Elizabeth M. Lord1 and Scott D. Russell2 · Abstract In ﬂowering plants, pollen grains germinate to form pollen tubes that transport male gametes (sperm cells) to the egg cell in

10 Sep 2002 13:22 AR AR170-CB18-04.tex AR170-CB18-04.sgm LaTeX2e(2002/01/18)P1: ILV10.1146/annurev.cellbio.18.012502.083438

Annu. Rev. Cell Dev. Biol. 2002. 18:81–105doi: 10.1146/annurev.cellbio.18.012502.083438

Copyright c© 2002 by Annual Reviews. All rights reservedFirst published online as a Review in Advance on May 9, 2002

THE MECHANISMS OF POLLINATION AND

FERTILIZATION IN PLANTS

Elizabeth M. Lord1 and Scott D. Russell21Department of Botany and Plant Sciences, University of California, Riverside,California 92521; e-mail: [email protected] of Botany and Microbiology, University of Oklahoma, Norman,Oklahoma 73019; e-mail: [email protected]

Key Words double fertilization, embryo sac, gametes, pistil, pollen tube guidance

■ Abstract In flowering plants, pollen grains germinate to form pollen tubes thattransport male gametes (sperm cells) to the egg cell in the embryo sac during sexualreproduction. Pollen tube biology is complex, presenting parallels with axon guidanceand moving cell systems in animals. Pollen tube cells elongate on an active extra-cellular matrix in the style, ultimately guided by stylar and embryo sac signals. Awell-documented recognition system occurs between pollen grains and the stigma insporophytic self-incompatibility, where both receptor kinases in the stigma and theirpeptide ligands from pollen are now known. Complex mechanisms act to preciselytarget the sperm cells into the embryo sac. These events initiate double fertilization inwhich the two sperm cells from one pollen tube fuse to produce distinctly differentproducts: one with the egg to produce the zygote and embryo and the other with thecentral cell to produce the endosperm.

CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82COMPATIBLE POLLEN/PISTIL INTERACTIONS. . . . . . . . . . . . . . . . . . . . . . . . . . 82

Adhesion and Hydration on the Stigma. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82Germination and Chemotropism on the Stigma. . . . . . . . . . . . . . . . . . . . . . . . . . . . 84Pollen Tube Cell Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85Pollen Tube Guidance in the Style and Ovary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

SELF-INCOMPATIBILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90Sporophytic—Brassicaceae. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90Gametophytic–Papaveraceae and Solanaceae. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

POLLINATION ECOLOGY AND EVOLUTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91Mate Choice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91Pollen Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

FERTILIZATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92Male Gamete Lineage Gene Expression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92In Vivo Studies of Fertilization in Flowering Plants. . . . . . . . . . . . . . . . . . . . . . . . . 93

1081-0706/02/1115-0081$14.00 81

10 Sep 2002 13:22 AR AR170-CB18-04.tex AR170-CB18-04.sgm LaTeX2e(2002/01/18)P1: ILV

82 LORD ¥ RUSSELL

In Vitro Studies of Fertilization in Flowering Plants. . . . . . . . . . . . . . . . . . . . . . . . . 96POST-FERTILIZATION EVENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Nuclear Migration and Fusion in Angiosperms. . . . . . . . . . . . . . . . . . . . . . . . . . . . 98Endosperm Initiation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

EVOLUTION OF DOUBLE FERTILIZATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99CONCLUSIONS AND PROSPECTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

INTRODUCTION

The earliest record we have suggesting a human interest in the mechanism of polli-nation and fertilization in plants is that of a relief carving from the palace of SyrianKing Ashuirnasir-pal II, 883–859 BC, which shows a winged man with a male inflo-rescence from a palm tree distributing pollen over a female inflorescence (Stanley &Linskens1974). It was not until 1824 that an Italian mathematician and astronomer,Amici, first observed pollen tubes germinating on a stigma. He later proposed thatthe pollen tube carried the sperm cells to the ovule where the egg resided. In 1884,Strasburger (Maheshwari 1950) demonstrated fertilization as fusion of egg andsperm nuclei in the flowering plants (angiosperms), a difficult task because of theenclosure of the egg inside the ovule within the ovary. By 1898, double fertilizationwas recognized as well, a feature unique to flowering plants (see below).

Our present effort is to provide an overview of the field of pollination and fertil-ization and propose directions for new research. Progress in our understanding ofplant reproduction has been particularly good in the past ten years. Here we attemptto cover events from pollination to fertilization in a single chapter, including somediscussion of the ecology and evolution of breeding systems. To do so, we mustfocus on the most recent literature. A major review covering the molecular and ge-netic basis of pollen/pistil interaction has just been published (Wheeler et al. 2001).

COMPATIBLE POLLEN/PISTIL INTERACTIONS

Adhesion and Hydration on the Stigma

Pollen, which contains the sperm cells, develops in the stamen, the male organof the flower. Here it undergoes dehydration to some extent before being releasedfrom the anther to the environment. The pistil, the female organ of the flower, iscomprised of the stigma (where pollen first alights), the style (a conduit between thestigma and ovary), and the ovary, which contains the ovules. When pollen comes incontact with the stigma, it adheres, hydrates, and germinates. The result is a pollentube cell that carries the sperm cells, endocytosed into the larger tube cell, thusconveying them to the embryo sac and egg in the ovule. A flower can produce pollenthat penetrates its own pistil and fertilizes its own ovules in most angiosperms,a predominantly self-compatible (SC) group. These early events of adhesion andhydration of pollen on the stigma have been studied intensively because they arethe initial steps of pollination and because they are readily observed.


PLANT POLLINATION AND FERTILIZATION 83

Stigmas are of two broad types, wet and dry, with respect to the amount ofsecreted matrices that the pollen encounters on their surface. The pollen grainhas a tough outer wall, like a spore, which typically is sculptured. It providesa porous surface containing the many substances that are deposited on it by theparent sporophyte while it is inside the anther or that are secreted into it by themale gametophyte during pollen development (Dickinson et al. 2000, Mayfieldet al. 2001). The function of these pollen coat substances is still a mystery forthe most part, even though they are the major allergens of humans and of medicalinterest. One exception to this is the recent discovery of the male recognition factorin self-incompatibility (SI) in the Brassicaceae that resides in the pollen coat (seebelow). This coat may have some role in adhesion of pollen to vectors such asinsects and birds because wind-pollinated species have thin pollen coats and mayalso have evolved as a food source for biotic vectors. The pollen coat is assumedto have a role in adhesion to the stigma (Doughty et al. 2000, Luu et al. 1999,Takayama et al. 2000), but data inArabidopsisleave this in doubt (Zinkl et al.1999). Hydration of the pollen grain is somehow aided by the presence of the coat,which inArabidopsiscontains predominantly lipases and oleosins (Mayfield et al.2001). The loss of one oleosin protein from the coat (GRP 17–1) impairs pollenhydration. These data from only one family, the Brassicaceae, suggest a complexseries of events in both adhesion and hydration of pollen on dry stigmas. Anotherrole for the pollen coat substances is that of facilitating penetration into the style.Enzymes residing in the coat play a role in modification of the stigma extracellularmatrix (ECM) to allow for penetration into the pistil (Bih et al. 1999, Wheeleret al. 2001).

On the stigma side, there is evidence for a plasma membrane–localizedaquaporin-like protein called MIP-MOD in the Brassicaceae (Dixit et al. 2001).The aquaporins are water-transporting molecular channels, and in plants they oc-cur in either the plasma membrane or the vacuole. A crucial first step in acceptanceof non-self pollen on dry stigmas occurs at pollen hydration, and presumably thisaquaporin-like protein is a major regulator of hydration by controlling water flowfrom the stigma to the pollen grain. On wet stigmas, lipids (trilinoleins) have beenimplicated in hydration as well (Wolters-Arts et al. 1998), but little is known aboutinitial steps in wet stigmas having exudates low in these lipids.

The pollen coat is full of proteins that could act as ligands for receptors in thestigma, preparing the way for adhesion, hydration, and germination of compatiblepollen, but so far the only known recognition event at the stigma surface is that ofSI, where active rejection ensues when self pollen lands on the stigma (Dixit &Nasrallah 2001). There are a plethora of receptor-like kinases in plants but a paucityof known ligands. Glycine-rich proteins (GRPs), of which oleosins are a member,are common cell wall proteins of unknown function. WAKs (wall-associated ki-nases) are receptors (serine-threonine kinases) (Kohorn 2001) that occur through-out the plant, and anArabidopsisGRP (AtGRP-3) regulates WAK-1 by bindingto its cell wall domain (Park et al. 2001). It would be interesting to know whetherWAKs and oleosins interact in the stigma. To date no receptors or ligands have


84 LORD ¥ RUSSELL

been described for self-compatible processes at the stigma surface even thoughthis first step in pollination is such a critical one for effective fertilization.

When the pollen grain is hydrated, the tube cell cytoplasm becomes activatedand some genes, such as MAP kinases of unknown function, may act specifically atthis stage (Heberle-Bors et al. 2001). Many of the genes necessary for germinationhave been transcribed, and the pollen tube cell is primed for the next events leadingto fertilization, which can take hours or days to occur (Taylor & Hepler 1997).

