SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
The angiosperm female gametophyte:
No longer the forgotten generation
Vladimir Brukhin*, Mark D. Curtis* and Ueli Grossniklaus#
Institute of Plant Biology and Zürich-Basel Plant Science Center, University of Zürich, Zollikerstrasse 107, CH-8008 Zürich, Switzerland
The plant life cycle alternates between the diploid
sporophyte (spore-producing organism) and the haploid gametophyte (gamete-producing organism). The
female gametophyte of flowering plants develops
within the ovule, a specialized structure within the ovary,
which gives rise to the seed after fertilization. The female gametophyte, or embryo sac, contains a small
number of clonally derived cell types and serves as a
model for the study of many fundamental processes
crucial to development. The female gametophyte develops coordinately with the sporophytic tissues of the
ovule, providing an ideal system to study axis determination, pattern formation, cell–cell communication,
and the role of cell lineage and position in cell specification and differentiation. Despite the important role
played by the gametophyte in reproduction, few studies have focused attention on the molecular or genetic
aspects of female gametophyte development and function. Consequently, the gametophyte was termed ‘the
forgotten generation’1. In this paper we illustrate the
developmental and structural diversity of gametophytes, studied to a large extent by the Indian school
influenced by P. Maheshwari, and review new insights
gained by genetic and molecular studies.
Keywords: Angiosperm embryology, Arabidopsis, cell
specification, developmental genetics, embryo sac.
Reduction of the gametophyte during land plant
evolution
THE appearance of alternating generations in the plant
life cycle occurred early in the evolution of plants. In
many lower plants, the gametophytes are the dominant
free-living generation, while in higher plants, the dominant generation is the sporophyte. This distinction has classically been used to separate lower plants from higher
plants. In many algae, for example, the predominant
phase of the life cycle is the haploid gametophyte, and
the diploid sporophytic phase is only transient2.
As plants became adapted to life on land,, the sporophytic phase became dominant – with the exception of
the bryophytes, in which the predominant generation remains gametophytic and reliant on wet land habitats. In
#
For correspondence. (e-mail: [email protected])
*These authors have contributed equally to this work.
1844
the lower vascular plants, such as ferns, the sporophyte is
clearly dominant, producing sporangia in clusters (sori)
on frond-like leaves. At maturity, these sporangia release
spores, which, if they settle on a damp surface, germinate
to produce a prothallus or gametophyte. The prothallus
produces chlorophyll and grows independently of the
sporophyte. It forms male (antheridia) and/or female (archegonia) reproductive organs, which produce male and
female gametes, respectively. Since the sperm cells are
motile, water is a requirement for fertilization.
As the land plants evolved to produce seed, the gametophytic phase became further reduced and more deeply
embedded in reproductive structures of the plant, relying
more on the sporophyte and less on water. In higher plants
this reduction culminates in the production of the male
gametophyte, or microgametophyte (i.e. the pollen grain).
The male gametophyte is typically a 3-celled structure,
comprising two sperm cells encased within a vegetative
cell3–6. The microgametophytes are derived from a diploid microspore mother cell that undergoes meiosis to
produce four haploid microspores. Each haploid microspore undergoes two mitotic divisions. The first division
produces a large vegetative cell (the primary function of
which is to deliver the sperm cells to the female gametophyte or embryo sac) and a smaller generative cell7,4. The
generative cell then undergoes a second mitotic division
to produce two sperm cells (for review see ref. 8).
The morphology of the female gametophyte, and the means
by which it is fertilized, are defining features that separate the angiosperms from the gymnosperms9. In gymnosperms, the female gametophyte is much larger than that
of angiosperms, consisting of several hundreds or even
thousands of cells with several archegonia enclosed within
jacket cells4. The large size of the female gametophyte
means that in gymnosperms there is no requirement for
endosperm to nourish the developing embryo. The socalled ‘endosperm’ of the gymnosperms is the haploid
gametophyte and, although typically two sperm cells are
produced, only one sperm participates in the fertilization
of the egg while the other degenerates.
In angiosperms, the female gametophyte, or megagametophyte, develops within the ovule, protected by the
nucellus and the integuments of the ovule. The structure
and organization of the female gametophyte varies between
angiosperms, the most common form being the Polygonum type, which is present in 70% of all angiosperms4.
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
a
b
c
cellularization and
cell specification
mitotic division cycles and
nuclear migrations
d
embryo sac
maturation
post-fertilization
development
Figure 1. a–d, Schematic representation of the development of a Polygonum type female gametophyte. a, Three free nuclear divisions
occur in a syncytium to form an 8-nucleate female gametophyte. One nucleus from each pole (the polar nuclei) migrates and will
eventually be enclosed by the central cell. b, Cellularization forms the typcial 7-celled, 8-nucleate female gametophyte with two
synergid cells (blue), one egg cell (red), a bi-nuleate central cell (green) and three antipodal cells (yellow). c, Before fertilization,
the female gametophytes fully differentiates: the two polar nuclei fuse to form the homo-diploid secondary endopsperm nucleus,
the antipodals undergo programmed cell death and one of the synergids degenerates as the pollen tube arrives. d, During double
fertilization one sperm cell fuses with the egg cell to form the diploid zygote (purple), while the second sperm fertilizes the central
cell to form the triploid endosperm (dark green). The primary endosperm nucleus divides prior to the zygote in a syncytium.
a
b
c
Figure 2 a–c. Selected examples illustrating the structural diversity of
angiosperm female gametophytes. a, Most common 7-celled 8-nucleate
Polygonum type embryo sac with an egg apparatus, consisting of an
egg cell and two synergids at the micropilar pole (top), two polar nuclei
in the centre, and three antipodes at the chalazal pole (adapted from
82). b, Penaea type embryo sac of Acalypha indica: a 16-nucleate embryo sac, with 8 free nuclei in the centre and four groups of two cells at
the periphery (adapted from 23). c, Embryo sacs of the Plumbagella
type, where synergids are missing (top), the antipodal cell is triploid,
and the central cell nucleus is tetraploid (adapted from 24).
