exclusively during embryogenesis. Distortion in embryo formation was observed only in ve transgenic lines (C1, C3, C5, C12 and C13) with the highest expression levels of the mutant cdc2a.n147 at the developing siliques. The observed phenotypes differed between the independent transgenic lines, and the majority of the lines exhibited a distorted apical±basal pattern. Here, we show that cell division events are essential for the elaboration of the apical±basal pattern and morphogenesis of embryos. In contrast, cell division might not govern cellular differentiation. Results and Discussion Arabidopsis plants expressing a dominant-negative cdc2aat mutant in embryos A dominant-negative mutant of the cell cycle regulator Cdc2aAt, containing a substitution of the residue D147 to N147, was used to prevent cell division during embryo formation in Arabidopsis. This mutation inactivates the kinase and arrests the cell cycle in eukaryotes (Mendenhall et al., 1988; van den Heuvel and Harlow, 1993). A construct with the dominant mutant cdc2a.n147 under control of a strong and constitutive promoter, named 35S-cdc2a.N147, was lethal when introduced into Arabidopsis plants. The mechanism of negative dominance of the mutant protein over the wild-type Cdc2a is dose-dependent and mediated by competition for regulators and/or substrates. In a heterologous system, transgenic tobacco plants, the mutant Cdc2a.N147 could not compete ef ciently with the wild-type protein at moderate levels, but was able to reduce kinase activity and cell division rates (Hemerly et al., 1995). cdc2aat had been shown to be expressed in all plant organs, including the embryo (Hemerly et al., 1993; Martinez et al., 1992), suggesting that the kinase might function in the control of cell division during embryo developmental processes. The promoter of the 2S2 albumin gene from Arabidopsis (da Silva ConceicËaÄ o and Krebbers, 1994; Guerche et al., 1990) was chosen to drive the expression of the mutant cdc2a.n147 during embryogenesis. Studies on Arabidopsis revealed that the At2S2 was expressed speci cally throughout the embryo body, including the meristems (Bies-Etheve et al., 1999). Although being preferentially expressed during late embryogenesis, low levels of At2S2 albumin mrnas are detected rst in late-heart to early torpedo stages (4±6 days after pollination (Guerche et al., 1990), when intense cell proliferation occurs in the developing embryo. Therefore, the dominant mutant cdc2a.n147 produced at low levels in Arabidopsis embryos was expected to compete with the wild-type protein and reduce cell division rates after the late-heart stage of embryo formation. The constructs containing the wildtype (cdc2a + ) and the mutant cdc2a.n147 under the control of the At2S2 promoter, named 2S2-cdc2a + and 2S2-cdc2a.N147, respectively, were introduced into Arabidopsis plants. Analysis of 2S2-cdc2a transgenic Arabidopsis seedlings Ten independent lines transformed with the 2S2-cdc2a + construct were analyzed in the T2 generation. No effects on embryo development and seedling germination were observed. However, during embryo formation, no plants with high levels of expression of the transgene were

3 Cell division and Arabidopsis embryogenesis 125 Figure 1. cdc2a.n147 expression in siliques of 5±10 mm length (4±6 days after pollination) of transgenic lines of Arabidopsis thaliana. Silique RNA extraction, gel blot analyses and hybridization reactions were performed as described in the Experimental procedures. RNA from siliques of wild-type plants was used as control. (Left) Representative transgenic lines transformed with the 2S2-cdc2a + construct. All lines showed very low levels of transgene expression, except for the B3 line, exhibiting moderate levels of expression. (Right) Transgenic lines transformed with the 2S2-cdc2a.N147 construct. The C3, C5 and C13 lines show the highest level and the C1 and C12 lines moderate levels of transgene expression. found, with the exception of the B3 line that