Plant Molecular 26: 1305-1313, 1994. © 1994 Kluwer Academic Publishers. Printed in Belgium. 1305

Signal molecules involved in plant embryogenesis

Ed D.L. Schmidt, Anke J. de Jong and Sacco C. de Vries* Department of Molecular Biology, Wageningen Agricultural University, Dreijenlaan 3, 6703 HA Wageningen, The Netherlands (* author for correspondence)

Received 10 May 1994; accepted 11 May 1994

Key words: zygotic embryogenesis, , chitinase, nod factor, signal molecules

Abstract

In plant embryogenesis, inductive interactions mediated by diffusable signal molecules are most likely of great importance. Evidence has been presented that at late globular stages in plant embryogenesis, perturbation of the polar transport results in abberrant morphology. Rhizobium lipo- oligosaccharides or Nod factors are a newly discovered class of bacterial molecules that are able to trigger initial steps in root nodule development in legumes. Part of the activity of Nod factors may be directed towards alteration of endogenous plant growth regulator balance. The same bacterial Nod factors promoted the formation of globular in the carrot line ts 11. Whether there exist plant ana- logues of the Nod factors and whether these molecules are active as a more universal control system perhaps designed to initiate and or mediate gradients in auxin and cytokinin remains to be determined.

Introduction is highly vacuolated. Dual fertilization of the dip- loid central cell and the haploid egg cell results in Currently there is a wide and increasing interest the and the zygote respectively. The in the molecular-genetic analysis of plant embryo- first zygotic division is asymmetrical and yields a genesis. Detailed descriptions of gametogenesis small apical cell and a large basal cell. The basal and embryogenesis have been the subject of cell remains positioned at the micropylar pole of many recent studies [1-8] as well as reviews the embryo sac, and undergoes a series of trans- [9-15] and therefore will only be recapitulated versal divisions to form the suspensor. From the briefly. uppermost cell of the suspensor, the hypophysial The male or grains are cell, the centre of the future root is formed in the anther, while the female gameto- formed [17]. The apical cell undergoes three di- phyte or embryo sac is formed in the pistil. In visions, resulting in the octant stage embryo angiosperms that exhibit the polygonum type [ 16 ] proper. Tangential divisions then set apart the it consists of seven cells: the egg cell, two syner- protoderm cells, from which the epidermis is de- gids, the central cell and three antipodal cells. The rived. Development of the Arabidopsis embryo polarized egg cell and synergids are positioned at from fertilization, through the octant, globular, the micropylar pole of the embryo sac. The po- triangular, heart, torpedo and bent- larity of the egg cell is reflected in the position of stages, to the mature desiccated embryo, has been the nucleus and most of the at the subdivided into a sequence of 20 different stages chalazal side of the cell, while the micropylar part [3].

[69] 1306

While the above-described developmental se- Signal molecules in zygotic embryogenesis quence that gives rise to the Arabidopsis embryo appears to involve a highly predictable series of Two mechanisms appear to be universally used in divisions, it is important to note that this is but animal embryogenesis to initiate cell differentia- one of the types of embryo development in plants tion. The first of these is a polarization of cellu- [ 16]. Many variations in the plants and positions lar determinants, sometimes, but not always, fol- of early cell divisions and in suspensor morphol- lowed by asymmetric [29]. In plants, ogy have been described, with apparently little or fertilization is followed by an extensive redistri- no consequence for the eventual seedling mor- bution of organelles resulting in polarization of phology. Recently the systematic genetic dissec- the zygote. Asymmetric cell division occurs fre- tion of the formation of the zygotic embryo has quently in plants, and the analysis of the Arabi- been initiated [18-21]. Based on the mutant dopsis gnom mutant [30], in which the mutant phenotypes obtained, a division of the young em- phenotype appears correlated with the inability to bryo along the longitudinal axis into an apical, perform a normal unequal division of the zygote, central and basal region was proposed [20, 22]. clearly shows that this mechanism is indeed of A second, radial pattern, superimposed on the crucial importance in plant embryogenesis. The apical-basal pattern and consisting of the vascu- second mechanism consists of the interaction be- lar, ground and epidermal tissues, appears to be tween an inducing cell or tissue and a responding established independently. cell or tissue brought about by specific signal mol- Somatic or asexual embryogenesis is the pro- ecules. Signal molecules that are produced out- cess by which somatic cells develop into plants side of a group of equivalent cells are defined as through the same characteristic morphological inducers [ 31 ]. Cellular communication between stages as their zygotic counterparts. For dicots adjacent cells can occur by cell-surface-located these are the globular, heart and torpedo stages. signal molecules able to act as inducers. Cells that The ability to form embryos that do not orignate are not in direct contact with an inducer- from a fertilized egg cell is quite widespread producing cell can be influenced if the signal mol- among plants. It may occur naturally as in Mal- ecule is diffusable, usually resulting in a concen- axis, where somatic embryos form spontanously tration gradient. Signal molecules are defined as on the leaf tips [23], or in the form of apomictic morphogens when the slope of their concentra- processes [24]. Under in vitro conditions somatic tion gradient provides reference points for the embryos can either form directly on the surface of formation of a pattern. The local concentration of an organized tissue such as a leaf or stem seg- the morphogen then determines the response of ment, from protoplasts or from microspores, or the cells. indirectly via an intermediary step of or Caenorhabditis elegans has served as an ex- suspension culture [25]. By virtue of their excel- ample for the existence of a series of segregating lent experimental accessibility, somatic embryo- cytoplasmic determinants that result in a rigid cell genesis is exploited to isolate plant-produced lineage as a means to generate patterns in the molecules that have promotive effects on the for- embryo and differentiated cells later on in devel- mation of somatic embryos [26]. Both somatic opment [32]. Recently described mutants pro- embryogenesis as well as in vitro cultured zygotic vide evidence that a number of maternally ex- embryos are being employed to try and answer pressed genes that encode nuclear proteins and long-standing questions concerning the role of cell surface proteins similar to the lin-12 family 'classical' plant growth regulators such as auxin are involved in cell-inductive processes [33]. in embryogenesis [27, 28]. In cell position rather than developmental history is considered to be essen- tial for the formation of the somatic tissues [34- 37 ]. This implies that cell-inductive processes and

