BIOLOGY OF REPRODUCTION 55, 1133-1146 (1996)
Regionalization and Redistribution of Membrane Phospholipids and Cholesterol in
Mouse Spermatozoa during In Vitro Capacitation
Yan Lin and Frederick W.K. Kan 2
Department of Anatomy and Cell Biology, Faculty of Medicine, Queen's University, Kingston,
Ontario, Canada K7L 3N6
ABSTRACT
Fracture-label, surface-replica, and routine freeze-fracture
techniques were used in combination with phospholipase A2colloidal gold (PLA 2-CG) and filipin as probes to study changes
inthe distribution of phospholipids and cholesterol, respectively,
inmorphologically defined plasma membrane domains of mouse
spermatozoa during in vitro capacitation. In noncapacitated
spermatozoa, quantitative analysis revealed that the fractured
plasma membrane overlying the equatorial segment carried the
highest PLA2-CG labeling density. The next highest labeling densities were found in the anterior acrosome region and the postacrosomal region. On the external surface of the plasma membrane revealed by surface replicas, a uniform distribution of
PLA2-CG was confined mainly to the acrosomal region of the
head. The plasma membrane of the sperm tail had a relatively
low labeling density for PLA 2-CG. In freeze-fracture replicas of
filipin-treated spermatozoa, the labeling density of filipin/sterol
complexes (FSCs) was high in the plasma membrane over the
acrosomal region where the FSCs were uniformly distributed.
The postacrosomal region was weakly labeled. After in vitro capacitation, the densities of PLA2-CG and FSCs were significantly
reduced in the fractured plasma membrane of the sperm head
and the middle piece of the tail. However, surface replicas revealed an increased PLA2-CG labeling on the external surface of
the plasma membrane covering the postacrosomal region, the
middle piece, and the principal piece. Another major change
detected in capacitated spermatozoa was the presence of small
aggregates and patches of elevated, membrane-associated particles on the surface-replicated plasma membrane in the upper
portion of the postacrosomal domain. Here the PLA 2-CG labeling density was found to be higher than in noncapacitated spermatozoa. These results provide new information with respect to
the reorganization and redistribution of phospholipids in specific regions of the plasma membrane during capacitation and
provide further support for the concept that removal or loss of
antifusigenic sterol from the sperm plasma membrane constitutes an important step of the capacitation process.
INTRODUCTION
Prior to fertilization, mammalian spermatozoa require a
period of residence in the female reproductive tract where
they undergo various physiological changes that allow the
sperm to acquire the ability to recognize and fertilize the
egg [1]. Capacitation, a process that takes place in the female reproductive tract, is believed to be a multifaceted
membrane phenomenon that results in dramatic changes in
plasma membrane lipid and protein composition as well as
in the lateral distribution or regionalization of these components [2-5]. For instance, membrane cholesterol depletion and a concomitant decrease of the cholesterol-to-phosAccepted July 2, 1996.
Received August 25, 1995.
'This work was supported by a grant (MT-10904) from the Medical
Research Council of Canada.
2
Correspondence. FAX: (613) 545-2566.
pholipid molar ratio in capacitated spermatozoa have been
documented [6, 7]. Biochemical techniques have also
shown changes in specific classes of plasma membrane
phospholipids, including a decrease in the amount of sphingomyelin and inositol phospholipids and an increase in the
amount of the diacylglycerol and free fatty acids [8].
Freeze-fracture studies of changes in sterol and anionic
phospholipid distribution in human sperm plasma membrane during in vitro capacitation have demonstrated that
capacitation induces the exclusion of filipin/sterol complexes (FSCs) from small areas of the plasma membrane covering the acrosomal region and causes a reduction in the
number of FSCs in the postacrosomal region [9]. The fluorescence recovery after photobleaching (FRAP) technique
showed that in vitro capacitation results in an increase in
the lipid analogue diffusion rate in the acrosomal region of
the head, the middle piece, and the principal piece as well
as a decrease in diffusing fraction in the acrosomal region
of the head [10].