The Polygonum type female gametophyte results from
precisely choreographed megasporogenesis and megagametogenesis. During megasporogenesis, a diploid megaspore mother cell (MMC) undergoes meiosis to produce
four haploid megaspores, one of which survives while the
other three degenerate. Then, during megagametogenesis,
the surviving functional megaspore undergoes three rounds of
mitosis, producing an 8-nucleate syncytium (Figure 1).
Cellularization follows to produce seven cells belonging
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
to four different cell types: an egg cell and two synergids
(the so-called egg apparatus, located at the micropylar
pole of the embryo sac), a bi-nucleate central cell, and three
antipodals (located at the chalazal pole of the embryo
sac). Despite the small number of cells and minute size,
the female gametophyte is a central structure in angiosperm reproduction. It is involved in pollen tube guidance
and reception, fertilization, egg activation, and the maternal control of seed development (for review see refs 10,
11). When the pollen tube reaches the embryo sac, the
two sperm cells within are released into the degenerating
synergid. One sperm cell then migrates to the central cell
and, after plasmogamy, the nuclei fuse to generate a triploid nucleus. The other sperm cell migrates to the egg cell,
with which it fuses to produce the diploid zygote. The
triploid central cell gives rise to the endosperm and the
diploid egg cell gives rise to the embryo. Viable seed formation depends on this double fertilization event, and the
coordinated development of the embryo, the endosperm,
and the maternal seed coat of sporophytic origin, to produce a
mature seed harbouring the next sporophytic generation.
Structural diversity among angiosperm female
gametophytes
Although the general features described above have been
used to define the angiosperms and the gymnosperms,
they do not clearly identify the evolutionary relationships
within these phyla. Difficulties arise due to the evolution
of different types of female gametophytes within the angiosperms. These differences can result from the number
of functional megaspore nuclei that take part in the formation of the embryo sac (i.e. monosporic, bisporic, or
1845
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
tetrasporic). There are two types of monosporic female
gametophytes, the Polygonum type, which produces an 8nucleate embryo sac12–15 (Figures 1 and 2 a), and the Oenothera type, which produces a 4-nucleate embryo sac, present only in the family Onagraceae4,12,16,17. Unlike the
Polygonum type, just two mitotic nuclear divisions occur
in the Oenothera type embryo sac. In the development of
a Polygonum type embryo sac the functional megaspore
nucleus undergoes an initial mitotic division. The two
daughter nuclei then migrate to opposite poles of the
syncytium, thereby establishing the proximal–distal axis
and polarity of the embryo sac (Figure 1). A large vacuole
forms between the two nuclei, which increases in size
during embryo sac development. Each of the nuclei undergoes two further mitotic divisions, prior to cellularization.
During cellularization, the three nuclei at the micropylar
pole give rise to the two synergids and the egg cell, and
the three nuclei at the chalazal pole form the three antipodal
cells. The remaining nucleus – one at each pole (polar nuclei) – migrates towards the center and often fuse (karyogamy), producing the homo-diploid nucleus of the central
cell (Figures 1 and 2 a). In the region where synergids, egg
cell and central cell meet, the cell wall is discontinuous18.
This is probably necessary to allow plasmogamy with the
sperm cells during fertilization. In the final stages of embryo sac development, the three antipodal cells usually
degenerate (Figure 1), but in some species, e.g. Zea mays,
they proliferate19. In Oenothera the micropylar megaspore develops into the embryo sac. After two mitotic divisions, one of the nuclei cellularizes to form the egg cell,
two nuclei form the synergids, and one constitutes the
single polar nucleus. The Oenothera type embryo sac is
monopolar, since all the nuclei are situated at the micropylar pole of the developing embryo sac4. It is interesting
to note that, in some apogametic complexes, these two
types of embryo sac coexist: e.g. the sexual forms of
Panicum maximum produce a bipolar 8-nucleate female
gametophyte, whereas the apomictic forms, which reproduce asexually through seeds, have a monopolar, 4nucleate embryo sac20,21.
Bisporic embryo sacs are derived from two megaspore
nuclei, which are not separated by cytokinesis and undergo two mitotic divisions each, to form an 8-nucleate,
7-celled mature bipolar embryo sac, the most common of
which is the Allium type. Further variation in the development of the female gametophyte is observed when all
four megaspores participate in the formation of the embryo
sac. For example, Peperomia type embryo sacs are 16nucleate, consisting of an egg and a synergid cell at the
micropylar pole, and a large secondary endosperm nucleus,
formed by the fusion of 8 of the remaining nuclei, whilst
the other 6 nuclei are cut off as antipodals22. In Penaea
type embryo sacs, the 16 nuclei form four distinct groups,
one at each pole, and one at either side of the embryo sac.
One nucleus from each quartet remains free (i.e. 4 polar
nuclei) and moves towards the center of the embryo sac,
1846
while the other three cellularize. Generally, one cell in the
group of three cells at the micropylar end is the functional egg
cell. Embryo sacs of Acalypha indica, although similar to
Penaea type embryo sacs, do not follow the same arrangement. Here, the 16 nuclei are arranged into four distinct groups at the four poles. However, 2 nuclei of each
quartet remain free and migrate to the center of the embryo
sac, while the others become organized into cells. This
produces four distinct groups of two cells each at the periphery and eight free nuclei in the center23 (Figure 2 b).
Some tetrasporic embryo sacs do not produce 16 nuclei.
In the Plumbagella type, for example, the 4 megaspores
migrate, one to the micropylar end, the remaining three to
the chalazal end, where they fuse to form a triploid nucleus. These two nuclei divide producing two haploid micropylar nuclei and two triploid chalazal nuclei. The
haploid nucleus nearest to the micropylar end becomes
organized into the egg and the triploid nucleus nearest the
chalazal end forms a single antipodal cell. The remaining
two nuclei fuse to form a tetraploid secondary endosperm
nucleus and no synergid cells are produced24 (Figure 2 c).
The enormous variety of developmental pathways that
lead to embryo sac formation provides a basis for studying
the evolution of these developmental programs. The continued characterization of genes that play key roles in
embryo sac development and function, together with cytological data, will provide insights into the evolution of
the various developmental pathways.