exhibited moderate levels of expression (Figure 1). Analyses of 13 independent lines of the T2 generation that were transformed with the 2S2-cdc2a.N147 construct revealed ve lines (C1, C3, C5, C12, C13) with altered embryo formation and with close to 25% T-DNA segregation ratios when self-fertilized heterozygously. Gel blot analysis of total RNA from developing siliques (4±6 days after pollination) of the independent transgenic lines showed that the C1, C3, C5, C12 and C13 lines have the highest levels of cdc2a.n147 expression (Figure 1). The same lines exhibited distinct patterns of distortion in seedling development. In the C5 transgenic line, approximately 10% of the segregating progeny showed a range of abnormal seedling phenotypes with a perturbed apical±basal pattern. Distortions in different structures of the apical±basal axis (cotyledons, shoot meristem, hypocotyl, root and root meristem) were found. The abnormal seedlings were divided in two large groups, according to the severity of the disturbed phenotypes. The most severe phenotypes were exhibited by seedlings consisting of one cotyledonlike structure (Figure 2j,k) or two cotyledon-like structures with distinct shapes (Figure 2i,l). Some seedlings lacked the hypocotyl and the root (Figure 2k), whereas in others the cotyledons appeared to be directly attached to an embryonic root (Figure 2l). The hypocotyl was present in some seedlings, either fused to an embryonic root (Figures 2i and 3a), or without any morphological traces of embryo root formation (Figure 3b). Cell differentiation was not affected in the cotyledon and hypocotyl because the three main tissue types, as well as the stomata, were present (Figure 3). In most cases, the upper part of the cotyledons exhibited a region consisting of polygonal isodiametric cells, resembling traces of the embryonic development (Figure 3c). In very few cases, when the roots continued growing (Figure 2j), the developing root contained the major cell types, thereby indicating that cell differentiation was not affected. Seedlings with less severe phenotypes amounted to 2% of the progeny in the C5 line and their apical region was more developed than in the previously described phenotypes. In most cases, the cotyledons still exhibited altered morphogenesis, but the rst leaves were formed, albeit with distortion in shape and phylotaxy. These observations suggest that the shoot apical meristem (SAM) might be affected, but that it still retains some activity. In most seedlings, initiation of root formation did not occur (Figure 2h). In few seedlings, traces of the embryonic root were present, whereas in others, the main root started to develop, but without growing any further, thus suggesting that they lacked a functional meristem. All plants failed to develop beyond the seedling stage. Cellular differentiation appeared not to be affected in the structures formed because all the major cell types could be recognized (data not shown). The C1 and C12 lines exhibited the same range of seedling phenotypes as the C5 line, albeit with a lower frequency (approximately 5%). In the segregating progeny of the C3 line, 5±10% of the seedlings had altered leaf primordia, suggesting that the SAM was established in the embryo, although its function was perturbed during vegetative growth. In general, the SAM of the affected seedlings produced multiple leaf primordia, ranging from two to eight in number and organized in an incorrect phylotactic pattern (Figure 2b±d). Most affected seedlings developed the rst eight leaf primordia, typical of long-day growing Arabidopsis plants with a timing similar to that of wild-type plants. The venation pattern and cellular differentiation was not altered in the leaf primordia, although distortion of the leaf shape was observed in some seedlings. Three weeks after germination, the leaf primordia did not grow further and the seedlings failed to develop beyond this stage. In more severe cases, defects on cotyledon formation