Proc. Natl. Acad. Sci. USA
Vol. 96, pp. 2554–2558, March 1999
Male gametic cell-specific gene expression in flowering plants
HUILING XU, INES SWOBODA, PREM L. BHALLA, AND MOHAN B. SINGH*
Plant Molecular Biology and Biotechnology Laboratory, Institute of Land and Food Resources, University of Melbourne, Parkville, Victoria 3052, Australia
Communicated by Rutherford N. Robertson, Australian National University, Yass, Australia, December 21, 1998 (received for review October 9,
cell—in the process of fertilization is to recognize, adhere to,
and fuse with the female gamete. These highly specialized
functions are expected to be controlled by activation of a
unique set of genes. However, male gametic cells traditionally
have been regarded as transcriptionally quiescent because of
highly condensed chromatin and a very reduced amount of
cytoplasm. Here, we provide evidence for male gamete-specific
gene expression in flowering plants. We identified and char-
acterized a gene, LGC1, which was shown to be expressed
exclusively in the male gametic cells. The gene product of
LGC1 was localized at the surface of male gametic cells,
suggesting a possible role in sperm–egg interactions. These
findings represent an important step toward defining the
molecular mechanisms of male gamete development and the
cellular processes involved in fertilization of flowering plants.
The role of the male gamete—the sperm
The process of male gametic cell development in higher plants
begins with a highly asymmetric mitotic division of the male
gametophyte (pollen). This division results in the formation of
two unequal cells—the larger vegetative and the smaller
generative cell—which have dramatically different structures
and functions. The small generative cell is wholly enclosed
within the much larger vegetative cell, forming a unique
‘‘cell-within-a-cell’’ structure. The generative cell, the progen-
itor of the male gametes (or sperm cells), has a very reduced
amount of cytoplasm containing relatively few organelles and
is surrounded by a double membrane. This cell undergoes a
mitotic division producing two sperm cells. In some plants,
such as maize, this division takes place in the mature pollen,
and, in other plants, such as tobacco and lily, it occurs after
pollination inside the pollen tube. One of the two sperm cells
resulting from this division fertilizes the egg cell, and the other
fuses with the central cell to produce the endosperm. In
contrast, the larger vegetative cell comprises the major part of
the pollen, including the pollen cytoplasm and the bulk of
stored mRNAs, proteins, lipids, and polysaccharides. It is
enclosed by a cellulosic inner wall, the intine, and a highly
sculptured thick outer sporopollenin wall, the exine. During
pollen germination, the vegetative cell wall extends, producing
a pollen tube through which the two sperm cells ultimately are
delivered to the female gamete.
The structural and functional differences between the veg-
etative and generative cells are likely to be controlled by
cell-specific gene activity. Indeed, a number of vegetative
cell-specific genes have been isolated, and, in some cases, a
biological function has been assigned (1). For example, re-
cently, a pollen-specific gene, Zea m1, was found to encode a
protein involved in loosening the cell walls of the stigma and
style and facilitating invasion of the pollen tube into the
maternal tissues (2). In contrast, little is known about tran-
scriptional activity of the cell of the male germline: the
generative and sperm cells. These cells are known to have
are transcriptionally quiescent (3). It was further proposed that
gametic cells are metabolically passive and wholly depend on
the vegetative cell for their development and function (4).
has been hampered by the inaccessibility of these cells. Re-
cently, researchers were able to isolate a sufficient number of
generative cells from lily (Lilium longiflorum) pollen for
biochemical analysis by using an enzymatic procedure (5).
Metabolic labeling with35S methionine demonstrated that
generative cells possess their own set of mRNAs and are
capable of synthesizing proteins independently from the veg-
etative cells (5). Comparative analysis of protein profiles of
generative cells and pollen grains showed the presence of
common as well as cell-specific proteins in the generative cell.
Here, we report the identification and characterization of a
male gamete-specific gene, LGC1, of flowering plants. This
study will facilitate our understanding of the molecular aspects
of the male gamete differentiation and function in flowering
MATERIALS AND METHODS
Construction and Screening of a cDNA Library. Generative
cells were isolated from lily pollen as described (5) and were
stored at ?70°C until use. mRNA extracted from ?1 ? 105
stored generative cells by using a mRNA purification kit
(Amersham Pharmacia) was reverse transcribed. The resultant
cDNA was amplified by PCR and was size fractionated and
cloned into the ?gt11 expression vector.
