Control of Male Gametophyte Development
Sheila McCormick1
Plant Gene Expression Center, United States Department of Agriculture, Agricultural Research Service,
and University of California Berkeley, Albany, California 94710
INTRODUCTION
In a previous review of male gametophyte development
(McCormick, 1993), it was noted that the two areas posed by
Mascarenhas (1975) as fruitful areas for future research were the
following. What are the differences in the two cytoplasms that
determine the different cell fates of the generative and vegetative
cells? And what are the functions of pollen-specific proteins?
Now, 10 years later, the genome sequences of Arabidopsis and
rice have been completed. There are extensive EST databases
for many plants and several data sets from microarray hybridizations.
There are extensive resources for disrupting the
functions of genes. The pollen research community has made
significant progress toward a deeper understanding of pollen
development using community resources as well as novel
techniques developed specifically for the analysis of pollen. This
review will provide an overview of these advances and prospects
for the future, focusing on male gametophyte development,
strictly defined as postmeiosis development, after the formation
of the haploid microspores. Meiosis will not be discussed in
detail. Anther development and the role of the tapetum in pollen
development are discussed by Dickinson and Scott in this issue.
The main features of pollen development are shown in Figure
1, which is based on an ultrastructural analysis of microsporogenesis
in Arabidopsis thaliana (Owen and Makaroff, 1995). The
male gametophyte, or pollen grain, is a three-celled organism
that is derived by stereotypical cell divisions. Our story starts
inside the anther, when sporogenous initial cells, also called
pollen mother cells, undergo meiosis to form a tetrad of cells.
Figure 2 shows Arabidopsis tetrads that have been extruded
from one anther. Each tetrad is enclosed in a thick callose wall.
The microspores in each tetrad are freed from their meiotic
brothers by the action of callase, an enzyme produced by the
tapetum. The tapetum is a nutritive cell layer that lines the locule
containing the developing microsporocytes. The tapetum disintegrates
in the later stages of pollen development. The microspores
enlarge and then each undergoes an asymmetric
mitosis. The mitosis is asymmetric because the dividing nucleus
is adjacent to the wall, and the spindle orientation is such that
after cytokinesis, one cell is much smaller than the other. The two
cells of this bicellular pollen grain have strikingly different fates.
The larger cell is called the vegetative cell, and the smaller cell is
called the generative cell. The larger vegetative cell does not
divide again but eventually will form the pollen tube. The
generative cell is engulfed inside the cytoplasm of the vegetative
cell. The generative cell undergoes mitosis, sometimes termed
a second or pollen mitosis, to form the two sperm cells. The
timing of this second pollen mitosis varies in different plant
families, sometimes occurring within the anther (as in grasses
and crucifers), although more commonly it occurs during pollen
tube growth. In most plants, mature pollen grains are released
from the anthers in a partially dehydrated state. Once on the
stigma, the pollen grains hydrate and the vegetative cell extends
a tube that grows by tip growth. As the tube extends, the
vegetative cell nucleus and the two closely associated sperm
cells move into the tube. Eventually, the entire vegetative cell
exits the pollen grain and travels at the tip of the rapidly growing
pollen tube. Pollen development is complete when the sperm
cells are released into the embryo sac.
MUTANT ANALYSES
A common way to dissect a developmental pathway is to isolate
mutants that disrupt the pathway. This approach has been used
extensively to dissect male meiosis and subsequent stages of
pollen development. Comprehensive reviews of meiotic stages
and cytology in Arabidopsis (Ross et al., 1996; Armstrong and
Jones, 2003) as well as a time course for meiotic stages
(Armstrong et al., 2003) now provide the necessary foundation
for the interpretation of phenotypes in mutants that disrupt
particular phases of meiotic development. Caryl et al. (2003)
comprehensively reviewed the mutants known to affect meiosis
in Arabidopsis. Some mutants that disrupt pollen development
affect the function of the tapetum. Because the tapetum is
sporophytic, such male-sterile mutants generally are recessive
and all of the pollen grains within an anther are affected. Mutants
are termed gametophytic if they disrupt genes that act after
meiosis, in the haploid phase of pollen development, and thus
only the pollen grains carrying the mutant allele are affected.
Male gametophytic mutants often exhibit segregation distortion
(i.e., reduced transmission through the male) if the gene that
is mutated is important for pollen development or function.
Indeed, this feature was used successfully as a first-pass
method in screens designed to identify gametophytic mutants
(Feldmann et al., 1997; Bonhomme et al., 1998; Howden et al.,
1998; Grini et al., 1999; Oh et al., 2003). The developmental stage
affected in each mutant was determined later. In some cases, the
phenotype was demonstrated to be attributable to a single gene
disruption. For example, several of the male transmissiondefective
mutants described by Bonhomme et al. (1998) had
1 To whom correspondence should be addressed. E-mail sheilamc@
nature.berkeley.edu; fax 510-559-5678.
Article, publication date, and citation information can be found at
www.plantcell.org/cgi/doi/10.1105/tpc.016659.
The Plant Cell, Vol. 16, S142­S153, Supplement 2004, www.plantcell.org  2004 American Society of Plant Biologists
Figure 1. Scheme of Microsporogenesis.
Control of Male Gametophyte Development S143
similar phenotypes, and upon further mapping and cloning they
were found to be alleles. The disrupted gene was renamed kinky
pollen because pollen grains harboring a mutant allele exhibited
aberrant tube growth (Procissi et al., 2003). However, frequently,
mutations identified from T-DNA insertion screens were unlinked
to the T-DNA insertion or exhibited complex molecular lesions. In
such cases, it was difficult to identify the molecular basis for the
transmission defect. For example, the T-DNA insertion mutant
named halfman (Oh et al., 2003) was demonstrated to have
a gametophytic defect with low penetrance. Molecular analysis
showed that the T-DNA insertion in halfman had induced an
adjacent deletion of 150 kb that encompassed 38 genes; the loss
of 1 or more of these genes might have been responsible for the
observed phenotype, in which 30% of the pollen aborted at
approximately the bicellular stage of development.
Assiduous screening for mutants based on morphology or
staining characteristics of pollen from individual M1 or M2
plants also has yielded very interesting phenotypes (Chen and
McCormick, 1996; Johnson and McCormick, 2001; Lalanne
and Twell, 2002). As illustrated in Figure 1, pollen development
undergoes stereotypical cell divisions to yield the three-celled
male gametophyte. To test whether the asymmetric division that
resulted in the formation of the vegetative cell and the generative
cell could be disrupted mutationally, or if the number of cell
divisions typical of pollen development could be disrupted
mutationally, pollen of mutagenized populations was examined
after 49,6-diamidino-2-phenylindole staining. Such screens
yielded, for example, the mutants sidecar pollen (Chen and
McCormick, 1996) and gemini pollen (Park et al., 1998). At the
mature pollen stage in sidecar pollen heterozygotes, 50% of
the pollen was normal, 43% was aborted, and 7% showed
the sidecar phenotype, namely an extra cell within the pollen
exine. Analyses of earlier stages of pollen development revealed
that microspores carrying the sidecar pollen mutant allele
frequently underwent a premature and symmetric cell division.
For unknown reasons, one of the resulting two cells then was
able to undergo the normal mitotic divisions to form the generative
cell and the two sperm cells. Despite the apparent
equal sizes of the two cells, this finding suggests that an inherent
asymmetry must persist. The molecular lesion in sidecar pollen is
still unknown.
In contrast to sidecar pollen, in gemini pollen, arrest occurs
after the first division, so that at the mature pollen stage the
affected pollen grains have two cells, each with decondensed
chromatin and each able to express a vegetative cell­specific
marker (Park et al., 1998). Gemini pollen encodes a microtubuleassociated
protein (MAP215 family), and the gemini pollen
phenotype is now believed to be attributable to a defect in the
correct positioning of the cytokinetic phragmoplast at pollen
mitosis I (Twell et al., 2002). It was hypothesized that the aberrant
partitioning of cytoplasmic factors led to the alterations in cell
fate. In another mutant, duo, the first mitotic division occurs
asymmetrically, yielding one larger cell with decondensed
chromatin and a vegetative cell fate and a smaller generative
cell with condensed chromatin. However, the second mitotic
division of the generative cell, to form two sperm cells, fails. The
molecular lesion in duo is not yet known. A gametophytic malesterile
mutant (gaMS-2) that affects this stage of development
was described in maize (Sari-Gorla et al., 1997); in microspores
carrying the mutant allele, the vegetative nucleus and the
generative nucleus exhibit altered identities and sometimes
undergo extra divisions.