were observed, such as disturbed shape (Figure 2e) and, eventually, fusion (Figure 2f). Cellular differentiation was unaffected (data not shown) and perturbations on root development were not detected in the C3 progeny. In the C13 line, approximately 12% of the seeds produced by heterozygous plants failed to germinate. Microscopic analysis revealed that cell proliferations occurred during embryo formation, although in a reduced number and disorganized pattern. Elaborated embryonic structures and basic tissue organization could not be recognized in the embryos formed (data not shown). In the experimental approach used here, the affected cell division events were selected randomly because the negative mutant cdc2a.n147 acts by competition with the wild-type protein. In addition, there are small variations in the timing of activation of the At2S2 promoter in embryos at the same silique developmental stage (A. da Silva

4 126 Adriana S. Hemerly et al. Figure 2. Phenotypes of 2S2-cdc2a.N147 seedlings of Arabidopsis thaliana. (a) Wild-type seedling, 3 weeks after germination. (b±f) Affected seedlings observed in the C3 transgenic line at the same developmental stage as that of the wild type. The apical region of these seedlings is perturbed. (b±d) Seedlings exhibiting less severe phenotypes develop the rst 6±8 leaf primordia, albeit with a distortion in the phylotactic pattern and leaf morphogenesis. (e,f) Seedlings showing the strongest phenotypes with few leaf primordia formed and altered cotyledons and leaf shape. (g) Wild-type seedling, 2 weeks after germination. (h±l) Affected seedlings observed in the C5 transgenic line at the same age as that of the wild type. (h) Seedling with a less severe phenotype, showing the apical region with the cotyledons and rst leaves. Initiation of root formation is not observed. (i±l) Seedlings exhibiting a drastically affected apical±basal pattern with two cotyledon-like structures, hypocotyl and traces of embryonic root (i), one cotyledon-like structure together with a normal root (j), one cotyledon-like structure without embryonic root formation (k), and two cotyledon-like structures with an embryonic root primordium (l). Bars = 500 mm.

5 Cell division and Arabidopsis embryogenesis 127 Figure 3. Photomicrographs of 2S2-cdc2a.N147 seedlings from the C5 transgenic line of Arabidopsis thaliana. Photomicrographs were taken from chlorallactophenol-cleared whole-mount seedlings using interference contrast optics (see Experimental procedures). (a) Abnormal seedling showing traces of embryonic root; (b) hypocotyl of an abnormal seedling without traces of root development; (c) upper part of the cotyledon-like structure presented in Figure 2k. Bars = 100 mm. ConceicËaÄ o, personal communication). Therefore, cytological analysis of embryo development of the independent transgenic lines was not performed because it would not be possible to identify and correlate a certain disturbed cell division with seedling phenotypes. An intriguing observation was that embryo formation of seedlings of independent lines was differentially affected, although they expressed the same 2S2-cdc2a.N147 construct. The distinct phenotypes cannot be explained by differences in the levels of transgene expression because the C1, C5 and C12 lines have similar phenotypes and show different levels of cdc2a.n147 mrna. The differences in phenotypes exhibited by the transgenic lines could result from temporal and/or spatial variations in At2S2 promoter activity possibly caused by position effects of the T-DNA insertions. Relationship between cell division and cell differentiation, pattern formation and morphogenesis during embryo formation There is increasing evidence that a degree of exibility guides the embryogenesis processes in plants. First, mutant studies demonstrate that the distinct developmental processes that occur during embryo formation are uncoupled, such as cell differentiation, apical±basal organization and morphogenesis (Goldberg et