[70] 1307 the use of signal molecules that act at short range exhibit a fused collar-like arrangement of their and over longer distances might be important in cotyledon primordia, as opposed to the normal the organization of the plant embryo. However, bilateral arrangement [27]. Application of the no direct evidence is available that cell-inductive polar auxin transport inhibitor to excised and in processes are indeed occurring in plants. A se- vitro cultured zygotic embryos of Brassica resulted quential and transient expression of a plasma in a similar fused-cotyledon phenotype as ob- membrane (AGP) served for the pin-1 mutant [27], and in the Ara- epitope, recognized by the monoclonal antibody bidopsis gnom mutant [30]. These results are in- JIM8, was observed in reproductive tissues and terpreted to suggest that polar auxin transport the suspensor of an early globular embryo of determines the transition from radial to bilateral Brassica. Pennell et al. [38] speculated that the symmetry in the globular dicot embryo. The amp-I JIM8 epitope may be a marker for a cell-inductive mutant [42], exhibiting an elevated level of cyto- process in plants. Whether AGPs themselves act kinins, has a complex and pleiotropic phenotype as signal molecules in zygotic embryogenesis is affecting photomorphogenesis and flowering time, not known. There is however some evidence from as well as an increase in the number of tricot and somatic embryogenesis (see next section) that tetracot seedlings. Multiplication of cotelydons is particular AGPs may indeed have a direct bio- also observed in embryo mutants of the hauptling logical function in embryogenesis. If so, they will type [15, 20]. It appears therefore that interfer- be likely to act as short-range inducer molecules ence with the local balance between auxin and over a distance of a few cells only. A fascinating cytokinin primarily influences the formation of result was recently reported for Fucus [39] where, the proper number and orientation of cotyledon upon laser microsurgery of two-celled embryos, it primordia in the late globular embryo. Because was established that prolonged contact of a thal- chemical inhibitors such as NPA and TIBA act lus cell protoplast with the wall of the ablated as inhibitors of auxin efflux carriers [43], appli- rhizoid cell resulted in the formation of cells with cation will result in elevated intracellular concen- the characteristics of the rhizoid cell. This experi- trations in all cells subjected to the inhibitor. ment provides evidence for the presence of stable Without the possibility to monitor the concentra- non-diffusible wall components able to change tion of (active) growth regulators at the individual the fate of algal cells, and it will be of great im- cell level, it is not easy to draw up conclusive portance to establish whether similar experimen- models of the way polar transport of auxin is tal systems are feasible in higher plants. The abil- involved in the formation of cotyledon primordia ity to perform in vitro fertilization [40] makes this [44]. However, all models assume that during the a realistic option. formation of the globular embryo gradients in A glimpse of the role that plant growth regula- these growth regulators are established by an un- tors may play as signal molecules during embryo- equal distribution of auxin-synthesizing cells. genesis has been provided by the characterization Whether this is the result of unequal distribution of Arabidopsis mutants, in which the normal bal- of cytoplasmic determinants in the zygote or the ance of auxin and cytokinins was disturbed. The effect of an earlier acting signal molecule remains pin-1 mutant was isolated as one of the Arabi- to be determined. In the next part of this review dopsis flower mutants, and found to have a se- we will describe some recent findings regarding verely reduced ability for auxin transport. The somatic embryogenesis, which may provide clues phenotype of pin-1 appears to be a gross abnor- to the identity of additional signal molecules in- mality of the inflorescence axis, as well as forma- volved in plant embryogenesis. tion of abnormal flowers and leaves. This pheno- type could be reproduced by germinating wild- type in the presence of polar auxin transport inhibitors [41]. The embryos of this mutant