In the present study, we investigated whether changes in
the distribution of phospholipids and cholesterol of the
plasma membrane occur during in vitro capacitation of
mouse spermatozoa, and if so, the specific localization of
these membrane components. Since sperm plasma membrane is known to be regionalized in both the distribution
of its membrane components and its functions [10], we employed fracture-label [11], surface-replica [12], and routine
freeze-fracture techniques [13] in conjunction with phospholipase A2 -colloidal gold (PLA 2-CG) [14] and filipin to
study the distribution of phospholipids and cholesterol in
morphologically defined regions of the plasma membrane
of mouse spermatozoa. In fracture-label replicas, both the
protoplasmic face (PMp) and the exoplasmic face (PMe) of
the plasma membrane are exposed and can be labeled. The
surface replicas reveal the overall organization of the external surface of the plasma membrane. In surface replicas,
coating materials derived from the male reproductive tract
are mainly localized on the external surface of the plasma
membrane of the sperm head [15]. It has been suggested
that removal or alteration of coating materials is an important step of capacitation [16, 17]. By quantitative analysis
of colloidal gold labeling and FSCs utilizing a combination
of the techniques mentioned above, we demonstrate the reorganization and redistribution of phospholipids in specific
regions of the sperm plasma membrane during in vitro capacitation.
MATERIALS AND METHODS
Retrieval of Sperm from the Epididymides and
Vas Deferens
Twenty male mice (CD1 strain, 8-10 wk old) were used.
The animals were killed by cervical dislocation. The caudae
epididymides were excised and punctured with a needle
(26-1/2 gauge) to extrude sperm. The sperm were squeezed
1133
1134
LIN AND KAN
out gently from the proximal portion of the vas deferens
by a pair of forceps and allowed to disperse for 5 min at
room temperature in a Krebs' Ringer bicarbonate medium
(KRB medium [pH 7.3]; 119.4 mM NaCl, 4.8 mM KC1,
1.7 mM CaC1 2, 1.2 mM KH 2PO 4, 1.2 mM Mg 2SO 4, 25.1
mM NaHCO 3, 25 mM sodium lactate, 1 mM sodium pyruvate, 5.6 mM glucose) [18]. The cell debris and large
aggregates of immotile sperm were removed by a two-step
discontinuous Percoll gradient.
Isolation of Motile Sperm and In Vitro Capacitation
Sperm suspensions were centrifuged through a two-step
discontinuous Percoll gradient (45% and 90%) [18]. Discontinuous Percoll gradients were made by deposition of 1
ml of each concentration of Percoll solution in a sterilized
16 x 125-mm tube, beginning at the bottom of the tube
with 90% Percoll and followed by the 45% Percoll fraction.
Once the gradient was prepared, 2 ml of the sperm suspension was placed onto the 45% Percoll fraction and centrifuged at 1000 x g for 30 min at room temperature. Sperm
were separated on the basis of differential densities into
three fractions: 1) the first interphase layer between the initial layer of sperm suspension and the 45% Percoll, 2) the
45% Percoll fraction, and 3) the second interphase layer
between the 45% Percoll and 90% Percoll. Sperm motility
was assessed under a light microscope. Sperm collected
from the second interphase layer were mainly motile,
whereas those from the 45% Percoll fraction and the first
interphase layer were largely aggregated and immotile. For
in vitro capacitation, motile cells were resuspended in KRB
medium supplemented with 15 mg/ml BSA (KRB-BSA)
and then incubated at 37°C under 5% CO2 for 1-1/2 h. At
the end of the incubation, capacitated sperm were washed
twice with PBS by resuspension and centrifugation at 600
x g for 5 min each; they were then fixed in 2.5% glutaraldehyde/0.1 M cacodylate buffer/0.2 M sucrose (final concentration) at room temperature for 1 h and washed twice
in PBS before being further processed. Motile, noncapacitated cells were used as controls and were also fixed with
glutaraldehyde as described above.
Fracture-Label Cytochemistry
Fixed noncapacitated and capacitated sperm were infiltrated with 30% glycerol in PBS and mixed in a solution
of 30% BSA in 30% glycerol. The sperm-BSA mixture (9
,ul) was dispersed on a Balzers-type copper disk, which was
then superimposed on another copper disk coated with 1 pIl
of 2.5% glutaraldehyde. The two copper disks with the
sperm cells sandwiched in between were strongly held together because of the cross-linking of BSA by glutaraldehyde. The copper disks were frozen rapidly in Freon 22
cooled at the temperature of liquid nitrogen. Freeze-fracture
of the BSA gels was performed by mechanical separation
of the complementary copper disks immersed in liquid nitrogen [19]. The copper disks, each carrying a complementary fractured half of the BSA gels, were thawed, deglycerinated [19], and incubated at room temperature for 1 h
in an undiluted solution of PLA 2-CG complex (colloidal
gold diameter, 11 nm) [14]. After labeling, the fractured
gels were osmicated, dehydrated in ethanol, and dried by
the Peldri II sublimation method as previously described
[20]. The fractured gels were replicated with platinum and
carbon in a Balzers BAF 400D (Balzers AG, Balzers,
Liechtenstein) freeze-fracture unit at ambient temperature
under a vacuum of 2 x 10-6 Torr. Replicas were obtained
after sequential shadowing, first with platinum at a fixed
angle of 45 ° and then with carbon at 90 °, using electron
beam gun evaporators and an oscillating quartz for monitoring thickness of replicas (approximately 2 nm for platinum and 25 nm for carbon). The replicated samples were
digested in a solution of sodium hypochlorite to remove the
cell debris. The replicas were washed three times in bi-distilled water, mounted on formvar-coated copper grids, and
examined with a Hitachi 7000 (Tokyo, Japan) transmission
electron microscope operated at 75 kV.