Genetic identification of gametophytically acting
genes
The Polygonum type embryo sac is the most common and
has been studied most extensively. Yet, despite our intimate knowledge of its morphology, very little is known
about the molecular genetic mechanisms that control its
development. The inaccessibility and small size of the female
gametophyte have possibly hampered progress and enthusiasm for identifying the molecular mechanisms in the
past. Although the Polygonum type female gametophyte
contains only seven cells, these cells perform specific
functions essential to reproduction. Mutational analysis of
the female gametophyte will allow the dissection of these
functions and their underlying developmental pathways.
The isolation of new gametophytic mutants, resulting from
large-scale insertional mutagenesis projects (reviewed in
ref. 25), as well as the establishment of novel tools to
analyse single cells at the molecular level (reviewed in
ref. 26), offer unique opportunities to dissect the molecular
mechanisms involved in gametophyte development and
function.
Polygonum type embryo sac development in wild-type
Arabidopsis has been very well described18,27–33 and has
been divided into seven morphologically distinct stages,
FG1 to FG7 (ref. 33). But even in Arabidopsis early reports
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
a
Seed set phenotype
b
Screening for distorted Kan
Normal fertility
res
sen
:Kan
segregation
Mendelian
inheritance
Zygotic
lethality
Female
gametophytic
lethality
3:1
2:1
1:1
Semisterility
Unfertilised ovules
Seed abortion
Figure 3 a,b. Strategy of the two-step screen for identifying gametophytic mutants. a, In the first step, siliques are assessed for a
reduction in seed set. In the case of a plant heterozygous for a gametophytic lethal mutation, about half of the ovules remain small
and do not develop into a seed. In plants heterozygous for a zygotic embryo lethal mutation, about 25% of the seeds abort, whereas
a gametophytic maternal effect mutation would show 50% seed abortion, which can be assayed in green or dry siliques. b, In a
second step, the segregation ratio of the insertion is analysed using the dominant marker contained within it. If the insertion has no
effect on transmission, it will segregate in a ratio of 3 : 1, if it causes embryonic lethality 2 : 1, and if it causes female gametophyte
lethality 1 : 1.
of gametophytic mutants were rare. In fact, until the mid1990s, only the gametophytic factor1 (gf1) mutant had
been described34. Over the last few years a large number
of gametophytic mutants have been isolated in Arabidopsis and their molecular analysis is underway33,35–39. This
has become possible thanks to the development of tools
for the production of insertional mutants and for their
rapid molecular analysis.
Mendel’s laws apply to traits expressed in the diploid
sporophyte, where a recessive mutation will segregate in
a 3 : 1 ratio in the F2 generation. By contrast, mutations
in genes that affect the haploid gametophyte prevent their
transmission through the egg and/or the sperm, resulting
in a distorted, non-Mendelian segregation ratio40. A mutation that affects both gametophytes and is fully penetrant
will not be transmitted to the next generation at all. However, partially penetrant mutations, or mutations affecting
only one sex, can be transmitted to subsequent generations and recovered as heterozygotes. Gametophytic mutants are difficult to identify because half of the pollen
and half of the embryo sacs carry a wild-type allele such
that the plants appear fertile. While male gameophytic
mutants produce enough wild-type pollen to fertilize all
ovules, female gametophytic mutants are expected to be
semi-sterile because half of the ovules will not form seed.
A half-filled fruit, however, looks normal and it has to be
opened to evaluate the reduction in seed set. Because most
plants have reduced fertility after a chemical mutagenesis
treatment, chemical mutagenesis followed by screening
for semi-sterility is not an appropriate method by which
to identify female gametophytic mutants. A new era in
gametophyte genetics began with the introduction of protocols for insertional mutagenesis. These methods did not
cause a general reduction in fertility and allowed segregation ratios to be determined by the simple analysis of
marker gene selection (present on the insertional element,
i.e. T-DNA or transposons).
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
The two hallmarks of gametophytic mutations, segregation ratio distortion and reduced fertility, can be used
to identify such mutants from collections of insertions. A
departure from the Mendelian segregation ratio of 3 : 1
indicates that either the homozygous zygotes (segregation
ratio of 2 : 1) or the haploid gametophytes are lethal (Figure
3). For a sex-specific, fully penetrant, gametophytic mutation,
a 1 : 1 segregation ratio is expected. Often, however, transmission through both gametophytes is reduced, resulting
in a more severe segregation ratio distortion. In order to
determine the extent to which each gametophyte is affected,
reciprocal crosses to the wild type are performed to
measure the transmission efficiency through each sex35,41.
If segregation ratio distortion is used for the identification of gametophytic mutants, mutants affecting either or
both the male and female gametophytes are recovered. To
enrich for mutants with an effect on embryo sac development or function, two-step screens have been devised
in which (i) fertility is analyzed to identify semisterile
plants, and (ii) the progeny of these plants is analyzed for
segregation ratio distortion25,35. Only true gametophytic
mutants show segregation ratio distortion, whereas all
other causes of a reduction in fertility, such as partially
penetrant sporophytic mutants, poor environmental conditions, or reciprocal translocations, do not cause a deviation from normal Mendelian segregation.
Instead of using a marked insertional mutagen, chromosomes carrying multiple markers can be used to identify
gametophytic mutants because linked markers will also
show some degree of segregation ratio distortion42. However, in addition to carrying a marker for direct assessment, insertional mutagens have the advantage that they
‘tag’ the mutations so that genomic sequences flanking
the insertion can be obtained by simple PCR-based methods
(see for instance, refs 43, 44). But insertional mutagens
can cause chromosomal rearrangements, which can hamper the analysis of such mutants45,46.