[71] 1308

Signal molecules in somatic embryogenesis members of the family of AGPs are involved in the formation of embryogenic clusters. Although The possibility of somatic plant cells in culture to the underlying mechanisms are unclear, these ob- acquire embryogenic potential has been exploited servations, together with earlier ones employing to establish experimental systems that allow the unfractionated conditioned medium [ 51 ], suggest identification of molecules able to promote or in- that molecules totally different from conventional fluence the formation of somatic embryos. An plant growth regulators are able to direct the tran- important and as yet unsolved question in these sition of somatic cells into embryogenic cells. studies is whether the findings are only relevant to Since cell-surface AGPs turn over very rapidly the somatic embryo system, or whether they are [52], and their expression is clearly developmen- also applicable to zygotic embryogenesis. tally regulated [53], they are likely candidates for In Daucus, the usual strategy to start an em- molecules able to mediate developmental pro- bryogenic suspension culture is to expose explants cesses in plants, perhaps by a cell-inductive to a high concentration of auxin. After reinitiation mechanism [38]. of cell division and a period of proliferation of the released explant cells in the presence of auxin, a Chitinases and chitin-containing signal molecules small subset of the cell population becomes em- bryogenic [45]. These embryogenic cells are usu- Recently de Jong et al. [26] have identified a ally in the form of clusters of small cytoplasmic 32 kDa endochitinase able to lift the arrest in cells, referred to as proembryogenic masses [46]. embryo development of the temperature-sensitive In contrast to the non-embryogenic cells, these carrot variant tsl 1. Later, de Jong et al. [54] have proembryogenic masses become insensitive to shown that chitin-containing bacterial signal mol- auxin [47]. By the time the embryos developed ecules, Nod factors, are able to mimic the effect from the proembryogenic masses reach the plant- of the 32 kDa endochitinase. These results sug- let stage, auxin sensitivity is regained. An impor- gest that the 32 kDa endochitinase is involved in tant question concerning the formation of em- the generation of plant analogues of these com- bryogenic cells in suspension cultures is whether pounds, but this hypothesis assumes that a plant they are formed continuously from non-embryo- substrate is indeed available for plant chitinases, genic cells or are derived from a subset of explant and that the phenotypical result of both the chiti- cells, which is propagated independently [48]. nase and the Nod factor act via the same mecha- Application of techniques to follow the develop- nism. mental fates of individual single cells present in Enzymes that catalyse hydrolysis of N-acetyl- embryogenic suspension cultures has revealed a fl-D-glucosaminide /3-1,4 linkages in chitin and striking heterogeneity in embryogenic single cell in chito-oligomers are classified as either exo- or types and early cell division patterns that was endochitinases. In 1965, Powning and Irzyk- nonetheless not readily apparent in the morphol- iewicz [55] proposed a role for chitinases in a ogy of the resulting somatic embryos [49]. defence mechanism against chitin-containing Recent evidence suggests that particular puri- parasites from the soil. As has been demonstrated fied AGPs isolated from the culture medium of repeatedly, plant chitinase activity is increased by embryogenic Daucus lines and from dry Daucus fungal elicitors [56-58], and is part of the hyper- seeds were able to promote the formation of pro- sensitive response to pathogen attack [59, 60]. embryogenic masses, even in previously non- For some plant-produced chitinases the ability to embryogenic Daucus cell lines, when added in na- inhibit fungal growth in vitro has been shown [61- nomolar concentrations. Other AGPs, isolated 63]. Some of the isolated plant chitinases only from the medium of a non-embryogenic line, acted possess antifungal activity in combination with negatively on the formation of proembryogenic /3-1,3 -gluc anase activity [ 64 ], while several others masses [50]. These results show that specific do not have antifungal activity in vitro at all [65]. [72] 1309