Mutants that affect these division patterns also were isolated
from the T-DNA screens. For example, in limpet pollen (Howden
et al., 1998), now named ingressus (http://www.le.ac.uk/biology/
research/pollen/ing.html), the generative cell divides to form two
sperm cells, but the sperm cells cannot migrate inward and
instead remain near the pollen tube wall. More recently, Lalanne
and Twell (2002) found that the association between the sperm
cells and the vegetative nucleus could be disrupted mutationally.
In gum (germ unit malformed), the vegetative nucleus stays near
the pollen wall, although the sperm cell pair associates and
migrates inward, as in the wild type. In mud (male germ unit
displaced), the sperm cells associate with the vegetative
nucleus, as in the wild type, but this group then remains near
the pollen wall. The phenotypes in double mutants allowed these
authors to infer that GUM acts earlier than MUD.
It is usually the case that gametophytic mutants that affect
pollen development or function must be maintained as heterozygotes.
If the female gametophyte also is affected (Howden
et al., 1998), the chance of obtaining homozygous lines is slight.
However, it was possible (Chen and McCormick, 1996) to obtain
a homozygous sidecar pollen line, albeit at a much lower than
expected frequency, indicating that pollen grains carrying the
sidecar pollen mutant allele could transmit the gene to the next
generation. The gum and mud mutants (Lalanne and Twell, 2002)
show only minor transmission defects through the male, and
homozygous lines of these mutants were obtained readily.
Consequently, visual screening facilitated the discovery of these
mutants, because they would not have been detected readily in a
segregation distortion screen. If homozygous lines can be
generated, it will be possible to test for phenotypes in the
sporophyte. The gum and mud homozygotes show no obvious
vegetative phenotypes.
Figure 2. Tetrads of Arabidopsis.
Male meiosis is essentially synchronous within an anther.
S144 The Plant Cell
Identifying additional alleles is a standard approach that
facilitates the positional cloning of genes. For a pollen mutant
whose phenotype is visible only after microscopic examination,
this task is not easy. Additional visual screens of mutagenized
populations will undoubtedly yield additional phenotypes of
some sort, but there is no guarantee that the desired phenotype
will be seen in the next 5000 plants screened. However,
screening for new alleles can be simplified for certain mutants
with striking phenotypes. For example, in anthers of raring-to-go
(rtg) (Johnson and McCormick, 2001), some of the pollen stains
with decolorized aniline blue, a stain specific for callose,
a component of the pollen tube wall. Because pollen does not
normally hydrate and begin germination until it contacts the
stigma, finding the precocious pollen germination phenotype in
rtg was serendipitous. To obtain additional rtg alleles, mutagenized
populations were screened by collecting pollen from 100
plants at a time and screening the pollen in bulk after staining with
decolorized aniline blue. This directed screen yielded several
probable alleles of rtg as well as other mutants (gift-wrapped
pollen, polka dot pollen, and emotionally fragile pollen) whose
pollen grains exhibited unusual staining patterns; the existence
of these staining patterns could not have been predicted
(Johnson and McCormick, 2001).
There are undoubtedly more mutants to find from such visual
screens. Libraries that generate cDNA­green fluorescent protein
(GFP) fusions can provide hints about the normal locations of
proteins of unknown function (Cutler et al., 2000). A highthroughput
screen for GFP localization in leaves that had been
infected transiently with a cDNA-GFP library was remarkably
successful at identifying proteins that are associated with
plasmodesmata (Escobar et al., 2003). Unfortunately, these
existing resources do not provide tools that can be used to study
subcellular locations for pollen proteins; Cutler et al. (2000) used
the 35S promoter of Cauliflower mosaic virus (CaMV), which is
expressed poorly during later stages of male gametophyte
development, whereas Escobar et al. (2003) used viral infection
to deliver the constructs to leaves. Nonetheless, it should be
straightforward to generate new transformed lines using a pollenexpressed
promoter driving cDNA-GFP libraries. In also might
prove fruitful to generate transgenic lines harboring GFP fusions
to selected pollen proteins and then to mutagenize these lines
and screen for mislocalization of the GFP fusion protein. Such
a screen was performed for mutants that mislocalized a GFP
fusion protein of the double-stranded RNA binding protein
Staufen during Drosophila embryo development (Martin et al.,
2003).
QUARTET AND TETRASPORE, USEFUL TOOLS
FOR MUTANT ANALYSIS
Two mutants in Arabidopsis, quartet1 and tetraspore, have
greatly facilitated the analysis of male gametophytic mutations.
quartet1 is a sporophytic recessive mutation. In quartet1/
quartet1 plants, all of the products of a single meiosis are held
together in a tetrad throughout pollen development (Preuss et al.,
1994), but each pollen grain is otherwise normal and can
germinate. Tetrad analysis is used to test whether pollen phenotypes
result from a gametophytic mutation or from a dominant
sporophytic mutation. This test is important because many
mutants that affect pollen phenotypes exhibit variable expressivity
and penetrance: an observation that 50% of the pollen
has an altered phenotype is not, in and of itself, proof of gametophytic
action. To perform tetrad analysis, mutant/1 plants
are crossed as females to homozygous quartet1 plants. Those
F1 progeny whose pollen exhibit the mutant phenotype are
self-pollinated, and the F2 progeny are scored to identify the
mutant/1 quartet1/quartet1 double mutants. If a mutation is
expressed sporophytically but has low expressivity, it might be
expected that the numbers of normal and affected pollen
resulting from meiosis would vary in each tetrad. However, if
a mutation is gametophytic, the ratio of normal to affected pollen
resulting from meiosis should be 2:2. Figures 3A to 3D show an
example of such a test: after staining with aniline blue, two of the
pollen grains in the tetrad are wild type and the other two show
tubules within the grain that are diagnostic of the gift-wrapped
pollen1 phenotype (Johnson and McCormick, 2001).
Because pollen is haploid, it is not straightforward to determine
if a mutation is dominant or recessive. To combine
a mutant allele with a wild-type allele in one cell for a dominance/
recessiveness test, the mutant can be crossed, as female, with
diploid pollen from a tetraploid plant. However, the resulting F1
progeny are triploid. Pollen from triploid plants typically have
a high degree of pollen abortion that might interfere with scoring
for the pollen phenotype of interest. Another alternative is to
introgress the mutant allele into a tetraploid background and
then examine the phenotype in the diploid pollen produced by
the tetraploid plant. This method has been used infrequently
(Kamps et al., 1996) in maize. A third option, in Arabidopsis, is to
cross the pollen mutant of interest with tetraspore. Because
plants that are homozygous for a tetraspore mutant allele fail to
undergo cytokinesis after meiosis, large multinucleate pollen
grains are formed (Hulskamp et al., 1997; Spielman et al., 1997;
Yang et al., 2003). This feature provides a way to determine
the dominance relationships of male gametophytic genes. In
a tetraspore (tes/tes) homozygote that also is heterozygous for
the gametophytic gene being tested, pollen grains carrying both
a mutant allele and a wild-type allele will exist.
Figure 3E shows anthers of polka dot pollen heterozygous
plants after staining with decolorized aniline blue, a stain specific
for callose (Johnson and McCormick, 2001); 50% of the pollen
grains show blobs of callose. A polka dot pollen heterozygote
(pdp/1) was crossed into a tes/tes line. The F1 plants had
normal-sized pollen, because tes is a sporophytic recessive
mutant. Pollen from each F1 plant were examined, and plants
with the pdp phenotype were self-pollinated. Then, tes/tes plants
were identified among the F2 progeny. If the mutant phenotype is
not observed in the large pollen grains, then the gametophytic
mutation is an effective recessive, because the wild-type allele
can compensate. However, some of the large pollen grains in the
tes/tes 1/pdp plants showed blobs of callose (Figures 3F and
3G). Because the mutant phenotype for pdp is observed in the
large pollen grains, we infer that the lesion in pdp is unlikely to be
a loss-of-function mutation. Instead, the mutant allele might
encode a nonfunctional protein that acts as a dominant negative,
or it might be a gain-of-function mutation. That a lesion in
a candidate gene is responsible for the observed mutant
Control of Male Gametophyte Development S145
phenotype usually is confirmed by transformation. For gametophytic
mutants, crosses with tes are useful because for transformation
it is important to know whether to introduce a wild-type
copy of the gene into the mutant background or to introduce the
mutant version of the gene into the wild-type background.