1847
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
Gametophytic mutants affect all stages of
gametophyte development
A large number of cellular processes are required for the
development and function of the angiosperm female gametophyte. Consequently, gametophytic mutants are recovered
at a high frequency in large-scale screens35–37,47,48 (VB &
UG, unpublished). Most mutants have been isolated from
Arabidopsis, but a few have also been described in maize
(Table 1). Female gametophyte development can be divided
into the following steps: (i) megaspore specification, (ii)
initiation of megagametogenesis, (iii) mitotic progression,
(iv) establishment of gametophyte polarity, (v) migration
of polar nuclei, (vi) fusion of polar nuclei (karyogamy),
(vii) cellularization, (viii) antipodal cell death, and (ix) degeneration of the synergids. The mature female gametophyte plays an important role in several developmental
processes: (i) pollen tube guidance, (ii) pollen tube reception, (iii) sperm migration, (iv) plasmogamy during double
fertilization, (v) karyogamy of male and female pronuclei, and (vi) maternal effects on early embryo and endosperm development. The gametopytic mutants isolated
to date are deficient in one or more of these processes and
have been classified according to the primary lesion they
cause in the developmental pathway (Table 1).
7-celled embryo sacs. In these embryo sacs, either the APC/C
complex remains active in the absence of the NOMEGA
protein, or sufficient NOMEGA protein remains in the
Table 1.
Types of the established female gametophytic mutations in
Arabidopsis
Mutant class
Mitotic
Karyogamy
gfa2, gfa3, gfa7
pri, nan
eda24-eda41
amon (amn), apis (aps)
Cellularization
gfa2, gfa3, fem4,
fem6-fem8, fem11, fem13,
fem15
jum, wlg, nja, dam
Mitotic class
Most gametophytic mutants fall into the mitotic class. This
class includes all mutants with a lesion in the initiation or
regulation of any of the three mitotic divisions that occur
during megagametogenesis (from FG1 to early FG5). The
mutants are characterized by an unusual number of nuclei,
or an aberrant distribution of nuclei in the developing
embryo sac35,37,47,48.
An interesting example of this class of mutant, which
illustrates how variable expressivity of a mutation influences the degree of segregation ratio distortion and seed
sterility, is the mutant nomega. In the nomega mutant the
embryo sac is arrested at the 2-nucleate stage, and although
this is a gametophytic mutant and 50% of the ovules should
contain mutant embryo sacs, only 30% of ovules were
aborted49. In fact, variable expressivity of the phenotype
appears to be suprisingly common among gametophytic
mutants36. This may be due to the fact that large populations of individual gametophytes are analyzed and, that
due to their haploid nature, they are more susceptible to
genetic and environmental modifiers.
The NOMEGA gene product has high homology to the
APC6/CDC16 subunit of the anaphase-promoting complex and is involved in post-metaphase events (separation
of chromosomes and cytokinesis). This gene is disrupted
by a Ds insertion, so that the mutant allele is unlikely to
produce a functional protein. Despite this mutation, however, some female gametophytes can develop into mature
1848
Mutant name
gf1
ig*
prl
hdd
gfa4, gfa5
ada, tya
fem2, fem3, fem5, fem9-fem16,
fem18-fem26, fem29-fem31,
fem33-fem38
hma, ana
cki1
apc2
nomega
rbr1
agp18
swa1
eda1-eda23
kupalo (kuo), astlik (alk)
References
34, 33
67, 68, 83
69
35
70
41
47
36
71, 72
73
49
74
75
76
48
(Brukhin,
unpublished)
70, 37
36
48
(Brukhin,
unpublished)
37, 47
47
36
Degeneration
fem1
fem14
cgfa2
nan
yarilo (yar)
37
47
47
36
(Brukhin,
unpublished)
Fertilization
fer
Srn
une1-une18
zmea1*
36, 53
54
48
77
Maternal-effect
fie
fis1 (mea), fis2, fis3 (fie)
mea
mel1*
prl
cap1, cap2
ctr1
zal, ash, aya, kem
dme
bga
mee1-mee70
lpat2
didilia (did)
58
57
44
78
79
38
47
36
80
62
48
81
(Brukhin,
unpublished)
*These are mutants in Zea mays; all others were described in Arabidopsis thaliana.
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
cytoplasm, which is derived from the heterozygote MMC,
allowing the three rounds of mitosis. The APC/C complex functions as an E3 ubiquitin ligase in the ubiquitinmediated proteolysis pathway, which controls several key
steps in the cell cycle. In nomega mutant embryo sacs,
cyclin B is not degraded, and this probably results in the
arrest of cell division at telophase.
Karyogamy class
The karyogamy class of mutants affects the fusion of polar
nuclei. For example, in the gfa2 mutant, the polar nuclei
migrate, but fail to fuse. This defect is accompanied by a
delay in antipodal cell degeneration and a defect in synergid degeneration (features of another class of mutants)47.
The GFA2 gene encodes a J-domain-containing protein, the
ortholog of which functions in budding yeast as a chaperone
in the mitochondrial matrix. This mutation provides evidence that nuclear fusion may be facilitated by mitochondrial
factors and that the processes of megaspore and antipodal
cell death are distinct from synergid degeneration and
may be somehow linked to mitochondrial functions.
Cellularization class
This class of embryo sac mutants exhibit defects in cellularization, cell polarity and cell shape37,36. In wild-type megagametophytes, cellularization begins immediately, following
the third mitosis in stages FG5 to FG7 (ref. 33). In mature
embryo sacs, the egg cell is pear-shaped and highly polarized, with the nucleus and cytoplasm at the chalazal end of
the cell with the remaining space filled by a large vacuole.
The synergids are also polarized, however, the orientation
is opposite to that of the egg cell, with the cytoplasm and
nucleus at the micropylar end of these cells. The gametophytic mutant fem4, is a good example of a mutant affecting
the shape and position of these cells37. Here, the egg cell is
not pear-shaped and the synergid cells show altered polarity
and shape, with their nuclei oriented towards the chalazal
end of the cells.
The cytological characterizations of hadad (hdd)35 and
of deletions in Zea mays50 have shown that cellularization
and mitotic progression are not coupled to each other. In
these mutants, which arrest during the mitotic divisions
and thus belong to the mitotic class, cell membranes can
form even though all cell division cycles have not been
completed. These findings suggest that the developmental
programs controlling nuclear division and cellularization
are independent.
