In vitro and in situ localization of Con A and WGA binding sites on the surface of female cells in Torenia fournieri L

Abstract

The interaction between lectins and their specific binding sites is believed to play a
critical role in fertilization in animals and some lower plants. However, for higher
plants there is no information on lectins or their binding sites related to female
gametes and fertilization. The present work was designed as a first attempt to reveal
the general pattern of lectin binding site distribution on the surface of female cells,
namely egg cells, central cells, and synergids of Torenia fournieri and, especially, to
investigate the possible effects of cell isolation procedure on the distribution of
lectin binding sites. Therefore, concanavalin A (Con A) and wheat germ agglutinin
(WAG) binding sites on the surface of both in vitro and in situ living female cells
were localized by using fluorescein isothiocyanate (FITC) conjugated Con A and
WGA as probes. We demonstrated that enzymatic treatment and isolation procedures
did not notably modify the surface character of the female cells and the
distribution of Con A and WGA binding sites. It was also found that Con A binding
sites were distributed differently on the surface of the female cells, with the strongest
fluorescent signal on central cells and the weakest on egg cells. Calcium could
greatly enhance the binding of Con A to the cell surface. A polar distribution pattern
of Con A binding sites in embryo sacs was observed. The binding sites were
obviously densest at the filiform apparatus of the synergids. The basic pattern of
WGA binding site distribution was similar to that of Con A’s. However, the
fluorescent signal of WGA was much weaker than that of Con A and fluorescent
patches were usually found on the cell surface.

Full text

© 2003 Laser Pages Publishing Ltd., Jerusalem
Israel Journal of Plant Sciences Vol. 51 2003 pp. 83–90
*Author to whom correspondence should be addressed. E-mail:
mxsun@whu.edu.cn
In vitro and in situ localization of concanavalin A and wheat germ agglutinin
binding sites on the surface of female cells in Torenia fournieri L.
KE-FENG FANG,a MENG-XIANG SUN,a,* ERHARD KRANZ,bAND CHANG ZHOUa
aKey Laboratory of MOE for Plant Developmental Biology, Wuhan University, Wuhan 430072, P.R. China
bCenter for Applied Plant Molecular Biology, AMP II, Institute for General Botany, University of Hamburg,
Ohnhorststrasse 18, D-22609, Hamburg, Germany
(Received 16 June 2002 and in revised form 20 October 2002)
ABSTRACT
The interaction between lectins and their specific binding sites is believed to play a
critical role in fertilization in animals and some lower plants. However, for higher
plants there is no information on lectins or their binding sites related to female
gametes and fertilization. The present work was designed as a first attempt to reveal
the general pattern of lectin binding site distribution on the surface of female cells,
namely egg cells, central cells, and synergids of Torenia fournieri and, especially, to
investigate the possible effects of cell isolation procedure on the distribution of
lectin binding sites. Therefore, concanavalin A (Con A) and wheat germ agglutinin
(WAG) binding sites on the surface of both in vitro and in situ living female cells
were localized by using fluorescein isothiocyanate (FITC) conjugated Con A and
WGA as probes. We demonstrated that enzymatic treatment and isolation proce-
dures did not notably modify the surface character of the female cells and the
distribution of Con A and WGA binding sites. It was also found that Con A binding
sites were distributed differently on the surface of the female cells, with the strongest
fluorescent signal on central cells and the weakest on egg cells. Calcium could
greatly enhance the binding of Con A to the cell surface. A polar distribution pattern
of Con A binding sites in embryo sacs was observed. The binding sites were
obviously densest at the filiform apparatus of the synergids. The basic pattern of
WGA binding site distribution was similar to that of Con A’s. However, the
fluorescent signal of WGA was much weaker than that of Con A and fluorescent
patches were usually found on the cell surface.
INTRODUCTION
Lectins are glycoproteins that bind reversibly to specific
mono- or oligosaccharides. Lectin binding sites (LBS)
are those glycoconjugates that possess a carbohydrate
moiety with a structure complementary to the LBS,
which are mainly glycoproteins, glycolipids, and poly-
saccharides. Glycoconjugates of a different nature but
with identical carbohydrates can act as binding sites for
the same lectin (Peumans and Van Damme, 1995).