Transgenic tobacco seedlings, expressing a bean in the form of glycolipids. Incubation of mem- chitinase gene, show an increased ability to sur- brane fractions from bean or pea stems with vive in soil infested with the fungal pathogen N-acetyl[14C]glucosamine and analysis of the Rhizoctonia solani [66]. Chitinase genes are also lipophilic fraction show incorporation of expressed in the absence of pathogens [67-70], a N-acetyl[14C]glucosamine only in the highly finding that is usually explained by assuming that charged dolichol pyrophosphate oligosaccha- they are part of a continuously present defense rides, the intermediates in the biosynthesis of gly- mechanism. coproteins [74, 75]. These intermediates contain The results presented by de Jong et al. [26] two N-acetylglucosamines at most per lipid mol- represent the first direct evidence for a role of at ecule. The minimal length of an N-acetylglu- least one plant chitinase in somatic embryogen- cosamine chain required to serve as substrate for esis. At the same time these results raise the ques- endochitinases is three [76], indicating that the tion what the natural substrate of these enzymes dolichol pyrophosphate oligosaccharides are un- might be, assuming that it is the catalytic property able to serve as substrate for chitinases. Another of the 32 kDa endochitinase which is required for class of glycolipids in plants that contain N- the rescue of ts 11 embryo development. Employ- acetylglucosamine are of the sphingolipid type. ing an indirect approach by testing putative prod- They are present in seeds and commonly contain ucts of chitinase activity for their effect on ts 11 complex oligosaccharides rather than a single embryo development, the Rhizobium Nod factor saccharide and a variety of sugar components NodRlv-V(Ac, C18:4) proved to be able to stim- including N-acetylglucosamine, glucuronic acid, ulate tsl 1 embryo formation with a similar effi- inositol, galactose, arabinose and fucose [77]. ciency as the 32 kDa endochitinase [54]. This However, a stretch of three N-acetylglucosamines result was interpreted to suggest that the 32 kDa has not been detected in such glycolipids. endochitinase acts by releasing Nod factor-like To test the hypothesis whether lipo-oligosac- signal molecules from plant-produced precursors. charides analogous to the Rhizobium Nod factors However, it must be kept in mind that presently occur in plants, Lathyrus flowers have been la- it is not known whether the 32 kDa endochitinase belled with N-acetyl[ 14C ]glucosamine [78 ]. Some and NodRlv-V(Ac, C18:4) act via the same or via of the extracted lipophilic compounds migrate different mechanisms. Therefore, it remains pos- similarly to the rhizobial lipo-oligosaccharides on sible that there is no direct relationship between TLC plates. To determine the nature of the oli- both observations. gosaccharide chain, the lipophilic compounds When boiled homogenates of bean plants are have been treated with commercial chitinase and treated with a bean endochitinase, no N-acetyl- analysed on TLC plates. The appearance of at glucosamine-containing fragments can be de- least three new spots after chitinase treatment tected, indicating that chitin-like substrates are suggests that lipophilic compounds can indeed be not present in the plant [71]. Employing cy- hydrolysed by chitinases. Indirect evidence for tochemical labelling, Benhamou and Asselin [72] the presence of chitin-like molecules in plants has have shown binding of chitinases and wheat germ been provided by the introduction of Rhizobium agglutinin, a plant lectin that is specific for oligo- nodulation genes nodA and nodB, involved in the mers of N-acetylglucosamine, to secondary cell biosynthesis of the Nod factors, into tobacco walls of a variety of Solanaceae. This binding can plants [79, 80]. Schmidt et al. [81] have shown be abolished by prior treatment with lipase. Al- that NodB together with the NodA protein though lipid-linked N-acetylglucosamine residues are sufficient to produce small, heat-stable have not been identified as structural components compounds that stimulate mitosis in legume and of plant cell walls [73], the results of Benhamou non-legume protoplasts. The NodB gene product and Asselin [72] suggest that the N-acetyl- is an oligosaccharide-modifying enzyme that glucosamine residues detected may be present deacetylates the non-reducing N-acetylglucos-