Germination and Chemotropism on the Stigma

After hydration, the pollen grain produces an outgrowth from an aperture or thinarea in the wall. This is the site of tip growth that results in the production of a pollentube, which will ultimately convey the sperm cells to the embryo sac. The initialprocess of germination is under complex control, but a major regulator is Rop, aGTPase that shares the same ancestor as Cdc42, Rac, and Rho in the RHO familyof small GTPases (Fu & Yang 2001). The function of Rop in germination wasstudied by using Rop1 overexpression and rop mutants inArabidopsis. Expressionof a dominant-negative (DN) rop1 mutant inhibits pollen germination (Li et al.1999), whereas overexpression of WT Rop1 promotes germination (V. Vernoud &Z. Yang, unpublished results).

Once germination occurs, maintenance of polar growth in the pollen tube isalso controlled by Rop in a complex signaling network that includes regulation ofa tip-focused Ca+2 gradient and F-actin polymerization (see below).

Hundreds of pollen grains are usually deposited on the stigma, and germinationthere is known to be density dependent. This is called the mentor effect, and hasrecently been attributed to a peptide growth factor, phytosulfokine-α (Chen et al.2000). This five amino acid–sulfated peptide, first confirmed as a mitogen in plantcell cultures, was subsequently found to be capable of promoting germination ofpollen cultured at low density. There are only four known peptides in higher plantsthat play a role in cell-cell communication via receptors, and phytosulfokine is arecent addition to this group (Matsubayashi et al. 2001).

After pollen germination, penetration into the style is the next step, and herethe processes vary considerably across species, stigma, and style types. Entranceis always into a specialized ECM of the pistil that appears to guide pollen tubesto the ovules (Lord 2000). In solid styles that have dry stigmas, pollen tubes haveto penetrate an outer lipidic cover of the stigmatic papilla, the cuticle, whereas inwet stigmas, entrance is directly into the ECM of a degenerating papilla. In openstyles, typically the stigma is also open and covered with a secreting epidermisthat is continuous with that of the stylar canal so no penetration is necessary, justguidance along a secretory dermal layer into the style and then to the ovary.

On the stigma, pollen tubes germinate in an ECM that is usually a combinationof both pollen coat secretions and stigmatic exudates. In the Solanaceae, lipids(triglycerides) in the stigma exudate were shown to be essential for pollen germi-nation oriented toward the style (Wolters-Arts et al. 1998). In this case, the cue fordirectional growth of the pollen tube is a physical factor, a gradient of water set up



by the stigmatic lipids over an aqueous component of the exudate. In other stigmatypes, pollen tube guidance is required for entry into the style. The open stigmaand style of lily is an example where pollen germinates over a large stigma surface,and the tubes must be guided toward the central stylar canal. An abundant, secretedpeptide in lily, SCA (stigma/stylar cysteine-rich adhesin) has been purified fromthe transmitting tract and shown to be involved, with a pectic polysaccharide, inpollen tube adhesion and guidance in the style (Mollet et al. 2000, Park et al. 2000).Binding of SCA to the pectin is necessary for adhesion of pollen tubes to this spe-cialized ECM. On the stigma, this peptide appears to act alone as a chemotropiccompound. Pollen tubes were shown to reorient their growth toward a chemical gra-dient of SCA in vitro establishing SCA as the first chemotropic peptide identifiedin plants (S. Kim, S.-Y. Park & E.M. Lord, unpublished results). Thus in the stigma,SCA may act alone as a chemotropic molecule, whereas in the style it acts witha pectin in haptotactic (adhesion-mediated) guidance. Allurin, a vertebrate spermchemo-attractant, is related to sperm adhesion molecules (Olson et al. 2001). Sothis combination of chemo-attractant and adhesion function is not unique to SCA.

These few studies document the interplay between the pollen tube and the stigmafor compatible pollinations. Pollen-stigma interactions that lead to the rejectionof self pollen is much better understood (see below). How recognition betweenpollen and stigma in a compatible situation regulates intracellular events is a farmore mysterious subject. Given the importance of Rop signaling in the activationof pollen germination and maintenance of pollen tube growth, it would be of greatinterest to see if recognition regulates intracellular Rop signaling.

Pollen Tube Cell Biology

The pollen tube cell is the fastest growing plant cell, with rates of up to 1 cm· h−1.It shows other unusual features that make it unique as a plant cell (Figure 1). Inparticular, it carries the sperm cells inside it as it grows and is guided by specializedECMs to the ovule where the egg resides. The pollen tube grows by tip growth,producing new membrane and wall at the front. The tube cell and sperm cells movealong with the advancing tip, a movement that is unique in plants where cell wallsnormally confine the protoplast in one location (Lord 2000).

The cell biology of tip-growing cells in general, and of pollen tubes in particular,has been studied intensively as systems of polarized growth. Pollen can be readilygrown in vitro, and the early stages of tube formation are now studied using a varietyof sophisticated tools. The exciting new findings on tip-focused calcium gradients,on proton, potassium, and chloride fluxes and oscillating growth have recentlybeen reviewed by Hepler et al. (2001), as are aspects of the pollen tube cytoskele-ton and membrane trafficking. Here, we focus on aspects of pollen cell biologyrelated to its function in sperm cell transport over long distances in vivo. The initialgrowth of the pollen tube is probably regulated primarily by internal cues. Oncethe tube cell has fully exited the pollen grain and the first cell wall (callose plug)has formed, external cues in the transmitting tract likely play a larger tole in pollentube growth. The most important recent discoveries about pollen tube cell biology


86 LORD ¥ RUSSELL

Figure 1 Pollination in the lily pistil. The pollen grains (PG) germinate on the stigma(St) and produce a pollen tube (PT) that carries the male germ unit (MGU), includingsperm cells, to the ovule (Ov). Lily has a hollow style lined with a transmitting tractepidermis (TTE) on which pollen tubes are guided to the ovary (O). An enlargementshows the pollen tube tip on the TTE, which secretes an extracellular matrix (ECM)including adhesion molecules. Golgi vesicles secrete wall material and membrane atthe tip where a dynamic F-actin occurs. Behind the tip, cytoplasmic streaming occurson F-actin cables, and a cortical cytoskeleton keeps the MGU near the tip.

relevant to growth and guidance in vivo include the discovery of a form of dynamicactin at the growing tip (Fu et al. 2001); the machinery for actin polymerization,including ARP2 (Klahre & Chua 1999, Hepler et al. 2001); and the master switch,Rop GTPase (Fu & Yang 2001). A prolonged debate on pollen tube cell biologyconcerned the presence or absence of F-actin at the leading edge or growing tip ofthe pollen tube. For various reasons, including fixation artifacts, the presence ofF-actin was questionable until an actin-binding domain of a mouse Talin taggedwith an enhanced GFP mutant was used (Fu et al. 2001). The same group had



discovered a subfamily of Rho GTPases in plants and Rop (Zheng & Yang2000) and localized a pollen-specific Rop to the pollen tube tip plasma mem-brane. They were able to block Rop signaling and induce constitutively active Ropto confirm that polar tip growth in pollen was controlled by this GTPase.

Because the RHO family of small GTPases is known to control the actin cyto-skeleton in mammals and yeast, the assertion that pollen tube tips were devoid ofF-actin was puzzling. The use of the enhanced GFP mutant was necessary to detectthe very dynamic form of tip-localized F-actin in pollen tubes (Fu et al. 2001; videoavailable at http://www.jcb.org/). This dynamic F-actin oscillates (as does growthat the pollen tube tip), appearing before the growth spurt, and its presence was foundto be controlled by Rop, as is the tip-focused cytosolic calcium gradient (Fu &Yang 2001).

The realization that a population of F-actin at the very tip of the pollen tube, notinvolved in cytoplasmic streaming, was essential for growth strongly suggests arole for the machinery of cell motility at the pollen tube tip (Lord 2000). A role forF-actin polymerization in animal cell movement is accepted, but the mechanismis only now being deciphered (Borisy & Svitkina 2000, Higgs & Pollard 2001)(http://www.cellmigration.org/). Several of the major players exist in pollen tubetips. The fact that the tube cell and the sperm cells it carries move through thestyle to the ovary is not debatable, but the mechanism of movement for this cellcomplex is still unknown. The discovery of the F-actin machinery for cell motilityat the tube cell tip begins to address mechanism, but the large size of the tube cell(1–5 mm) and the presence of other cytoskeletal components such as myosin andmicrotubules (MTs) probably means a large cast of characters is involved. A rolefor MTs in pollen tube growth was dismissed initially because colchicine, whichbreaks down MTs, had no effect on pollen tube tip growth in young in vitro–grownpollen tubes. Subsequent studies using older pollen tubes and performed in vivoestablished that MTs were essential for maintenance of polarity in the tube celland particularly for movement of the male germ unit (tube cell nucleus and twosperm cells), which insures its retention at the tip (Joos et al. 1994). The MTs areoriented longitudinally in the tube cell cortex, and several motor proteins (kinesinsand dyneins) are associated with them (Cai et al. 2001). Breakup of MTs leads to alack of forward movement of the male germ unit and displacement of the vacuoleinto the tip, so these MTs may play a role in movement of the tube cell protoplastalong the pollen tube wall. The motor molecules associated with the MTs areprominent in the cortex and may be involved in this movement. This importantaspect of pollination has been neglected because protoplast movement is mostapparent only after hours of in vitro culture when the first callose plug, whichsequesters the living part of the pollen tube toward the tip, is formed at the rear ofthe tube cell. In culture, these pollen tubes are considered old and difficult to workwith, but they represent just a fraction of the mature tube lengths that occur in vivo.

Vesicle secretion rates at the tip of the pollen tube may also be a controlling factorfor growth rate in concert with endocytosis behind the tip (Parton et al. 2001). Otherimportant findings indicate that cAMP occurs in pollen tubes, that modulation of its


88 LORD ¥ RUSSELL

concentration affects growth (Moutinho et al. 2001), and that transient adhesionsbetween the plasma membrane and wall occur at the tip (Parton et al. 2001).