Inserts of cDNA clones randomly picked from the genera-
tive cell cDNA library were labeled with
priming (Bresatec, Adelaide, Australia) and were used for
probing of RNA slot blots that contained ?300 ng of mRNAs
from various tissues, including leaf, stem, petal, stigma?style,
ovary, pollen, and generative cells. cDNA clones showing
preferential or specific hybridization to generative cell mRNA
Sequence Analysis. DNA sequencing was performed on
an Applied Biosystems PRISM dye terminator cycle sequenc-
ing kit with an automated DNA sequencer. DNA sequence
analysis was performed with DNA STRIDER software and the
BLAST network service at the National Centre for Biotechnol-
ogy Information (National Institutes of Health). A hydropathy
profile of LGC1 protein was obtained according to Kyte and
Doolittle (6). Secondary structure prediction was done ac-
cording to Chou and Fasman (7).
RNA Gel Blot and Reverse Transcription (RT)–PCR Anal-
ysis. Total RNA was isolated from generative cells and various
32P by random
The publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked ‘‘advertisement’’ in
accordance with 18 U.S.C. §1734 solely to indicate this fact.
PNAS is available online at www.pnas.org.
Abbreviations: RT, reverse transcription; DIG, digoxigenin.
Data deposition: The sequence reported in this paper has been
deposited in the GenBank database (accession no. AF110779).
*To whom reprint requests should be addressed. e-mail: m.singh@
tissues by using the SNAP RNA purification kit (Invitrogen).
Ten micrograms of total RNA were separated on a 1% agarose
gel containing formaldehyde, were transferred to a Hybond
N?nylon membrane (Amersham Pharmacia), and were
probed with a32P-labeled LGC1 cDNA insert. Hybridization
was performed under high stringency conditions as described
by Xu et al (8). Filters subsequently were probed with lily
ribosomal RNA to verify the amount of total RNA loaded
from each tissue.
For RT-PCR, mRNAs from generative cells and various
tissues were reverse transcribed and amplified by PCR with a
pair of sequence-specific primers by using the Access RT-PCR
System (Promega). For each tissue, mRNA was subjected to
serial 2-fold dilutions. Based on the signal intensity of the
amplified products, the relative amount of LGC1 mRNA in
each tissue was estimated.
In Situ Hybridization. Nonradioactive whole mount in situ
hybridization was performed in both developing and mature
pollen based on published protocols (9–11). Fresh pollen at
various developmental stages was fixed [1% glutaraldehyde in
50 mM Pipes buffer (pH 7.4)] for 2 h at room temperature. The
fixed pollen was washed in buffer and was stored in 70%
ethanol at 4°C until use. Both sense and antisense riboprobes
labeled with digoxigenin (DIG)–UTP were generated from
linearized DNA templates. Hybridization signal was detected
with an alkaline phosphatase-conjugated anti-DIG antibody
by using a DIG nucleic acid detection kit (Boehringer Mann-
heim). To obtain better resolution, protoplasts of developing
pollen were released from the exine by treatment with an
enzyme solution (1% macerozyme?0.5% cellulase?0.5% BSA)
as described (5). Vegetative and generative nuclei within the
pollen were visualized by counter-staining with 4?, 6?-
In lily, generative cell division occurs in the pollen tube
during its growth in the female stylar tissue. In situ hybridiza-
tion of mRNA in sperm cells, therefore, only can be performed
in a pollen tube. Pollen tubes were grown in vivo by hand
pollinating pistils with freshly collected pollen. After 48 h, a
1-cm segment was taken from the base of the style and was cut
into two symmetrical halves. Pollen tubes growing in the
hollow stylar canal were teased out, were fixed, and then were
used for in situ hybridization.