al., 1994). Secondly, the highly reproducible cell division pattern of Arabidopsis embryo is not absolutely rigid in plant embryogenesis because extremely variable patterns of cell division are found during embryo formation in some plant species (for instance, Daucus carota). To address the question whether accurate cell division and normal formation of embryos are causally related, we perturbed cell division directly during embryo formation. The majority of the affected transgenic lines (C1, C5 and C12) presented a disturbed apical±basal pattern. The At2S2 promoter becomes active at late-heart stage, when the major embryo structures (the primary root, the hypocotyl and the cotyledons) and the main tissue types (protoderm, ground and vascular tissues) can already be distinguished. At that stage, the rate of cell division is extremely high in all tissues, promoting embryo growth (Berleth, 1998; JuÈ rgens and Mayer, 1994; Mans eld and Briarty, 1991). The mutant phenotypes suggest that apical±basal pattern elaboration depends on cell division, despite all pattern elements and cell types being produced correctly. A shortage of cells could not be sensed by upstream mechanisms nor be xed by neighboring cells, even when the information on pattern speci cation was not disrupted. A different situation was found in distinct developmental contexts. The suspensor cells of the twin mutants can change their developmental program and initiate the formation of a secondary embryo when the embryo proper is defective (Vernon and Meinke, 1994; Zhang and Somerville, 1997). During iterative development, apical meristems with fewer cells can promote normal organogenesis when pattern speci cation signals are not disrupted (Hemerly et al., 1995; van den Berg et al., 1995). Although cell division events appear to be essential for the elaboration of embryo structures, the formation of the basic body plan does not seem to be governed by the orientation of the cells in the early embryo. For example, in the fass/tonneau and hydra mutants, apical±basal structures are found, despite irregular cellular shapes (Topping et al., 1997; Torres-Ruiz and JuÈ rgens, 1994; Traas et al., 1995). Cell types are not missing in any of the C1, C5 and C12 seedlings. In contrast with the apical±basal patterning, the elaboration of the radial pattern does not depend on cell divisions. The embryo cell differentiation might be dependent on position and cell fate might be determined gradually, as already indicated by extensive analysis of mutants in the apical±basal pattern of Arabidopsis (for a

6 128 Adriana S. Hemerly et al. review see Berleth, 1998; Laux and JuÈ rgens, 1997). Nevertheless, the root mutants scr and wol have been suggested to fail in the formation of speci c cell types because of a cell shortage (Scheres et al., 1995). Therefore, a minimum number of cells at meristem establishment might reasonably be expected to form all cell types. Alternatively, we can also speculate that the mutated SCR and WOL genes can be part of an upstream cell-to-cell communication machinery. The relationship between cell division within the embryo and seedling morphogenesis should also be addressed. Most of the mutations in pattern formation affect overall embryo and seedling morphogenesis. This effect has also been observed in the C1, C5 and C12 lines. The affected C3 seedlings can provide some clues on morphogenetic controls because their initial development is normal, except for defects on SAM functioning and leaf morphogenesis. The fate map of the embryonic SAM of Arabidopsis suggests a degree of exibility in the number of cells specifying each rosette leaf (Irish and Sussex, 1992). A reduction in cell number can possibly be tolerated by the SAM of the C3 embryo, which can still produce leaves, albeit with altered shapes. When cell division rates were reduced at the vegetative SAM in the 35S-cdc2a.N147 plants (Hemerly et al., 1995), implementation of increased and polarized cell growth gradually