Tissue-specific transcriptome analysis reveals cell wall metabolism, flavonol biosynthesis and defense responses are activated in the endosperm of germinating Arabidopsis thaliana seeds

Seed germination is a result of the competition of embryonic growth potential and mechanical constraint by surrounding tissues such as the endosperm. To understand the processes occurring in the endosperm during germination, we analyzed tiling array expression data on dissected endosperm and embryo from 6 and 24 h-imbibed Arabidopsis seeds. The genes preferentially expressed in the endosperm of both 6 and 24 h-imbibed seeds were enriched for those related to cell wall biosynthesis/modifications, flavonol biosynthesis, defense responses and cellular transport. Loss of function of AtXTH31/XTR8, an endosperm-specific gene for a putative xyloglucan endotransglycosylase/hydrolase, led to faster germination. This suggests that AtXTH31/XTR8 is involved in the reinforcement of the cell wall of the endosperm during germination. In vivo flavonol staining by diphenyl boric acid aminoethyl ester (DPBA) showed flavonols accumulated in the endosperm of both dormant and non-dormant seeds, suggesting that this event is independent of germination. Notably, DPBA fluorescence was also intense in the embryo, but the fluorescent region was diminished around the radicle and lower half of the hypocotyl during germination. DPBA fluorescence was localized in the vacuoles during germination. Vacuolation was not seen in imbibed dormant seeds, suggesting that vacuolation is associated with germination. A gene for δVPE (vacuolar processing enzyme), a caspase-1-like cysteine proteinase involved in cell death, is expressed specifically in endosperms of 24 h-imbibed seeds. The δvpe mutant showed retardation of vacuolation, but this mutation did not affect the kinetics of germination. This suggests that vacuolation is a consequence, and not a trigger, of germination.     

A mutational approach to the study of seed development in maize

The maize seed comprises two major compartments, the embryo and the endosperm, both originating from the double fertilization event. The embryogenetic process allows the formation of a well-differentiated embryonic axis, surrounded by a single massive cotyledon, the scutellum. The mature endosperm constitutes the bulk of the seed and comprises specific regions containing reserve proteins, complex carbohydrates, and oils. To gain more insight into molecular events that underlie seed development, three monogenic mutants were characterized, referred to as emp (empty pericarp) on the basis of their extreme endosperm reduction, first recognizable at about 12 d after pollination. Their histological analysis reveals a partial development of the endosperm domains as well as loss of adhesion between pedicel tissues and the basal transfer layer. In the endosperm, programmed cell death (PCD) is delayed. The embryo appears retarded in its growth, but not impaired in its morphogenesis. The mutants can be rescued by culturing immature embryos, even though the seedlings appear retarded in their growth. The analysis of seeds with discordant embryo-endosperm phenotype (mutant embryo, normal endosperm and vice-versa), obtained using B-A translocations, suggests that emp expression in the embryo is necessary, but not sufficient, for proper seed development. In all three mutants the picture emerging is one of a general delay in processes related to growth, as a result of a mutation affecting endosperm development as a primary event.     

Family life at close quarters: communication and constraint in angiosperm seed development

The formation of viable angiosperm seeds involves the co-ordinated growth and development of three genetically distinct organisms, the maternally derived seed coat and the zygotic embryo and endosperm. The physical relationships of these tissues are initially established during the specification and differentiation of the female gametophyte within the tissues of the developing ovule. The molecular programmes implicated in both ovule and seed development involve elements of globally important pathways (such as auxin signalling), as well as ovule- and seed-specific pathways. Recurrent themes, such as the precisely controlled death of specific cell types and the regulation of cell–cell communication and nutrition by the selective establishment of symplastic and apoplastic barriers, appear to play key roles in both pre- and post-fertilization seed development. Much of post-fertilization seed growth occurs during a key developmental window shortly after fertilization and involves the dramatic expansion of the young endosperm, constrained by surrounding maternal tissues. The complex tissue-specific regulation of carbohydrate metabolism in specific seed compartments has been shown to provide a driving force for this early seed expansion. The embryo, which is arguably the most important component of the seed, appears to be only minimally involved in early seed development. Given the evolutionary and agronomic importance of angiosperm seeds, the complex combination of communication pathways which co-ordinate their growth and development remains remarkably poorly understood.     