To assess the specificity of the labeling, controls were
performed by incubating the fractured BSA gels in a solution of free PLA 2 (0.5 mg/ml) for 1 h at 37 ° followed by
labeling with a PLA 2-CG complex (11-nm colloidal gold
diameter) for 1 h at room temperature before they were
further processed for freeze-fracture as described above.
Preparation of Surface Replicas
Fixed noncapacitated and capacitated sperm were directly incubated in an undiluted solution of PLA 2-CG (colloidal gold diameter, 11 nm) at room temperature for 1 h
and then washed twice in PBS. A drop of the labeled sperm
in suspension was placed on a formvar-coated nickel grid.
The grids were air dried for 10 min and replicated with
platinum (2 nm) and carbon (25 nm) in a Balzers 400D
freeze-fracture unit at ambient temperature as described
above. Formvar support film was dissolved by immersing
the grid in chloroform for 2 min; the replicas were cleaned
by floating the grid on sodium hypochlorite solution and
then washing three times in bi-distilled water. The replicas
were examined with a Hitachi 7000 transmission electron
microscope operated at 75 kV.
Filipin Treatment and Freeze-Fracture
Filipin (Sigma Chemical Co., St. Louis, MO) was initially dissolved in 10 1 of dimethyl sulfoxide; it was then
added to I ml of 0.01 M PBS to make a final concentration
of 0.01%. Glutaraldehyde-fixed noncapacitated and capacitated sperm were each incubated with a freshly prepared
filipin solution overnight at 4°C. At the end of the incubation, the cells were washed twice by resuspension and
centrifugation at 600 x g for 5 min each; this was followed
by equilibration with 30% glycerol/0.01 M PBS. The cells
were sandwiched between two Balzers-type copper disks
and frozen as described above. Freeze-fracture was carried
out at -130°C in a Balzers freeze-etch unit. The replicas
were obtained by shadowing with platinum (2 nm) and carbon (25 nm) in a manner similar to that described above.
Control samples were incubated in 0.01 mM PBS instead
of filipin solution for 1 h before they were further processed
for freeze-fracture.
Quantitative Evaluation
The labeling densities of the PLA2-CG complex and
FSCs over various plasma membrane domains of the sperm
were evaluated on positive electron micrographs enlarged
to x28 000. The mean labeling density, evaluated as the
number of gold particles and FSCs per micrometer square
surface area, was determined by using a Zeiss Mop-3 (Carl
Zeiss, Thornwood, NY) modular system. At least 20 photomicrographs were analyzed for each domain. Similar
quantitative evaluation was also performed on control samples. Background labeling was evaluated on 20 photomicrographs in randomly selected regions of the replicas away
SPERM MEMBRANE PHOSPHOLIPIDS AND CHOLESTEROL
1135
from labeled structures of the sperm. The statistical analysis
of the difference in the mean labeling densities for
PLA2 -CG and FSCs in various membrane domains was undertaken using Kruskal-Wallis one-way ANOVA with P
values < 0.001. Pairwise comparison analysis of different
groups was carried out by using the Student-NewmanKeuls method with P values < 0.05. The difference in the
labeling density for PLA 2-CG and FSCs before and after
capacitation was determined by Student's t-test with P values < 0.05.
RESULTS
Isolation of Motile Sperm and In Vitro Capacitation
Freshly retrieved caudal epididymal sperm suspension
comprised both motile and immotile sperm populations.
