Plant Physiol, January 2001, Vol. 125, pp. 102-104

Fertilization in Flowering plants. New Approaches for an Old Story

Ecole Normale Supérieure de Lyon, Laboratory of Plant Reproduction and Development, Unité Mixte de Recherche 5667, Centre National de la Recherche Scientifique-Institut National de la Recherche Agronomique-Ecole Normale Supérieure, Lyon-Université Claude-Bernard Lyon I, Lyon 69364 cedex 07, France

	INTRODUCTION

TOP INTRODUCTION PROGRESS FROM CELL AND... NEW DEVELOPMENTS FROM... AT RANDOM OR PREFERENTIAL? THE ORIGIN OF DOUBLE... CONCLUSIONS LITERATURE CITED

Our understanding of how plant fertilization operates, because of the complexity of the reproductive structures and the highly diverse nature of the cellular processes involved, has historically been very closely linked to the developments in microscopy. In the 19th century, the development of light microscopy equipment and protocols allowed S. Nawaschin and L. Guignard (8) to discover independently that two fertilization events take place when the pollen tube meets the embryo sac. One event involves the union of a sperm with the central cell, from which the endosperm develops. The other event involves the union of sperm and egg, giving rise to the embryo. The use of the electron microscope led to a second series of discoveries, in particular by the group of Jensen (10). Sperm were shown to be real cells, despite their having been considered for decades to be no more than naked nuclei. Observations also led to the idea that the two fertilization events consist of plasmogamy and karyogamy steps. However, further progress was strongly limited by the inaccessibility of the gametes due to their encasement in parental tissues.

	PROGRESS FROM CELL AND MOLECULAR BIOLOGY APPROACHES

TOP INTRODUCTION PROGRESS FROM CELL AND... NEW DEVELOPMENTS FROM... AT RANDOM OR PREFERENTIAL? THE ORIGIN OF DOUBLE... CONCLUSIONS LITERATURE CITED

New information came from the development of techniques to isolate sperm cells from pollen grains and the female gametes from the maternal tissues (5, 12). In vitro fusions of sperm cells with egg cells or central cells were achieved by electrofusion (11) or in Ca²⁺-containing media (6). Regeneration of the sperm-egg fusion products into fertile plants in addition to in vitro divisions of the fertilized central cells was obtained using feeder cells (12, 13). A major benefit of in vitro fusion is that fertilization can be directly visualized and manipulated. Isolated cells can also be collected for molecular studies. The first currently detectable cellular events that take place after gamete fusion are an increase of the concentration of cytosolic Ca²⁺ (4) as in animal gamete fusion. An influx of extracellular Ca²⁺ may contribute to this cytosolic increase (1). However, an important question that was extensively studied in animal fertilization still remains unanswered in plants: Is this Ca²⁺ elevation necessary and/or sufficient to trigger egg activation and the initiation of development? Another similarity to animals is the establishment of a block to polyspermy: Maize (Zea mays) sperm cannot fuse with zygotes in vitro (6). This barrier is established as early as 45 s after the initial fusion in maize. Cell wall deposition may mechanically contribute to this block to polyspermy in analogy to the fertilization envelope that leads to slow block to polyspermy in several animal species. However, this remains to be demonstrated. The electrical properties of the gamete membranes also need to be characterized to understand if an electrical block exists, i.e. a fast block to polyspermy as occurs in several animal species.

Unlike in animals (19), a molecule with a possible role in sperm-egg interactions at the plasma membrane level has not yet been identified. Analogies with other organisms will probably not be sufficient to isolate these components because molecules involved in the reproduction process seem to have diverged to a large extent between different taxonomic groups (19). The identification of such molecules is important for understanding how development is activated and for identifying any possible difference in fertilization between the egg and central cells. It should also help us to understand how far the fertilization mechanisms have diverged between animal and plant species. New approaches hopefully should allow the identification of such molecules in the future. cDNA libraries have been prepared from isolated maize gametes and in vitro zygotes. Subsequent differential screening has already provided a few clones specifically expressed, or more highly expressed, after fertilization (12). Libraries have also been prepared from the mother cells of the two sperms, i.e. the generative cells, in Lilium longiflorum (22). Differential screening has led to the identification of LGC1, a gene expressed specifically in the generative and sperm cells (22). It is interesting that the product of LGC1 has been shown to localize at the surface of the male gametic cells. Further studies will be required to understand the role of all these newly isolated genes. In addition, a major challenge in the upcoming years will be to understand how far the in vitro situation resembles the in vivo fertilization and how far in vitro procedures can be used to understand the cellular events during gamete fusion.