DETERMINING PROTEIN FUNCTION
In addition to mutant screens, there are several alternative
ways to determine the roles of genes that are expressed in the
male gametophyte. These include the analysis of T-DNA insertion
lines (Alonso et al., 2003) in particular genes and analysis of
RNA interference or antisense lines for particular transcripts (for
example, Muschietti et al., 1994; Gupta et al., 2002). For
example, because calcium levels and calmodulin were known
to be important for pollen function, it was reasonable that
Golovkin and Reddy (2003) chose to study the Arabidopsis
homologs of a pollen-specific calmodulin binding protein (no
pollen germination1 [NPG1]) from maize. There are three closely
related members in the Arabidopsis genome; NPG1 was
expressed only in pollen, whereas the two related genes were
expressed in pollen as well as in other tissues. Homozygotes for
the T-DNA insertion in NPG1 could not be obtained. Reciprocal
crosses indicated that there was a problem with transmission
through the male. Further examination revealed that although
NPG1 was not required during pollen development, it was
essential for pollen germination. Another example involves
GTPases. An important role for Rop (Rho-like GTPases of plants)
proteins in pollen tube growth was discovered by analyzing lines
in which mutant versions of Rop were expressed in pollen
(reviewed by Yang, 2002). A surprise came from the analysis of
a maize mutant that is disrupted in Rop2, one of two very closely
related Rops that are expressed in maize pollen (Arthur et al.,
2003). The rop2 mutant allele is poorly transmitted through the
male in heterozygotes, implying that the mutant pollen is at
a competitive disadvantage. Surprisingly, pollen tube growth in
the mutant appeared normal. Further analysis might reveal
a previously unsuspected role for Rops, perhaps during targeting
of the pollen tube to the ovule.
Such approaches have serendipitously allowed the identification
of gene products that must play a crucial, but previously
unpredicted, role in pollen development. For example, Kang et al.
(2003) found that they could not identify homozygous T-DNA
insertions in a dynamin-like protein, ADL1C, and subsequently
noticed that in heterozygotes, 50% of the pollen grains were
collapsed and failed to germinate. Further examination revealed
that this defect was first evident during microspore maturation
and that the affected microspores appeared to have plasma
membrane defects. Further work will be required to determine
precisely why the absence of ADL1C leads to pollen abortion.
Another recent example involves apyrases (Steinebrunner et al.,
2003). Apyrases are known to hydrolyze nucleoside phosphates;
in animals, they are important in several signaling pathways, but
their roles in plants are not well understood. In an effort to study
the roles of apyrases in plants, gene disruptions were generated.
There are two apyrase genes in Arabidopsis, and disruption of
either gene alone had no discernible phenotype. However, the
double knockout could not be obtained, because pollen grains
carrying both disrupted alleles could not germinate. Again,
further work will be required to determine what processes during
pollen germination are affected by the absence of apyrases.
In many multigene families, only a few isoforms are expressed
in pollen, and mutant phenotypes can sometimes be obtained
Figure 3. Crosses with quartet1 or tetraspore: Useful Tools for the Analysis of Gametophytic Mutants.
(A) to (D) Tetrads from a giftwrapped pollen1/1 quartet1/quartet1 plant.
(E) Pollen grains within an anther of a polka dot pollen/1 plant.
(F) and (G) Large pollen grains from a pdp/1 tes1/tes1 plant.
S146 The Plant Cell
even when the expression of one isoform might in principle
compensate for the lack of another (Arthur et al., 2003; Golovkin
and Reddy, 2003). Sometimes, however, redundancy does
thwart efforts to obtain a phenotype and thus infer a function
for a protein. For example, there are several monosaccharide
transporters that are expressed late during pollen maturation;
a homozygous T-DNA gene disruption in one of them, AtSTP6,
showed no apparent defects in pollen germination or transmission
through the male (Scholz-Starke et al., 2003). Does this
mean that monosaccharide transporters (or any other genes
whose mutants lack a discernible phenotype) are not important?
Not necessarily. Edelman and Gally (2001) suggest that gene
degeneracy should be considered in addition to redundancy,
that important developmental processes may be ``covered'' by
several proteins, any of which can accomplish the task in the
absence of another.
GENE REGULATION
Two groups (Becker et al., 2003; Honys and Twell, 2003) have
recently used microarrays to analyze gene expression patterns in
mature pollen of Arabidopsis. The results are not yet global (only
8000 of the predicted 27,000 genes of Arabidopsis were
analyzed), but they can be extrapolated to the whole genome.
The two studies differ quantitatively and qualitatively, probably
because of differences in sample preparation and methods of
data analysis. Nonetheless, it is noteworthy that the estimates for
the numbers of genes expressed in pollen and for the numbers of
genes that might be pollen specific are similar to those estimated
from cDNA­poly(A) RNA hybridization kinetics (Mascarenhas,
1975, 1990, 1993). Microarray experiments are extremely useful
in identifying targets for further analyses, especially for genes
currently annotated as hypothetical or of unknown function. But
such experiments are only a start and must be confirmed by
other independent tests. Many transcripts have low abundance
and their expression will not be assayable by global expression
methods such as microarray analysis. Indeed, for several genes
called ``Absent'' in the microarray experiments (Becker et al.,
2003; Honys and Twell, 2003), their probable homologs are
indeed present in tomato pollen cDNA libraries (Tang et al., 2002;
Van der Hoeven et al., 2002) or in a maize sperm cell cDNA library
(Engel et al., 2003). A different approach, serial analysis of gene
expression (SAGE), was used by Lee and Lee (2003) to obtain
gene expression data from pollen from Arabidopsis plants grown
under normal or chilling conditions in an effort to understand why
plants raised under cold conditions exhibit impaired male fertility.
Although most transcript levels were not affected by cold
treatment, the SAGE analysis provided a possible explanation
for why pollen of some species is not cold tolerant: transcripts
involved in cold tolerance in sporophytic cells were poorly
represented in pollen mRNA. In plants for which there are limited
sequence data, techniques such as cDNA­amplified fragment
length polymorphism transcript profiling have proved useful for
identifying genes for further study. For example, Cnudde et al.
(2003) used this technique to characterize gene expression
patterns in Petunia anthers at different developmental stages.
What are the roles of the mRNAs that are present in pollen?
Based on results from early experiments with transcriptional and
translational inhibitors, it was widely accepted that pollen would
contain mRNAs that are synthesized during pollen development
and stored for translation during pollen germination (for review,
see Mascarenhas, 1993). However, many proteins corresponding
to such mRNAs already are present in mature, ungerminated
pollen, as was discovered once antibodies were generated
(Muschietti et al., 1994, 1998; Kim et al., 2002). The results from
the microarray experiments (Becker et al., 2003; Honys and
Twell, 2003) indicate that mRNAs that encode signal transduction
and cell wall biosynthesis proteins were highly represented,
whereas transcription and translation proteins were
underrepresented. SAGE analysis (Lee and Lee, 2003) confirmed
that mRNAs that encode cell wall biogenesis were highly
expressed in pollen. The potential roles for one category of
genes expressed prevalently in pollen were tested collectively
by Lalanne et al. (2004) by examining the pollen phenotypes in
T-DNA insertion lines that disrupted the SETH1 and SETH2
genes. SETH1 and SETH2 encode proteins required for the
addition of glycosylphosphatidylinositol (GPI) anchors to proteins.
Lalanne et al. (2004) found that preventing the addition of
GPI anchors to proteins resulted in defects in pollen tube growth;
further studies now can be focused on determining precisely
which GPI-anchored proteins are important and why. Similar
experiments (Alfieri et al., 2003) were used to show that GPIanchored
proteins were important for sperm­egg adhesion in
mammals.
A few groups have tried to identify mRNAs that are specific to
the germination phase of pollen development. In Petunia,
flavonoids are required for germination. Guyon et al. (2000) took
advantage of this requirement to precisely control the initiation of
germination in flavonoid-deficient pollen. Thus, they performed
a differential screen for cDNAs that were transcribed or
upregulated soon after the addition of flavonoids to the
germination medium. They identified 20 such cDNAs; almost
all were pollen specific, and many corresponded to lowabundance
mRNAs. Despite such intricate cDNA selection
schemes (Guyon et al., 2000; see also Mu et al., 1994), it has
been difficult to identify genes that are transcribed only upon
pollen germination or to identify mRNAs that are essentially not
translated until after pollen germination (Wittink et al., 2000).