Arabidopsis GLAUCE promotes fertilization-independent endosperm development and expression of paternally inherited alleles

Early seed development of sexually reproducing plants requires both maternal and paternal genomes but is prominently maternally influenced. A novel gametophytic maternal-effect mutant defective in early embryo and endosperm development, glauce (glc), has been isolated from a population of Arabidopsis Ds transposon insertion lines. The glc mutation results from a deletion at the Ds insertion site, and the molecular identity of GLC is not known. glc embryos can develop up to the globular stage in the absence of endosperm and glc central cells appear to be unfertilized. glc suppresses autonomous endosperm development observed in the fertilization-independent seed (fis) class mutants. glc is also epistatic to mea, one of the fis class mutants, in fertilized seeds, and is essential for the biparental embryonic expression of PHE1, a repressed downstream target of MEA. In addition, maternal GLC function is required for the paternal embryonic expression of the ribosome protein gene RPS5a and the AMP deaminase gene FAC1, both of which are essential for early embryo and endosperm development. These results indicate that factors derived from the female gametophyte activate a subset of the paternal genome of fertilized seeds.     

A Seed Coat Bedding Assay to Genetically Explore In Vitro How the Endosperm Controls Seed Germination in Arabidopsis thaliana

The Arabidopsis endosperm consists of a single cell layer surrounding the mature embryo and playing an essential role to prevent the germination of dormant seeds or that of nondormant seeds irradiated by a far red (FR) light pulse. In order to further gain insight into the molecular genetic mechanisms underlying the germination repressive activity exerted by the endosperm, a "seed coat bedding" assay (SCBA) was devised. The SCBA is a dissection procedure physically separating seed coats and embryos from seeds, which allows monitoring the growth of embryos on an underlying layer of seed coats. Remarkably, the SCBA reconstitutes the germination repressive activities of the seed coat in the context of seed dormancy and FR-dependent control of seed germination. Since the SCBA allows the combinatorial use of dormant, nondormant and genetically modified seed coat and embryonic materials, the genetic pathways controlling germination and specifically operating in the endosperm and embryo can be dissected. Here we detail the procedure to assemble a SCBA.     

Embryonic cuticle establishment: The great (apoplastic) divide

The plant cuticle, a dynamic interface between plants and their environment, is formed by the secretion of hydrophobic lipids and waxes into the outer wall of aerial epidermal cells. Cuticle formation is such a ubiquitous feature of epidermal cells, and is of such fundamental importance for plant survival, that identifying and understanding specific developmental roles for this structure has been a major challenge for plant scientists. In recent work, we have tried to understand the functional relationships between a signaling feedback loop required for epidermal cell specification in developing plant embryos, and a seed specific signaling cascade, involving components localized both in the embryo and in the embryo surrounding endosperm, and necessary for embryo cuticle function. Analysis of the strongly synergistic genetic relationships between these 2 independent pathways, combined with mathematical simulations of the behavior of the signaling feedback loop, have allowed us to propose an important, and hitherto unsuspected, role for the embryonic cuticle as an apoplastic diffusion barrier, necessary for preventing the excessive diffusion of developmentally important signaling molecules away from developing embryo into surrounding tissues.     

Effects of APETALA2 on embryo, endosperm, and seed coat development determine seed size in Arabidopsis