compensated for the lower cell number, originating leaves with a correct shape. However, cell size was normal in C3 leaves. Possibly, perturbations on the SAM during its establishment at the embryonary stage could trigger serious distortions in its organization that could not be xed later during plant development. The perturbed phylotaxy reinforces that possibility. Alternatively, the developmental programs governing cell divisions during embryogenesis and vegetative development differ, and likewise the morphogenetic controls are also distinct. By affecting exclusively cell division events during embryo formation, we could elucidate some aspects of the relationship between cell division and developmental pathways of embryo formation. Comparison of embryogenesis and vegetative growth suggests that cell division is not a major determinant of cell differentiation at both plant developmental stages. However, cell division events are essential for the elaboration of the embryo structures, although they do not govern vegetative organogenesis. Finally, whereas cell division is uncoupled from iterative plant development, its perturbation during embryo formation appears to be instrumental in disturbing embryo and seedling morphogenesis. The construction of cdc2a dominant-negative mutants under control of promoters driving expression in speci c embryo structures or cell types could provide a clearer picture of the role played by cell division during distinct developmental programs of the embryo. Experimental procedures Plasmid construction Construction of the cdc2a.n147 cdna of Arabidopsis thaliana (L.) Heynh. has been reported previously (Hemerly et al., 1995). The albumin At2S2 promoter was transferred from the plasmid phb14 (Plant Genetic Systems N.V., Gent, Belgium) to the vector phph1 (W. Boerjan, personal communication) that contains the 3 nopaline synthase, originating from vector ph2s2. The cdc2a.n147 and cdc2a + cdnas were removed from puc18 by NcoI and BglII digestion and cloned into the NcoI and BamHI restriction sites of the vector ph2s2, generating the plasmids phdn2s and phbn2s, respectively. The cassettes 2S2-cdc2a NOS and 2S2-cdc2a.N147±3 NOS were cloned into the binary vector pgsv4 (He rouart et al., 1994) generating the vectors pvbn2s and pvdn2s, respectively. The binary vectors were mobilized into Agrobacterium tumefaciens C58C1Rif R (pgv2260) using a triparental mating system (Deblaere et al., 1985). Growth conditions and phenotypic analysis Arabidopsis ecotype C24 plants transformed with pvbn2s and pvdn2s were obtained by using the root transformation procedure (Valvekens et al., 1988). Seeds of 10 independent cdc2a + (T0) and 13 independent cdc2a.n147 transgenic lines were harvested under sterile conditions and germinated in selective germination medium, as described previously (Valvekens et al., 1988). One month after germination, two individuals (A and B) of each T1 progeny were transferred to soil for self-fertilization and seed setting. For segregation and phenotypical analysis, approximately 200 seeds of each T2 progeny were sterilized and germinated in sterile selective GM medium. Plants were grown in vitro at 20 C and under 18 h light/6 h dark. The progenies A and B of lines C1, C3, C5, C12 and C13 of cdc2a.n147 showed a segregation ratio close to 3 : 1, except for the C2B line, which was homozygous. T2 seedlings of the cdc2a + and cdc2a.n147 AandB lines were analyzed phenotypically and exhibited the same phenotypes as those observed previously in the T1 progenies. Photographs were taken with a M3Z stereomicroscope (Wild, Heerbrugg, Switzerland), equipped with an MPS 51 camera. Cell types were analyzed microscopically. Whole-mount Arabidopsis seedlings were visualized with a Jenalumar microscope (Zeiss, Jena, Germany) after clearing with chlorallactophenol (Beeckman and Engler, 1994). RNA analysis Siliques of 5±10 mm (4±6 days after pollination) from the A and B individuals of each T1 progeny (see above) were collected