	NEW DEVELOPMENTS FROM ARABIDOPSIS GENETICS

TOP INTRODUCTION PROGRESS FROM CELL AND... NEW DEVELOPMENTS FROM... AT RANDOM OR PREFERENTIAL? THE ORIGIN OF DOUBLE... CONCLUSIONS LITERATURE CITED

Screens to identify Arabidopsis mutants with fertilization defects have been done or are under progress in several laboratories. The group of A.M. Chaudhury has isolated three mutants, fis1, fis2, and fis3 (for fertilization-independent seed), that show some aspects of seed development without fertilization (3, 14). The group of R.M. Fischer similarly has isolated fie, a fertilization-independent endosperm mutant (14, 15) allelic to fis3. They have also isolated f644, a mutant allelic to fis1, to an embryo-defective mutant emb173, and to the medea mutant from U. Grossniklaus' group (9, 14). The three genes corresponding to these described mutants are now often referred as FIE (for fie or fis3), MEA (for medea, fis1, f644, or emb173), and FIS2 (for fis2). MEA and FIE were shown to encode proteins from the polycomb group, whereas FIS2 encodes a zinc finger protein (9, 14). The products of the three genes are thought to form a complex that represses the genes involved in seed development (3). The phenotype of the mutants is complex. It is interesting that if the female gametophyte has a mutant medea allele, for example, embryo development does not proceed normally (9). Therefore, the corresponding genes are thought to have a gametophytic maternal effect on the development of the embryo. Although the precise role of these master genes needs to be identified, as does their integration into the signaling that leads to seed development, these data have important consequences for our understanding of fertilization. They outline the importance of the female gametophyte in embryo development, in contrast with the long-held belief that genes newly expressed in zygotes were more important in early development. Besides this important new perspective, it should be noted that this could constitute, in particular, a limitation to the strategies of differential screening described above to identify genes involved in early zygote development.

It has been suggested that the observed maternal effect for the MEA gene is actually due to the silencing of the paternal alleles by genomic imprinting (21). It is possible that FIS2 and FIE are also regulated in that manner. Methylation may be the basis for such a differential expression because crosses with pollen from the ddm1 mutant or MET1 antisense plants rescue seeds with a maternal copy of medea (7, 21). The group of U. Grossniklaus recently suggested that the whole paternal genome is actually transiently silenced after fertilization and that embryogenesis and endosperm development is mainly under maternal control (20). These authors focused on 20 genes expressed early in the endosperm and/or embryo. They studied the expression of the maternal or paternal copies of these genes using -glucuronidase and allele-specific reverse transcriptase-PCR and proposed that the paternal copies are silenced during the first few days following fertilization. From these data they argued that an unequal contribution of the parental genomes is not restricted to MEA, FIS2, or FIE, but is true for the whole genome. These data are exciting but will need further examination. It is difficult to extend these observations made with a few genes to the whole genome. There may exist very subtle and critical differences from one gene to another, and also between the embryo and the endosperm. In addition, one of the 20 genes studied by the group of U. Grossniklaus was PROLIFERA, but data on the expression of this gene just published by Springer et al. (18) have been interpreted in a conflicting way.

	AT RANDOM OR PREFERENTIAL?

TOP INTRODUCTION PROGRESS FROM CELL AND... NEW DEVELOPMENTS FROM... AT RANDOM OR PREFERENTIAL? THE ORIGIN OF DOUBLE... CONCLUSIONS LITERATURE CITED

The two male gametes have two different target cells, the egg and the central cells, with two very different developmental fates, the embryo and the endosperm. In addition, observations suggest sperm cell dimorphism in several species (17). It is therefore reasonable to ask if the sperm cells fuse at random or selectively, i.e. if there is a preferential fertilization, and if this could be important for further development. The possibility of a preferential fertilization of one sperm with the egg cell has been suggested from a very small number of studies performed in maize and Plumbago zeylanica (16). In vitro fertilization may now be used to test if the two male gametes from the same pollen grain can fuse with egg cells. However, the data supporting a general paternal silencing would argue against any differential role of the sperm in early development, except the possible inheritance of specific organelles or molecules.

	THE ORIGIN OF DOUBLE FERTILIZATION?

TOP INTRODUCTION PROGRESS FROM CELL AND... NEW DEVELOPMENTS FROM... AT RANDOM OR PREFERENTIAL? THE ORIGIN OF DOUBLE... CONCLUSIONS LITERATURE CITED

The evolutionary significance of double fertilization in flowering plants has been questioned during the last ten years. Original works from the early 1900s, suggesting that two karyogamy events occur in parallel in the female gametophyte of some gnetales, were reinvestigated and confirmed by Friedman (8). A rudimentary process of double fertilization was observed in Ephedra nevadensis, Ephedra trifurca, and Gnetum gnemon (8). It consists of the fusion of one sperm nucleus with the egg cell nucleus and of a second sperm nucleus brought by the same pollen tube with the mitotic sister nucleus of the egg. The author proposed that this process developed in a common ancestor of the gnetales and the flowering plants and that one of the fusion products later evolved into the endosperm of angiosperms. However, recent data separate angiosperms from all gymnosperms and put the gnetales as the closest relatives to conifers (2). This implies that double fertilization arose independently in gnetales and angiosperms. Molecular evidence is therefore now required to determine whether the model proposed for the origin of double fertilization and the evolutionary equivalence of the two fertilization events is valid, or if double fertilization emerged differently in the angiosperms.