Degenerative class
The degenerative class of embryo sac mutants shows defects
in the degeneration of the three non-functional megaspore
cells, and/or the synergids and/or the antipodal cells. This
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
class includes the gfa2 mutation that affects synergid cell
degeneration, as described earlier47, and may result simply from their failure to attract pollen tubes. Many other
mutants show degenerating embryo sacs, but this may often be a secondary consequence of the failure to complete
megagametogenesis. In the case of gfa2 and nantosuetta
(nan), both degeneration and a failure in the fusion of polar nuclei are observed36. In nan embryo sacs, the first
abnormality observed is an enlargement and distortion of
the nucleolus in polar nuclei, which may or may not fuse.
Fertilization class
The female gametophyte produces chemotactic signals that
guide the growing pollen tube towards the micropyle51,52.
One of the two synergids becomes receptive to pollen tube
entry and, just before or during contact with the pollen
tube, starts to disintegrate. The growth of the pollen tube
arrests and the tip ruptures to release the sperm cells. After
the sperm cells are released, they migrate to their female
target cells and undergo plasmogamy. The mutants feronia and sirene are examples of this mutant class53,54. In
mutant gametophytes, the pollen tubes continue to grow
after entering the receptive synergid, fail to rupture and
do not release the sperm cells. Frequently, numerous pollen tubes enter the embryo sac, but none effects fertilization, leading to semisterility (50% unfertilized ovules in
the silique). This evidence, together with cell ablation
studies in Torenia fournieri55, shows that the synergid
cells control both pollen tube guidance and reception.
The absence of synergids in some classes of angiosperm
embryo sacs, such as the Plumbagella type (Figure 2 c),
indicates that there must be alternative modes of pollen
tube attraction among the angiosperms.
Maternal-effect class
After double fertilization the zygote develops into a diploid embryo (new sporophyte originating from the fertilized egg cell) and is nourished by the triploid endosperm
(originating from the fertilized central cell). The two fertilization products are surrounded by the integuments of
the ovule (tissues of sporopytic origin), which contribute
to the development of the seed coat. Gametophytic mutants
can also affect post-fertilization development. A mutant
phenotype of this kind that depends on the genotype of the
female gametophyte, but is independent of the paternal contribution, is referred to as a gametophytic maternal effect10.
The mechanistic basis of maternal effects is varied and can
be caused by: (i) mutations in genes that are expressed
during embryo sac development, but whose products are
required after fertilization for embryo and/or endosperm
development, (ii) abnormal mitochondria or plastids that
are usually inherited from the mother plant, (iii) alterations in gene dosage, or (iv) genomic imprinting.
1849
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
The first example of such a maternal effect to be described
in detail was medea (mea). Embryo and endosperm derived from a mea mutant embryo sac show abnormal cell
proliferation, independent of the parental allele and gene
dosage, leading to the formation of giant embryos and an
enlarged chalazal cyst in the endosperm44. It was shown
later that this was due to genomic imprinting, where only
the maternally inherited allele was active after fertilization56. The MEA protein forms part of a Polycomb group
complex that suppresses cell proliferation, not only after
fertilization, but also in the absence of fertilization, preventing fertilization-independent endosperm development. Three other components of the complex are known, all
of which share this interesting phenotype39,44,57–62. Whether
the genes encoding other components of the Polycomb
group complex are also regulated by genomic imprinting,
or show a maternal effect because they are cytoplasmically stored, is currently unknown63.
Many gametophytic mutants show defects after fertlization. In fact, about half of the mutants described by Moore36
show post-fertilization defects at a certain frequency. Outcrossing with wild-type pollen has shown that these defects
were indeed under maternal control36. A similar situation
was found recently by Pagnussat et al.48 who also reported
that half of the gametophytic mutants they isolated showed
post-fertilization defects. The genes disrupted in these
mutants come from diverse families that have been implicated in a wide variety of cellular functions, including protein
degradation, transcriptional regulation, signal transduction
and secondary metabolism48. However, many genes with
unknown function have also been identified, presenting
new challenges in the characterization and assignment of
a function to these genes. In summary, gametophytic maternal effects seem to play a prominent role in seed development (see also ref. 64). However, with the exception
of mea, the underlying nature of these maternal effects is
still unclear63.
Conclusion
Over the last century, a vast amount of knowledge has been
acquired about the cytology and ultrastructure of the angiosperm female gametophyte. However, our understanding
of the molecular mechanisms that determine gametophyte
development and how these developmental programs
have evolved is not complete. During evolution several
different developmental pathways have been adopted that
account for the diversity of embryo sac structures. Despite
these differences, the female gametophyte, in all its forms,
plays a pivotal role in reproduction. Although most molecular
and genetic analyses have been carried out in model organisms with a Polygonum type embryo sac, these studies
will inevitably shed light on the processes that have led to
the evolution of the other, less common, types of embryo
sacs. Comparative, functional genetics will help elucidate the
1850
evolutionary relationships among the different embryo
sacs providing new insights into the evolution of developmental processes.
In recent years, studies in Arabidopsis and maize have
identified interesting genes that play a role in megagametogenesis and seed development, but a great deal is still to
be learned. By taking advantage of the molecular-genetic
tools available in model plants, a combination of forwardand reverse-genetics approaches can be applied to unravel
the mechanisms that orchestrate the development and reproductive functions of the female gametophyte. Genetic
screens directed to isolate mutants with disruptions in genes
of importance to the gametophyte have already identified
genes that play a role in many fundamental processes. Cytological observations show that gametophytic mutants
typically have variable and incomplete phenotypes. Their
defects include reduced male and female fertility, due to
gametophyte lethality, the production of non-viable zygotes,
and gametophytic maternal-effect seed abortion.
Understandably, the female gametophyte is physiologically
very active and probably expresses thousands of genes.
Certainly, the genes identified so far have diverse roles in
cellular processes such as protein degradation, transcriptional regulation and signal transduction. Functional redundancy no doubt hinders the identification of some
important genes that are involved in female gametophyte
development and function. To identify these, second site
mutagenesis or the construction of multiple mutations within
single plants may be required, presenting great challenges
to investigators. Strategies such as inducible gene silencing
by RNA interference that specifically targets and silences
gene families, could help resolve these problems65,66. New
approaches such as microgenomics also hold much promise. The isolation of single cells by laser capture microdissection, used in conjunction with transcriptome analysis, will
help to identify cell type-specifically expressed genes that
are important for the identity and function of gametophytic
cells26. Certainly, the application of such new technologies to this interesting phase of the plant life cycle will
lead to new insights in the near future.