Given that most mRNAs are presynthesized before pollen
maturation, it is curious that only a few studies have examined
the potential for post-transcriptional control of gene expression
during pollen development. Ylstra and McCormick (1999) tested
the long-held assumption that pollen mRNAs were long-lived.
For 10 messages, they confirmed that the mRNA half-life was
very long. However, two mRNAs were relatively short-lived in
pollen, and an mRNA that was very unstable in somatic cells was
relatively stable in pollen. Bate et al. (1996) used transient
expression assays and stable transgenic plants to demonstrate
that the 59 untranslated region (UTR) of the late anther tomato52
(LAT52) gene acted as a translational enhancer in pollen. No
significant increase in translational efficiency was seen when
constructs with the LAT52 59 UTR were introduced into
sporophytic cells. In pollen, the translational enhancement was
maximal during the last stages of pollen development. Although
LAT52 transcripts can be detected at approximately the time of
first microspore mitosis (Eady et al., 1994), there might be
Control of Male Gametophyte Development S147
mechanisms in place to prevent significant accumulation of late
gene products such as LAT52 until near pollen maturity. Bate
et al. (1996) suggested that pollen-specific translation initiation
factors, such as eiF-4a (Brander and Kuhlemeier, 1995), might
play a role in this regulation. Curie and McCormick (1997) studied
the 59 UTR of the LAT59 gene, which, when removed from
constructs, enhanced transgene expression in transient assays.
In this case, the effect was not pollen specific: including the
LAT59 59 UTR in gene constructs reduced gene expression of the
reporter gene in both pollen and in somatic cells.
The most detailed analyses of pollen-specific promoters were
performed with the LAT52 (Eyal et al., 1995; Bate and Twell,
1998) and the LAT59 (Eyal et al., 1995) genes. Gain-of-function
and loss-of-function gene constructs were tested by transient
assays in pollen and in somatic cells as well as in stably
transformed plants. In this way, 30-bp elements that could confer
pollen specificity to a heterologous promoter (CaMV35S) were
defined. Competition experiments using constructs with concatamers
of these elements suggested that the promoter
elements might compete for binding to a shared trans-acting
factor. We still do not know what transcription factors bind to
these identified promoter elements, so our understanding of the
determinants for pollen-specific gene expression is incomplete.
It may be informative to use microarrays to identify all of the
genes with similar expression profiles in pollen and then use
bioinformatics tools (Rombauts et al., 2003) to examine their
promoter regions for common elements. However, the critical
sequence elements identified in the tomato LAT59 promoter
(Eyal et al., 1995) are not well conserved in the promoter of Nt59,
its tobacco homolog (Kulikauskas and McCormick, 1997). This
result suggests that sequence identities between promoter
regions might not be sufficient to explain the regulation of
pollen-expressed genes.
The lack of defined transcription factors for well-characterized
pollen-specific promoters does not mean that pollen-specific
transcription factors are not known. For example, Zachgo et al.
(1997) characterized a pollen-specific MADS box transcription
factor in Antirrhinum, as did Heuer et al. (2000) in maize; because
expression of the maize MADS box gene continued during pollen
tube growth, it was surmised that this factor might be needed
during the later phases of pollen tube growth. Lohrmann et al.
(2001) demonstrated the pollen-specific expression of a transcription
factor that regulates nuclear genes required for
mitochondrial function. Kobayashi et al. (1998) reported that
seven different zinc-finger transcription factors in Petunia were
expressed transiently and sequentially at different stages of
pollen development. They suggested that such transcription
factors might constitute a regulatory cascade and that each
might have specific target genes, including the next-acting
transcription factor in the cascade.
The extensively characterized LAT52 promoter (Twell et al.,
1991; Eyal et al., 1995) is the workhouse for gene expression
analyses in pollen (Muschietti et al., 1994; Lee et al., 1996;
Cheung, 2001). However, there is a need for other promoters of
differing strengths, and certain experimental designs require
multiple promoters. For example, coexpression of a reporter
gene probably should not use the same promoter as the gene
whose function is being assayed, in case the transcription factors
required for promoter activation are limiting. Gene expression
driven by the LAT52 promoter can be detected in uninuclear
microspores in Arabidopsis (Eady et al., 1994), and this promoter
drives strong gene expression at all developmental stages
occurring after the bicellular pollen stage, even after delivery
into mature pollen via particle bombardment (Twell et al., 1991;
Eyal et al., 1995). If a narrower window of gene expression of
a transgene is desired, the LAT52 promoter may not be the best
choice. For example, Lee et al. (1996) were unable to assess
whether the pollen receptor kinase PRK1 played a role during
pollen germination, because antisense expression driven by the
LAT52 promoter resulted in pollen abortion at the unicellular
microspore stage.
Several groups have made progress in filling in the gaps
for gene expression at other critical stages during pollen
development. For example, Klimyuk and Jones (1997) characterized
a meiosis-specific promoter, AtDMC1, and showed that it
directs reporter gene expression exclusively in meiocytes.
Custers et al. (1997) characterized the promoter of the tobacco
NTM19 gene and showed that it will direct gene expression only
in microspores. They also characterized the Bp4 promoter and
showed that it was active a bit later, when bicellular pollen is
being formed. The transcripts for the tomato receptor kinases
LePRK1, LePRK2, and LePRK3 are detected only in mature
pollen (Muschietti et al., 1998; Kim et al., 2002). Although not yet
tested, it is likely that these promoters, or the promoters of their
Arabidopsis homologs (Kim et al., 2002), might provide a way
to express transgenes only at the latest stages of pollen development.
Numerous other candidates for precisely timed
promoters may become obvious as microarray analyses (Becker
et al., 2003; Honys and Twell, 2003) are extended to other stages
of pollen development. Despite the small size of the anthers in
which meiosis occurs, pollen mother cells can be extruded from
anthers at the appropriate stage. It is reasonable to anticipate
that molecular analyses of isolated meiocytes will be achievable
and that analyses with such material might eventually provide
promoters that will be suitable for directing gene expression at
precise stages of meiotic development.
Molecular analyses of the generative cell were first accomplished
with lily. Lily has a large generative cell that is easily
isolated free of the vegetative cell, and generative cell­specific
histones were characterized (Ueda and Tanaka, 1995; Xu et al.,
1999a; Ueda et al., 2000). Xu et al. (1999b) used a cDNA library
prepared from lily generative cells to isolate and characterize
a generative cell­specific gene, lily generative cell1 (LGC1). The
LGC1 protein was localized to the surface of the generative cell
and sperm cells, suggesting that it might play a role in gamete
interactions. Subsequently, Singh et al. (2003) characterized the
LGC1 promoter in transgenic tobacco plants. From their
preliminary deletion analyses, they suggest that generative
cell­specific expression driven by the LGC1 promoter is attributable
to a repressor element, because when the promoter
fragment was truncated, the reporter gene was expressed more
widely. Xu et al. (2002) constructed a cDNA library from partially
purified tobacco sperm and, after differential screening of this
library and analysis of 400 cDNAs, reported that two cDNAs
were sperm specific: one was similar to a polygalacturonase and
one was similar to a protein of unknown function (At3g23860).
S148 The Plant Cell
More recently, Engel et al. (2003) sequenced 5000 ESTs from
a cDNA library prepared from sperm of maize that had been
purified away from vegetative cytoplasm by fluorescenceactivated
cell sorting and used reverse transcriptase­mediated
PCR and in situ hybridizations to characterize the expression
patterns for some of these mRNAs.
For 4% of the maize sperm ESTs, the best database
matches were to proteins that were annotated as hypothetical
(i.e., not in other available EST databases). We reasoned that
those might be good candidates for sperm-specific messages.