Plant lectins were usually applied as molecular
probes to study surface properties of a variety of cells,
providing useful data on the organization of oligo-
saccharides on the cell surface. Specific localization of
LBS on gametes has long been considered to be associ-
ated with their special function in fertilization and early
embryogenesis. As early as 1975, it was reported that
Con A binding sites showed mosaic distribution on the
membrane surface of unfertilized mouse egg cells.
Based on this, the membrane surface could be divided
into two distinct regions, one stained with Con A and

Israel Journal of Plant Sciences 51 2003
84
constituting most of the surface, the other on which little
or no fluorescence was detected, overlying the second
metaphase spindle and specifically isolated as the mem-
brane of the second polar body (Johnson et al., 1975).
Con A binding sites on the plasma membrane of the
fertilized sea urchin egg were uniformly distributed.
However, with the onset of first division, fluorescent
lectin-binding site complexes were found concentrated
in the developing cleavage furrow. Between the comple-
tion of the first cleavage and the start of the second
cleavage, the pattern of LBS asymmetry changed: fewer
LBS were found in the cleavage furrow, while most
fluorescence was found at the poles of the daughter cells
(McCaig and Robinson, 1982). In the red alga (Agla-
othamnion oosumiense) soybean agglutinin (SBA) and
Dolichos biflorus agglutinin (DBA) bound to the sper-
matial appendages; Con A bound to all of the spermatial
surface except the spermatial appendages; WGA labeled
a narrow region that connected the spermatial body and
appendages. The experiments illustrated that SBA and
Con A binding sites on the spermatial surface might be
involved in gamete recognition (Kim and Kim, 1999).
LBS were also found on the surface of somatic proto-
plasts of higher plants (Burgess and Linstead, 1976;
Larkin, 1978; Walko et al., 1987). Recently we devel-
oped a reliable protocol to test the LBS on the surface of
sperm cells in angiosperm plants (Sun et al., 2002).
Unevenly distributed Con A binding sites were ob-
served on the membrane of maize sperms. However,
LBS distribution on the surface of female gametes in
angiosperm plants has not been reported. One of the
main reasons for this is that female gametes are deeply
embedded in various plant tissues. It is therefore diffi-
cult to isolate them from those tissues. Although some
techniques are now available for the isolation of female
gametes from a few species of higher plants, it is usually
necessary to apply enzymatic treatment to release the
cells. Some early works suggested that the treatment of
most enzymes did not modify the LBS on the membrane
surface of plant somatic protoplasts (Walko et al.,
1987). However, others showed a combination of
cellulysin and pectolyase Y-23 affected the peanut (Ara-
chis hypogaea) agglutinin (PNA) binding sites, but not
Ricinus communis agglutinin (RCA) binding sites (Sun
et al., 1992). It is still questionable whether such an
isolation process can change the basic pattern of LBS
distribution on the gamete cells.
In Torenia fournieri, the embryo sac is naturally half
naked. Its egg cell, two synergids, and approximately
half of the central cell are clearly observed without any
treatment. Therefore, it is convenient to study the char-
acters of female cells and visualize in situ fertilization
events. Taking advantage of the material, we set up a
procedure for probing LBS on the surface of female
cells both in situ and in vitro. A careful comparison be-
tween the labeling of isolated and in situ gametes was also
carried out to investigate the basic pattern of LBS distribu-
tion on the female cell surface and mainly evaluate the
influence of isolation procedure on lectin labeling. Con A
and WGA were chosen as probes in the present work.
MATERIALS AND METHODS
Materials
Plants of Torenia fournieri L. were grown in a greenhouse
under standard conditions, 16-h photoperiods and 25 ºC,
in Wuhan University. Unpollinated floral buds were har-
vested 2 days after anthesis for preparation, in which
embryo sacs were fully mature but not yet fertilized.
Con A and WGA conjugated with FITC (Sigma)
were dissolved in 13% mannitol solution adjusted to
0.72 mol/l. The concentrations of the FITC-Con A and
FITC-WGA stock solution were 400 and 1000 μg/ml,
respectively.