Expression of Recombinant LGC1 Protein in Escherichia
coli and Antibody Production. The coding region of LGC1
cDNA was cloned into the bacterial expression vector pQE-30
(Qiagen, Chatsworth, CA). The recombinant plasmid was
transformed into E. coli strain M15. LGC1 protein expression
was induced by the addition of 1 mM isopropyl ?-D-
thiogalactopyranoside. The recombinant protein was purified
by using Ni-nitrilotriacetic acid agarose affinity chromatogra-
phy according to the manufacturer’s instruction (Qiagen,
Chatsworth, CA). Purified recombinant protein was used to
raise polyclonal antibodies in rabbits following the standard
protocols (12). Antiserum obtained after the third booster was
affinity-purified as described (13).
SDS?PAGE, Immunoblotting, and Immunoprecipitation.
Protein samples denatured under reducing conditions by boil-
SDS, 2% DTT?10% glycerol, pH 6.8) were subjected to
SDS?PAGE according to the method described by Laemmli
(14) and then were blotted onto nitrocellulose membranes
(Schleicher & Schuell). Membranes were blocked with 10%
milk powder in PBS for 1 h, were washed with PBS, and were
incubated with affinity-purified anti-LGC1 antibody (1:200
dilution in PBS containing 1% BSA) for 3 h at room temper-
ature. After washing twice in PBS containing 0.1% Tween 20
and twice in PBS, the blots were incubated in the anti-rabbit
antibody conjugated with horseradish peroxidase (1:2,000
dilution in PBS containing 1% BSA) for 2 h. The blots were
developed by using 4-chloro-1-naphthol as color substrate.
Immunoprecipitation was performed according to Harlow
and Lane (12). Generative cells were resuspended in cell lysis
buffer (10 mM K2PO4?2 mM EDTA?150 mM NaCl?0.1 mM
phenylmethylsulfonyl fluoride?1% Triton X-100), were incu-
bated for 10 min at 4°C, and were centrifuged for 5 min. The
supernatant was incubated with affinity-purified anti-LGC1
antibody for 2 h at 4°C. Fifty microliters of anti-rabbit IgG
agarose beads (Sigma) then were added to the mixture and
were incubated for 1 h at 4°C. Immunoprecipitates were
washed three times with cell lysis buffer and were resuspended
in 30 ?l of 2 ? SDS-sample buffer. The mixture was boiled for
5 min and was centrifuged briefly to precipitate the beads. The
immunoprecipitated proteins were subjected to SDS?PAGE
and were visualized by staining with Coomassie brilliant blue
Immunocytochemistry. Isolated generative cells were
placed on the coverslips coated with poly-L-lysine and were
fixed in cold methanol for 2 min. After washing with PBS, the
cells were incubated with either anti-LGC1 antibody or pre-
immune serum diluted in PBS containing 1% BSA for 6–8 h
in a moist chamber at 4°C. Fluorescein isothiocyanate-labeled
secondary antibody was used for the detection of antibody
binding. The specificity of the antibodies was confirmed by
using recombinant LGC1 protein as a competitor for antibody
RESULTS AND DISCUSSION
Molecular Analysis of the LGC1 cDNA Clone. Pollen is a
two-celled structure in which the generative cell occupies a
very small portion of the volume. So far, all of the genes
isolated from pollen cDNA libraries have been shown to be
specific to the larger vegetative cell (15, 16). To maximize the
likelihood of obtaining genes specifically expressed in the
generative cell, we constructed a PCR-based cDNA library of
generative cells isolated from lily (L. longiflorum) pollen.
Inserts of cDNA clones picked randomly from the library were
used as probes for differential hybridization to RNA from
generative cells, pollen, pistil, petal, ovary, leaf, and stem. One
cDNA clone, LGC1, that showed strong hybridization to RNA
of generative cells, weak hybridization to RNA of pollen
(containing generative cells), and no detectable hybridization
to RNA of other tissues was considered as a putative gener-
ative cell-specific clone.
LGC1 contains a cDNA insert of 618 bp [nucleotide se-
quence deposited in the GenBank database (accession no.