Arabidopsis APETALA2 (AP2) controls seed mass maternally, with ap2 mutants producing larger seeds than wild type. Here, we show that AP2 influences development of the three major seed compartments: embryo, endosperm, and seed coat. AP2 appears to have a significant effect on endosperm development. ap2 mutant seeds undergo an extended period of rapid endosperm growth early in development relative to wild type. This early expanded growth period in ap2 seeds is associated with delayed endosperm cellularization and overgrowth of the endosperm central vacuole. The subsequent period of moderate endosperm growth is also extended in ap2 seeds largely due to persistent cell divisions at the endosperm periphery. The effect of AP2 on endosperm development is mediated by different mechanisms than parent-of-origin effects on seed size observed in interploidy crosses. Seed coat development is affected; integument cells of ap2 mutants are more elongated than wild type. We conclude that endosperm overgrowth and/or integument cell elongation create a larger postfertilization embryo sac into which the ap2 embryo can grow. Morphological development of the embryo is initially delayed in ap2 compared with wild-type seeds, but ap2 embryos become larger than wild type after the bent-cotyledon stage of development. ap2 embryos are able to fill the enlarged postfertilization embryo sac, because they undergo extended periods of cell proliferation and seed filling. We discuss potential mechanisms by which maternally acting AP2 influences development of the zygotic embryo and endosperm to repress seed size.     

Imprinting of the MEDEA polycomb gene in the Arabidopsis endosperm.

In flowering plants, two cells are fertilized in the haploid female gametophyte. Egg and sperm nuclei fuse to form the embryo. A second sperm nucleus fuses with the central cell nucleus that replicates to generate the endosperm, which is a tissue that supports embryo development. MEDEA (MEA) encodes an Arabidopsis SET domain Polycomb protein. Inheritance of a maternal loss-of-function mea allele results in embryo abortion and prolonged endosperm production, irrespective of the genotype of the paternal allele. Thus, only the maternal wild-type MEA allele is required for proper embryo and endosperm development. To understand the molecular mechanism responsible for the parent-of-origin effects of mea mutations on seed development, we compared the expression of maternal and paternal MEA alleles in the progeny of crosses between two Arabidopsis ecotypes. Only the maternal MEA mRNA was detected in the endosperm from seeds at the torpedo stage and later. By contrast, expression of both maternal and paternal MEA alleles was observed in the embryo from seeds at the torpedo stage and later, in seedling, leaf, stem, and root. Thus, MEA is an imprinted gene that displays parent-of-origin-dependent monoallelic expression specifically in the endosperm. These results suggest that the embryo abortion observed in mutant mea seeds is due, at least in part, to a defect in endosperm function. Silencing of the paternal MEA allele in the endosperm and the phenotype of mutant mea seeds supports the parental conflict theory for the evolution of imprinting in plants and mammals.     

Endosperm cellularization defines an important developmental transition for embryo development

The endosperm is a terminal seed tissue that is destined to support embryo development. In most angiosperms, the endosperm develops initially as a syncytium to facilitate rapid seed growth. The transition from the syncytial to the cellularized state occurs at a defined time point during seed development. Manipulating the timing of endosperm cellularization through interploidy crosses negatively impacts on embryo growth, suggesting that endosperm cellularization is a critical step during seed development. In this study, we show that failure of endosperm cellularization in fertilization independent seed 2 (fis2) and endosperm defective 1 (ede1) Arabidopsis mutants correlates with impaired embryo development. Restoration of endosperm cellularization in fis2 seeds by reducing expression of the MADS-box gene AGAMOUS-LIKE 62 (AGL62) promotes embryo development, strongly supporting an essential role of endosperm cellularization for viable seed formation. Endosperm cellularization failure in fis2 seeds correlates with increased hexose levels, suggesting that arrest of embryo development is a consequence of failed nutrient translocation to the developing embryo. Finally, we demonstrate that AGL62 is a direct target gene of FIS Polycomb group repressive complex 2 (PRC2), establishing the molecular basis for FIS PRC2-mediated endosperm cellularization.     