Isolation of embryo sacs and female cells
The isolation process was performed as described by
Mól (1986) with some modifications. Ovaries were col-
lected and ovules were then carefully dissected from the
ovaries with handmade glass needles. The ovules were
put into 2 ml enzyme solution composed of 13% manni-
tol, 1.5% cellulase R-10 (Yakult Honsha Co. Ltd),
0.75% hemicellulase (Sigma), 0.375% macerozyme
R-10 (Yakult Honsha Co. Ltd), and 0.0375% pectolyase
Y-23 (Sheishin Pharmaceutical, Tokyo), pH 5.4. After
incubation in the solution for 1 h in the dark at 25 ºC,
female cells were released from the ovules. Calcofluor
white ST (Sigma) was used to detect whether any cell
wall remained on the surface of the cells. Fluorescein
diacetate (Sigma) was applied for a viability test.
Lectin labeling on isolated female cells
Labeling manipulation was carried out as previously de-
scribed (Johnson et al., 1975; Sun et al., 2002) with some
modifications. Isolated embryo sacs or female cells were
collected individually and washed in a solution of 13%
mannitol and then placed in a droplet of the same solution,
which was covered with mineral oil. The volume of the
droplet was about 200 nl. After 200 nl FITC-Con A or
FITC-WGA stock solution had been added to the drop-
lets containing female cells, they were incubated for
15–90 min at room temperature. 50–400 μg/ml of FITC-
Con A or 400–1000 μg/ml FITC-WGA was used as the
final concentration to select the best condition. After
incubation the female cells were washed in several

Fang et al. / Lectin binding sites on female cells
85
changes of 13% mannitol to remove the remaining un-
bound lectin before observation.
Lectin labeling on female cells in situ
Ovules were dissected from ovaries and placed into a
drop of low concentration enzyme solution composed of
13% mannitol, 0.2% cellulase R-10, 0.1% hemicellu-
lase, 0.05% macerozyme R-10, and 0.005% pectolyase
Y-23, pH 5.4. The droplets were also covered with a thin
layer of mineral oil. At the same time, the same volume
of lectin stock solution was added. The final concentra-
tion of FITC-Con A or FITC-WGA was 200 or
500 μg/ml, respectively. The ovules were incubated for
1 h. After incubation, the ovules were washed at least
three times with 13% mannitol solution. In this case, the
cell wall of female cells in the embryo sac was digested,
but the positions of the cells in situ remained un-
changed. Since the cells were still protected in the em-
bryo sacs, possible mechanical damage was avoided.
The ovules were also directly incubated in lectin
solution without any enzyme treatment, following the
same procedure described above except that the incuba-
tion time was prolonged to 3 h.
Controls
In the case of either isolated or in situ female cells,
controls were set, in which the material was treated by the
same methods described above but without lectin staining.
Mechanically isolated female cells were also labeled in the
same way to test the influence of enzymatic treatment on
LBS distribution. The binding specificity was tested fol-
lowing the same procedure described above, but the incu-
bation medium contained competing monosaccharides,
namely D-mannose for Con A or D-N-acetyl-glucosamine
for WGA. In this case the material was pre-incubated with
0.1 M mannose or D-N-acetyl-glucosamine for at least
30 min before the labeling (following Walko et al.,
1987); 0.1 mM calcium chloride was added to the solution
to test its influence on lectin binding to the cell surface.
Observation, image collection, and analyses
Isolation of ovules and female cells and labeling manipu-
lation were performed under an inverted microscope.
Fluorescent signal was observed using a Leica DM IRB
inverted microscope. Images were viewed and recorded
by a Cooled CCD (Charge Couple Device, MicroMAX
Princeton Instruments, Inc.). The relevant data were
collected and the relative intensity of fluorescence on
different cells was calculated and compared using
MetaMorph software (Universal Imaging Corporation
Inc., El-Husseini et al., 2002). On the labeled membrane,
6 sites were chosen randomly for calculating average
fluorescence intensity of a cell. Ten repeats were
performed for calculating the average fluorescence
intensity of each kind of labeled cell.