When these sperm were centrifuged through a two-step discontinuous Percoll gradient (45% and 90%), the sperm collected from the first interphase fraction (Fig. la) (between
the sperm suspension and the 45% Percoll) were mainly
immotile and co-sedimented with epididymal epithelial
cells and appeared to be abnormal in ultrastructure. The
sperm collected from the 45% fraction (Fig. lb) were also
mainly immotile and their tails were enclosed by membranous vesicles. On the other hand, the sperm collected from
the second interphase (Fig. c) between the 45% and the
90% Percoll were largely motile upon resuspension in
KRB-BSA. The cells of this motile fraction showed all the
morphological features characteristic of normal sperm.
These motile sperm were capacitated in KRB-BSA for 11/2 h. At this incubation time point, most of the sperm cells
possessed a hyperactivated motility pattern typical of that
of capacitated sperm.
General Topography of Mouse Sperm
The falciform head of the mouse sperm (Figs. 2 and 3)
is usually divided into two regions: the acrosomal region
and the postacrosomal region [15, 21, 22]. The acrosomal
region can be further subdivided into 1) the anteriorly located anterior acrosome region (acrosomal cap) and 2) the
posteriorly located equatorial segment. The sperm tail consists of the middle piece, the annulus, and the principal
piece. The annulus is a specialized membrane structure separating the middle piece from the principal piece. In the
middle piece, small intramembranous particles run in diagonal rows in some areas. In the principal piece, rectilinear
double strands of intramembranous particles, also known as
"zippers," run parallel to the longitudinal axis (Fig. 4a,
small arrows).
Fracture-LabelCytochemistry
After being washed, noncapacitated epididymal sperm
were subjected to freeze-fracture and then to PLA 2-CG
probe; gold particles were found to be differentially distributed among the three major morphologically defined do-
FIG. 1. Transmission electron micrographs of mouse spermatozoa collected from three different fractions after separation of motile from nonmotile spermatozoa by a discontinuous Percoll gradient: a)the first Percoll
gradient interphase (between the initial layer of sperm suspension and the
45% Percoll fraction) comprises sperm, disrupted epididymal epithelial
cells, and membranous vesicles (small arrows); (b) the tails of spermatozoa collected from the 45% Percoll fraction are enclosed in membrane
folds (small arrows) that form outpocketing with many vesicles inside; (c)
sperm collected from the second Percoll gradient interphase (between the
45% and 90% Percoll fraction) show normal morphological features that
are devoid of membranous vesicles and epididymal epithelial cells. a, Bar
= 1 pim; b, bar = 0.5 Ipm; c, bar = 1 Am.
LIN AND KAN
1136
Acrosomal
region
I/\
equatorial
segment
Postacrosomal
Annulus
anterior
acrosome
region
Middle
piece
Subacrosomal
ring
Principal
FIG. 2. Schematic drawing of the general topography of a mouse spermatozoon. The major domains on the sperm head are the acrosomal region and the postacrosomal region. They are separated from each other
by the subacrosomal ring. The acrosomal region can be further subdivided
into the anterior acrosome region (acrosomal cap) and the equatorial segment. The plasma membrane of the sperm tail consists of the middle
piece, the annulus, and the principal piece.
mains of the PMp of the sperm head (Fig. 3a). The equatorial segment carried the highest PLA 2-CG labeling density, which was significantly higher than that of either the
anterior acrosome region or the postacrosomal region (Fig.
5). The latter two membrane domains had a moderate
PLA2-CG labeling density, and no significant difference in
the number of gold particles was detected between these
two domains. In the tail, the PMp of the annulus also had
a moderate PLA2-CG labeling density (Fig. 4a), which was
significantly higher than that of the middle piece and the
principal piece (Fig. 5). When comparative analysis was
carried out between the two fractured leaflets of the plasma
membrane of the sperm head, it was found that the PMp
was four times more heavily labeled than the PMe (Fig. 6).
When motile sperm were cultured in KRB-BSA medium, capacitation did not change the overall distribution pattern of the PLA 2-CG labeling over various membrane domains (Fig. 3b). However, quantitative analysis showed that
the PLA 2-CG labeling was significantly reduced in the PMp
over the acrosomal region, the postacrosomal region, and
the middle piece (Fig. 5). No changes were detected in the
annulus and the principal piece (Fig. 4b). Thus, in vitro
capacitation resulted in a decrease of phospholipid concentration over the protoplasmic leaflet of the plasma membrane overlying the sperm head and the middle piece of the
tail.
In fracture-label replicas of control samples, no gold particles were detected in various membranous structures after
incubation of the BSA gels in a solution of free PLA2 prior
to labeling with a PLA2-CG complex (Fig. 3c), confirming
the specificity of the PLA2-CG labeling.