Pollen Tube Guidance in the Style and Ovary

A prevalent model for pollen tube guidance is based on an analogy with adhesionand cell movement in animals, in particular with the cell biology of neuronal nav-igation (Lord 2000, Palanivelu & Preuss 2000, Wheeler et al. 2001). Migration ofeither neurons or their nerve growth cones occurs by mechanisms similar to thoseresponsible for chemotaxis in non-neuronal cells (Song & Poo 2001). The impor-tance for proper guidance of adhesion molecules in the ECM and their receptors onthe moving cell is now well documented. These extracellular factors occur alongthe path and, by selective adhesion and intracellular signaling, are responsible forguiding cells or their extensions to a goal. A hierarchy of guideposts is proposed sothat these target-derived factors can act in succession. Some are diffusible, othersbind to matrix molecules, but all are secreted into a specialized ECM at the righttime and place to effect growth and/or guidance. How these ligand-receptor com-plexes control axon guidance is not known, but signaling is thought to occur throughthe actin cytoskeleton via the Rho family of small GTPases (Song & Poo 2001).

The cast of characters is present in both the pollen tube and the pistil for asimilar mechanism to be operating in pollen tube guidance. The ECM moleculesinvolved are no doubt unique to plants, as no animal ECM homologs were foundin the Arabidopsisgenome (theArabidopsisGenome Initiative 2000); however,many of the pollen tube cytoskeleton components and their associated signalingcomplexes have animal homologs or at least have conserved functional domains(Wu et al. 2001, Hepler et al. 2001).

The components of the stylar-transmitting tract ECM are a complex mixtureand vary depending on the species studied (Wheeler et al. 2001). AGPs (arabino-galactan proteins) are abundant in many styles, and in one case, the TTS proteinoccurs in a glycosylation gradient in the tobacco style. TTS is one of the very fewmolecules described that forms a gradient in vivo and induces chemotropism ofpollen tubes in vitro (Wheeler et al. 2001).

On the pollen tube side, there are pollen-specific glycoproteins in the wallsof several species, the Pex proteins, that have conserved leucine-rich repeats andN-terminal variable regions suggestive of specific ligand binding (Stratford et al.2001). Receptor kinases in tomato pollen tubes that are phosphorylated by con-tact with the stylar matrices are also good candidates for pollen/pistil cross talk(Wheeler et al. 2001). We have very little information about the cell biology ofpollen tubes growing in vivo. The use of GFP as a reporter system allows liveimaging of pollen tubes, and two-photon microscopy should be capable of pen-etrating the pistil tissues, which until now have obscured the pollen tube path invivo, without having to resort to fixations and staining (Cheung 2001).

Pollen tubes adhere to transmitting tract surfaces and appear to be guided alongthem, especially in hollow styles such as those in lily, but also in solid styles asin Arabidopsis(Lord 2000). In lily, two molecules, a pectic polysaccharide and a



peptide, SCA, were found to be necessary for pollen tube adhesion to and growthon an in vitro stylar matrix (Park et al. 2000, Mollet et al. 2000). Both moleculesare secreted into the lily stylar ECM where pollen tubes adhere. In the assay, SCAbinds to the pectin in a pH-dependent manner, and this binding is necessary forpollen tube adhesion to the SCA/pectin matrix. Recombinant SCA was shownto be active in the adhesion assay (Park & Lord 2002). Pollen tubes grown invitro do not adhere to one another. Only when they are in contact with the pistilSCA/pectin matrix in vitro will they adhere to the matrix and to each other andgrow. A mechanism for this adhesion is not known, but the assumption is thatSCA binds a receptor in the pollen tube plasma membrane. Growth, which occursnear the pollen tube tip where plasma membrane may be exposed to the stylarmatrix, is necessary for adhesion. SCA’s activity in chemotropism on the stigma(S. Kim, S.-Y. Park & E.M. Lord, unpublished results), where free diffusion wouldbe necessary to set up a gradient, and its activity in adhesion in the style, wherebinding to pectin is necessary, suggests a dual role for this peptide in the lily pistil.The pollen receptors for SCA in the stigma and in the style may be different. Pollentubes adhere best in the in vitro assay when they are at least two hours old andhave achieved the length they would in vivo as they enter the style, which suggestsdevelopmental regulation of the capacity to respond to the SCA/pectin complex inthe style (Mollet et al. 2000). The effect on the cytoskeleton of pollen tube adhesionto the stylar ECM needs to be studied in vivo and preferably with live pollen tubes.

An exciting new finding about pollen tube guidance concerns the last step inthe process, entrance into the ovule. A variety of genetic studies have revealedthe importance of the ovule and, in particular, the female gametophyte or embryosac in guiding a single pollen tube into the ovule opening or micropyle wherecontact can occur with the embryo sac and sperm can be released (Higashiyamaet al. 2001). In vitro assays using isolated ovules demonstrated clearly that a cueattractive to pollen tubes emanated from the ovule, and laser ablation studies onthe embryo sac show definitively that the cue comes from the embryo sac synergidcells that flank the egg cell. These elegant studies provide experimental supportfor an old proposal that when the pollen tube is in the vicinity of the ovule,the secretory synergid cells, one of which will receive the pollen tube contentsincluding the sperm cells (see below), guide the pollen tube to the micropyle. Thesignal is not yet known, but heat treatment of the ovule prevents it from attractingpollen tubes. This suggests that either active secretion, a protein factor, or both arenecessary for chemotropism. One intriguing aspect of this work is that pollen tubesmust pass through the stigma and style in order to be capable of chemotropism inthe isolated ovule assay (Higashiyama et al. 1998). When pollen is cultured in thepresence of the ovules, no chemotropism to the micropyle occurs even though thegenerative cell has divided and produced sperm cells (T. Higashiyama, personalcommunication). This means that contact with the pistil, starting with the stigmaand then the style, is essential in some way to prepare the pollen tube to receivethe signal from the ovule. This supports the premise that a hierarchy of controlpoints exists in pollination and that study must be done in vivo to decipher thecomplexities of these signaling pathways. Different receptors may form on the


90 LORD ¥ RUSSELL

pollen tube cell at different stages in its passage from stigma to ovary, and thesemay be induced to form by contact with specific ligands in the transmitting tracttissues. The complexities here may rival those in the animal nervous system inaxon guidance.

SELF-INCOMPATIBILITY

The most studied and understood process of pollination in the flowering plants isthat of recognition and rejection of self pollen. Several groups have independentlyevolved unique mechanisms for self-incompatibility (SI) in the angiosperms. Thereare some recent excellent reviews on this topic (Wheeler et al. 2001) so we coveronly the major findings here.

Sporophytic—Brassicaceae

The Brassicaceae (mustard family) has sporophytic SI where the outcome of thepollination is determined by the genotype of the pollen parent (Dixit & Nasrallah2001). The S-gene products in the pistil were discovered some time ago and foundto reside in the stigmatic papillae in this dry stigma type, where the site of self-pollen arrest occurs. The essential component is a membrane-spanning serine-threonine kinase (S receptor kinase; SRK), which has a large extracellular domainin the papillar ECM that was presumed to interact with an S-gene-specific pollenligand. Recently, the long-sought-after male component was discovered and foundto be a small, cysteine-rich protein (SCR) that resides in the pollen coat (Schopferet al. 1999, Shiba et al. 2001). The model now is that interaction between thetransmembrane glycoprotein SRK and the peptide ligand SCR induces a cascade ofsignaling events that lead to rejection of self pollen. Indeed, using tagged proteins,the Nasrallah group and others have shown that SRK and SCR bind to each otherand that the binding is allele specific (Kachroo et al. 2001, Takayama et al. 2001).This work on SI inBrassicais one of the few cases where both the ligand and thereceptor have been identified for a well-known cell-cell signaling event in plants.The discovery of the pollen component in this story is a major advance in the fieldof plant cell biology.

Gametophytic–Papaveraceae and Solanaceae

In the other major type of SI, gametophytic SI, the outcome of pollination isdetermined by the haploid pollen genotype (Wheeler et al. 2001). Advances havebeen made in two families that have evolved quite separate mechanisms for thistype of SI: the Papaveraceae (poppy family) and the Solanaceae (tobacco family).In poppy, small pistil S proteins secreted by the stigma interact with the S-geneproduct in the pollen tube, thought to be a plasma membrane receptor, causing thearrest of incompatible pollen on the stigma (poppy has no style). An in vitro assaysystem has facilitated study of the cell biology of SI, and a fascinating cascadeof events has been described that occur within minutes after contact of pollen



tubes with the stigmatic S-gene protein. Inhibition is mediated by a calcium-basedsignal transduction pathway in the pollen tube that results in fragmentation ofits F-actin during a process of programmed cell death. Downstream of the Ca+2

signals induced by SI, several proteins are phosphorylated in an S-specific manner,and a MAPK is activated (Wheeler et al. 2001). Positive identification of the maleS-gene product would complete the story.