1. Heslop-Harrison, J., The forgotten generation: some thoughts on
the genetics and physiology of angiosperm gametophytes. In The
Bateson Lecture: Proceedings of the Fourth John Innes Symposium, 1979, pp. 1–14.
2. Pickett-Heaps, J. D., Green Algae: Structure, Reproduction and
Evolution in Selected Genera, Sinauer, Sunderland, MA, USA,
1975.
3. Coulter, J. M. and Chamberlain, C. J., Morphology of Spermatophytes, D. Appleton and Company publisher, New York, 1901.
4. Maheshwari, P., An Introduction to the Embryology of Angiosperms, McGraw-Hill, New York, 1950, pp. 453.
5. da Costa-Nunes, J. A. and Grossniklaus, U., Unveiling the geneexpression profile of pollen. Genome Biol., 2003, 5, 205.1–205.3.
6. McCormick, S., Control of male gametophyte development. Plant
Cell, 2004, 16, 142–153.
7. Nawaschin, S. G., Resultate einer Revision der Befruchtungsvorgänge
bei Lilium martagon und Fritillaria tenella. Bul. Acad. Imp. Des
Sci. St. Petersburg, 1898, 9, 377–382.
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
8. Ma, H., Molecular genetic analyses of microsporogenesis and microgametogenesis in flowering plants. Annu. Rev. Plant Biol.,
2005, 56, 393–434.
9. Baroux, C., Spillane, C. and Grossniklaus, U., Evolutionary origins
of the endosperm in flowering plants. Genome Biol., 2002, 3,
1026.1–1026.5.
10. Grossniklaus, U. and Schneitz, K., The molecular and genetic basis
of ovule and megagametophyte development. Semin. Cell Dev.
Biol., 1998, 9, 227–238.
11. Yadegari, R. and Drews, G. N., Female gametophyte development.
Plant Cell, 2004, 16, S133–141.
12. Hofmeister, W., Untersuchungen des Vorgangs bei der Befruchtung
der Oenotheren. Bot. Ztg., 1847, 5, 785–792.
13. Strassburger, E., Die Angiospermen und die Gymnospermen, Jena,
Germany, 1879.
14. Palm, B., Studien über Konstruktionstypen und Entwicklungswege
des Embryosackes der Angiospermen, Diss. Stockholm, 1915.
15. Maheshwari, P., The angiosperm embryo sac. Bot. Rev., 1948, 14,
1–56.
16. Geerts, J. M., Beiträge zur Kenntnis der cytologischen Entwicklung
von Oenothera lamarckiana. Ber. Deutsch. Bot. Gesell., 1908,
26A, 608–614.
17. Modilewski, J., Zur Embryobildung von einigen Onagraceen. Ber.
Deutsch. Bot. Gesell., 1909, 27, 287–291.
18. Mansfield, S. G., Briarty, L. G. and Erni, S., Early embryogenesis
in Arabidopsis thaliana. I. The mature embryo sac. Can. J. Bot.,
1991, 69, 447–460.
19. Kiesselbach, T. A., The structure and reproduction of corn. Univ.
Nebraska Coll. Agric. Exp. Station Res Bull. 161, 50th Anniversary Edition, Cold Spring Harbor Laboratory Press, USA, 1949.
20. Savidan, Y., Chromosomal and embryological analyses in sexual ×
apomicitic hybrids of Panicum maximum. Jacq. Theor. Appl.
Genet., 1980, 57, 153–156.
21. Grossniklaus, U., From sexuality to apomixis: molecular and genetic approaches. In Advances in Apomixis Research (eds Savidan,
Y., Carman, J. and Dresselhaus, T.), CYMMIT Press, Mexico,
2001, pp. 168–211.
22. Johnson, D. S., On the endosperm and embryo of Peperomia pellucida. Bot. Gaz., 1900, 80, 1–11.
23. Maheshwari, P. and Johri, B. M., The embryo sac of Acalypha indica L. Beihefte Bot. Centbl., 1941, 61A, 125–136.
24. Fagerlind, F., Embryosack von Plumbagella und Plumbago. Arkiv.
För Bot., 1938, 29B, 1–8.
25. Page, D. R. and Grossniklaus, U., The art and design of genetic
screens: Arabidopsis thaliana. Nature Rev Genet., 2002, 3, 124–36.
26. Day, R. C., Grossniklaus, U. and Macknight, R. C., Be more specific! Laser-assisted microdissection of plant cells. Trends Plant
Sci., 2005, 10, 397–406.
27. Misra, R. C., Contribution to the embryology of Arabidopsis
thaliana (Gay and Monn.). Agra Univ. J. Res. Sci., 1962, 11, 191–199.
28. Poliakova, T. F., Development of the male and female gametophytes
of Arabidopsis thaliana (L) Heynh. Issledov Genet. USSR, 1964,
2, 125–133.
29. Webb, M. C. and Gunning, B. E. S., Embryo sac development in
Arabidopsis. Sex. Plant Rep., 1990, 3, 244–256.
30. Murgia, M., Huang, B.-Q., Tucker, S. C. and Musgrave, M. E.,
Embryo sac lacking antipodal cells in Arabidopsis thaliana (Brassicaceae). Am. J. Bot., 1993, 80, 824–838.
31. Webb, M. C. and Gunning, B. E. S., Embryo sac development in
Arabidopsis thaliana. II. The cytoskeleton during megagametogenesis.
Sex. Plant Reprod., 1994, 7, 153–163.
32. Schneitz, K., Hülskamp, M. and Pruitt, R. E., Wild-type ovule development in Arabidopsis thaliana: A light microscope study of
cleared whole-mount tissue. Plant J., 1995, 7, 731–749.
33. Christensen, C. A., King, E. J., Jordan, J. R. and Drews, G. N.,
Megagametogenesis in Arabidopsis wild type and the Gf mutant.