AF110779)] encoding a predicted gene product of 128 amino
acids with a calculated molecular mass of 13.8 kDa and a pI of
5.33. Analysis of the deduced protein sequence revealed an
N-glycosylation motif Asn-X-Ser at the C-terminal and a
highly hydrophobic region at the N-terminal end. Secondary
structure prediction according to Chou and Fasman (7)
showed that the hydrophobic region corresponds to an ?-helix,
indicating that this region may function as a transmembrane
anchor and that LGC1 protein might be associated with the
LGC1 was investigated by using RNA gel blot analysis. LGC1
hybridized to a transcript of 0.6 kilobase, which is present at a
high level in generative cells isolated from mature pollen (Fig.
1A). A faint signal was visible in pollen containing generative
petal, pistil, and ovule.
We further examined the tissue specificity of LGC1 by using
the more sensitive RT-PCR analysis. RT-PCR amplifications
tissues and two gene-specific primers that amplified a 0.3-
kilobase portion of the coding region. A PCR product of the
expected size (0.3 kilobases) was obtained in generative cells
and pollen but not in any other parts of the plant tested,
Plant Biology: Xu et al.Proc. Natl. Acad. Sci. USA 96 (1999)2555
including vegetative tissues such as leaf and stem as well as
reproductive tissues such as petal, female stigma?style, and
ovary (Fig. 1B). Based on the signal intensity, we estimated
that ?20-fold more PCR product was obtained when gener-
ative cell mRNA was used as compared with pollen mRNA.
The results from RT-PCR confirmed the data obtained by
RNA gel blot analysis. In both cases, LGC1 transcription
product was only detectable in the male gametophyte, with
much higher levels of expression in the generative cell as
compared with that in the entire pollen. We considered that
the LGC1 mRNA detected in pollen owes its origin to the
generative cell that constitutes a small portion of the pollen
In Situ Localization of LGC1 Transcripts in Generative and
Sperm Cells. We further analyzed the differential expression
of LGC1 in the vegetative and the generative cell by in situ
hybridization. The results clearly showed that LGC1 mRNA is
confined to the generative cell but is not present in the
vegetative cell of mature pollen (Fig. 2). The strong hybrid-
ization signal detected in the cytoplasm of the generative cell
appeared as a distinct spindle-shaped form in the pollen grain.
The generative cell is the product of a highly asymmetrical
division. It is possible that mRNA present in the generative
cell may originate from differential RNA localization and
partitioning before generative cell formation, as known
during Drosophila and Fucus embryogenesis (10, 17). To
answer this question, the temporal expression pattern of
LGC1 during the process of male gametogenesis was ana-
lyzed. We monitored accumulation of LGC1 mRNA before
and after the formation of the generative cell by in situ
hybridization. LGC1 mRNA was not detected at any of three
stages before microspore division, including microsporocyte,
tetrad, and unicellular microspore stages (data not shown).
We further examined six different stages after the formation
of generative cells. At the earliest stage, the newly formed
generative cell is attached at one pole of pollen with the
vegetative nucleus located in its vicinity (Fig. 3 A and F). As
development progresses, the generative cell starts to detach
from the intine (inner cell wall of pollen) while the vegetative
nucleus moves toward the center of the pollen (Fig. 3 B and
G). No detectable signal was observed in either these two
early developmental stages (Fig. 3 A and B). Concomitant
with the rapid size expansion of the pollen, the generative
cell detaches completely from the intine and is suspended
freely within the vegetative cell cytoplasm. Its shape be-
comes elongated, with a large nucleus in the center and most
of the cytoplasm at both ends of the cell (Fig. 3 C and H).
The blot was reprobed with lily rRNA to verify the relative amount of RNA in each lane. (B) RT-PCR of different tissues as indicated. Pollen
mRNA includes contributions of both generative cell and vegetative cell. Numbers 16, 32, and 64 represent 1?16, 1?32, and 1?64, etc., for mRNA
input in each lane, respectively. Molecular sizes are indicated on the left.