In the Solanaceae, the S-gene product on the female side is an RNase that actson SI pollen tubes in the style by degrading ribosomes (McCubbin & Kao 2000).The male S-gene product is as yet unknown. Pollen tubes endocytose the S-RNase,which is secreted into the stylar ECM, whether they are compatible or not, and invitro the RNase enters and kills both compatible and incompatible pollen. A recentstudy in vivo shows that this uptake also occurs in the style, but only SI pollen tubesare arrested in vivo (Luu et al. 2000). It has been shown that the RNase capacityof the S-gene product in the style is necessary and sufficient to cause SI in theSolanaceae. The model for SI in this family is that the pollen S-gene product is anRNase inhibitor that recognizes and inhibits all S-RNases except those of the sameallelic specificity. The in vivo data are good evidence that the pistil environmentcan profoundly affect pollen tube cell biology. In some as yet unknown mannerthe compatible tube cell can endocytose an RNase that in vitro would arrest itsgrowth, but to which it is immune when growing in vivo. Discovery of the maleS-gene products in both the Papaveraceae and Solanaceae will certainly advancethe field of pollen tube cell biology in general.

POLLINATION ECOLOGY AND EVOLUTION

Mate Choice

Two aspects of pollination ecology that have gained attention lately are pollencompetition and female choice, both components of sexual selection in plants.Mate choice in plants certainly occurs in cases of self-incompatibility, but whethercompatible pollination and fertilization involves a component of female choiceby the pistil is still debated. One mechanism for sexual selection would be pollencompetition, with the fastest pollen tubes achieving the most seed set. Anothermechanism could be choice by the maternal plant, with selection of particularpollen tubes by the pistil, thus causing facilitated tracking to the ovary (Marshall& Diggle 2001). The recent work on rapid evolution of reproductive proteins fromboth the female and male sides supports the concept of sexual selection in plants(Swanson & Vacquier 2002).

It is known that pollen tube growth rates vary in vivo and in vitro and thatthe ability of a pollen donor to sire seeds varies with the maternal plant in self-compatible species. Thus nonrandom mating occurs in plants. It is obvious thatinteractions between pollen tubes of varying genetic background in one pistil, aswell as pollen/pistil-specific interactions, create a complex environment so thatcellular mechanisms that produce nonrandom matings will be difficult to deci-pher. The major lesson for cell biologists from these population studies is clear,


92 LORD ¥ RUSSELL

however; the pollen tube’s function in delivering viable sperm cells to the embryosac involves a close partnership with the transmitting tract of the pistil that mustinclude a myriad of recognition and signaling events.

Pollen Flow

Gene flow in the environment affects genetic structure of populations, a fact ofsome importance to evolutionists, conservationists, and agriculturalists (Ellstrand2001, Schulke & Waser 2001). Pollen-mediated gene movement can be studiedusing paternity analysis. This method has allowed us to study the effects of polli-nator activity, both biotic and abiotic. Recently, due to concerns about gene flowfrom transgenic crop plants, pollen flow has become a hot topic. Earlier concerns,which are ongoing, regard how crop-to-wild-plant gene flow could endanger nativeplants and create new weeds. Basic knowledge of pollination is essential today toadequately address the social and economic problems that are inevitable with newtechnologies, especially those involving food crops.

Evolution

Angiosperms, literally meaning vessel seed, are defined by several features, but theone most defining feature is the pistil or vessel. The carpels that form the pistil arisefrom a floral meristem, and during development some form of epidermal fusionstake place to close the carpels around the ovules. Typically, the carpel fusion eventresults in formation of the transmitting tract of the style with the stigma formed atthe top. Thus the transmitting tract, including the placental epidermis in the ovarywhere the ovules form, is formed from a dermal layer. In primitive angiosperms, thereceptive surface of the sometimes open carpels can be less distinct, with no styleformed. Primitive flowers are typically bisexual (though the extantAmborella, atthe base of all angiosperms, is unisexual), insect pollinated (beetles, flies, and bees),and self compatible, although SI occurs in some taxa (Thien et al. 2000). A morethorough study of the mechanisms of pollination in the primitive angiosperms,including genetic analyses, is needed especially now that we have the results ofthe Deep Green Consortium, the Angiosperm Phylogeny Group (1998).

FERTILIZATION

Male Gamete Lineage Gene Expression

Genetic expression in the male gametophyte has increasingly focused on a dif-ferentiation of cellular programs between the pollen vegetative or tube cell andthe male germ lineage within it. The male gametes themselves are nonmotile, andthus the metabolic needs and control of their position are ceded to the surroundingpollen. Sperm cells move through the pollen tube by association with cytoskeletalcomponents on the perispermatic envelope (inner pollen plasma membrane thatencloses the sperm cells) and cortical arrays of actin filaments and MTs that line



the interior of the pollen tube. Superficial myosin on the vegetative nucleus andperispermatic envelope may have a role in translocating the male germ unit in thepollen tube; the sperm cells, although moving, remain passive. A lower number ofnuclear pores in the sperm (Straatman et al. 2000) would also suggest passivity,but interestingly this lower number does not prevent the male germ cells fromexpressing their own genetic program.

Expression studies of pollen indicate a diversity of wall-producing genes and avariety of housekeeping genes that presumably provide for the metabolic needs ofthe male gametophyte. Genes expressed in the male germ (generative and sperm)cells emphasize the unique nature of these cells. Histones of the generative celllineage are substituted with germ-line-specific gene products ofH2AandH3 (Xuet al. 1999a). The vegetative nucleus, in contrast, displays decreased quantities ofH1 protein (Tanaka et al. 1998), hypoacetylation of H4, and increased methyla-tion, especially prior to germination (Janousek et al. 2000), presumably related toheterochromatin formation. Nuclei of the male germ lineage become increasinglydistinct from the vegetative nucleus. Differences in chromatin condensation dis-tinguish the vegetative and germ line nuclei, presumably reflecting commitment toa reproductive program. Differentiation of expression programs is provided by oc-currence of germ-line-specific promoters, for instance theLGC1promoter, whichis apparently activated only after the separation of the cell from the intine (Xu et al.1999b). The actual divergence in the developmental program, however, appearsto coincide with the eccentric placement of the cell wall during generative cellformation (Eady et al. 1995).

Additional germ-line-specific expression is evidenced in the activation of cyclin,DNA repair, and ubiquitin genes in the generative and sperm cells (Xu et al. 1998,Singh et al. 2002). In addition to germ-line-unique polyubiquitins, there is alsoRT-PCR and in situ hybridization evidence of differences in the two sperm cellsof Plumbago(Singh et al. 2002). These specific ubiquitin genes are turned on ineach of these cells in succession prior to their final differentiation, presumablypurging remaining cytoplasmic signals. If there is a specific surface protein orglycoprotein displayed on the surface of the cell, it is likely to be formed in suchcells or under their control. Further evidence for an independent developmentalprogram in sperm cells is the presence of specific arabinogalactan proteins on thesperm cell surface (Southworth & Kwiatkowski 1996) and heterogeneity of cellsurface determinants in different flowering plant species (Southworth et al. 1999).Nonrandom positioning of B chromosomes in the nuclei of maize sperm cells alsopoints to independent developmental control (Rusche et al. 2001).

In Vivo Studies of Fertilization in Flowering Plants

The past several years have seen a resurgence of in vivo approaches using theremarkable plantTorenia fournieri(Scrophulariaceae), in which the embryo sacprotrudes from the tip of the ovule (Erdelská 1974) exposing the egg appara-tus. Higashiyama et al. (1997) recorded the kinetics of fertilization by direct


94 LORD ¥ RUSSELL

observation of the two synergids, egg cell and central cell (which form the so-called female germ unit). One item of contention given the apparent participationof sperm-associated myosin with a pre-existing actin corona is how long the un-fused sperm cells remain in the synergid. According to Higashiyama et al. (1997),both sperm cells are present only transiently inside the degenerated synergid, butWallwork & Sedgley (2000) report durations in excess of an hour. Reconcilingthese observations is problematical. The rapid kinetics of nuclear changes may re-flect decay of the synergid and vegetative nuclei in the degenerated synergid afterpollen tube discharge. The release of enzymes, potentially including nucleases,may be a consequence of synergid penetration that may protect the interior of theembryo sac from transmission of foreign, including viral, DNA (Heslop-Harrisonet al. 1999).

Nuclear migration within the central cell precedes fertilization. Initially, thepolar fusion nucleus occupies a mid-chalazal position, migrating toward the eggapparatus just before pollen tube arrival. After the sperm nucleus enters the centralcell, it adheres to the polar fusion nucleus, fuses, and migrates to a far chalazalposition where division of the primary endosperm nucleus occurs at 15 h afterpollination. Consistent with prior microcinematographic observations, the spermnucleolus enlarges to the size of the polar nucleoli and fuses during migration(Higashiyama et al. 1997). Migration of the polar fusion nucleus will also occurwithout pollination, but it is delayed 24 h.

Symplastic communication between the cells of the female germ unit changesdramatically during embryo sac maturation and early embryogenesis as well. Oneday prior to anthesis, rhodamine-labeled dextran of 10 kDa passes freely betweenthe central cell and egg apparatus cells. At anthesis,∼40% of ovules still permitpassage of the 10-kDa tracer, but near fertilization, the exclusion barrier dropsbelow 3 kDa in 80% of the ovules (12 h after pollination) and 90% in the zygote(24 h after pollination). Without pollination, the exclusion barrier drops to 3 kDain nearly half of the ovules at two days after anthesis (Han et al. 2000). Growthof the integument and multiplication of endosperm nuclei unfortunately inhibit amore direct observation of fertilization.