RESULTS
Female cell isolation and lectin labeling
Female cells were successfully isolated from the flowers
two days after anthesis. After incubation in enzyme
solution for 1 h, the central cell and egg apparatus were
gradually released from the ovules. Usually, four female
cells or three cells of egg apparatus were released to-
gether as an aggregate. They later separated further from
each other in the solution. Quite often, a pair of syner-
gids were linked. Sometimes, component cells of egg
apparatus fused into one big cell. Isolated central cells
(Pl. I, 1a) were about 72 μm in diameter, with a large
vacuole occupying most of the cell. The nucleus
and cytoplasm were located in the center. The egg cell
(Pl. I, 2a) was about 29 μm in diameter while the syner-
gid (Pl. I, 3a) was approximately 25 μm in diameter. In
an egg cell, dense cytoplasm and cytoplasm strands
crossing the vacuole were observed, whereas in the
synergid most of the cytoplasm accumulated at one end,
and a large vacuole at the other end. Three female cells
could be easily distinguished. CW staining and cell
fusion experiments confirmed they were real proto-
plasts. Strong FDA fluorescence and active cytoplasmic
streaming indicated the viability of the cells.
Different concentrations of lectins and incubation
times were applied and compared. 200 μg/ml FITC-Con
A for 40 min or 500 μg/ml WGA for 60 min incubation
was suitable for keeping cell viability and having a good
fluorescent signal (data not shown). Therefore this
procedure was adopted in all our later experiments.
Calcium could obviously influence both FITC-Con A
and FITC-WGA binding to their binding sites. 0.1 mM
calcium could notably enhance the fluorescent signal on
the surface of female cells. Calcium, however, usually
made female cells stick to cover slips and thus made
further manipulation difficult. Without enzymatic treat-
ment egg cells could be mechanically isolated. CW
staining and fusion test verified their protoplast nature.
Comparison of mechanically isolated and enzymatically
isolated egg cells showed no significant differences be-
tween them in fluorescence intensity and distribution
pattern, indicating that enzymatic treatment used in this
study did not significantly modify Con A and WGA
binding sites on the membrane of egg cells. When
FITC-Con A or FITC-WGA solution was pre-incubated
with D-mannose or D-N-acetyl-glucosamine, respec-
tively, before labeling the cells, no fluorescent signal
was found on the surface of the cells. The results
confirmed that the fluorescent signal observed on the

Israel Journal of Plant Sciences 51 2003
86

Fang et al. / Lectin binding sites on female cells
87
cells was due to specific binding. No autofluorescence
was found on the isolated female cells under the same
condition in control experiments.
Distribution characters of LBS on different
female cells
In our experiments, 66 egg cells, 198 central cells, 34
synergids, 30 pairs of linked synergids, and 22 aggrega-
tions of egg apparatus were used in Con A labeling. Con
A binding sites were found on the plasma membrane of
all female cells (Pl. I, 1b,2b,3b). Among 66 labeled egg
cells, 90% showed bright fluorescence of the Con A-
binding site complex in a smooth ring around the egg
cells, but some of them showed uneven distribution of
Plate I previous page. Fluorescence of LBS on the surface of embryo sacs and female cells in Torenia fournieri. Bar = 11 μm in
1, 6, 7, 8, 9, and 10. Bar = 6.5 μm in 2, 3, 4, and 5.
1a. A freshly isolated central cell. 1b. The fluorescence image of the same cell labeled by FITC-Con A. 2a. An isolated egg cell.
Notice its cytoplasm. 2b. The fluorescence image of the same cell labeled by FITC-Con A. 3a. An isolated synergid. 3b. The
fluorescence image of the same cell labeled by FITC-Con A. 4a. A pair of isolated synergids showing filiform apparatus (arrow).
4b. The same synergids labeled by FITC-Con A. Notice the strong fluorescence of the filiform apparatus (arrow). 5a. An isolated
egg cell. 5b. The fluorescence image of the same cell labeled by FITC-WGA, showing weak fluorescence. 6a. An embryo sac
protuberance after enzymatic treatment, showing that the membrane of the embryo sac was still intact and the egg cell inside had
become protoplast (arrow). 6b. The same embryo sac labeled by FITC-Con A. Long arrow indicates weak fluorescence on the
egg cell membrane and short one indicates the fluorescence on central cell membrane. Micropyle end of the embryo sac and the
synergid (star) with filiform apparatus show the strongest fluorescence. 7a. A broken embryo sac after brief dissection without
any enzymatic treatment. The egg cell (arrow) was released from its cell wall and clearly observed at this focus. A star indicates
a wounded synergid. 7b. The fluorescence image of 7a labeled by FITC-Con A. The egg cell (arrow) shows clear fluorescence.