Expression of LGC1 mRNA in different tissues of lily. (A) RNA gel blot of the indicated tissues probed with32P-labeled LGC1 probe.
pollen. Dark staining in the generative cell (arrowhead) represents
hybridization signal detected by using an alkaline phosphatase con-
jugated anti-DIG antibody. The outer wall of pollen, exine, appears as
a sculptured pattern. (A) Pollen probed with a DIG-UTP-labeled
LGC1 antisense riboprobe. (B) Control pollen probed with a sense
2556Plant Biology: Xu et al. Proc. Natl. Acad. Sci. USA 96 (1999)
At this stage, a weak signal was detected at both ends of the
generative cell, indicating the initiation of LGC1 mRNA
transcription (Fig. 3C). As development continues, the gen-
erative cell becomes spindle-shaped (Fig. 3 D and I), and
accumulation of LGC1 mRNA in the generative cell be-
comes more evident (Fig. 3D). At the time of pollen matu-
rity, a very high level of LGC1 mRNA was observed in the
generative cell (Figs. 2A and 3 E and J). After pollination,
pollen germination occurs on the female stigma, and pollen
tubes grow inside the female stylar tissue. During this
process, the generative cell moves into the pollen tube and
undergoes a mitotic division producing two male gametes,
the sperm cells (Fig. 3 K and L). LGC1 mRNA was clearly
detectable in the two sperm cells inside the pollen tubes (Fig.
of developing pollen were released from the sculptured exine. Developing pollen (A–E) and pollen tube (K) were probed with a DIG-UTP-labeled
riboprobe and then were counterstained with 4?, 6?-diamidino-2-phenylindole to visualize the vegetative and generative nuclei within the pollen
(F–J) and to visualize sperm nuclei in the pollen tube (L). Arrowheads indicate the generative cell at early developmental stages. GN, generative
nucleus; VN, vegetative nucleus; SC, sperm cell; SN, sperm nucleus.
In situ hybridization of LGC1 mRNA to whole-mounted lily pollen at different developmental stages. For better resolution, protoplasts
Immunoblot probed with anti-LGC1 antiserum showing specific binding of antibody to LGC1 protein. (C) LGC1 protein immunoprecipitated from
total protein extract of generative cells. Two strong bands represent the heavy chain (HC) and light chain (LC) of Ig, respectively.
Immunoblot analysis of LGC1 protein. (A) Coomassie blue-stained SDS?PAGE gel showing the recombinant LGC1 protein. (B)
Plant Biology: Xu et al.Proc. Natl. Acad. Sci. USA 96 (1999)2557
3K). No hybridization signal was detected in the vegetative Download full-text
cell at any stage of pollen development.
If the generative cell-specific presence of LGC1 transcript is
the result of asymmetric RNA localization and partitioning, we
would expect the transcripts to be present in unicellular
microspores and early generative cells. Our in situ hybridiza-
tion results indicated the absence of LGC1 transcript at any
developmental stages before the first pollen mitosis. The
activation of LGC1 gene appears to occur at the late bicellular
the vegetative cell cytoplasm. The expression increases during
pollen maturation, reaches the highest level at the mature
pollen stage, and remains active in the two sperm cells after
division of the generative cell. Based on these results, we
conclude that differential localization of LGC1 transcript in
the generative cell is attributable to gene activation in the
Characterization of LGC1 Gene Product. To characterize
the LGC1 gene product, we fused the coding region in frame
to an affinity tag of six histidines and overexpressed the
resulting construct in E. coli. The fusion protein had the
expected molecular mass of ?19 kDa (Fig. 4A). Polyclonal
antibodies raised against the purified protein recognized the
recombinant LGC1 polypeptide on immunoblots, confirming
the specificity of the antiserum (Fig. 4B). The purified anti-
LGC1-antibody then was used to analyze the LGC1 gene
product in the plant. A single protein band of ?30 kDa was
purified from total protein extract of generative cells by
immunoprecipitation (Fig. 4C). The molecular mass of the
identified LGC1 protein was higher than that predicted from
the deduced amino acid sequence. This discrepancy might
be attributable to glycosylation of the protein, which could
have occurred at the predicted N-glycosylation site at the C
To determine the cellular distribution of the LGC1 gene
product, anti-LGC1 antibody was used to localize the protein
in the generative cell. From the fluorescence pattern shown in
Fig. 5, we conclude that LGC1 polypeptide is present at the
surface of the generative cell membrane. The observed local-
ization on the membrane might be attributed to the N-terminal
hydrophobic region, which acts as a transmembrane ?-helix
Cell surface localization of LGC1 protein is of considerable
interest. Plant male gametic cells lack cell walls, and the first
contact between sperm and egg cells during the process of
fertilization occurs at plasma membranes. Recently, an in vitro
system for adhesion and fusion of maize gametes has been
reported (18). Various combinations including sperm–sperm,
sperm–egg, and sperm–mesophyll protoplasts were allowed to
fuse to test the specificity of fusion. Fusion was found to be
mainly restricted to sperm–egg pairs. This result suggests the
presence of specific and potentially complimentary fusogenic
determinants on the gamete membrane surface. In case of
mammalian fertilization, several membrane surface proteins
involved in sperm–egg adhesion, fusion, and signaling during
the events of fertilization have been identified (19, 20). The
strict male gametic cell-specificity of LGC1 gene expression
and localization of its gene product on the membrane surface
points toward a putative role in sperm–egg fusion events.