CELL CYCLE AND FERTILIZATION The role of the cell cycle during fertilizationhas recently gained renewed interest. Animal fertilization, with rare exceptions,involves gametes with a 1C complement of DNA corresponding to the G1 phaseand without intervening DNA synthesis. Early results in angiosperms, however,provided a baffling variety of data and DNA complements. Recent evidence sug-gests three major patterns of pollen dissemination in the bicellular (generativecell–containing) and tricellular (sperm cell–containing) pollen of flowering plants:1C, 2C, and intermediate values (Friedman 1999). Furthermore, fusion occurs intwo conditions: 1C (G1 fusion) or 2C (G2 fusion). Only inArabidopsisand thegnetophytes have data on both the egg cell and sperm cell been followed throughdevelopment to fusion, although these may be inferred in others by zygotic mea-surements (Friedman 1999).



Phase synchrony appears to be an important factor governing gamete fusion.Gametes that are in S phase, even if appressed to one another, will presumably notfuse until both reach a 2C complement, indicative of G2. In tobacco, 1C sperm in thestyle remain at a G1 until arriving near the ovule. Only when the sperm completesS phase in the degenerate synergid and enters G2 does fusion occur (H.Q. Tian,T. Yuan & S.D. Russell, unpublished data). Interestingly, progression in the cellcycle in the male and female gametes occurs more or less in concert, as femalegametes require an extra 24 h to reach G2 without pollination. Complex commu-nication between the male and female gametes is also evidenced by receptivityrelated responses in maize egg cells (Mol et al. 2000) and tobacco female germcells (Tian & Russell 1997a) in response to the presence of pollen tubes in the style.

ATTRACTION AND DELIVERY OF MALE GAMETES WITHIN THE EMBRYO SAC Evi-dence for pollen tube attraction has been an elusive target until recently, and effortsto demonstrate it using living male and female gametophytes have been inconclu-sive. Pollen tubes inArabidopsiswere attracted over distances of little more than100µm to cross the septum and fertilize the ovule (Hülskamp et al. 1995). Noattraction was observed in incompletely formed ovules or those lacking embryosacs. These, and observations with other mutants (Ray et al. 1997), suggest thatembryo sac cells may themselves govern pollen tube guidance.

A direct demonstration of this (see above) required use of a remarkable mem-ber of the Scrophulariaceae family,Torenia fournieri. A semi in vitro system ofculturing mature ovules and pollen tubes was painstakingly developed so thatfertilization could be directly observed in some ovules (Higashiyama et al. 1998).

Higashiyama et al. (2000), using only light microscopy, described the pollentube entering the embryo sac at the micropylar end “thrusting its way betweenthe two synergid cells.” Although an exact mode of entry will presumably revealthat the pollen tube enters one synergid directly (van der Pluijm 1964), ruptureclearly occurs at the pollen tube tip, with the discharged contents of the pollentube restricted to the degenerated synergid (Figure 2). When the contents of thepollen tube were discharged into the synergid, they arrived at a peak velocity es-timated to be 12,000± 5800µm3 s−1, decreasing rapidly during the first 0.1 s(Higashiyama et al. 2000). A remarkable video of this event is available as supple-mentary data on the journal web site (http://www.plantphysiol.org/cgi/content/full/122/1/11/DC1).

MIGRATION OF MALE GAMETES WITHIN THE EMBRYO SAC Upon the discharge ofthe pollen tube within the embryo sac, pollen tube cytoskeletal arrays are rapidlydepolymerized, the myosin-coated perispermatic envelope disintegrates, and thesperm are typically immersed in the micropylar end of the degenerate synergid.Although neither sperm cell is typically in contact with the egg or central cell atthis stage, the male gametes nonetheless are conveyed chalazally to their respec-tive fusion site. Male gametes often are in contact with both the egg and centralcell immediately prior to fusion (Russell 1992). The means of this movement—a


96 LORD ¥ RUSSELL

nonmotile cell traversing the lifeless cytoplasm of two former cells—defies con-ventional mechanisms, but both chemical elements of an actomyosin movementsystem are present. Actin becomes organized as a corona, tracing the anticipatedpathway of gamete travel. Coronas extend from the middle of the synergid, oneterminating near the egg nucleus and the other near the polar nuclei (Huang &Russell 1994, Huang & Sheridan 1998, Huang et al. 1999, Fu et al. 2000, Yeet al. 2002). Myosin released from the pollen tube binds with the surface of thesperm cells (Zhang et al. 1999), presumably through electrostatic force (Zhang &Russell 1999). Microinjected Alexa 488-phalloidin in living embryo sacs ofTore-nia reveals a dramatic reorganization of F-actin in the synergid and egg cells duringpollen tube approach, arrival, and discharge. An actin cap is located at the base ofthe synergid, and with degeneration, fluorescent label accumulates in patches onthe cortex of the egg cell. An actin corona becomes conspicuous upon pollen tubearrival and discharge, when sperm cells are deposited near the base of the corona.Sperm cells in the synergid subsequently migrate chalazally toward the point offusion. The actin corona disappears soon after male gamete transmission (Figure3). That actin coronas have a role in fertilization is indicated by the associationof actin coronas with each fertilized egg in maize plants with supernumerary fe-male gametes (Huang & Sheridan 1998). Recent work on the nun orchid providesevidence that living egg and central cells control these events; cellular indenta-tions complementary in shape to the sperm cells are presented as target sites nearthe termination of coronas (Ye et al. 2002). In tobacco, nun orchid, and possiblyTorenia, transit in the synergid appears to be a protracted process.

In Vitro Studies of Fertilization in Flowering Plants

Isolation of living, viable gametes late in the 1980s led to the combination of maleand female gametes using electrofusion (Kranz et al. 1991), chemical induction(Kranz & Lorz 1994a), and more subtle means (Faure et al. 1994). Ultimately,some products resulted in the regeneration of plants (Kranz & Lörz 1994b) andformation of in vitro endosperm (Kranz et al. 1998). These studies have focusedon a small group of monocots that includes grasses such asZea, Coix, Sorghum,Hordeum, andTriticum (reviewed previously by Kranz & Kumlehn 1999).

ADHESION AND CHARACTERISTICS OF IN VITRO FUSION In electrofusion andchemical induction, the combination of gametes is coerced; however, in an ex-ample by Faure et al. (1994), the medium, which included low concentrations ofcalcium, did not appear to coerce fusion. Polyspermy was apparently successfullyblocked, and chemical signals, including a calcium influx, have been recorded(see below). Attempts to combine dicot gametes have met with some success inNicotiana tabacum. (Sun et al. 1995, 2000, 2001; Tian & Russell 1997b). In eachof these cases, fusion has been stimulated using chemical means with polyethy-lene glycol (PEG). Despite the fusigenic nature of PEG, however, multiple fusionsseemed to be inhibited within 30 min after fusion. This inhibition of fusion applied



to various combinations of gametes, somatic cells, and gametic-somatic combi-nations, as well. This refractory period is attributed to a requirement of the cellsto complete cytoplasmic reorganization prior to further fusion (Sun et al. 2000),casting doubts on claims of polyspermy block. The greater apparent efficiency ofgamete-to-gamete fusion has also been challenged. Recently, Sun et al. (2001),fusing somatic cells (and even chloroplasts) with cells of various sizes, found thatefficiency of fusion in each size class did not appear to differ with that foundin male gamete lineage cells of similar size. Disparities in volume could also bea mechanism for preferentiality during fertilization when sperm cells differ involume (Tian et al. 2001).

Combination of distant hybrids and genetic modification of the artificial zygoteshave been a frequently cited goal of in vitro fertilization. This goal may not bedistant, as Scholten & Kranz (2001) have now demonstrated the expression oftransgenes in gametes and in vitro fertilized zygotes. With the use of germ-line-specific promoters (for instance, theLCG1promoter; Xu et al. 1999b), alterationscould be rapidly incorporated in the germ line for transmission or reimplantationin host plants to produce seed.

GAMETE MATURATION, POLARITY DETERMINATION AND REGENERATION Gametematuration and the competence of gametes to respond appropriately to gametesof the opposite sex are incompletely known in angiosperms. Maturation of thesperm cell surface can be demonstrated by attempting to induce fusion of spermcells during maturation (Tian & Russell 1998). When placed in apposition, newlyformed sperm cells of tobacco often fuse spontaneously, but within minutes theylose this tendency. At 20 h after sperm formation, fusion could be induced onlyby exposing cells to dilute solutions of cellulase and pectinase (interestingly bothwere required). Apparently, sperm maturation involves important carbohydratedeterminants on the cell surface, but not a wall in any conventional sense becausethis would impede fusion. Polarity of the male germ unit appears to be initiatedprior to generative cell separation from the intine and determined by the positionof the vegetative nucleus (Russell et al. 1996).

For female gametes, the best understood model is maize, in which egg matura-tion is categorized into three classes (Mol et al. 2000). Type A egg cells are small,densely cytoplasmic and have a central nucleus, suggestive of a nonpolar condi-tion. Type B egg cells are larger and have numerous small vacuoles surrounding anearly centrally located perinuclear cytoplasm. Type C cells are highly polarized,containing the largest cells, a prominent apical vacuole, and a mid-basal perinu-clear cytoplasm. As silks emerge from their husks, type A egg cells are prevalentin the ovules (88%), but they decrease to nearly half (58%) at optimum silk length,and types B and C become more common near the time of fertilization.

Cross talk from pollen seems to stimulate maturation of female gametes in waysthat include calcium accumulation (Tian & Russell 1997a), cell organization (Molet al. 2000), nuclear migration (Higashiyama et al. 1997), and cell cycle progression(H.Q. Tian, T. Yuan, S.D. Russell unpublished data). Conversely, signals from the


98 LORD ¥ RUSSELL

embryo sac serve to stimulate pollen tube growth and maturation. However, thenature of this communication is unknown.