The wounded synergid (star) and cytoplasm mass attached to the embryo sac wall also show strong fluorescence. 8a. An embryo
sac showing a pair of synergids (arrows). 8b. The fluorescence image of 8a labeled by FITC-Con A. Notice that fluorescence in
the filiform apparatus is stronger than that of other part (arrow). 9a. An embryo sac without enzymatic treatment showing an egg
cell and a degenerated synergid (arrow) on this focus. 9b. The same embryo sac labeled by FITC-Con A showing strong fluorescence
in the degenerated synergid (arrow). 10a. An embryo sac without treatment with FITC-lectins. 10b. The same embryo sac as in 10a
showing the autofluorescence from integument part but no fluorescent signal from the embryo sac and female cells.
0
20
40
60
80
100
120
140
Central
cell
Syergid Egg cell
Fluorescent strength (Pixel)
0
10
20
30
40
50
60
Central
cell
Syergid Egg cell
Fluorescent strength (Pixel)
Fig. 1. Comparison of fluorescence intensity among different
isolated female cells labeled by FITC-Con A in Torenia
fournieri. Each value in this figure is the average intensity of
10 cells selected randomly. The comparison is made under the
same condition of 200 μg/ml Con A dissolved in 13% manni-
tol and 40 min incubation. White columns represent the fluo-
rescence value of the cell indicated. Black columns represent
the value (= 0) of the control of corresponding cells. The error
bars indicate the standard deviation from the means of ten
independent experiments.
the fluorescent signal on the plasma membrane. A simi-
lar phenomenon was also found on the central cell.
There was no significant difference between the fluores-
cence intensity of a pair of synergids, but stronger fluo-
rescence was observed at the filiform apparatus and
where two cells connected (Pl. I, 4b).
Image analyses showed that the intensity of fluores-
cence on different female cells varied. Among them the
central cell showed the strongest fluorescent signal. The
signal on the synergid was stronger than that on the egg
cell (Fig. 1). Among 90 central cells, 42 synergids, and
31 egg cells incubated with FITC-WGA, 90% of cells
were clearly labeled and showed binding site distribution
similar to the patterns of FITC-Con A labeling (Pl. I,
Fig. 2. Comparison of fluorescence intensity among different
isolated female cells labeled by FITC-WGA in Torenia
fournieri. Each value in this figure is the average intensity of
10 cells selected randomly. The comparison is made under the
same condition of 500 μg/ml WGA dissolved in 13% manni-
tol and 60 min incubation. White columns represent the fluo-
rescence value of the cell indicated. Black columns represent
the value (= 0) of the control of corresponding cells. The error
bars indicate the standard deviation from the means of ten
independent experiments.

Israel Journal of Plant Sciences 51 2003
88
5b), but the fluorescence intensity was much lower
(Fig. 2) than that labeled by FITC-Con A. Some fluores-
cent patches were often found on the membrane of
labeled cells.
In all controls, in which there was no Con A or WGA
in the solution, there was no fluorescent signal on any
female cells (Pl. I, 10b). The results further demon-
strated that labeling on the plasma membrane of the
protoplasts was due to specific binding of Con A to D-
mannose or WGA to β-D-N-acetyl-glucosamine resi-
dues located on the surface of female cells.
Distribution characters of LBS on the female cells
in situ
In order to examine the possible influence of isolation
manipulation itself on the distribution of LBS on the
membrane of female cells, intact embryo sacs were
treated with enzymes to digest cell walls and, at the
same time, stained by FITC-Con A or FITC-WGA so
that the female cells were labeled in situ and any me-
chanical injury was avoided. Compared with isolated
ones, the in situ female cells usually had to be incubated
longer in lectin solution for better binding. Our experi-
ments showed that after 1 h incubation the female cells
were viable and well-labeled by FITC-Con A. But for
FITC-WGA, it took longer to get a visible fluorescent
image and 1.5 h was usually adopted. Ca2+ was necessary
for the labeling.