Accordingly, the promoters of the Arabidopsis homologs were
tested (M. Engel and S. McCormick, unpublished data) to determine
if any would confer sperm-specific expression to a reporter
gene, eGFP. In transgenic lines harboring one such construct,
the two sperm cells are brightly fluorescent. Figure 4 shows
a pollen grain from one line: note that one sperm has a long
cytoplasmic extension that connects it to the vegetative nucleus
(M. Engel and S. McCormick, unpublished data). A cytoplasmic
extension between one of the two sperm cells and the vegetative
nucleus was already documented from transmission electron
microscopy analyses in Plumbago (Russell, 1984), and extensions
of the vegetative nucleus toward one of the two sperm cells
are visible in 49,6-diamidino-2-phenylindole­stained pollen
(Lalanne and Twell, 2002). It will be interesting to determine if
the sperm cell that is ``in communication'' with the vegetative
nucleus is destined for a particular fusion partner (egg or central
cell), a situation that has been documented in Plumbago (Russell,
1985). Genetic evidence for fusion partner selectivity comes from
studies of maize lines that carry supernumerary (so-called B)
chromosomes. In maize lines carrying B chromosomes, the B
centromere frequently undergoes nondisjunction at the second
mitosis, so that one sperm cell of a pair acquires two B
centromeres and the other acquires none (Rusche et al., 1997).
Genetic markers linked to the B centromere revealed preferential
transmission to the embryo. But Faure et al. (2003) were not able
to discern any preference of the B chromosome­containing
sperm when sperm pairs from one pollen grain were used for in
vitro fertilization. However, when sperm are isolated from pollen
grains, connections to the vegetative nucleus are necessarily
disrupted. Sperm cytoplasm that is marked with eGFP, in conjunction
with improvements for live-cell imaging (Feijo and Cox,
2001), should allow a more detailed analysis of sperm dynamics
at all stages of their transit through the pollen tube for delivery
to the embryo sac.
Given that generative cells and sperm cells are thought to be
relatively quiescent transcriptionally, it is unclear why sperm
have diverse mRNAs (Engel et al., 2003). Perhaps some of
these mRNAs are not translated until they are delivered to the
central cell or the egg cell. In Arabidopsis, numerous genes have
been tested for expression from the paternal genome, and most
(Vielle-Calzada et al., 2000, 2001) but not all (Baroux et al., 2001;
Weijers et al., 2001) were not expressed from the paternal
genome until a few days after fertilization. However, Scholten
et al. (2002) used a transgenic line of maize harboring a CaMV35S
promoter driving GFP expression to test whether there was
expression from the paternal allele soon after fertilization. There
were no detectable GFP transcripts in sperm of this line, but
when sperm carrying the transgene were fused in vitro with wildtype
eggs, paternally derived GFP transcripts were detectable
within the first few hours after fertilization and GFP was detectable
soon thereafter. It will be important to test how widespread
such paternal expression might be, preferably using endogenous,
rather than virally derived, promoters to drive transgene
expression.
Although transcriptome analyses are useful (Becker et al.,
2003; Honys and Twell, 2003) for selecting candidates for functional
analyses, it is equally important to determine the protein
complement of the cells of the male gametophyte. To begin to
develop information about the proteins present at different
stages of pollen development, Kerim et al. (2003) used twodimensional
gel electrophoresis and mass spectrophotometry to
analyze the protein complement of rice anthers. This study was
not comprehensive, but it was able to identify 150 protein spots
that varied between different stages of development. Because
multiple isoforms for several of these proteins were found, the
authors inferred that post-translational modifications might play
a significant role during pollen development. Proteomics efforts
also are warranted in plants whose genomes are not yet
sequenced, because there are extensive EST databases available
from male gametophytes of several crops (for example, Van
der Hoeven et al., 2002). Nonetheless, for many plants, it will be
difficult to collect large quantities of male gametophytes or of
their component cells, so future proteomics efforts probably will
Figure 4. Arabidopsis Pollen Grain from a Line Expressing a Sperm
Promoter:eGFP Construct.
Image is a maximum projection of the central Z-sections acquired with
a two-photon microscope. The GFP image is inverted to more clearly
show that one of the two sperm cells has a long cytoplasmic extension
that appears to partially encircle the vegetative cell nucleus.
Control of Male Gametophyte Development S149
use more sensitive methods (Koller et al., 2002) that do not rely
on the need for two-dimensional gels.
EXPECTATIONS FOR THE NEXT 10 YEARS
Because comparative studies had indicated that many of the
structural and molecular aspects of microsporogenesis are
conserved throughout evolution (McCormick, 1993), a concerted
molecular, morphological, and cell biological description of
microsporogenesis in a few plant species was suggested
(McCormick, 1993) to provide a framework for future studies. It
also was suggested that we take advantage of the increasing
knowledge and methodology available from nonplant organisms
to make more rapid progress in understanding meiosis, cell fate
determinants, and cell­cell interactions. Such efforts have paid
off. Researchers studying pollen can select a plant species most
suitable for the question they are asking and then rather easily
move to other species to apply different methodologies. And new
methods continue to be developed or adapted. For example,
Touraev et al. (1997) pioneered the biolistic introduction of gene
constructs into unicellular microspores of tobacco, followed by
in vitro maturation of these microspores to mature pollen. When
the resulting pollen grains were used for pollination, some
transgenic plants were obtained, although the efficiency was
quite low (<103). Aziz and Machray (2003) recently optimized
this protocol for tobacco and reported up to 15% transgenic
progeny. If this efficiency can be achieved routinely, it will be an
attractive way to generate transgenic plants, especially if in vitro
maturation of microspores can be developed for other plant
species.
The ambitious goal of the NSF2010 project is to elucidate the
functions of all of the proteins of Arabidopsis by the year 2010.
Can we meet this goal for the proteins in the male gametophyte?
From gene disruptions and the other methods discussed in this
review, we have learned about particular proteins that are
important for pollen development. But much of our knowledge is
still discrete; how various proteins interact is largely unknown. In
some fully sequenced organisms, protein­protein interaction
networks are beginning to be elucidated (reviewed by Zhu and
Snyder, 2002). Yeast two-hybrid screens have not yet been used
extensively in pollen biology. But where such screens have been
performed (Skirpan et al., 2001; Tang et al., 2002), it is already
clear that some proteins identified from such screens also are
pollen specific. Thus, signaling pathways already thought to be
elucidated in sporophytic cells might have different players in the
male gametophyte, and proteins that are expressed widely might
have unique roles in the male gametophyte.
Two recent studies (Levsky et al., 2002; Jongeneel et al., 2003)
offer inspirations for the future: truly global expression analyses
for all of the developmental stages of the male gametophyte, and
true single-cell analysis of the male gametophyte transcriptome.
Jongeneel et al. (2003) used the sensitive technique called
massively parallel signature sequencing to analyze gene expression
in two different mammalian cell lines. They found that
almost 60% of the genes were expressed at such a low level that
they would not have been detected by standard techniques such
as microarray analysis. Furthermore, the complexity of their
expression profiles was limited; one cell line expressed 10,000
genes, and the other expressed 15,000 genes. And there were
2000 mRNAs identified that corresponded to no known (i.e.,
annotated) genes. Most methods used to analyze gene expression
necessarily rely on RNA samples prepared from many cells.
Levsky et al. (2002) pioneered a method to examine gene
expression of multiple genes at the true single-cell level. In this
method, sites of de novo transcription are visualized by
fluorescence in situ hybridization; up to 11 genes can be assayed
at once. By using spectrally distinct fluorophore labels for the
gene-specific probes, the authors were able to have a unique
bar code identify each transcript. In addition to transcriptional
changes, a gene expression difference deduced from a microarray
experiment encompasses the abundance of the message
before the start of the experiment and the message stability over
the course of the experiment. Thus, our lesson is that microarray
analyses (Becker et al., 2003; Honys and Twell, 2003) are only
a start to a global understanding of male gametophyte development.
Single-cell transcriptome analyses might even allow the
detection of subtle differences in gene expression in different
pollen grains in a tetrad. The recent availability of GFP-tagged
sperm (Figure 4) offers the promise of global transcriptome (and
proteome) analyses of the gametes, assuming that large enough
quantities can be purified for such analyses.
In the last 10 years, more and more emphasis has been put on
model systems such as Arabidopsis. But there is still a place for
nonmodel systems. Some very interesting phenomena are not
known or their analysis is not tractable in the well-studied model
plants. As just one example, a type of segregation distortion
mediated by so-called gametocidal genes is known in several
species (wheat, maize, rice, and tomato). In wheat lines that are
hemizygous for a gametocidal gene (Gc2/), gametophytes that
lack the Gc2 allele undergo DNA fragmentation, presumably
induced during meiosis by allelic interaction with the Gc2 allele.