In many biological systems, the polarity of the zygote can be altered by illumi-nation source or site where gamete fusion occurred and can be induced by factorsas subtle as electrical charge, touch, or changes in membrane permeability. In an-giosperms, however, there is no evidence that such manipulations have succeededin reorienting or establishing polarity, nor is there evidence of a particular regionof the egg cell being more receptive to fusion than another (Antoine et al. 2001).If egg cells possess inborn polarity signals, their placement presumably occurs inreference with pre-cellular stages of embryo sac development. Normally, egg cellsoccur within the embryo sac, which presumably provides a structural context forthe determination of the egg’s cellular polarity.

CALCIUM WAVES, SITE OF FUSION AND EGG ACTIVATION As in animal fertiliza-tion, a calcium influx is associated with in vitro fertilization in plants. Usingmaize, Antoine et al. (2000) demonstrate influx rates peaking at more than10 pmol· cm−2 · s−1 that result in a homogeneous influx over the surface of the cellthat may continue up to 2 h. Although the calcium influx is initiated at the fusionsite and propagated in a wave-like progression, unlike that in some other models,polarity is not evident past the first several minutes after fusion. The most visiblemanifestation of egg activation is a contraction of the egg cell/zygote, causing awrinkling of the cell surface that apparently accompanies calcium influx (Figure 4).This is followed by resumption of a smooth cell surface with turgid appearance andthe subsequent deposition of cell wall as detected by Calcofluor white. Artificialinflux induction can be triggered by 20µM A23187 ionophore or 5µM ionomycin,resulting in calcium influx and egg events that mimic activation. In contrast, influxcan be inhibited by the addition of gadolinium at a functional concentration of10µM GdCl3, thereby inhibiting cell wall formation.

POST-FERTILIZATION EVENTS

Nuclear Migration and Fusion in Angiosperms

The matter of nuclear migration once the zygote has fused is still an understudiedphenomenon of zygote activation. Cytoskeletal elements such as microtubulesand F-actin will undoubtedly play a role in the positioning and movement of themale and female nuclei. Studies to date, however, seem to suggest that rather thanextensive cytoskeletal rebuilding immediately prior to nuclear fusion, there is aphase of cytoplasmic re-establishment coinciding with the establishment of thecell wall (Antoine et al. 2001). Only with greater cellular stability are the usualmechanisms of nuclear positioning conspicuous.

Endosperm Initiation

Endosperm forms a determinate, polarized structure with considerable complexityand a distinct developmental program, as demonstrated by the fate of cultured



Figure 4 The chart represents Ca2+ flux measurements during maize in vitro fer-tilization (time 0 is gametic fusion). Arrow indicates the onset of a Ca2+ influx afterfusion. Figures illustrate (A) the egg cell before fusion (male gamete position is shownby a white arrowhead); (B) egg cell after fusion, after cell contraction has started;(C) strong egg cell contraction; and (D) post-fusion egg cell reshaping. Figure fromAntoine et al. (2000) reproduced with permission.

endosperm derived from in vitro fertilization of the central cell (Kranz et al. 1998).The uniqueness of cell origins and architecture in free-nuclear endosperm hasbeen recognized just recently to consist of nuclear cytoplasmic domains (Brownet al. 1999) that form compartments without a conventional cell wall (Ortegui &Staehelin 2000) in a highly atypical manner compared with the other generations.In the scheme of alternation of generations, endosperm truly does form a thirdgeneration of flowering plants.

EVOLUTION OF DOUBLE FERTILIZATION

Apparently, angiosperm predecessors may have produced female gametophyteswith more than one fertilizable nucleus. Presumably, each sperm fused with afemale target nucleus. The genetic benefit of the second nutritive fertilizationincreasingly outweighed the terminal nature of the endosperm lineage sufficientlythat the central cell no longer competes with the egg for survival in modern plants.Instead, endosperm development precedes the zygote, providing an aggressivelynutritive tissue that assures the success of its sibling embryo (Friedman 1995).


100 LORD ¥ RUSSELL

Based on examinations of basal angiosperms, the following seem to be ple-siomorphic characteristics of endosperm: (a) initial division of the primary en-dosperm nucleus to form two cells; (b) polar development of micropylar andchalazal domains with different patterns of development (initially uniseriate de-velopment of the micropylar domain and more random cellular development ofthe chalazal domain), with the micropylar domain being the larger of the two; and(c) copious nutritive development, with proteins and lipids of endosperm formingthe primary food source for the embryo (Floyd & Friedman 2000). In fact, earlyendosperm lineages can sometimes display strikingly similar patterns in polarityand organization compared with early embryo lineages.

CONCLUSIONS AND PROSPECTS

In comparison with animals, sexual reproduction in seed plants appears far morecomplex, subtle, and painstakingly accurate in the production and delivery ofgametes. Reproductive interactions in flowering plants are largely restricted tothose between the pistil (stigma, style, and ovary, which are actually sporophytetissues) and pollen tubes, which are a separate three-celled male gametophyticstage of the life cycle. Just before fertilization, the female gametophyte exerts aneffect on the pollen tube, guiding it to the micropyle. These differences make themechanism of pollination and fertilization in plants a fundamentally interestingproblem in cell biology that should stimulate research for many years to come.

The Annual Review of Cell and Developmental Biologyis online athttp://cellbio.annualreviews.org

LITERATURE CITED

Angiosperm Phylogeny Group. 1998. An ordi-nal classification for the families of floweringplants.Ann. MO. Bot. Gard.85:531–53

Antoine AF, Dumas C, Faure JE, Feijo JA,Rougier M. 2001. Egg activation in floweringplants.Sex. Plant Reprod.14:21–26

Antoine AF, Faure JE, Cordeiro S, Dumas C,Rougier M, et al. 2000. A calcium influxis triggered and propagates in the zygote asa wavefront during in vitro fertilization offlowering plants.Proc. Natl. Acad. Sci. USA97:10643–48

ArabidopsisGenome Initiative. 2000. Analy-sis of the genome sequence of the floweringplantArabidopsis thaliana. Nature408:796–813

Bih FY, Wu SSH, Ratnayayake C, Walling LL,

Nothnagel EA, et al. 1999. The predominantprotein on the surface of maize pollen isan endoxylanase synthesized by a tapetummRNA with a long 5′ leader.J. Biol. Chem.274:22884–94

Borisy GG, Svitkina TM. 2000. Actin machin-ery: pushing the envelope.Curr. Opin. CellBiol. 12:104–12

Brown RC, Lemmon BE, Nguyen H, Olsen OA.1999. Development of endosperm inAra-bidopsis thaliana. Sex. Plant Reprod.12:32–42

Cai G, Romagnoli S, Cresti M. 2001. Micro-tubule motor proteins and the organizationof the pollen tube cytoplasm.Sex. Plant Re-prod.14:27–34

Chen Y-F, Matsubayashi Y, Sakagami Y. 2000.



Peptide growth factor phytosulfokine-α con-tributes to the pollen population effect.Planta211:752–55

Cheung AY. 2001. Imaging elongating pollentubes by green fluorescent protein.Sex. PlantReprod.14:9–14

Dickinson HG, Elleman CJ, Doughty J. 2000.Pollen coatings—chimaeric genetics andnew functions.Sex. Plant Reprod.12:302–9

Dixit R, Nasrallah JB. 2001. Recognizing selfin the self-incompatibility response.PlantPhysiol.125:105–8

Dixit R, Rizzo C, Nasrallah M, Nasrallah J.2001. TheBrassica MIP-MODgene encodesa functional water channel that is expressed inthe stigma epidermis.Plant Mol. Biol.45:51–62

Doughty J, Wong HY, Dickinson HG. 2000.Cysteine-rich pollen coat proteins (PCPs)and their interactions with stigmatic S (in-compatibility) and S-related proteins inBras-sica: putative roles in SI and pollination.Ann.Bot.85:161–69

Eady C, Lindsey K, Twell D. 1995. The signif-icance of microspore division and divisionsymmetry for vegetative cell-specific tran-scription and generative cell differentiation.Plant Cell7:65–74

Ellstrand NC. 2001. When transgenes wander,should we worry?Plant Physiol.125:1543–45

Erdelska O. 1974. Contribution to the study offertilization in the living embryo sac. InFer-tilization in Higher Plants, ed. HF Linskens,pp. 191–95. Amsterdam: North Holland

Faure JE, Digonnet C, Dumas C. 1994. An invitro system for adhesion and fusion of maizegametes.Science263:1598–600

Floyd SK, Friedman WE. 2000. Evolution ofendosperm developmental patterns amongbasal flowering plants.Int. J. Plant Sci.161:S57–81

Friedman WE. 1995. Organismal duplication,inclusive fitness theory, and altruism: under-standing the evolution of endosperm and theangiosperm reproductive syndrome.Proc.Natl. Acad. Sci. USA92:3913–17

Friedman WE. 1999. Expression of the cell cy-cle in sperm ofArabidopsis: implications forunderstanding patterns of gametogenesis andfertilization in plants and other eukaryotes.Development126:1065–75

Fu Y, Wu G, Yang Z. 2001. Rop GTPase-depen-dent dynamics of tip-localized F-actin con-trols tip growth in pollen tubes.J. Cell Biol.152:1019–32

Fu Y, Yang Z. 2001. Rop GTPase: a masterswitch of cell polarity development in plants.Trends Plant Sci.6:545–47