Among 35 ovules observed, around 64% of central
cells, 77% of egg cells, and 82% of synergids in the
ovules were clearly labeled by FITC-Con A. The results
showed that Con A binding sites were evenly distributed
on the surface of embryo sacs except at the micropylar
end, where the filiform apparatus was located. The fluo-
rescence at the micropylar end and on synergids was
much stronger than that at other parts of the embryo sac
surface and also stronger than other female cells (Pl. I,
6a,6b). The fluorescence on these female cells in situ
showed no obvious difference from that on isolated cells
concerning the distribution pattern and signal strength.
Both central cells and egg cells showed smooth fluores-
cent rings (Pl. I, 6b) similar to that seen in vitro (Pl. I,
1b,2b). Thus, the embryo sac showed an obvious polar-
ity of Con A binding site distribution since its filiform
apparatus showed much stronger fluorescence than the
other part of the embryo sac.
Ovules without any enzyme treatment were also in-
cubated with FITC-Con A. The FITC-Con A solution
could enter the embryo sac through its micropyle and
label the female cells inside. No central cells were la-
beled in this case. The labeling could be improved by
brief dissection to open the embryo sac at the micropylar
end. Since all female cells in situ still had a cell wall,
FITC-Con A seemed to bind mainly to the cell wall
surface rather than to the plasma membrane of the cells.
By a brief dissection, some of the egg cells released
from their cell wall; the outline of the cell could be
clearly observed (Pl. I, 7a), and the fluorescent signal on
these egg cells was also similar to that on isolated ones,
indicated by its smooth fluorescent ring (Pl. I, 7b, com-
pare with 6b and 2b). However, the fluorescence of egg
cell and synergids with the cell wall was no longer a
uniform ring. In this case, fluorescence at the micropy-
lar end was much stronger than that at the other end
(Pl. I, 8b). Without enzymatic treatment the surface of
an embryo sac also showed a bright fluorescent signal,
especially on its micropylar part. Sometimes, one of two
synergids degenerated before fertilization and the fluo-
rescent signal from the degenerated synergid was very
strong (Pl. I, 9b). In the control of unlabeled embryo
sacs, there was no fluorescent signal on any female
cells; only the integument showed weak autofluor-
escence (Pl. I, 10b).
After FITC-WGA labeling in both enzymatically
treated and non-enzymatically treated embryo sacs, egg
cells and synergids showed very weak fluorescent sig-
nals and it was difficult to get clear images showing the
distribution pattern of WGA binding sites. However, very
strong fluorescence was also present on the filiform appa-
ratus in non-enzymatically treated embryo sacs.
DISCUSSION
The factors influencing lectin labeling have been exten-
sively studied with regard to the possible role of enzy-
matic treatment. Walko et al. (1987) demonstrated that
among enzymes tested, including cellulysin and pecto-
lyase Y-23, only driselase reduced the binding of lectin
to the surface of plant somatic protoplasts. Burgess and
Linstead (1976) also used macerozyme R-10 and cellu-
lase to isolate tobacco and vine protoplast, and found no
obvious influence on lectin labeling. Sun et al. (1992)
showed that cellulysin and pectolyase Y-23 modified
PNA binding sites. However, by their method for proto-
plast isolation, less than 50% of isolated protoplasts
were viable and nearly half of them were actually cyto-
protoplasts. Mechanical effects during the isolation pro-
cedure might be more serious than enzymatic treatment.
Walko et al. (1987) also reported that plant somatic
protoplast surface showed uniform fluorescence when
labeled by fluorescein-lectins in the presence of cal-
cium, even incubated for 60 min. But if calcium was
replaced by strontium or magnesium, the fluorescence
of the surface was slightly speckled. Burgess and

Fang et al. / Lectin binding sites on female cells
89
Linstead (1976) showed that protoplasts labeled by
Fe-Con A showed a patchy distribution, which was also
observed in protoplasts treated by Au-Con A with a
relatively high ratio of Con A:Au. Decreasing the ratio
of Con A:Au showed a uniform distribution of colloidal
gold on the protoplast. The results revealed that certain
concentration of ions could influence the basic pattern
of LBS distribution.