In conclusion, identification of a male gamete-specific gene
LGC1 has opened up opportunities for further investigations
into molecular mechanisms involved in fertilization of higher
plants. Although still speculative in nature, models for sperm–
egg recognition and fusion during double fertilization are
emerging from experimental studies using in vitro systems (18).
The availability of the LGC1 gene should make it possible to
egg recognition and fusion at the molecular level.
We thank the Australian Research Council for financial support,
Prof. W. H. Sawyer for analyzing the protein secondary structure, and
Prof. Scott Russell and Prof. Rudolf Valenta for critical reading of the
1.Mascarenhas, J. P. (1992) in International Review of Cytology, eds.
Russell, S. & Dumas, C. (Academic, San Diego), Vol. 140, pp.
Cosgrove, D. J., Bedinger, P. & Durachko, D. M. (1997) Proc.
Natl. Acad. Sci. USA 94, 6559–6564.
McCormick, S. (1993) Plant Cell 5, 1265–1275.
Eady, C., Lindsey, K. & Twell, D. (1995) Plant Cell 7, 65–74.
Blomstedt, C. K., Knox, R. B. & Singh, M. B. (1996) Plant Mol.
Biol. 31, 1083–1086.
Kyte, J. & Doolittle, R. F. (1982) J. Mol. Biol. 157, 105–132.
Chou, P. Y. & Fasman, G. D. (1975) J. Mol. Biol. 96, 29–45.
Xu, H., Knox, R. B., Taylor, P. E. & Singh, M. B. (1995) Proc.
Natl. Acad. Sci. USA 92, 2106–2110.
Bouget, F., Gerttula, S. & Quatrano, R. S. (1995) Dev. Biol. 171,
Bouget, F., Gerttula, S., Shaw, S. L. & Quatrano, R. S. (1996)
Plant Cell 8, 189–201.
Torres, M., Rigau, J., Puigdomenech, P. & Stiefel, V. (1995)Plant
J. 8, 317–321.
Harlow, E. & Lane, D. (1988) Antibodies: A Laboratory Manual
(Cold Spring Harbor Lab. Press, Plainview, NY).
Gu, J., Stepheson, C. G. & Iadarola, M. J. (1994) BioTechniques
Laemmli, U. K. (1970) Nature (London) 277, 680–685.
Twell, D., Wing, R., Yamaguchi, J. & McCormick, S. (1989) Mol.
Gen. Genet. 217, 240–245.
Theerakulpisuit, P., Xu, H., Singh, M. B., Pettitt, J. M. & Knox,
R. B. (1990) Plant Cell 3, 1073–1084.
Wilhelm J. E. & Vale, R. D. (1993) J. Cell Biol. 123, 269–274.
Faure, J.-E., Digonnet, C. & Dumas, C. (1994) Science 263,
Wassarman, P. M. (1995) Curr. Opin. Cell Biol. 7, 658–664.
Snell, W. J. & White, J. M. (1996) Cell 85, 629–637.
ative cell membrane. Generative cells were incubated with either
purified anti-LGC1 polyclonal antibody (A) or preimmune serum (B).
Binding of anti-LGC1 antibody was detected by using an anti-rabbit
fluorescein isothiocyanate-conjugated secondary antibody.
Localization of LGC1 protein at the surface of the gener-
2558Plant Biology: Xu et al.Proc. Natl. Acad. Sci. USA 96 (1999)