Sex. Plant Reprod., 1997, 10, 49–64.
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005
34. Redei, G. P., Non-Mendelian megagametogenesis in Arabidopsis.
Genetics, 1965, 51, 857–872.
35. Moore, J. M., Vielle-Calzada, J.-P., Gagliano, W. and Grossniklaus, U.,
Genetic characterization of hadad, a mutant disrupting megagametogenesis in Arabidopsis thaliana. Cold Spring Harbor Symp.
Quant. Biol., 1997, 62, 35–47.
36. Moore, J. M., Isolation and characterization of gametophytic mutants
in Arabidopsis thaliana. Ph.D. Thesis, State University of New
York at Stony Brook, USA, 2002.
37. Christensen, C. A., Subramanian, S. and Drews, G. N., Identification of gametophytic mutations affecting female gametophyte development in Arabidopsis. Dev. Biol., 1998, 202, 136–151.
38. Grini, P. E., Jürgens, G. and Hülskamp, M., Embryo and endosperm
development is disrupted in the female gametophytic capulet mutants of Arabidopsis. Genetics, 2002, 162, 1911–1925.
39. Ohad, N. et al., Mutations in FIE, a WD Polycomb group gene, allow endosperm development without fertilization. Plant Cell,
1999, 11, 407–416.
40. Brink, R. A., MacGillivray, Segregation for the waxy character in
maize pollen and differential development of the male gametophyte. Am. J. Bot., 1924, 11, 465–469.
41. Howden, R., Park, S. K., Moore, J. M., Orme, J., Grossniklaus, U.
and Twell, D., Selection of T-DNA-tagged male and female gametophytic mutants by segregation distortion in Arabidopsis. Genetics,
1998, 149, 621–631.
42. Grini, P. E., Schnittger, A., Schwarz, H., Zimmermann, I., Schwab,
B., Jürgens, G. and Hülskamp, M., Isolation of ethyl methanesulfonate-induced gametophytic mutants in Arabidopsis thaliana by a
segregation distortion assay using the multimarker chromosome 1.
Genetics, 1999, 151, 849–863.
43. Sundaresan, V., Springer, P., Volpe, T, Haward, S., Jones, J. D.,
Dean, C., Ma, H. and Martienssen, R., Patterns of gene-action in
plant development revealed by enhancer trap and gene trap transposable elements. Genes Dev., 1995, 9, 1797–1810.
44. Grossniklaus, U., Vielle-Calzada, J.-P., Hoeppner, M. A. and Gagliano, W., Maternal control of embryogenesis by MEDEA, a
Polycomb group gene in Arabiodpsis. Science, 1998, 280, 446–
450.
45. May, B. P. and Martienssen, R. A., Transposon mutagenesis in
the study of plant development. Crit. Rev. Plant Sci., 2003, 22, 1–
35.
46. Page, D. R., Köhler, C., da Costa-Nunes, J. A., Baroux, C., Moore,
J. M. and Grossniklaus, U., Intrachromosomal excision of a hybrid
Ds element induces large genomic deletions in Arabidopsis. Proc.
Natl. Acad. Sci. USA, 2004, 101, 2969–2974.
47. Christensen, C. A., Gorsich, S. W., Brown, R. H., Jones, L. G.,
Brown, J., Shaw, J. M. and Drews, G. N., Mitochondrial GFA2 is
required for synergid cell death in Arabidopsis. Plant Cell, 2002,
14, 2215–2232.
48. Pagnussat, G. C. et al., Genetic and molecular identification of
genes required for female gametophyte development and function
in Arabidopsis. Development, 2005, 132, 603–614.
49. Kwee, H. S. and Sundaresan, V., The NOMEGA gene required for
female gametophyte development encodes the putative APC6/CDC16
component of the Anaphase Promoting Complex in Arabidopsis.
Plant J., 2003, 36, 853–866.
50. Vollbrecht, E. and S. Hake, S., Deficiency analysis of female gametogenesis in maize. Dev. Genet., 1995, 16, 44–63.
51. Hülskamp, M., Schneitz, K. and Pruitt, R. E., Genetic evidence for
a long-range activity that directs pollen tube guidance in Arabidopsis. Plant Cell, 1995 7, 57–64.
52. Ray, S. M., Park, S. S. and Ray, A., Pollen tube guidance by the
female gametophyte. Development, 1997, 124, 2489–2498.
53. Huck, N., Moore, J. M., Federer, M. and Grossniklaus, U., The
female gametophytic control of pollen tube reception is disrupted
in the Arabidopsis mutant feronia. Development, 2003, 130, 2149–
2159.
1851
SPECIAL SECTION: EMBRYOLOGY OF FLOWERING PLANTS
54. Rotman, N., Rozier, F., Boavida, L., Dumas, C., Berger, F. and
Faure, J. E., Female control of male gamete delivery during fertilization in Arabidopsis thaliana. Curr. Biol., 2003, 13, 432–436.
55. Higashiyama, T., Yabe, S., Sasaki, N., Nishimura, Y., Shin-ya Miyagishima, S., Kuroiwa, H. and Kuroiwa, T., Pollen tube attraction
by the synergid cell. Science, 2001, 293, 1480–1483.
56. Vielle-Calzada, J.-P., Thomas, J., Spillane, C., Coluccio, A.,
Hoeppner, M. A. and Grossniklaus, U., Maintenance of genomic
imprinting at the Arabidopsis medea locus requires zygotic DDM1
activity. Genes Dev., 1999, 13, 2971–2982.
57. Chaudhury, A. M., Ming, L., Craig, S., Dennis, E. S. and Peacock,
W. J., Fertilization-independent seed development in Arabidopsis
thaliana. Proc. Natl. Acad. Sci. USA, 1997, 94, 223–228.
58. Ohad, N., Margossian, L., Hsu, Y. C., Williams, C., Repetti, P.
and Fischer, R. L., A mutation that allows endosperm development without fertilization. Proc. Natl. Acad. Sci. USA, 1996, 93,
5319–5324.