The incubation of rat liver cells with Con A and
horseradish peroxidase (HRP) without further re-
incubation resulted in a continuous labeling at the
plasma membrane, but the internalization of the plasma
membrane component was observed in labeled cells
reincubated with PBS (Roth, 1978). Obviously, pro-
longed incubation after lectin labeling could also
modify the distribution of LBS on plasma membrane
due to internalization.
In our experiments, all the cells were labeled under a
standard condition by means of micromanipulation. The
cells were incubated with FITC-Con A or FITC-WGA
in an ion-free and sugar-free solution, but with a cal-
cium supplement, which is necessary for WGA binding.
Fluorescent signals were observed and measured di-
rectly after labeling without further reincubation in cell
culture medium. Therefore, the fluorescent signal on the
membrane was stable and comparable in repeated ex-
periments. The smooth signal ring was observed on both
enzymatically isolated and mechanically released egg
cells, indicating the enzyme treatment and isolation ma-
nipulation did not notably influence the lectin binding to
the egg cell surface. Furthermore, the female cells re-
leased from ovules after maceration were protoplasts as
assayed by CW staining and cell fusion. Therefore the
labeled binding sites must have been located on the
membrane surface of the cells. Competitive inhibitor
treatment demonstrated that Con A and WGA labeling
were due to specific binding.
To sum up, our experiments revealed that (a) Con A
and WGA binding sites were found clearly on all the
female cells, on both their membrane and cell wall
surface; (b) The Con A binding sites were differently
distributed on different cells. Image analyses showed
that there were more binding sites on central cells and
synergids than on egg cells; (c) Female cells in vitro and
in situ showed similar intensity of fluorescent signal on
the surface of the membrane, indicating that the isola-
tion manipulation itself did not notably vary the surface
feature of their membrane. The distribution pattern of
the binding sites on the cells in two cases was also
similar; (d) The distribution of Con A and WGA binding
sites in an embryo sac showed obvious polarity. The
binding sites were located more at the micropylar part
and filiform apparatus where the pollen tube penetrates
into the embryo sac during fertilization.
Data from lower plants and animals showed that
lectins and their binding sites were involved in gametic
recognition and adhesion (Wassarman, 1992; Freeman,
1996; Kim and Kim, 1999). In higher plants, lectins
binding glycoproteins were found specific for different
stages of microspore and pollen development in tobacco
(Hrubá and Tupy´, 1999). Pretreatment of Gladiolus
gandavensis stigma with Con A decreased the adhesive
capacity for pollen grain and prevented compatible pol-
len tubes from penetrating. Pollen tube penetration
through the papillary cuticle was prevented when the
binding sites for Con A on the surface of stigma were
occupied (Knox et al., 1976; Clarke et al., 1979). The
results revealed a close relationship between LBS and
pollen tube growth and penetration. Recently Con A
binding sites were found on the surface of maize sperm
cells (Sun et al., 2002). According to our present results,
Con A binding sites are located on the membrane of all
female cells that are involved in fertilization. Whether
these sites play a role in gamete interaction or fusion
remains to be determined. It is interesting that Con A
binding sites are specially located at the micropylar end
of the embryo sac and the filiform apparatus of syner-
gids. This lectin–binding site interaction may be in-
volved in the interaction between pollen tubes and mi-
cropylar tissue and/or the embryo sacs, as also happens
on the surface of the stigma. The filiform apparatus is a
thickened area of proliferated cell wall of mature syner-
gids and is present in most species, suggesting a central
role in the attraction and acceptance of the pollen tube
into the embryo sac (Russell, 1992). A recent study of
Torenia fournieri using laser cell ablation confirmed
that it was the synergid that causes the pollen tube to
grow toward the embryo sac (Higashiyama et al., 2001).
However the mechanism of this attraction is not known
yet. The fact that there are more Con A and WGA
binding sites in the filiform apparatus than other female
cells may offer a clue for lectin–binding site interaction
regarding pollen tube guidance and penetration into the
embryo sac. The dynamics of LBS distribution in the
embryo sac or along the pathway of the pollen tube
during fertilization deserve further studies.