and pooled for RNA analyses at different developmental stages. RNA extraction, gel blot analyses and hybridization reactions were performed essentially as described by Hemerly et al. (1993). The same amount of total RNA was loaded on the gel, as con rmed by visualization with ethidium bromide. Radiolabelled antisense RNA (Riboprobe Gemini II Core System; Promega, Madison, WI, USA) synthesized from the full-length cdc2a cdna was used as probe. Acknowledgements The authors are grateful to Dr Alexandre da Silva ConceicËaÄ o for helpful discussions on the At2S2 promoter, Dr James Dat for critical reading of the manuscript, Martine De Cock for help in

7 Cell division and Arabidopsis embryogenesis 129 preparing it, Plant Genetic Systems N.V. for the phb14 plasmid containing the At2S2 promoter, Chris Genetello for technical assistance, and Rebecca Verbanck for photographs. This work was supported by grants from the Interuniversity Poles of Attraction Programme (Belgian State, Prime Minister's Of ce ± Federal Of ce for Scienti c, Technical and Cultural Affairs; P4/15) and the European Union BIOTECH program (BIO4-CT96±0217). A.S.H and P.C.G.F are indebted to the Conselho Nacional de Desenvolvimento Cientõ co e Tecnolo gico (CNPq) for nancial support. References Beeckman, T. and Engler, G. (1994) An easy technique for the clearing of histochemically stained plant tissue. Plant Mol. Biol. Report, 12, 37±42. Berleth, T. (1998) Experimental approaches to Arabidopsis embryogenesis. Plant Physiol. Biochem. 36, 69±82. Bies-Etheve, N., da Silva ConceicËaÄ o, A., Giraudat, J., Koornneef, M., Le on-kloosterziel, K., Valon, C. and Delseny, M. (1999) Importance of the B2 domain of the Arabidopsis ABI3 protein for Em and 2S albumin gene regulation. Plant Mol. Biol. 40, 1045±1054. da Silva ConceicËaÄ o, A. and Krebbers, E. (1994) A cotyledon regulatory region is responsible for the different spatial expression patterns of Arabidopsis 2S albumin genes. Plant J. 5, 493±505. Deblaere, R., Bytebier, B., De Greve, H., Deboeck, F., Schell, J., Van Montagu, M. and Leemans, J. (1985) Ef cient octopine Ti plasmid-derived vectors for Agrobacterium-mediated gene transfer to plants. Nucl. Acids Res. 13, 4777±4788. Doerner, P., Jùrgensen, J.-E., You, R., Steppuhn, J. and Lamb, C. (1996) Control of root growth and development by cyclin expression. Nature, 380, 520±523. Goldberg, R.B., de Paiva, G. and Yadegari, R. (1994) Plant embryogenesis: zygote to seed. Science, 266, 605±614. Guerche, P., TireÂ, C., Grossi de Sa, F., De Clercq, A., Van Montagu, M. and Krebbers, E. (1990) Differential expression of the Arabidopsis 2S albumin genes and the effect of increasing gene family size. Plant Cell, 2, 469±478. Hardtke, C.S. and Berleth, T. (1998) The Arabidopsis gene MONOPTEROS encodes a transcription factor mediating embryo axis formation and vascular development. EMBO J. 17, 1405±1411. Hemerly, A.S., Ferreira, P., de Almeida Engler, J., Van Montagu, M., Engler, G. and InzeÂ, D. (1993) cdc2a expression in Arabidopsis is linked with competence for cell division. Plant Cell, 5, 1711±1723. Hemerly, A., de Almeida Engler, J., Bergounioux, C., Van Montagu, M., Engler, G., InzeÂ, D. and Ferreira, P. (1995) Dominant negative mutants of the Cdc2 kinase uncouple cell division from iterative plant development. EMBO J. 14, 3925±3936. Hemerly, A.S., Ferreira, P.C.G., Van Montagu, M. and InzeÂ, D. (1999) Cell cycle control and plant morphogenesis: is there an essential link? BioEssays, 21, 29±37. He rouart, D., Van Montagu, M. and InzeÂ, D.(1994) Developmental and environmental regulation of the Nicotiana plumbaginifolia cytosolic Cu/Zn-superoxide dismutase promoter in transgenic tobacco. Plant Physiol. 