Surface Replicas of Whole Sperm Labeled with PLA-CG
Surface replicas were used in this study to map the distribution of phospholipids over the external surface of various plasma membrane domains. In noncapacitated sperm,
surface replicas revealed the presence of coating materials
in the acrosomal region (Fig. 7). After sperm were directly
incubated with a PLA2 -CG solution, a uniform distribution
of gold particles was detected in the acrosomal region of
the sperm head (Fig. 7). No significant difference in the
PLA2 -CG labeling density was detected between the ante-
rior acrosome region and the equatorial segment (Fig. 8).
However, the labeling intensity of PLA 2 -CG over the postacrosomal region was significantly lower than that found
in the anterior acrosome region and the equatorial segment.
Only a few gold particles were found over the plasma membrane of the postacrosomal region (Fig. 7). The margin between the acrosomal and postacrosomal regions is delimited
by a narrow serrated band, also named the subacrosomal
ring (Fig. 7). The subacrosomal ring separated the strongly
labeled acrosomal region from the poorly labeled postacrosomal region. In the tail the external surface of the plasma
membrane over the middle piece, the annulus, and the principal piece had a relatively low PLA 2-CG labeling density
(Fig. 9). No significant difference in the concentration of
gold particles was found among these three membrane domains (Fig. 8).
After in vitro capacitation, capacitated sperm exhibited
morphological changes on the external surface of the plasma membrane, especially over the equatorial segment and
the postacrosomal region. Two types of capacitated sperm
were identified based on the presence (Fig. 10) and absence
(Fig. 11) of coating materials covering the equatorial segment. As mentioned earlier, coating materials seen on the
external surface of the plasma membrane of the sperm head
are derived from the male reproductive tract, and removal
of these coating materials is a characteristic of capacitation.
In both types of capacitated sperm, however, small aggregates and patches of elevated particles, which were absent
in noncapacitated sperm, were found in the upper portion
of the postacrosomal region. These aggregates and patches
of elevated particles were concentrated immediately below
the subacrosomal ring on the lateral surface of the sperm
head (Figs. 10 and 11, see insets). When capacitated sperm
were labeled with PLA 2-CG, a high concentration of
PLA 2-CG labeling was detected below the subacrosomal
ring in the same region where aggregates of membrane
patches were found (Figs. 10 and 11, insets). Quantitative
analysis indicated that the number of gold particles in the
postacrosomal region of capacitated sperm was significantly higher than that for noncapacitated sperm (Fig. 8). It was
interesting to note that the occurrence of a decrease in
PLA 2-CG labeling in the equatorial segment was accompanied by the removal of the coating materials from the
same region (Fig. 11). In the tail, in contrast, PLA 2 -CG
labeling over the middle piece and the principal piece was
significantly elevated in comparison with that seen in noncapacitated sperm (Figs. 8 and 12). No statistically significant difference in the mean labeling density for PLA 2-CG
was found in the annulus before and after capacitation.
Therefore, in vitro capacitation not only induced morphological changes on the cell surface, but also resulted in the
reorganization and redistribution of phospholipids over the
equatorial segment and the postacrosomal region.
Freeze-Fracture of Filipin- Treated Sperm
The FSCs created hemispherical bulges (20-30 nm in
diameter) in freeze-fractured sperm membranes as previously reported [2, 23, 24]. When noncapacitated sperm
were treated with filipin, a heterogeneous distribution of
FSCs was detected over the fractured plasma membrane
(Fig. 13a). The PMp over the acrosomal region was strongly labeled by filipin. No significant difference in the number
of FSCs was found between the anterior acrosome region
(fusigenic region) and the equatorial segment (nonfusigenic
region) (see Fig. 15 for a statistical comparison). However,
SPERM MEMBRANE PHOSPHOLIPIDS AND CHOLESTEROL
1137
FIG. 3. a)Fracture-label preparation showing the distribution of PLA 2-CG labeling over the protoplasmic fractured face of the PMp of a noncapacitated mouse
sperm. The equatorial segment has the highest labeling density; the next highest densities are seen in the anterior acrosome region and the postacrosomal
region. b) PLA 2-CG labeling in the PMp overlying the three domains of the head of an in vitro-capacitated sperm. While gold particles are abundant,
morphometric evaluation revealed a significant decrease in labeling compared to that seen in noncapacitated sperm (see quantitative data in Fig. 5). c) Control
fracture-label replica of mouse sperm showing the absence of PLA 2-CG labeling over the plasma membrane after digestion of BSA gels for 1 h at 37°C in a solution
of free PLA 2 (0.5 mg/ml) before labeling with PLA 2-CG complex. Aa, anterior acrosome region; Eq, equatorial segment Pa, postacrosomal region. Bar = 1 m.