[73] 1310 amine residues of chito-oligosaccharides [82], Purified Nod factors, when applied to legume while the NodA protein is involved in the seedlings at concentrations as low as 10 -12 M, N-acylation of the chito-oligosaccharide, essential stimulate differentiation of epidermal ceils into for the coupling of the sugar and fatty acid parts root hairs, deformation of root hairs, induction of of the active Nod factor [ 83 ]. Expression of nodA early nodulin genes related to the infection pro- and nodB in tobacco results in plants with mor- cess in root epidermal cells, and induction of cell phological abnormalities such as wrinkled leaves, divisions in the inner cortex of the roots [see for reduced growth and compact inflorescences. reviews 91-93]. These events normally occur These experiments suggest that tobacco is able to during the early nodulation process. Induction of produce substrate molecules that can be used by early nodulin gene expression, cortical cell divi- the nodA- and nodB-encoded proteins to synthe- sion and subsequent meristem formation can also size growth-controlling factors. Cultured tomato be triggered by auxin transport inhibitors [94, 95 ], cells also respond to both chitin fragments as well suggesting that a change in the endogenous bal- as Nod factors by a rapid and transient alkalin- ance of growth regulators plays a role in the ini- ization of the culture medium [84]. tiation of nodule formation. A cytokinin-produc- Taken together, although no substrate mol- ing gene, cloned into Escherichia coli, is sufficient ecules for chitinases have been unequivocally to provide this bacterium with the ability to in- identified in plants, several lines of circumstantial duce cortical cell divisions in alfalfa roots [96]. evidence indicate that substrates for chitinases Although there are still several possibilities to ex- may indeed be present. It will now be of great plain these findings, one attractive hypothesis is interest to identify plant substrates for chitinases, that bacterial Nod factors are able to modulate because they may act as inactive precursors of the endogenous auxin-cytokinin balance in the lipo-oligosaccharides, with chitinases in the role root cortex. A plausible but unproven explanation of enzymes releasing these molecules. If plant for the effect of Nod factors on carrot cells may chitinases indeed are able to release signal mol- be that they alter the endogenous growth regula- ecules from plant-produced precursor molecules, tor balance of the susceptible cells. they may not only be able to release but also to inactivate lipo-oligosaccharide signal molecules, Acknowledgements like they may do with bacterial Nod factors [85, 86]. We thank Ton Bisseling and Gerd Jtlrgens for valuable comments on the manuscript. We are Do lipo-oligosaccharides represent a novel class grateful to our colleagues for communicating un- of plant growth regulators? published results. Primary research in our labo- ratory is supported by the Foundation for Bio- Besides the Rhizobium lipo-oligosaccharides, logical Research, subsidized by the Netherlands other oligosaccharides can also influence plant Organization for Scientific Research (A.J.d.J.), growth and development at concentrations sev- the Technology Foundation, subsidized by the eral orders of magnitude below those of more Netherlands Organization for Scientific Research conventional plant growth regulators. These oli- (E.D.L.S.) and the Biotech-PTP programme of gosaccharides are termed oligosaccharins [87, the European Community. 88]. One of these, XXFG, inhibits 2,4-D-stim- ulated elongation of pea stem segments in a dose- dependent fashion [89, 90] Hence, oligosaccha- References rins can act either as intermediates in the growth 1, Castle LA, Meinke DW: Embryo-defective mutants as hormone-regulating mechanism or interact with a tools to study essential functions and regulatory processes consecutive site somewhere in the cascade of in plant embryo development. Semin Devel Biol 4:31-39 events normally triggered by auxin. (1993).