	CONCLUSIONS

TOP INTRODUCTION PROGRESS FROM CELL AND... NEW DEVELOPMENTS FROM... AT RANDOM OR PREFERENTIAL? THE ORIGIN OF DOUBLE... CONCLUSIONS LITERATURE CITED

The last 15 years have been characterized by a burst of new information from cell biology and genetics. Important new ideas have emerged: (a) Some fertilization steps and cellular processes, such as the Ca²⁺ increase, seem to be very similar to the equivalent events in animals, at least in the egg. Future investigations will be required to identify the molecules involved, and to understand if these steps are similar in the central cell; (b) Genes with maternal effects seem to be important for early seed development; and (c) There may be a broad imprinting of genes and an early silencing of the paternal genome. Therefore, the precise contribution of the male gametes to the very early development will have to be identified. In the future, a greater understanding of fertilization should be gained from further technological advances, such as confocal microscopy techniques and the use of green fluorescent proteins to visualize structures in living materials.

	ACKNOWLEDGMENTS

We are grateful to Dr. Charlie Scutt (Centre National de la Recherche Scientifique, Lyon, France) and Professor Sheila McCormick (University of California, Berkeley) for their critical reading of the manuscript.

	FOOTNOTES

^* Corresponding author; e-mail Christian.Dumas{at}ens-lyon.fr; fax 33-4-72-72-86-00.

	LITERATURE CITED

TOP INTRODUCTION PROGRESS FROM CELL AND... NEW DEVELOPMENTS FROM... AT RANDOM OR PREFERENTIAL? THE ORIGIN OF DOUBLE... CONCLUSIONS LITERATURE CITED

• Antoine A-F, Faure J-E, Cordeiro S, Dumas C, Rougier M, Feijo JA (2000) Proc Natl Acad Sci USA 97: 10643-10648 [Abstract/Free Full Text])
• Bowes LM, Coat G, de Pamphilis CW (2000) Proc Natl Acad Sci USA 97: 4092-4097 [Abstract/Free Full Text]
• Chaudhury AM, Ming L, Miller C, Craig S, Dennis ES, Peacock WJ (1997) Proc Natl Acad Sci USA 94: 4223-4228 [Abstract/Free Full Text]
• Digonnet C, Aldon D, Leduc N, Dumas C, Rougier M (1997) Development 124: 2867-2874 [Abstract]
• Dumas C, Faure J-E (1995) Curr Opin Biotechnol 6: 183-188
• Faure J-E, Digonnet C, Dumas C (1994) Science 263: 1598-1600 [Abstract/Free Full Text]
• Finnegan EJ, Peacock WJ, Dennis ES (2000) Curr Opin Genet Dev 10: 217-223 [CrossRef][Web of Science][Medline]
• Friedman WE (1998) Sex Plant Reprod 11: 6-16 [CrossRef]
• Grossniklaus U, Vielle-Calzada J-P, Hoeppner MA, Gagliano WB (1998) Science 280: 446-450 [Abstract/Free Full Text]
• Jensen WA (1998) Sex Plant Reprod 11: 1-5
• Kranz E, Bautor J, Lörz H (1991) Sex Plant Reprod 4: 12-16
• Kranz E, Dresselhaus T (1996) Trends Plant Sci 1: 82-89
• Kranz E, von Wiegen P, Quader H, Lörz H (1998) Plant Cell 10: 511-524 [Abstract/Free Full Text]
• Ma H (1999) Curr Biol 9: 636-639 [CrossRef]
• Ohad N, Margossian L, Hsu Y-C, Williams C, Repetti P, Fischer RL (1996) Proc Natl Acad Sci USA 93: 5319-5324 [Abstract/Free Full Text]
• Russell SD (1993) Plant Cell 5: 1349-1359 [Free Full Text]
• Saito C, Nagata N, Sakai A, Mori K, Kuroiwa H, Kuroiwa T (2000) Sex Plant Reprod 12: 296-301 [CrossRef]
• Springer PS, Holding DR, Groover A, Yordan C, Martienssen RA (2000) Development 127: 1815-1822 [Abstract]
• Vacquier VD (1998) Science 281: 1995-1998 [Abstract/Free Full Text]
• Vielle-Calzada J-P, Baskar R, Grossniklaus U (2000) Nature 404: 91-94 [CrossRef][Medline]
• Vielle-Calzada J-P, Thomas J, Coluccio A, Hoeppner MA, Grossniklaus U (1999) Genes Dev 13: 2971-2982 [Abstract/Free Full Text]
• Xu H, Swoboda I, Bhalla PL, Singh MB (1999) Proc Natl Acad Sci USA 96: 2554-2558 [Abstract/Free Full Text]