Fu Y, Yuan M, Huang BQ, Yang HY, Zee SY,O’Brien TP. 2000. Changes in actin organi-zation in the living egg apparatus ofToreniafournieri during fertilization.Sex. Plant Re-prod.12:315–22

Han YZ, Huang BQ, Zee SY, Yuan M. 2000.Symplastic communication between the cen-tral cell and the egg apparatus cells in theembryo sac ofTorenia fournieriLind beforeand during fertilization.Planta211:158–62

Heberle-Bors E, Voronin V, Touraev A, SanchezTestillano P, Risueno MC, et al. 2001. MAPkinase signaling during pollen development.Sex. Plant Reprod.14:15–19

Hepler PK, Vidali L, Cheung AY. 2001. Polar-ized cell growth in higher plants.Annu. Rev.Cell Dev. Biol.17:159–87

Heslop-Harrison J, Heslop-Harrison JS, Hes-lop-Harrison Y. 1999. The structure and pro-phylactic role of the angiosperm embryo sacand its associated tissues:Zea maysas amodel.Protoplasma209:256–72

Higashiyama T, Kuroiwa H, Kawano S,Kuroiwa T. 1997. Kinetics of double fertili-zation in Torenia fournieribased on directobservations of the naked embryo sac.Planta203:101–10

Higashiyama T, Kuroiwa H, Kawano S,Kuroiwa T. 1998. Guidance in vitro of thepollen tube to the naked embryo sac ofTore-nia fournieri. Plant Cell10:2019–31

Higashiyama T, Kuroiwa H, Kawano S,Kuroiwa T. 2000. Explosive discharge ofpollen tube contents inTorenia fournieri.Plant Physiol.122:11–13

Higashiyama T, Yabe S, Sasaki N, Nishimura


102 LORD ¥ RUSSELL

Y, Miyagishima SY, et al. 2001. Pollen tubeattraction by the synergid cell.Science293:1480–83

Higgs HN, Pollard TD. 2001. Regulation ofactin filament network formation throughARP2/3 complex: activation by a diverse ar-ray of proteins.Annu. Rev. Biochem.70:649–76

Huang BQ, Fu Y, Zee SY, Hepler PK. 1999.Three-dimensional organization and dyna-mic changes of the actin cytoskeleton in em-bryo sacs ofZea maysandTorenia fournieri.Protoplasma209:105–19

Huang BQ, Russell SD. 1994. Fertilization inNicotiana tabacum. Cytoskeletal modifica-tions in the embryo sac during synergid de-generation. A hypothesis for short-distancetransport of sperm cells prior to gamete fu-sion.Planta194:200–14

Huang BQ, Sheridan WF. 1998. Actin coronasin normal andindeterminate gametophyte lembryo sacs of maize.Sex. Plant Reprod.11:257–64

Hulskamp M, Schneitz K, Pruitt RE. 1995. Ge-netic evidence for a long-range activity thatdirects pollen tube guidance inArabidopsis.Plant Cell7:57–64

Janousek B, Zluvova J, Vyskot B. 2000. His-tone H4 acetylation and DNA methylationdynamics during pollen development.Proto-plasma211:116–22

Joos U, van Aken J, Kristen U. 1994. Micro-tubules are involved in maintaining the cel-lular polarity in pollen tubes ofNicotianasylvestris. Protoplasma179:5–15

Kachroo A, Schopfer CR, Nasrallah ME, Nas-rallah JB. 2001. Allele-specific receptor-ligand interactions inBrassicaself-incompa-tibility. Science293:1824–26

Klahre U, Chua N-H. 1999. TheArabidop-sis ACTIN-RELATED PROTEIN 2(AtARP2)promoter directs expression in xylem precur-sor cells and pollen.Plant Mol. Biol.41:65–73

Kohorn BD. 2001. WAKs; cell wall associatedkinases.Curr. Opin. Cell Biol.13:529–33

Kranz E, Bautor J, Lörz H. 1991. In vitro fertil-ization of single, isolated gametes of maize

mediated by electrofusion.Sex. Plant Re-prod.4:12–16

Kranz E, Kumlehn J. 1999. Angiosperm fertil-ization, embryo and endosperm developmentin vitro. Plant Sci.142:183–97

Kranz E, Lorz H. 1994a. In vitro fertilizationin maize by single egg and sperm cell pro-toplast fusion mediated by high calcium andhigh pH.Zygote2:125–28

Kranz E, Lorz H. 1994b. In vitro fertilizationwith isolated single gametes results in zy-gotic embryogenesis and fertile maize plants.Plant Cell5:739–46

Kranz E, Von Wiegen P, Quader H, Lörz H.1998. Endosperm development after fusionof isolated, single maize sperm and centralcells in vitro.Plant Cell10:511–24

Li H, Lin Y, Heath RM, Zhu MX, Yang Z.1999. Control of pollen tube tip growth bya Rop GTPase-dependent pathway that leadsto the tip-localized calcium influx.Plant Cell11:1731–42

Lord EM. 2000. Adhesion and cell move-ment during pollination: cherchez la femme.Trends Plant Sci.5:68–73

Luu D-T, Marty-Mazars D, Trick M, Dumas C,et al. 1999. Pollen-stigma adhesion inBras-sicaspp. involves SLG and SLR-1 glycopro-teins.Plant Cell11:251–62

Luu D-T, Qin X, Morse D, Cappadocia M.2000. S-RNase uptake by compatible pollentubes in gametophytic self-incompatibility.Nature407:649–51

Maheshwari P. 1950.An Introduction to theEmbryology of Angiosperms.New York:McGraw-Hill

Marshall DL, Diggle PK. 2001. Mechanismsof differential pollen donor performancein wild radish, Raphanus sativus(Brassi-caceae).Am. J. Bot.88:242–57

Matsubayashi Y, Yang H, Sakagami Y. 2001.Peptide signals and their receptors in higherplants.Trends Plant Sci.6:573–77

Mayfield JA, Fiebig A, Johnstone SE, PreussD. 2001. Gene families from theArabidop-sis thalianapollen coat proteome.Science292:2482–85

McCubbin AG, Kao T-H. 2000. Molecular



recognition and response in pollen and pis-til interactions.Annu. Rev. Cell Dev. Biol.16:333–64

Mol R, Idzikowska K, Dumas C, Matthys-Rochon E. 2000. Late steps of egg cell dif-ferentiation are accelerated by pollination inZea maysL. Planta210:749–57

Mollet J-C, Park S-Y, Nothnagel EA, Lord EM.2000. A lily stylar pectin is necessary forpollen tube adhesion to an in vitro stylar ma-trix. Plant Cell12:1737–49

Moutinho A, Hussey PJ, Trewavas AJ, MalhoR. 2001. cAMP acts as a second messengerin pollen tube growth and reorientation.Proc.Natl. Acad. Sci. USA98:10481–86

Olson JH, Xiang X, Ziegert T, Kittelson A,Rawls A, et al. 2001. Allurin, a 21-kDa spermchemoattractant fromXenopusegg jelly, isrelated to mammalian sperm-binding pro-teins.Proc. Natl. Acad. Sci. USA98:11205–10

Ortegui M, Staehelin LA. 2000. Syncytial-typecell plates: a novel kind of cell plate in-volved in endosperm characterization.PlantCell 12:933–47

Palanivelu R, Preuss D. 2000. Pollen tube tar-geting and axon guidance: parallels in tipgrowth mechanisms.Trends Cell Biol.10:517–24

Park AR, Cho SK, Yun UJ, Jin MY, Lee SH,et al. 2001. Interaction of theArabidopsisre-ceptor protein kinase Wak1 with a glycine-rich protein, AtGRP-3.J. Biol. Chem.276:26688–93

Park S-Y, Jauh G-Y, Mollet J-C, Eckard KJ,Nothnagel EA, et al. 2000. A lipid transfer-like protein is necessary for lily pollen tubeadhesion to an in vitro stylar matrix.PlantCell 12:151–63

Park S-Y, Lord EM. 2002. Expression studiesof SCA in lily and confirmation of its rolein pollen tube adhesion.Plant Mol. Biol. Inpress

Parton RM, Fischer-Parton S, Watahiki MK,Trewavas AJ. 2001. Dynamics of the apicalvesicle accumulation and the rate of growthare related in individual pollen tubes.J. CellSci.114:2685–95

Ray S, Park SS, Ray A. 1997. Pollen tube guid-ance by the female gametophyte.Develop-ment124:2489–98

Rusche ML, Mogensen HL, Chaboud A, FaureJE, Rougier M, et al. 2000. B chromosomesof maize (Zea maysL.) are positioned non-randomly within sperm nuclei.Sex. Plant Re-prod.13:231–34

Russell SD. 1992. Double fertilization.Int. Rev.Cytol.140:357–88

Russell SD, Strout GW, Stramski AK, MislanTW, Thompson RA, et al. 1996. Develop-ment polarization and morphogenesis of thegenerative and sperm cells ofPlumbago zey-lanica. 1. Descriptive cytology and three-dimensional organization.Am. J. Bot.83:1435–53

Scholten S, Kranz E. 2001. In vitro fertilizationand expression of transgenes in gametes andzygotes.Sex. Plant Reprod.14:35–40

Schopfer CR, Nasrallah ME, Nasrallah JB.1999. The male determinant of self-incompa-tibility in Brassica. Science286:1697–700

Schulke B, Waser NM. 2001. Long-distancepollinator flights and pollen dispersal be-tween populations ofDelphinium nuttallia-num. Oecologia127:239–45

Shiba H, Takayama S, Iwano M, Shimosato H,Funato M, et al. 2001. A pollen coat protein,SP11/SCR, determines the pollenS-specifi-city in the self-incompatibility ofBrassicaspecies.Plant Physiol.125:2095–103