Characterization of the two maize embryo-lethal defective kernel mutants rgh*-1210 and fl*-1253b: effects on embryo and gametophyte development

We have examined the effects on embryonic and gametophytic development of two nonallelic defective-kernel mutants of maize. Earlier studies indicated that both mutants are abnormal in embryonic morphogenesis as well as in the formation of their endosperm. Mutant rgh*-1210 embryos depart from the normal embryogenic pathway at the proembryo and transition stage, by developing meristematic lobes and losing bilateral symmetry. They continue growth as irregular cell masses that enlarge and become necrotic. Somatic embryos arising in rgh*-1210 callus cultures display the rgh*-1210 mutant phenotype. Mutant fl*-1253B embryos are variably blocked from the coleoptilar stage through stage 2. Following formation of the shoot apex in the mutant embryos the leaf primordia and tissues surrounding the embryonic axis continue growth and cell division, while the scutellum ceases development and becomes hypertrophied. Mutant fl*-1253B embryos are unable to germinate, either in mutant kernels or as immature embryos in culture, and the mutant scutellar tissue does not produce regenerable callus. Expression of the fl*-1253B locus during male gametophytic development is revealed by a marked reduction in pollen transmission as a result of mutant expression during the interval between meiosis and the initiation of pollen tube growth. In both mutants, there is considerable proliferation of the aleurone cells of the endosperm. Mutant expression of rgh*-1210 in the female gametophyte is revealed by the abnormal antipodal cells of the embryo sac. These results show that these two gene loci play unique and crucial roles in normal morphogenesis of the embryo. In addition, it is evident that both mutants are pleiotropic in affecting the development of the endosperm and gametophyte as well as the embryo. These pleiotropisms suggest some commonality in the gene regulation of development in these three tissues.     

ABNORMAL DEVELOPMENT OF THE SUSPENSOR IN AN EMBRYO-LETHAL MUTANT OF ARABIDOPSIS THALIANA

Developmental arrest of the embryo proper in aborted seeds from mutant 50B, a recessive embryo-lethal mutant of Arabidopsis thaliana, was shown to be followed by abnormal growth of the suspensor. Each of the 12 aborted seeds examined in sectioned material contained an abnormally large suspensor and an embryo proper arrested at a preglobular stage of development. Analysis of serial sections revealed that mutant suspensors contained 15–150 cells whereas wild-type suspensors were composed of only six to eight cells. Development of the mutant endosperm continued to a late nuclear or early cellular stage even in the absence of further development of the embryo proper. These results suggest that the missing gene product in mutant 50B is required for development of the embryo proper but not for continued growth of the suspensor or endosperm tissue. The pattern of abnormal development observed in this mutant provides further evidence that continued growth of the suspensor during normal development is inhibited by the developing embryo proper and that the full developmental potential of cells in the suspensor is expressed only when this inhibitory effect is removed through a mutation or experimental treatment that is lethal only to cells of the embryo proper.     

Control by the Embryo Axis of the Breakdown of Storage Proteins in the Endosperm of Germinated Castor BeanSeed: A Role for Gibberellic Acid

The embryo axis is required for the rapid breakdown of the crystalloid,albumin and lectin protein storage reserve in the endospermof castor bean (Ricinus communis L. cv. Hale) seeds, and forthe attainment of high specific activities of several endospermicproteolytic enzymes: one carboxy-peptidase and two -SH- dependentaminopeptidases. The embryo axis must be present to initiatestorage protein breakdown but it is not required to maintainthis process. We suggest that the embryo axis controls storageprotein breakdown through the release of promoters, which canbe replaced by gibberellins. Storage protein breakdown is notinfluenced by source-sink effects. However, the endosperm becomessensitive to gibberellin only after an imbibition period forup to 24 h. Key words: Castor bean, Protein breakdown, Storage protein, Embryo control, Gibberellin, Seed germination     

Endosperm: food for humankind and fodder for scientific discoveries

The endosperm is an essential constituent of seeds in flowering plants. It originates from a fertilization event parallel to the fertilization that gives rise to the embryo. The endosperm nurtures embryo development and, in some species including cereals, stores the seed reserves and represents a major source of food for humankind. Endosperm biology is characterized by specific features, including idiosyncratic cellular controls of cell division and epigenetic controls associated with parental genomic imprinting. This review attempts a comprehensive summary of our current knowledge of endosperm development and highlights recent advances in this field.     