104, 873±880. Irish, V.F. and Sussex, I.M. (1992) A fate map of the Arabidopsis embryonic shoot apical meristem. Development, 115, 745±753. JuÈ rgens, G. and Mayer, U. (1994) Arabidopsis.InA Colour Atlas of Developing Embryos (Bard, J., ed.). London: Wolfe Publishing, pp. 7±21. Laux, T. and JuÈ rgens, G. (1997) Embryogenesis: a new start in life. Plant Cell, 9, 989±1000. Long, J.A., Moan, E.I., Medford, J.I. and Barton, M.K. (1996) A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature, 379, 66±69. Lynn, K., Fernandez, A., Aida, M., Sedbrook, J., Tasaka, M., Masson. P. and Barton, M.K. (1999) The PINHEAD/ZWILLE gene acts pleiotropically in Arabidopsis development and has overlapping functions with the ARGONAUTE1 gene. Development, 126, 469±481. Mans eld, S.G. and Briarty, L.G. (1991) Early embryogenesis in Arabidopsis thaliana. II. The developing embryo. Can. J. Bot. 69, 461±476. Martinez, M.C., Jùrgensen, J.-E., Lawton, M.A., Lamb, C.J. and Doerner, P.W. (1992) Spatial pattern of cdc2 expression in relation to meristem activity and cell proliferation during plant development. Proc. Natl Acad. Sci. USA, 89, 7360±7364. Mayer, K.F., Schoof, H., Haecker, A., Lenhard, M., JuÈ rgens, G. and Laux, T. (1998) Role of WUSCHEL in regulating stem cell fate in the Arabidopsis shoot meristem. Cell, 95, 805±815. Mendenhall, M.D., Richardson, H.E. and Reed, S.I. (1988) Dominant negative protein kinase mutations that confer a G 1 arrest phenotype. Proc. Natl Acad. Sci. USA, 85, 4426±4430. Meyerowitz, E.M. (1996) Plant development: local control, global patterning. Curr. Opin. Genet. Dev. 6, 475±479. Mironov, V., De Veylder, L., Van Montagu, M. and InzeÂ, D.(1999) Cyclin-dependent kinases and cell division in higher plants ± the nexus. Plant Cell, 11, 509±521. Reynolds, J.O., Eisses, J.F. and Sylvester, A.W. (1998) Balancing division and expansion during maize leaf morphogenesis: analysis of the mutant, warty-1. Development, 125, 259±268. Scheres, B. and Heidstra, R. (1999) Digging out roots: pattern formation, cell division, and morphogenesis in plants. Curr. Top. Dev. Biol. 45, 207±247. Scheres, B., Di Laurenzio, L., Willemsen, V., Hauser, M.-T., Janmaat, K., Weisbeek, P. and Benfey, P.N. (1995) Mutations affecting the radial organisation of the Arabidopsis root display speci c defects throughout the embryonic axis. Development, 121, 53±62. Shevell, D.E., Leu, W.-M., Gillmor, C.S., Xia, G., Feldmann, K.A. and Chua, N.-H. (1994) EMB30 is essential for normal cell division, cell expansion, and cell adhesion in Arabidopsis and encodes a protein that has similarity to Sec7. Cell, 77, 1051± Smith, L.G., Hake, S. and Sylvester, A.W. (1996) The tangled-1 mutation alters cell division orientations throughout maize leaf development without altering leaf shape. Development, 122, 481±489. Topping, J.F., May, V.J., Muskett, P.R. and Lindsey, K. (1997) Mutations in the HYDRA1 gene of Arabidopsis perturb cell shape and disrupt embryonic and seedling morphogenesis. Development, 124, 4415±4424. Torres-Ruiz, R.A. and JuÈ rgens, G. (1994) Mutations in the FASS gene uncouple pattern formation and morphogenesis in Arabidopsis development. Development, 120, 2967±2978. Traas, J., Bellini, C., Nacry, P., Kronenberger, J., Bouchez, D. and Caboche, M. (1995) Normal differentiation patterns in plants lacking microtubular preprophase bands. Nature, 375, 676±677. Valvekens, D., Van Montagu, M. and Van Lijsebettens, M. (1988) Agrobacterium tumefaciens-mediated transformation of Arabidopsis thaliana root explants by using kanamycin selection. Proc. Natl Acad. Sci. USA, 85, 5536±5540.

8 130 Adriana S. Hemerly et al. van den Berg, C., Willemsen, V., Hage, W., Weisbeek, P. and Scheres, B. (1995) Cell fate in the Arabidopsis root meristem determined by directional signalling. Nature, 378, 62±65. van den Heuvel, S. and Harlow, E. (1993) Distinct roles for cyclindependent kinases in cell cycle control. Science, 262, 2050±2054. Vernon, D.M. and Meinke, D.W. (1994) Embryogenic transformation of the suspensor in twin, a polyembryonic mutant of Arabidopsis. Dev. Biol. 165, 566±573. Zhang, J. and Somerville, C.R. (1997) Suspensor-derived polyembryony caused by altered expression of valyl-trnasynthase in the twn2 mutant of Arabidopsis. Proc. Natl Acad. Sci. USA, 94, 7349±7355.