[741 1311

2. Cresti M, Blackmore S, van Went JL: Atlas of Sexual 21. Meinke DW: Embryonic mutants ofArabidopsis thaliana. Reproduction in Flowering Plants. Springer-Verlag, Ber- Devel Genet 12:382-392 (1991). lin (1992). 22. Mayer U, Torres Ruiz RA, Berleth T, Mistra S, Jflrgens 3. Jtlrgens G, Mayer U: Arabidopsis. In: Bard J (ed) A G: Mutations affecting body organization in the Arabi- Colour Atlas of Developing Embryos. Wolfe, London dopsis embryo. Nature 353:402-407 (1991). (1994). 23. Taylor RL: The foliar embryos of Malaxispaludosa. Can 4. Mansfield SG, Briarty LG: Development of the free- J Bot 45:1553-1556 (1967). nuclear endosperm in Arabidopsis thaliana L. Arabidopsis 24. Koltunow AM: : embryo sacs and embryos Inf Serv 27:53-64 (1990). formed without meiosis or fertilization in . Plant 5. Mansfield SG, Briarty LG: Endosperm cellularization in Cell 5:1425-1437 (1993). Arabidopsis thaliana L. Arabidopsis lnf Serv 27:65-72 25. Williams EG, Maheswaran G: Somatic embryogenesis: (1990). factors influencing coordinate behavior of cells as an em- 6. Mansfield SG, Briarty LG: Early embryogenesis in Ara- bryogenetic group. Ann Bot 57:443-462 (1986). bidopsis thaliana. II. The developing embryo. Can J Bot 26. de Jong AJ, Cordewener J, Lo Schiavo F, Terzi M, 69:461-476 (1991). Vandekerckhove J, van Kammen A, de Vries SC: A car- 7. Mansfield SG, Briarty LG, Erni S: Early embryogenesis rot somatic embryo mutant is rescued by chitinase. Plant in Arabidopsis thaliana. 1. The mature embryo sac. Can J Cell 4:425-433 (1992). Bot 69:447-460 (1991). 27. Liu C-m, Xu Z-h, Chua N-h: Auxin polar transport is 8. Webb MC, Gunning BES: Embryo sac development in essential for the establishment of bilateral symmetry dur- Arabidopsis thaliana. I. Megasporogenesis, including the ing early plant embryogenesis. Plant Cell 5:621-630 microtubular cytoskeleton. Sex Plant Reprod 3:244-256 (1993). (1990). 28. Schiavone FM, Cooke TJ: Unusual patterns of somatic 9. de Jong AJ, Schmidt EDL, de Vries SC: Early events in embryogenesis in domesticated carrot: developmental ef- higher-plant embryogenesis. Plant Mol Biol 22:367-377 fects of exogenous and auxin transport inhibitors. (1993). Cell Diff 21:53-62 (1987). 10. Thomas TL: Gene expression during plant embryogen- 29. Gurdon JB: The generation of diversity and pattern in esis and : an overview. Plant Cell 5: 1401- animal cells. Cell 68:185-199 (1992). 1410 (1993). 30. Mayer U, Bflttner G, Jflrgens G: Apical-basal pattern 11. Zimmerman JL: Somatic embryogenesis: a model for fromation in the Arabidopsis embryo: studies on the early development in higher plants. Plant Cell 5:1411- role of the gnom gene. Development 117:149-162 1423 (1993). (1993). 12. Yeung EC, Meinke DW: Embryogenesis in angiosperms: 31. Greenwald I, Rubin GM: Making a difference: the role of development of the suspensor. Plant Cell 5:1371-1381 cell-cell interactions in establishing separate identities for (1993). equivalent cells. Cell 68:271-281 (1992). 13. Lindsey K, Topping JF: Embryogenesis: a question of 32. Sulton JE, Schierenberg E, White JG, Thomson JN: The pattern. J Exp Bot 44:359-374 (1993). of Caenorhabditis elegans. Devel 14. West MAL, Harada JJ: Embryogenesis in higher plants: Biol 100:64-119 (1983). an overview. Plant Cell 5:1361-1369 (1993). 33. Wood WB, Edgar LG: Patterning in the C. elegans em- 15. Jgrgens G, Tortes Ruiz RA, Berleth T: Embryonic pat- bryo. Trends Genet 10:49-54 (1994). tem formation in flowering plants. Annu Rev Genet 28, 34. Dawe RK, Freeling M: Cell lineage and its consequences in press (1994). in higher plants. Plant J 1:3-8 (1991). 16. Johri BM: Embryology of Angiosperms. Springer-Verlag, 35. Furner IJ, Pumfrey JE: Cell fate in the shoot apical mer- Berlin (1984). istem ofArabidopsis thaliana. Development 115:755-764 17. Dolan L, Janmaat K, Willemsen V, Linstead P, Poethig (1992). S, Roberts K, Scheres B: Cellular organization of the Ara- 36. Irish VF, Sussex IM: A fate map of the Arabidopsis em- bidopsis thaliana root. Development 119:71-84 (1993). bryonic shoot apical meristem. Development 115: 745- 18. Patton DA, Meinke DW: Ultrastructure of arrested em- 753 (1992). bryos from lethal mutants of Arabidopsis thaliana. Am J 37. Poethig RS: Genetic mosaic and cell lineage analysis in Bot 55:807-819 (1990). plants. Trends Genet 5:273-277 (1989). 19. Errempalli D, Patton D, Castle L, Mickelson L, Hansen 38. PenneU RI, Janniche L, Kjellbom P, Scofield GN, Peart K, Schnall J, Feldmann K, Meinke D: Embryonic lethals JM, Roberts K: Developmental regulation of a plasma and T-DNA insertional mutagenesis in Arabidopsis. Plant membrane arabinogalactan protein epitope in oilseed rape Cell 3:149-157 (1991). flowers. Plant Cell 3:1317-1326 (1991). 20. Jargens G, Mayer U, Torres Ruiz RA, Berleth T, Mistra 39. Berger F, Taylor A, Brownlee C: S: Genetic analysis of pattern formation in the Arabidopsis by the cell wall in early Fucus development. Science 263: embryo. Development, Suppl 91.1:27-38 (1991). 1421-1423 (1994).