Brassinosteroids and gibberellins promote tobacco seed germination by distinct pathways

Seed germination of Nicotiana tabacum L. cv. Havana 425 is determined by the balance of forces between the growth potential of the embryo and the mechanical restraint of the micropylar endosperm. In contrast to the gibberellin GA4, the brassinosteroid (BR) brassinolide (BL) did not release photodormancy of dark-imbibed photodormant seeds. Brassinolide promoted seedling elongation and germination of non-photodormant seeds, but did not appreciably affect the induction of class I beta-1,3-glucanase (betaGLU I) in the micropylar endosperm. Brassinolide, but not GA4, accelerated endosperm rupture of tobacco seeds imbibed in the light. Brassinolide and GA4 promoted endosperm rupture of dark-imbibed non-photodormant seeds, but only GA4 enhanced betaGLU I induction. Promotion of endosperm rupture by BL was dose-dependent and 0.01 microM BL was most effective. Brassinolide and GA4 promoted abscisic acid (ABA)-inhibited dark-germination of non-photodormant seeds, but only GA4 replaced light in inducing betaGLU I. These results indicate that BRs and GAs promote tobacco seed germination by distinct signal transduction pathways and distinct mechanisms. Gibberellins and light seem to act in a common pathway to release photodormancy, whereas BRs do not release photodormancy. Induction of betaGLU I in the micropylar endosperm and promotion of release of 'coat-enhanced' dormancy seem to be associated with the GA-dependent pathway, but not with BR signalling. It is proposed that BRs promote seed germination by directly enhancing the growth potential of the emerging embryo in a GA- and betaGLU I-independent manner.     

Effect of Salicylhydroxamic Acid on Endosperm Strength and Embryo Growth of Lactuca sativa L. cv Waldmann''s Green Seeds

Salicylhydroxamic acid (SHAM) stimulated germination of photosensitive lettuce (Lactuca sativa L. cv Waldmann's Green) seeds in darkness. To determine whether SHAM acts on the embryo or the endosperm, we investigated separately effects of SHAM on growth potential of isolated embryos as well as on endosperm strength. Embryo growth potential was quantified by incubating decoated embryos in various concentrations of osmoticum and measuring subsequent radicle elongation. Growth potential of embryos isolated from seeds pretreated with 4 millimolar SHAM was equal to that of untreated controls. Rupture strength of endosperm tissue excised from seeds pretreated with SHAM was 33% less than that of controls in the micropylar region. To determine if the embryo must be in contact with the endosperm for SHAM to weaken the endosperm, some endosperms were incubated with SHAM only after dissection from seeds. Rupture strength of SHAM-treated, isolated endosperms in the micropylar region was 25% less than that of untreated controls. There was no difference in rupture strength in the cotyledonary region of endosperm isolated from seeds treated with SHAM in buffer or buffer alone. SHAM therefore stimulates germination not by enhancing embryo growth potential, but by weakening the micropylar region of the endosperm enclosing the embryo.     

Hypomethylation promotes autonomous endosperm development and rescues postfertilization lethality in fie mutants

In most flowering plants, fertilization is necessary for development of the central cell into endosperm, but in the fie-1 mutant of Arabidopsis, the central cell can proliferate autonomously. However, autonomous fie-1 endosperms do not develop completely: They have fewer nuclei than sexually produced endosperms, cellularization does not take place, and no clear distinction is seen between the different endosperm compartments. Here, we show that autonomous endosperm develop much further in hypomethylated than normally methylated fie-1 mutants, undergoing cellularization and regional specification to resemble endosperm in sexually produced wild-type seeds. Therefore, the combination of maternal hypomethylation and loss of FIE function enables formation of differentiated endosperm without fertilization. A maternal fie-1 mutation is also lethal to sexual seeds, even if the pollen donor is wild type. We report that sexual mutant fie-1 endosperms fail to cellularize and overproliferate, consistent with the hypothesis that embryo abortion may be due, at least in part, to a defect in endosperm development. Finally, we show that pollen from hypomethylated plants rescues fie-1 mutant seeds provided that it also donates a wild-type paternal FIE allele. These results are discussed in light of models for parent-of-origin effects on seed development.