59. Kinoshita, T., Yadegari, R., Harada, J. J., Goldberg, R. B. and
Fischer, R. L., Imprinting of the MEDEA Polycomb gene in the
Arabidopsis endosperm. Plant Cell, 1999, 11, 1945–1952.
60. Yadegari, R., Kinoshita, T., Lotan, O., Cohen, G. and Katz, A.
Choi, Y., Mutations in the FIE and MEA genes that encode interacting Polycomb proteins cause parent-of-origin effects on seed
development by distinct mechanisms. Plant Cell, 2000, 12, 2367–
2381.
61. Köhler, C., Hennig, L., Bouveret, R., Gheyselinck, J., Grossniklaus, U.
and Gruissem, W., Arabidopsis MSI1 is a component of the
MEA/FIE Polycomb group complex and required for seed development. EMBO J., 2003, 22, 4804–4814.
62. Pischke, M. S., Jones, L. G., Otsuga, D., Fernandez, D. E., Drews,
G. N. and Sussman, M. R., An Arabidopsis histidine kinase is essential for megagametogenesis. Proc. Natl. Acad. Sci. USA, 2002,
99, 15800–15805.
63. Guitton, A. E., Page, D. R., Chambrier, P., Lionnet, C., Faure, J.
E., Grossniklaus, U. and Berger F., Identification of new members
of fertilisation independent seed Polycomb group pathway involved in the control of seed development in Arabidopsis thaliana.
Development, 2004, 131, 2971–2981.
64. Grossniklaus, U., Genomic imprinting in plants: a predominantly
maternal affair. In Annual Plant Reviews: Plant Epigenetics (ed.
Meyer, P.), Blackwell, Sheffield, UK, 2005, pp. 174–200.
65. Vielle-Calzada, J.-P., Baskar, R. and Grossniklaus, U., Delayed
activation of the paternal genome during seed development. Nature, 2000, 404, 91–94.
66. Chen, S., Hofius, D., Sonnewald, U. and Börnke, F., Temporal and
spatial control of gene silencing in transgenic plants by inducible
expression of double-stranded RNA. Plant J., 2003, 36, 731–740.
67. Guo, H. S., Fei, J. F., Xie, Q. and Chua, N-H., A chemicallyregulated inducible RNAi system in plants. Plant J., 2003, 34, 383–
392.
68. Kermicle, J. L., Pleiotropic effects on seed development of the indeterminate gametophyte gene in maize. Am. J. Bot., 1971, 58, 1–7.
69. Springer, P. S., McCombie, W. R., Sundaresan, V. and Martinssen,
R. A., Gene trap tagging of prolifera, an essential Mcm2-3-5-Like
gene in Arabidopsis. Science, 1995, 268, 877–880.
1852
70. Feldmann, K. A., Coury, D. A. and Christianson, M. L., Exceptional
segregation of a selectable marker (KanR) in Arabidopsis identifies genes important for gametophytic growth and development.
Genetics, 1997, 147, 1411–1422.
71. Drews, G. N., Lee, D. and. Christensen, C. A., Genetic analysis of
female gametophyte development and function. Plant Cell, 1998,
10, 5–18.
72. Hejatko, J., Pernisova, M., Eneva, T., Palme, K. and Brzobohaty,
B., The putative sensor histidine kinase CKI1 is involved in female
gametophyte development in Arabidopsis. Mol. Genet. Genomics,
2003, 269, 443–453.
73. Capron, A. et al., The Arabidopsis anaphase-promoting complex
or cyclosome: Molecular and genetic characterization of the APC2
subunit. Plant Cell, 2003, 15, 2370–2382.
74. Ebel, C., Mariconti, L. and Gruissem, W., Plant retinoblastoma
homologues control nuclear proliferation in the female gametophyte.
Nature, 2004, 429, 777–780.
75. Acosta-Garcia, G. and Vielle-Calzada, J.-P., A classical arabinogalactan
protein is essential for the initiation of female gametogenesis in
Arabidopsis. Plant Cell, 2004, 16, 2614–2628.
76. Shi, D.-Q., Liu, J.; Xiang, Y.-H., Ye D., Sundaresan, V. and Yang,
W.C., SLOW WALKER1, essential for gametogenesis in Arabidopsis, encodes a WD40 protein involved in 18S ribosomal RNA biogenesis. Plant Cell, 2005, 17, 2340–2354.
77. Márton, M. L., Cordts, S., Broadhvest, J. and Dresselhaus, T., Micropylar pollen tube guidance by egg apparatus1 of maize. Science, 2005, 307, 573–576.
78. Evans, M. M. S. and Kermicle, J. L., Interaction between maternal
effect and zygotic effect mutations during maize seed development. Genetics, 2001, 159, 303–315.
79. Holding, D. R. and Springer, P. S., The Arabidopsis gene PROLIFERA
is required for proper cytokinesis during seed development.
Planta, 2002, 214, 373–82.
80. Choi, Y. et al., DEMETER, a DNA glycosylase domain protein, is
required for endosperm imprinting and seed viability in Arabidopsis.
Cell, 2002, 110, 33–42.
81. Kim, H. U., Li, Y. and Huang, A. H., Ubiquitous and endoplasmic
reticulum-located lysophosphatidyl acyltransferase, LPAT2, is essential for female but not male gametophyte development in
Arabidopsis. Plant Cell, 2005, 17, 1073–1089.
82. Gerasimova-Nawaschina, E. N., Fertilization as an ontogenetic
process. Bot. Zhurnal USSR, 1957, 42, 1654–1673.
83. Huang, B.-Q. and Sheridan, W. F., Embryo sac development in the
maize indeterminate gametophyte1 mutant: Abnormal nuclear behaviour and defective microtubule organization. Plant Cell, 1996,
8, 1391–1407.
ACKNOWLEDGEMENTS. Many laboratories around the world have
been working to extend our understanding of the development of the female gametophyte in angiosperms. We have made every effort to record
all the female gametophytic mutations in Arabidopsis described to date
and apologize to any colleague(s) whose work has not been covered in
this review.
CURRENT SCIENCE, VOL. 89, NO. 11, 10 DECEMBER 2005