[75] 1312

40. Dumas C, Mogensen HL: Gametes and fertilization: 55. Powning RF, Irzykiewicz H: Studies on the chitinase maize as a model for experimental embryogenesis in flow- systems in bean and other seeds. Comp Biochem Physiol ering plants. Plant Cell 5:1337-1348 (1993). 14:127-133 (1965). 41. Okada K, Ueda J, Komaki MK, Bell CJ, Shimura Y: 56. Ishige F, Mori H, Yamazaki K, Imaseki H: Cloning of a Requirement of the auxin polar transport system in early complementary DNA that encodes an acidic chitinase stages of Arabidopsis floral formation. Plant Cell 3: which is induced by ethylene and expression of the cor- 677-684 (1991). responding gene. Plant Cell Physiol 34:103-111 (1993). 42. Chaudhury AM, Letham S, Craig S, Dennis ES: Amp-l: 57. Kurosaki F, Tashiro N, Nishi A: Chitinase induction in a mutant with high cytokinin levels and altered embryonic carrot cell cultures treated with various fungal com- pattern, faster vegetative growth, constitutive photomor- pounds. Biochem Internatl 20:99-106 (1990). phogenesis, and precocious flowering. Plant J 4:907-916 58. Kirsch C, Hahlbrock K, Kombrink E: Purification and (1993). characterization of extracellular, acidic chitinase isoen- 43. Rubery PH: Auxin transport. In: Davies PJ (ed) Plant zymes from elicitor-stimulated parsley cells. Eur J Bio- Hormones and their Role in Plant Growth and Develop- chem 213:419-425 (1993). ment, pp. 341-362. Martinus Nijhoff, Hingham, MA 59. Metraux JP, Boiler T: Local and systemic induction of (1987). chitinase in cucumber plants in response to viral, bacte- 44. Cooke TJ, Racusen RH, Cohen JD: The role of auxin in rial and fungal infections. Physiol Mol Plant Path 28: plant embryogenesis. Plant Cell 5:1494-1495 (1993) 161-169 (1986). 45. de Vries SC, Booij H, Meyerink P, Huisman G, Wilde 60. Margis-Pinheiro M, Metz-Boutigue MH, Awade A, De DH, Thomas TL, van Kammen A: Acquisition of em- Tapia M, Le Ret M, Burkard G: Isolation of a comple- bryogenic potential in carrot cell-suspension cultures. mentary DNA encoding the bean PR4 chitinase: an acidic Planta 176:196-204 (1988). enzyme with an amino-terminus cysteine-rich domain. 46. Halperin W: Alternative morphogenetic events in cell sus- Plant Mol Biol 17:243-253 (1991). pensions. Am J Bot 53:443-453 (1966). 61. Schlumbaum A, Mauch F, Vogeli U, Boiler T: Plant chiti- 47. Lo Schiavo F, Filippini F, Cozzani F, Vallone D, Terzi nases are potent inhibitors of fungal growth. Nature 324: M: Modulation of auxin-binding proteins in cell suspen- 365-367 (1986). sions. I. Differential responses of carrot embryo cultures. 62. Broekaert WF, van Parijs J, Allen AK, Peumans WJ: Plant Physiol 97:60-64 (1991). Comparison of some molecular, enzymatic and antifun- 48. van Engelen FA, de Vries SC: Extracellular proteins in gal properties ofchitinases from thorn-apple, tobacco and plant embryogenesis. Trends Genet 8:66-70 (1992). wheat. Physiol Mol Plant Path 33:319-331 (1988). 49. Toonen MAJ, Hendriks T, Schmidt EDL, Verhoeven 63. Huynh QK, Hironaka CM, Levine EB, Smith CE, Borg- HA, van Kammen A, de Vries SC: Description of somatic meyer JR, Shah DM: Antifungal proteins from plants. embryo-forming single cells in carrot suspension cultures Purification, molecular cloning, and antifungal properties employing video cell tracking. Planta 194:565-572 of chitinases from maize . J Biol Chem 267: 6635- (1994). 6640 (1992). 50. Kreuger M, van Holst GJ: Arabinogalactan proteins are 64. Sela-Buurlage MB, Ponstein AS, Bres-Vloemans SA, essential in somatic embryogenesis of Daucus carota L. Melchers LS, van den Elzen PJM, Cornelissen BJC: Only Planta 189:243-248 (1993). specific tobacco (Nicotiana tabacum) chitinases and 8-1,3- 51. de Vries SC, Booij H, Janssens R, Vogels R, Saris L, Lo glucanases exhibit antifungal activity. Plant Physiol 101: Schiavo F, Terzi M, van Kammen A: Carrot somatic 857-863 (1993). embryogenesis depends on the phytohormone-controlled 65. Woloshuk CP, MeulenhoffJS, Sela-Buurlage M, van den presence of correctly glycosylated extracellular proteins. Elzen PJM, Cornelissen BJC: Pathogen-induced proteins Genes Devel 2:462-476 (1988). with inhibitory activity toward Phytophthora infestans. 52. van Hoist GJ, Klis FM, de Wildt PJM, Hazenberg CAM, Plant Cell 3:619-628 (1991). Buijs J, Stegwee D: Arabinogalactan protein from a crude 66. Broglie K, Chet I, Holliday M, Cressman R, Biddle P, cell organelle fraction of Phaseolus vulgaris L. Plant Knowlton, S, Mauvais CJ, Broglie R: Transgenic plants Physiol 68:910-913 (1981). with enhanced resistance to the fungal pathogen Rhizoc- 53. Knox JP, Day S, Roberts K: A set of cell surface glyco- tonia solani. Science 254:1194-1197 (1991). proteins forms an early marker of cell position, but not 67. Shinshi H, Mohnen D, Meins F Jr: Regulation of a plant cell type in the root apical meristem of Daucus carota L. pathogenesis-related enzyme: inhibition of chitinase and Development 106:47-56 (1989). chitinase mRNA accumulation in cultured tobacco tissue 54. de Jong AJ, Heidstra R, Spaink HP, Hartog MV, Meijer by auxin and cytokinin. Proc Natl Acad Sci USA 84: EA, Hendriks T, Lo Schiavo F, Terzi M, Bisseling T, van 89-93 (1987). Kammen A, de Vries, SC: Rhizobium lipooligosaccha- 68. Lotan T, Ori N, Fluhr R: Pathogenesis-related proteins rides rescue a carrot somatic embryo mutant. Plant Cell are developmentally regulated in tobacco flowers. Plant 5:615-620 (1993). Cell 1:881-887 (1989).