Singh MB, Xu H, Bhalla P, Zhang Z, SwobodaI, et al. 2002. Developmental expression ofpolyubiquitin genes and distribution of ubiq-uitinated proteins in generative and spermcells.Sex. Plant Reprod.In press

Song H-J, Poo M-M. 2001. The cell biologyof neuronal navigation.Nat. Cell Biol.E81–88

Southworth D, Kwiatkowski S. 1996. Arabino-galactan proteins at the cell surface ofBras-sicasperm andLilium sperm and generativecells.Sex. Plant Reprod.9:269–72

Southworth D, Kwiatkowski S, Smith AR,Sharpless H, Merwin J, et al. 1999. Antibod-ies to flowering-plant sperm.Protoplasma208:115–22


104 LORD ¥ RUSSELL

Stanley RG, Linskens HF. 1974.Pollen: Bio-logy, Biochemistry, and Management.Berlin/Heidelberg: Springer-Verlag

Straatman KR, Nijsse J, Kieft H, Van Aelst AC,Schel JHN. 2000. Nuclear pore dynamicsduring pollen development and androgenesisin Brassica napus. Sex. Plant Reprod.13:43–51

Stratford S, Barnes W, Hohorst DL, Sagert DL,Cotter R, et al. 2001. A leucine-rich repeatregion is conserved in pollen extensin-like(Pex) proteins in monocots and dicots.PlantMol. Biol. 46:43–56

Sun MX, Moscatelli A, Yang HY, Cresti M.2000. In vitro double fertilization inNico-tiana tabacum (L.): polygamy comparedwith selected single pair somatic protoplastand chloroplast fusions.Sex. Plant Reprod.13:113–17

Sun MX, Moscatelli A, Yang HY, Cresti M.2001. In vitro double fertilization inNico-tiana tabacum(L.): the role of cell volumein cell division.Sex. Plant Reprod.13:225–29

Sun MX, Yang HY, Zhou C. 1995. Single-pairfusion of various combinations between fe-male gametoplasts and other protoplasts inNicotiana tabacum. Acta Bot. Sin.36:489–93

Swanson WJ, Vacquier VD. 2002. The rapidevolution of reproductive proteins.Nat.Genet.3:137–44

Takayama S, Shiba H, Iwano M, Asano K, HaraM, et al. 2000. Isolation and characterizationof pollen coat proteins ofBrassica campes-tris that interact with S locus-related glyco-protein 1 involved in pollen stigma adhesion.Proc. Natl. Acad. Sci. USA97:3765–70

Takayama S, Shimosato H, Shiba H, Funato M,Che F-S, et al. 2001. Direct ligand-receptorcomplex interaction controlsBrassicaself-incompatibility.Nature413:534–38

Tanaka I, Ono K, Fukuda T. 1998. The develop-mental fate of angiosperm pollen is asso-ciated with a preferential decrease in thelevel of histone H1 in the vegetative nucleus.Planta206:561–69

Taylor LP, Hepler PK. 1997. Pollen germination

and tube growth.Annu. Rev. Plant Physiol.Plant Mol. Biol.48:461–91

Thien LB, Azuma H, Kawano S. 2000. New per-spectives on the pollination biology of basalangiosperms.Int. J. Plant Sci.161:S225–35

Tian H-Q, Russell SD. 1997a. Calcium distri-bution in fertilized ovules and embryo sacsof Nicotiana tabacumL. Planta202:93–105

Tian H-Q, Russell SD. 1997b. Micromanipula-tion of male and female gametes ofNicotianatabacum. II. Preliminary attempts for in vitrofertilization and egg cell culture.Plant CellRep.16:657–61

Tian HQ, Russell SD. 1998. The fusion of spermcells and the function of male germ unit(MGU) of tobacco (Nicotiana tabacumL).Sex. Plant Reprod.11:171–76

Tian HQ, Zhang ZJ, Russell SD. 2001. Spermdimorphism inNicotiana tabacumL. Sex.Plant Reprod.14:123–25

van der Pluijm JE. 1964. An electron micro-scopic investigation of the filiform apparatusin the embryo sac ofTorenia fournieri.In Pollen Physiology and Fertilization, ed.HF Linskens, pp. 8–16. Amsterdam: North-Holland

Wallwork MAB, Sedgley M. 2000. Early eventsin the penetration of the embryo sac inTore-nia fournieri (Lind.). Ann. Bot.85:447–54

Wheeler MJ, Franklin-Tong VE, Franklin FCH.2001. The molecular and genetic basis ofpollen-pistil interactions.New Phytol.151:565–84

Wolters-Arts M, Lush WM, Mariani C. 1998.Lipids are required for directional pollen tubegrowth.Nature392:818–21

Wu G, Gu Y, Li S, Yang Z. 2001. A genome-wide analysis of Arabidopsis Rop-inter-active CRIB motif-containing proteins thatact as Rop GTPase targets.Plant Cell 13:2841–56

Xu H, Swoboda I, Bhalla PL, Sijbers AM, ZhaoC, et al. 1998. Plant homologue of human ex-cision repair geneERCC1points to conser-vation of DNA repair mechanisms.Plant J.13:823–29

Xu H, Swoboda I, Bhalla PL, Singh MB. 1999a.Male gametic cell-specific expression of



H2A andH3 histone genes.Plant Mol. Biol.39:607–14

Xu H, Swoboda I, Bhalla PL, Singh MB. 1999b.Male gametic cell-specific gene expressionin flowering plants.Proc. Natl. Acad. Sci.USA96:2554–58

Ye XL, Yeung EC, Zee SY. 2002. Sperm move-ment during double fertilization of a flow-ering plant,Phaius tankervilliae. Planta. Inpress

Zhang Z, Russell SD. 1999. Sperm cell surfacecharacteristics ofPlumbago zeylanicaL. inrelation to transport in the embryo sac.Planta208:539–44

Zhang Z, Tian HQ, Russell SD. 1999. Immuno-gold scanning electron microscopic localiza-tion of myosin on isolated sperm cells of to-bacco (Nicotiana tabacumL.). Protoplasma208:123–28

Zheng Z-L, Yang Z. 2000. The Rop GTPaseswitch turns on polar growth in pollen.TrendsPlant Sci.5:298–303

Zinkl GM, Zweibel BI, Grier DG, PreussD. 1999. Pollen-stigma adhesion inAra-bidopsis: a species-specific interactionmediated by hydrophobic molecules inthe pollen exine.Development126:5431–40

21 Sep 2002 14:13 AR AR170-04-COLOR.tex AR170-04-COLOR.SGM LaTeX2e(2002/01/18)P1: GDL

Fig

ure

2D

isch

arge

ofth

eco

nten

tsof

apo

llen

tube

into

the

embr

yosa

cin

Tore

nia

fou

rnie

riwith

time

afte

rini

tiatio

nof

disc

harg

e.Arr

ow

sin

dica

teth

eex

pand

ing

cont

ents

ofth

epo

llen

tube

inth

eem

bryo

sac.

The

rmal

pseu

doco

lor

imag

esdi

spla

ych

ange

inim

ages

.S

ma

lla

rro

win

dica

tes

the

rupt

urin

gpl

asm

am

embr

ane

ofth

esy

nerg

idce

ll.B

aris

10µ

m;e

c,eg

gce

ll;es

,em

bryo

sac;

pt,p

olle

ntu

be;s

y,sy

nerg

idce

ll.F

igur

efr

omH

igas

hiya

ma

etal

.(20

00)

repr

oduc

edw

ithpe

rmis

sion

ofA

mer

ican

Soc

iety

ofP

lant

Bio

logi

sts.

21 Sep 2002 14:13 AR AR170-04-COLOR.tex AR170-04-COLOR.SGM LaTeX2e(2002/01/18)P1: GDL

Fig

ure

3D

iagr

amm

atic

reco

nstr

uctio

nof

actin

orga

niza

tion

(g

ree

n)an

dnu

clea

rpo

sitio

n(red

)du

ring

fert

iliza

tion

inTo

ren

iafo

urn

ieri.

(a)

Mat

ure

embr

yosa

cw

ithde

nse

actin

pres

entn

ear

filifo

rmap

para

tus

(a

rro

ws)

;(b)

onse

tofs

yner

gid

dege

nera

tion

with

first

evid

ence

ofco

rona

form

atio

n;(c)

actin

reor

gani

zatio

nju

stpr

ior

topo

llen

tube

pene

trat

ion;

(d

)po

llen

tube

arriv

al,

disc

harg

e,an

dtr

ansp

ort

ofsp

erm

nucl

ei(a

rro

wh

ea

ds);

and

(e)co

mpl

etio

nof

fusi

on,d

isso

lutio

nof

actin

coro

na,i

nitia

tion

ofpo

st-f

ertil

izat

ion

actin

patte

rn.C

C,c

entr

alce

ll;dS

y,de

gene

rate

dsy

nerg

id;E

,egg

cell;

EN

,end

ospe

rmnu

cleu

s;P

T,po

llen

tube

;SN

,sec

onda

rynu

cleu

s;Z

,zyg

ote.

Fig

ure

from

Fu

etal

.(2

000)

repr

oduc

edw

ithpe

rmis

sion

ofS

prin

ger-

Verla

g.

Documents

Elizabeth M. Lord1 and Scott D. Russell2 · Abstract In ﬂowering plants, pollen grains germinate to form pollen tubes that transport male gametes (sperm cells) to the egg cell in