Friebe et al. (2003) recently obtained an ethyl methanesulfonateinduced
mutant at Gc2; the mutant plants are fully fertile because
the () gametophytes do not die. Map-based cloning of Gc2, and
molecular analyses of this interesting phenomenon, should soon
be possible.
That vertebrates have >1000 genes for olfactory receptor
proteins was no surprise, but that some olfactory receptors
were expressed in human spermatogenic cells was considered
bizarre. Now, Spehr et al. (2003) have shown that one of these
odorant receptors mediates human sperm chemotaxis in vitro,
detecting an odorant similar to a compound found in rose petals.
Kim et al. (2003) recently demonstrated that a small protein that is
secreted from the style of lily flowers acts as a chemoattractant
for pollen tube growth. There is every reason to believe that plant
reproductive biology will tempt us with such surprises for years
to come.
ACKNOWLEDGMENTS
I thank all of the members of my laboratory for discussion, especially
Sheila Johnson. I thank Dior Kelley for the lovely artwork in Figure 1.
Thanks to Ming Yang (Oklahoma State University, Stillwater) for Figure
2. Thanks to Sheila Johnson for Figure 3. Thanks to Michele Engel for
characterizing the sperm-specific promoter and generating the transgenic
lines, and to Jose Feijo, Nuno Moreno, and Leonor Boavida
S150 The Plant Cell
(Institute Gulbenkian Ciencia, Oeiras, Portugal) and Paul Herzmark
(University of California San Francisco) for their help in acquiring and
processing the two-photon microscope image shown in Figure 4. I thank
Betty Lord (University of California Riverside) and Loverine Taylor
(Washington State University, Pullman) for comments on the manuscript.
I dedicate this article to the memory of my wonderful parents,
Harold and Lorraine McCormick. The work in my laboratory is supported
by U.S. Department of Agriculture, Agricultural Research Service
Current Research Information System Grant 5335-21000-020-00D and
by National Science Foundation Plant Genome Grant 0211742.
Received August 25, 2003; accepted January 17, 2004.
REFERENCES
Alfieri, J.A., Martin, A.D., Takeda, J., Kondoh, G., Myles, D.G., and
Primakoff, P. (2003). Infertility in female mice with an oocyte-specific
knockout of GPI-anchored proteins. J. Cell Sci. 116, 2149­2155.
Alonso, J.M., et al. (2003). Genome-wide insertional mutagenesis of
Arabidopsis thaliana. Science 301, 653­657.
Armstrong, S.J., Franklin, F.C.H., and Jones, G.H. (2003). A meiotic
time course for Arabidopsis thaliana. Sex. Plant Reprod. 16, 141­149.
Armstrong, S.J., and Jones, G.H. (2003). Meiotic cytology and
chromosome behaviour in wild-type Arabidopsis thaliana. J. Exp.
Bot. 54, 1­10.
Arthur, K.M., Vejlupkova, Z., Meeley, R.B., and Fowler, J.E. (2003).
Maize ROP2 GTPase provides a competitive advantage to the male
gametophyte. Genetics 165, 2137­2151.
Aziz, N., and Machray, G.C. (2003). Efficient male germ line transformation
for transgenic tobacco production without selection. Plant
Mol. Biol. 51, 203­211.
Baroux, C., Blanvillain, R., and Gallois, P. (2001). Paternally inherited
transgenes are down-regulated but retain low activity during early
embryogenesis in Arabidopsis. FEBS Lett. 509, 11­16.
Bate, N., Spurr, C., Foster, G.D., and Twell, D. (1996). Maturationspecific
translational enhancement mediated by the 59-UTR of a late
pollen transcript. Plant J. 10, 613­623.
Bate, N., and Twell, D. (1998). Functional architecture of a late pollen
promoter: Pollen-specific transcription is developmentally regulated
by multiple stage-specific and co-dependent activator elements.
Plant Mol. Biol. 37, 859­869.
Becker, J.D., Boavida, L.C., Carneiro, J., Haury, M., and Feijo, J.A.
(2003). Transcriptional profiling of Arabidopsis tissues reveals the
unique characteristics of the pollen transcriptome. Plant Physiol. 133,
713­725.
Bonhomme, S., Horlow, C., Vezon, D., de Laissardiere, S., Guyon,
A., Ferault, M., Marchand, M., Bechtold, N., and Pelletier, G.
(1998). T-DNA mediated disruption of essential gametophytic genes in
Arabidopsis is unexpectedly rare and cannot be inferred from
segregation distortion alone. Mol. Gen. Genet. 260, 444­452.
Brander, K.A., and Kuhlemeier, C. (1995). A pollen-specific DEAD-box
protein related to translation initiation factor eIF-4a from tobacco.
Plant Mol. Biol. 27, 637­649.
Caryl, A.P., Jones, G.H., and Franklin, F.C. (2003). Dissecting plant
meiosis using Arabidopsis thaliana mutants. J. Exp. Bot. 54, 25­38.
Chen, Y.-C.S., and McCormick, S. (1996). sidecar pollen, an
Arabidopsis thaliana male gametophytic mutant with aberrant cell
divisions during pollen development. Development 122, 3243­3253.
Cheung, A. (2001). Imaging elongating pollen tubes by green
fluorescent protein. Sex. Plant Reprod. 14, 9­14.
Cnudde, F., Moretti, C., Porceddu, A., Pezzotti, M., and Gerats, T.
(2003). Transcript profiling on developing Petunia hybrida floral
organs. Sex. Plant Reprod. 16, 77­85.
Curie, C., and McCormick, S. (1997). A strong inhibitor of gene
expression in the 59 untranslated region of the pollen-specific LAT59
gene of tomato. Plant Cell 9, 2025­2036.
Custers, J.B., Oldenhof, M.T., Schrauwen, J.A., Cordewener, J.H.,
Wullems, G.J., and van Lookeren Campagne,, M.M. (1997).
Analysis of microspore-specific promoters in transgenic tobacco.
Plant Mol. Biol. 35, 689­699.
Cutler, S.R., Ehrhardt, D.W., Griffitts, J.S., and Somerville, C.R.
(2000). Random GFP::cDNA fusions enable visualization of subcellular
structures in cells of Arabidopsis at a high frequency. Proc. Natl.
Acad. Sci. USA 97, 3718­3723.
Eady, C., Lindsey, K., and Twell, D. (1994). Differential activation and
conserved vegetative cell-specific activity of a late pollen promoter in
species with bicellular and tricellular pollen. Plant J. 5, 543­550.
Edelman, G.M., and Gally, J.A. (2001). Degeneracy and complexity in
biological systems. Proc. Natl. Acad. Sci. USA 98, 13763­13768.
Engel, M., Chaboud, A., Dumas, C., and McCormick, S. (2003).
Sperm cells of Zea mays have a complex complement of mRNAs.
Plant J. 34, 697­707.
Escobar, N.M., Haupt, S., Thow, G., Boevink, P., Chapman, S., and
Oparka, K. (2003). High-throughput viral expression of cDNA­green
fluorescent protein fusions reveals novel subcellular addresses and
identifies unique proteins that interact with plasmodesmata. Plant Cell
15, 1507­1523.
Eyal, Y., Curie, C., and McCormick, S. (1995). Pollen specificity
elements reside in 30 bp of the proximal promoters of two pollenexpressed
genes. Plant Cell 7, 373­384.
Faure, J.E., Rusche, M.L., Thomas, A., Keim, P., Dumas, C.,
Mogensen, H.L., Rougier, M., and Chaboud, A. (2003). Double
fertilization in maize: The two male gametes from a pollen grain have
the ability to fuse with egg cells. Plant J. 33, 1051­1062.
Feijo, J.A., and Cox, G. (2001). Visualization of meiotic events in intact
living anthers by means of two-photon microscopy. Micron 32,
679­684.
Feldmann, K.A., Coury, D.A., and Christianson, M.L. (1997).
Exceptional segregation of a selectable marker (KanR) in Arabidopsis
identifies genes important for gametophytic growth and development.
Genetics 147, 1411­1422.
Friebe, B., Zhang, P., Nasuda, S., and Gill, B.S. (2003). Characterization
of a knock-out mutation at the Gc2 locus in wheat. Chromosoma
111, 509­517.
Golovkin, M., and Reddy, A.S.N. (2003). A calmodulin-binding protein
from Arabidopsis has an essential role in pollen germination. Proc.
Natl. Acad. Sci. USA 100, 10558­10563.