[76] 1313

69. Herget T, Schell J, Schreier PH: Elicitor-specific induc- synthesis is a chitooligosaccharide deacetylase. Proc Natl tion of one member of the chitinase gene family in Ara- Acad Sci USA 90:625-629 (1993). this hypogaea. Mol Gen Genet 224:469-476 (1990). 83. R6hrig H, Schmidt J, Wieneke U, Kondorosi E, Barlier 70. Kaufmann H, Kirch H, Wemmer T, Peil, A, Lottspeich I, Schell J, John M: Biosynthesis of lipooligosaccharide F, Uhrig H, Salamini F, Thompson R: Sporophytic and nodulation factors: Rhizobium NodA protein is involved gametophytic self-incompatibility. In: Cresti M, Tiezzi A in N-acylation of the chitooligosaccharide backbone. Proc (eds) Sexual Plant Reproduction, pp. 115-125. Springer- Natl Acad Sci USA, in press (1994). Verlag, Berlin (1992). 84. Staehelin C, Granado J, Mtlller J, Wiemken A, Mellor, 71. Boller T, Gehri A, Mauch F, Vogeli U: Chitinase in bean RB, Felix G, Regenass M, Broughton WJ, Boller T: Per- leaves: induction by ethylene, purification, properties and ception of Rhizobium nodulation factors by tomato cells possible function. Planta 157:22-31 (1983). and inactivation by root chitinases. Proc Natl Acad Sci 72. Benhamou N, Asselin A: Attempted localization of a sub- USA 91:2196-2200 (1994). strate for chitinases in plant cells reveals abundant 85. Vasse J, De Billy F, Truchet G: Abortion of infection N-acetyl-D-glucosamine residues in secondary walls. Biol during the Rhizobium meliloti-alfalfa symbiotic interaction Cell 67:341-350 (1989). is accompanied by a hypersensitive reaction. Plant J 4: 73. Bolwell GP: Synthesis of cell wall components: aspects 555-566 (1993). of control. Phytochemistry 27:1235-1253 (1988). 86. Staehelin C, Schultze M, Kondoros E, Mellor RB, Boiler 74. Lehle L, Fartaczek F, Tanner W, Kauss H: Formation of T, Kondorosi A: Structural modifications in Rhizobium polyprenol-linked mono- and oligosaccharides in Phaseo- meliloti Nod factors influence their stability against hy- lus aureus. Arch Biochem Biophys 175:419-426 (1976). drolysis by root chitinases. Plant J 5:319-330 (1994). 75. Durr M, Bailey DS, Maclachlan G: Subcellular distribu- 87. Darvill AG, Augur C, Bergmann C, Carlson RW, Cheong tion of membrane-bound glycosyltransferases from pea J-j, Eberhard S, Hahn MG, L6 V-m, Marf~ V, Meyer B, stems. Eur J Biochem 97:445-453 (1979). Mohnen D, O'Neill MA, Spiro MD, van Halbeek H, 76. Usui T, Matsui H, Isobe K: Enzymic synthesis of useful York WS, Albersheim P: Oligosaccharins-oligosaccha- chito-oligosaccarides utilizing transglycosylation by chiti- rides that regulate growth, development and defence re- nolytic enzymes in a buffer containing ammonium sulfate. sponses in plants. Glycobiology 2:181-198 (1992). Carbohydr Res 203:65-77 (1990). 88. Aldington S, Fry SC: Oligosaccharins. Adv Bot Res 19: 77. Carter HE, Betts BE, Strobach DR: Biochemistry of the 1-101 (1993). sphingolipids. XVII. The nature of the oligosaccharide 89. York WS, Darvill AG, Albersheim P: Inhibition of 2,4- component of phytoglycolipid. Biochemistry 3: 1103- dichlorophenoxyacetic acid-stimulated elongation of pea 1107 (1964). stem segments by a xyloglucan oligosaccharide. Plant 78. Spaink HP, Aarts A, Bloemberg GV, Folch J, Geiger O, Physiol 75:295-297 (1984). Schlaman HRM, Oates JE, van de Snade K, van Spron- 90. McDougall GJ, Fry SC: Structure-activity relationships sen P, van Brussel AAN, Wijfjes AHM, Lugtenberg B J J: for xyloglucan oligosaccharides with antiauxin activity. Rhizobial lipo-oligosaccharides: their biosynthesis and Plant Physiol 89:883-887 (1989). their role in the plant. In: Nester E (ed) Advances in 91. Fisher RF, Long SR: Rhizobium-plant signal exchange. Molecular Genetics of Plant-Microbe Interactions, vol 2, Nature 357:655-660 (1992). pp. 151-162. Kluwer Academic Publishers, Dordrecht 92. Spaink HP: Rhizobial lipo-oligosaccharides: answers and (1993). questions. Plant Mol Biol 20:977-986 (1992). 79. Schmidt J, ROhrig H, John M, Wieneke U, Stacey G, 93. Vijn I, Das Neves L, van Kammen A, Franssen H, Bis- Koncz C, Schell J: Alteration of plant growth and devel- seling T: Nod factors and nodulation in plants. Science opment by Rhizobium nodA and nodB genes involved in 260: 1764-1765(1993). the synthesis of oligosaccharide signal molecules. Plant J 94. Hirsch AM, Bhuvaneswari TV, Torrey JG, Bisseling T: 4:651-658 (1993). Early nodulin genes are induced in alfalfa root outgrowths 80. Spaink HP, Sheeley DM, van Brussel AAN, Glushka J, elicited by auxin transport inhibitors. Proc Natl Acad Sci York WS, Tak T, Geiger O, Kennedy EP, Reinhold, VN, USA 86:1244-1248 (1989). Lugtenberg BJJ: A novel highly unsaturated fatty acid 95. Van de Wiel C, Norris JH, Bochenek B, Dickstein R, moiety of lipo-oligosaccharide signals determines host Bisseling T, Hirsch AM: Nodulin expression and ENOD2 specificity of Rhizobium. Nature 354:125-130 (1991). localization in effective, nitrogenfixing and ineffective, 81. Schmidt J, Wingender R, John M, Wieneke U, Schell J: bacteria-free nodules of alfalfa (Medicago sativa). Plant Rhizobium meliloti nodA and nodB genes are involved in Cell 2:1009-1017 (1990). generating compounds that stimulate mitosis of plant 96. Cooper JB, Long SR: Morphogenetic rescue of Rhizo- cells. Proc Natl Acad Sci USA 85:8578-8582 (1988). bium meliloti nodulation mutants by trans-zeatin secre- 82. John M, R0hrig H, Schmidt J, Wieneke U, Schell J: tion. Plant Cell 6:215-225 (1994). Rhizobium NodB protein involved in nodulation signal

[77]