ACKNOWLEDGMENTS
This research is supported by the National Natural Sci-
ence Foundation of China (39970368) and State Key
Basic Research and Development Plan of China
(G1999053909). The project was also supported by the
National Outstanding Youth Science Fund (30225006).

Israel Journal of Plant Sciences 51 2003
90
REFERENCES
Burgess, J., Linstead, P.J. 1976. Ultrastructural studies of the
binding of concanavalin A to the plasmalemma of higher
plant protoplasts. Planta 130: 73–79.
Clarke, A., Gleeson, P., Harrison, S., Knox, R.B. 1979. Pollen–
stigma interactions: identification and characterization of
surface components with recognition potential. Proc. Natl.
Acad. Sci. U.S.A. 76: 3358–3362.
El-Husseini, A. E-D., Schnell, E., Dakoji, S., Sweeney, N.,
Zhou, Q., Prange, O., Gauthier-Campbell, C., Aguilera-
Moreno, A., Nicoll, R.A., Bredt, D.S. 2002. Synaptic
strength regulated by palmitate cycling on PSD-95. Cell
108: 849–863.
Freeman, G. 1996. The role of localized cell surface-associ-
ated glycoproteins during fertilization in the Hydrozoan
Aequorea. Dev. Biol. 179: 17–26.
Higashiyama, T., Yabe, S., Sasaki, N., Nishimura, Y.,
Miyagishima, S., Kuroiwa, H., Kuroiwa, T. 2001. Pollen
tube attraction by the synergid cell. Science 293: 1480–1483.
Hrubá, P., Tupy, J. 1999. N-glycoproteins specific for differ-
ent stages of microspore and pollen development in to-
bacco. Plant Sci. 141: 29–40.
Johnson, M.H., Eager, D., Muggleton-Harris, A., Grave, H.M.
1975. Mosaicism in organization of concanavalin A recep-
tors on surface membrane of mouse egg. Nature 257: 321–
322.
Kim, S.H., Kim, G.H. 1999. Cell-cell recognition during fer-
tilization in the red alga, Aglaothamnion oosumiense
(Ceramiaceae, Rhodophyta). Hydrobiologia 398/399: 81–
89.
Knox, R.B., Clarke, A., Harrison, S., Smith, P., Marchalonis,
J.J. 1976. Cell recognition in plants: determinants of the
stigma surface and their pollen interactions. Proc. Natl.
Acad. Sci. U.S.A. 73: 2788–2792.
Larkin, P.J. 1978. Plant protoplast agglutination by lectins.
Plant Physiol. 61: 626–629.
McCaig, C.D., Robinson, K.R. 1982. The distribution of lectin
receptors on the plasma membrane of the fertilized sea
urchin egg during first and second cleavage. Dev. Biol. 92:
197–202.
Mól, R. 1986. Isolation of protoplasts from female gameto-
phytes of Torenia fournieri. Plant Cell. Rep. 3: 202–206.
Peumans, W.J., Van Damme, E.J.M. 1995. Lectins as plant
defense proteins. Plant Physiol. 109: 347–352.
Roth, J. 1978. Compensatory membrane biogenesis and exo-
cytosis as a result of concanavalin A-induced membrane
internalization. Exp. Cell. Res. 114: 31–38.
Russell, S.D. 1992. Double fertilization. Int. Rev. Cytol. 140:
357–388.
Sun, S., Furtula, V., Nothnagel, E.A. 1992. Mechanical release
and lectin labeling of maize root protoplasts. Protoplasma
169: 49–56.
Sun, M.X., Kranz, E., Moscatelli, A., Yang, H.Y., Lorz, H.,
Cresti, M. 2002. A reliable protocol for direct detection of
lectin receptors on the plasma membrane of a single living
sperm cell in maize. Sex. Plant. Reprod. 15: 53–55.
Walko, R.M., Furtula, V., Nothnagel, E.A. 1987. Analysis of
labeling of plant protoplast surface by fluorophore-conju-
gated lectins. Protoplasma 141: 33–46.
Wassarman, P.M. 1992. Mouse gamete adhesion molecules.
Biol. Reprod. 46: 186–191.