Grini, P.E., Schnittger, A., Schwarz, H., Zimmermann, I., Schwab, B.,
Jurgens, G., and Hulskamp, M. (1999). Isolation of ethyl methanesulfonate-induced
gametophytic mutants in Arabidopsis thaliana by
a segregation distortion assay using the multimarker chromosome 1.
Genetics 151, 849­863.
Gupta, R., Ting, J.T., Sokolov, L.N., Johnson, S.A., and Luan, S.
(2002). A tumor suppressor homolog, AtPTEN1, is essential for pollen
development in Arabidopsis. Plant Cell 14, 2495­2507.
Guyon, V.N., Astwood, J.D., Garner, E.C., Dunker, A.K., and Taylor,
L.P. (2000). Isolation and characterization of cDNAs expressed in the
early stages of flavonol-induced pollen germination in petunia. Plant
Physiol. 123, 699­710.
Heuer, S., Lorz, H., and Dresselhaus, T. (2000). The MADS box gene
ZmMADS2 is specifically expressed in maize pollen and during maize
pollen tube growth. Sex. Plant Reprod. 13, 21­27.
Control of Male Gametophyte Development S151
Honys, D., and Twell, D. (2003). Comparative analysis of the
Arabidopsis pollen transcriptome. Plant Physiol. 132, 640­652.
Howden, R., Park, S.K., Moore, J.M., Orme, J., Grossniklaus, U., and
Twell, D. (1998). Selection of T-DNA-tagged male and female
gametophytic mutants by segregation distortion in Arabidopsis.
Genetics 149, 621­631.
Hulskamp, M., Parekh, N.S., Grini, P., Schneitz, K., Zimmermann, I.,
Lolle, S.J., and Pruitt, R.E. (1997). The STUD gene is required for
male-specific cytokinesis after telophase II of meiosis in Arabidopsis
thaliana. Dev. Biol. 187, 114­124.
Johnson, S.A., and McCormick, S. (2001). Pollen germinates precociously
in the anthers of raring-to-go, an Arabidopsis gametophytic
mutant. Plant Physiol. 126, 685­695.
Jongeneel, C.V., Iseli, C., Stevenson, B.J., Riggins, G.J., Lal, A.,
Mackay, A., Harris, R.A., O'Hare, M.J., Neville, A.M., Simpson,
A.J.G., and Strausberg, R.L. (2003). Comprehensive sampling of
gene expression in human cell lines with massively parallel signature
sequencing. Proc. Natl. Acad. Sci. USA 100, 4702­4705.
Kamps, T.L., McCarty, D.R., and Chase, C.D. (1996). Gametophyte
genetics in Zea mays L.: Dominance of a restoration-of-fertility allele
(Rf3) in diploid pollen. Genetics 142, 1001­1007.
Kang, B.-H., Rancour, D.M., and Bednarek, S.Y. (2003). The dynaminlike
protein ADL1C is essential for plasma membrane maintenance
during pollen maturation. Plant J. 35, 1­15.
Kerim, T., Imin, N., Weinman, J.J., and Rolfe, B.G. (2003). Proteome
analysis of male gametophyte development in rice anthers. Proteomics
3, 738­751.
Kim, H.-U., Cotter, R., Johnson, S., Senda, M., Dodds, P.,
Kulikauskas, R., Tang, W., Ezcurra, I., Herzmark, P., and
McCormick, S. (2002). New pollen-specific receptor kinases identified
in tomato, maize and Arabidopsis: The tomato kinases show
overlapping but distinct localization patterns on pollen tubes. Plant
Mol. Biol. 50, 1­16.
Kim, S., Mollet, J.-C., Dong, J., Zhang, K., Park, S.-Y., and Lord, E.M.
(2003). Chemocyanin, a small basic protein from the lily stigma,
induces pollen tube chemotropism. Proc. Natl. Acad. Sci. USA 100,
16125­16130.
Klimyuk, V.I., and Jones, J.D. (1997). AtDMC1, the Arabidopsis
homologue of the yeast DMC1 gene: Characterization, transposoninduced
allelic variation and meiosis-associated expression. Plant J.
11, 1­14.
Kobayashi, A., Sakamoto, A., Kubo, K., Rybka, Z., Kanno, Y., and
Takatsuji, H. (1998). Seven zinc-finger transcription factors are
expressed sequentially during the development of anthers in petunia.
Plant J. 13, 571­576.
Koller, A., Washburn, M.P., Lange, B.M., Andon, N.L., Deciu, C.,
Haynes, P.A., Hays, L., Schieltz, D., Ulaszek, R., Wei, J., Wolters,
D., and Yates, J.R. (2002). Proteomic survey of metabolic pathways
in rice. Proc. Natl. Acad. Sci. USA 99, 11969­11974.
Kulikauskas, R., and McCormick, S. (1997). Identification of the
tobacco and Arabidopsis homologues of the pollen-expressed LAT59
gene of tomato. Plant Mol. Biol. 34, 809­814.
Lalanne, E., Honys, D., Johnson, A., Borner, G.H.H., Lilley, K.S.,
Dupree, P., Grossniklaus, U., and Twell, D. (2004). SETH1 and
SETH2, two components of the glycosylphosphatidylinositol anchor
biosynthetic pathway, are required for pollen germination and tube
growth in Arabidopsis. Plant Cell 16, 229­240.
Lalanne, E., and Twell, D. (2002). Genetic control of male germ unit
organization in Arabidopsis. Plant Physiol. 129, 865­875.
Lee, H.-S., Karunanandaa, B., McCubbin, A., Gilroy, S., and Kao,
T.-h. (1996). PRK1, a receptor-like kinase of Petunia inflata is essential
for post-meiotic development of pollen. Plant J. 9, 613­624.
Lee, J.Y., and Lee, D.H. (2003). Use of serial analysis of gene expression
technology to reveal changes in gene expression in Arabidopsis
pollen undergoing cold stress. Plant Physiol. 132, 517­529.
Levsky, J.M., Shenoy, S.M., Pezo, R.C., and Singer, R.H. (2002).
Single-cell gene expression profiling. Science 297, 836­840.
Lohrmann, J., Sweere, U., Zabaleta, E., Ba¨ urle, I., Keitel, C., KozmaBognar,
L., Brennicke, A., Scha¨ fer, E., Kudla, J., and Harter, K.
(2001). The response regulator ARR2: A pollen-specific transcription
factor involved in the expression of nuclear genes for components of
mitochondrial complex I in Arabidopsis. Mol. Genet. Genomics 265,
2­13.
Martin, S.G., Leclerc, V., Smith-Letiere, K., and St. Johnston,, D.
(2003). The identification of novel genes required for Drosophila
anteroposterior axis formation in a germline clone screen using GFPStaufen.
Development 130, 4201­4215.
Mascarenhas, J.P. (1975). The biochemistry of angiosperm pollen
development. Bot. Rev. 41, 259­314.
Mascarenhas, J.P. (1990). Gene activity during pollen development.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 41, 317­338.
Mascarenhas, J.P. (1993). Molecular mechanisms of pollen tube
growth and differentiation. Plant Cell 5, 1303­1314.
McCormick, S. (1993). Male gametophyte development. Plant Cell 5,
1265­1275.
Mu, J.H., Lee, H.S., and Kao, T.H. (1994). Characterization of a pollenexpressed
receptor-like kinase gene of Petunia inflata and the activity
of its encoded kinase. Plant Cell 6, 709­721.
Muschietti, J., Dircks, L., Vancanneyt, G., and McCormick, S. (1994).
LAT52 protein is essential for tomato pollen development: Pollen
expressing antisense LAT52 RNA hydrates and germinates abnormally
and cannot achieve fertilization. Plant J. 6, 321­338.
Muschietti, J.P., Eyal, Y., and McCormick, S. (1998). Pollen tube
localization implies a role in pollen-pistil interactions for the tomato
receptor-like protein kinases LePRK1 and LePRK2. Plant Cell 10,
319­330.
Oh, S.-A., Park, S.K., Jang, I., Howden, R., Moore, J.M.,
Grossniklaus, U., and Twell, D. (2003). halfman, an Arabidopsis
male gametophytic mutant associated with a 150 kb chromosomal
deletion adjacent to an introduced Ds transposable element. Sex.
Plant Reprod. 16, 99­102.
Owen, H.A., and Makaroff, C.A. (1995). Ultrastructure of microsporogenesis
and microgametogenesis in Arabidopsis thaliana (L.) Heynh.
ecotype Wassilewskija (Brassicaceae). Protoplasma 185, 7­21.
Park, S.K., Howden, R., and Twell, D. (1998). The Arabidopsis thaliana
gametophytic mutation gemini pollen1 disrupts microspore polarity,
division asymmetry and pollen cell fate. Development 125, 3789­
3799.
Preuss, D., Rhee, S.Y., and Davis, R.W. (1994). Tetrad analysis
possible in Arabidopsis with mutation of the QUARTET (QRT) genes.
Science 264, 1458­1460.
Procissi, A., Guyon, A., Pierson, E.S., Giritch, A., Knuiman, B.,
Grandjean, O., Tonelli, C., Derksen, J., Pelletier, G., and
Bonhomme, S. (2003). Kinky pollen encodes a SABRE-like protein
required for tip growth in Arabidopsis and conserved among
eukaryotes. Plant J. 36, 894­904.
Rombauts, S., Florquin, K., Lescot, M., Marchal, K., Rouze, P., and
Van De Peer, Y. (2003). Computational approaches to identify
promoters and cis-regulatory elements in plant genomes. Plant
Physiol. 132, 1162­1176.
Ross, K.J., Fransz, P., and Jones, G.H. (1996). A light microscopic
atlas of meiosis in Arabidopsis thaliana. Chromosome Res. 4,
507­516.
Rusche, M.L., Mogensen, H.L., Shi, L., Keim, P., Rougier, M.,
Chaboud, A., and Dumas, C. (1997). B chromosome behavior in
S152 The Plant Cell
maize pollen as determined by a molecular probe. Genetics 147,
1915­1921.
Russell, S.D. (1984). Ultrastructure of the sperm of Plumbago zeylanica.
II. Quantitative cytology and three-dimensional organization. Planta
162, 385­391.
Russell, S.D. (1985). Preferential fertilization in Plumbago: Ultrastructural
evidence for gamete-level recognition in an angiosperm. Proc.
Natl. Acad. Sci. USA 82, 6129­6132.
Sari-Gorla, M., Gatti, E., Villa, M., and Pe, M.E. (1997). A multinucleate
male-sterile mutant for maize with gametophytic expression.
Sex. Plant Reprod. 10, 22­26.
Scholten, S., Lo¨ rz, H., and Kranz, E. (2002). Paternal mRNA and
protein synthesis coincides with male chromatin decondensation in
maize zygotes. Plant J. 32, 221­231.
Scholz-Starke, J., Buttner, M., and Sauer, N. (2003). AtSTP6, a new
pollen-specific H1-monosaccharide symporter from Arabidopsis.
Plant Physiol. 131, 70­77.
Singh, M., Bhalla, P.L., Xu, H., and Singh, M.B. (2003). Isolation and
characterization of a flowering plant male gametic cell-specific
promoter. FEBS Lett. 542, 47­52.
Skirpan, A.L., McCubbin, A.G., Ishimizu, T., Wang, X., Hu, Y., Dowd,
P.E., Ma, H., and Kao, T.-h. (2001). Isolation and characterization of
kinase interacting protein 1, a pollen protein that interacts with the
kinase domain of PRK1, a receptor-like kinase of petunia. Plant
Physiol. 126, 1480­1492.
Spehr, M., Gisselmann, G., Poplawski, A., Riffell, J.A., Wetzel, C.H.,
Zimmer, R.K., and Hatt, H. (2003). Identification of a testicular
odorant receptor mediating human sperm chemotaxis. Science 299,
2054­2058.
Spielman, M., Preuss, D., Li, F.L., Browne, W.E., Scott, R.J.,
and Dickinson, H.G. (1997). TETRASPORE is required for male
meiotic cytokinesis in Arabidopsis thaliana. Development 124, 2645­
2657.
Steinebrunner, I., Wu, J., Sun, Y., Corbett, A., and Roux, S.J. (2003).
Disruption of apyrases inhibits pollen germination in Arabidopsis.
Plant Physiol. 131, 1638­1647.
Tang, W.-H., Ezcurra, I., Muschietti, J., and McCormick, S. (2002). A
cysteine-rich extracellular protein, LAT52, interacts with the extracellular
domain of the pollen receptor kinase LePRK2. Plant Cell 14,
2277­2287.
Touraev, A., Stoger, E., Voronin, V., and Heberle-Bors, E. (1997).
Plant male germ line transformation. Plant J. 12, 949­956.
Twell, D., Park, S.K., Hawkins, T.J., Schubert, D., Schmidt, R.,
Smertenko, A., and Hussey, P.J. (2002). Mor1/Gem1 has an
essential role in the plant-specific cytokinetic phragmoplast. Nat. Cell
Biol. 4, 711­714.
Twell, D., Yamaguchi, J., Wing, R.A., Ushiba, J., and McCormick, S.
(1991). Promoter analysis of genes that are coordinately expressed
during pollen development reveals pollen-specific enhancer sequences
and shared regulatory elements. Genes Dev. 5, 496­507.
Ueda, K., Kinoshita, Y., Xu, Z.J., Ide, N., Ono, M., Akahori, Y.,
Tanaka, I., and Inoue, M. (2000). Unusual core histones specifically
expressed in male gametic cells of Lilium longiflorum. Chromosoma
108, 491­500.
Ueda, K., and Tanaka, I. (1995). The appearance of male gametespecific
histones gH2B and gH3 during pollen development in Lilium
longiflorum. Dev. Biol. 169, 210­217.
Van der Hoeven,, R., Ronning, C., Giovannoni, J., Martin, G., and
Tanksley, S. (2002). Deductions about the number, organization, and
evolution of genes in the tomato genome based on analysis of large
expressed sequence tag collection and selective genomic sequencing.
Plant Cell 14, 1441­1456.
Vielle-Calzada, J.P., Baskar, R., and Grossniklaus, U. (2000). Delayed
activation of the paternal genome during seed development. Nature
404, 91­94.
Vielle-Calzada, J.-P., Baskar, R., and Grossniklaus, U. (2001). Early
paternal gene activity in Arabidopsis: A reply. Nature 414, 710.
Weijers, D., Geldner, N., Offringa, R., and Jurgens, G. (2001). Early
paternal gene activity in Arabidopsis. Nature 414, 709­710.
Wittink, F.R.A., Knuiman, B., Derksen, J., Capkova, V., Twell, D.,
Schrauwen, J.A.M., and Wullems, G.J. (2000). The pollen-specific
gene Ntp303 encodes a 69-kDa glycoprotein associated with
the vegetative membranes and the cell wall. Sex. Plant Reprod. 12,
276­284.
Xu, H., Swoboda, I., Bhalla, P.L., and Singh, M.B. (1999a). Male
gametic cell-specific expression of H2A and H3 histone genes. Plant
Mol. Biol. 39, 607­614.
Xu, H., Swoboda, I., Bhalla, P.L., and Singh, M.B. (1999b). Male
gametic-cell-specific gene expression in flowering plants. Proc. Natl.
Acad. Sci. USA 96, 2554­2558.
Xu, H., Weterings, K., Vriezen, W., Feron, R., Xue, Y., Derksen, J.,
and Mariani, C. (2002). Isolation and characterization of male-germcell
transcripts in Nicotiana tabacum. Sex. Plant Reprod. 14, 339­346.
Yang, C.Y., Spielman, M., Coles, J.P., Li, Y., Ghelani, S., Bourdon, V.,
Brown, R.C., Lemmon, B.E., Scott, R.J., and Dickinson, H.G.
(2003). TETRASPORE encodes a kinesin required for male meiotic
cytokinesis in Arabidopsis. Plant J. 34, 229­240.
Yang, Z. (2002). Small GTPases: Versatile signaling switches in plants.
Plant Cell 14 (suppl.), S375­S388.
Ylstra, B., and McCormick, S. (1999). Analysis of mRNA stabilities
during pollen development and in BY2 cells. Plant J. 20, 101­108.
Zachgo, S., Saedler, H., and Schwarz-Sommer, Z. (1997). Pollenspecific
expression of DEFH125, a MADS-box transcription factor in
Antirrhinum with unusual features. Plant J. 11, 1043­1050.
Zhu, H., and Snyder, M. (2002). `Omic' approaches for unraveling
signaling networks. Curr. Opin. Cell Biol. 14, 173­179.
Control of Male Gametophyte Development S153