1138
LIN AND KAN
FIG. 4. a) Fracture-label replica showing a portion of the tail of a noncapacitated sperm tail. The labeling density of PLA 2-CG in the PMp overlying
the annulus (A) is significantly higher than that of the middle piece (MP) and the principal piece (PP). The small arrows indicate the zipper structures
in the principal piece. b) After capacitation, PLA2 -CG labeling is significantly reduced in the PMp overlying the middle piece. No significant change
in the number of gold particles was detected in the annulus and the principal piece. Bar = 0.5 jIm.
200
d
I
Qa
6
i
2
+1
N 150
140
II
nnn--anritted
EKI
capacitated
T
120
1E
e
+
+1
o
100
eec.
*C CC
N
T
I-
o
0
o
50
ID
.0
Z
2
z
0
_
I
K
K
K
K
K
T
T
"I
Hrg
middle
anterior equatorial postecrosoml piece
serosome segmenL
region
region
* eecc
* eec.
* eecc
80
0,
T
0
n
60
* CeeCI
* C ce
* . CC
* . C.
* CCC,
S..
o
,0
CCC..
* C *.
CCC..
* . ee
40
o
annulus
principal
piece
FIG. 5. Comparative analysis of PLA 2-CG labeling over various fractured
plasma membrane domains of the mouse sperm before and after in vitro
capacitation. In noncapacitated sperm, there was a statistically significant
difference in the mean labeling density of PLA 2-CG among the six different plasma membrane domains as determined by Kruskal-Wallis one-way
ANOVA with p values < 0.001. The desnsities were also significantly
different from one another (except among the anterior acrosome region,
the postacrosomal region, and the annulus and between the middle piece
and the principal piece) as evaluated by the pairwise multiple comparison
method with p values < 0.05. After in vitro capacitation, the labeling
density of PLA 2-CG was significantly reduced in various membrane domains (except for the annulus and the principal piece) as evaluated by
Student's t-test with p values < 0.05.
20
z
_
CC...
* eC·a
* . .e
* . Ce
* . CC
..
0
PMp(H)
D.U I
.
PMe(H)
Background
FIG. 6. PLA 2 -CG labeling over the two freeze-fractured leaflets of the
plasma membrane covering the head of noncapacitated sperm, and background labeling. The mean labeling density of PLA 2-CG in the PMp of
the sperm head was significantly higher than that of the PMe, and the
labeling density of PLA 2-CG of both PMp and PMe of the head was statistically higher than that of the background as evaluated, respectively, by
a Kruskal-Wallis one-way ANOVA with p values < 0.001 and a pairwise
multiple comparison method with p values < 0.05. PMp(H): Protoplasmic
leaflet of the plasma membrane of the sperm head. PMe(H), exoplasmic
leaflet of the plasma membrane of the sperm head.
SPERM MEMBRANE PHOSPHOLIPIDS AND CHOLESTEROL
1139
<
o
E
E
N
2o u
-o
'
E0
"-
o
U,
-
0
=)
5§0
SO
3
o
0 o-
O
e
_C
00
w. 0)"
° X c
COM
'g
OOICO0
2
-g
CO.e
uC
0
o,
0)
m
ES
0)m
om
-1
00-l
0)C
COn~~
U )O
Cg
00u
~"
8
CO0
LIN AND KAN
1140
100
non-capacitated
=
a
4)
I
80
capacitated
kX
+1
N
4)
0
I
T
60
IW
.
10
40
0
Ar
8
0
·FI
Z
4)
T
sl
T
20
0
LY
-
i-
anterior
acroesome
region
T
gr A:
I
211
i
equatorial postacroeemal
segment
region
l
s
l
middle
plece
I
I
I
LY
annulus
1
-
k\
l
principal
piece
FIG. 8. Comparative analysis of PLA 2-CG labeling over various plasma
membrane domains of surface-replicated mouse sperm before and after
in vitro capacitation. In noncapacitated sperm, there was a statistically
significant difference in the mean labeling density of PLA 2-CG among the
six different plasma membrane domains as determined by Kruskal-Wallis
one-way ANOVA with p values < 0.001. Pairwise multiple comparison
analysis with p values < 0.05 showed that the PLA 2-CG labeling was not
significantly different between the anterior acrosome region and the equatorial segment; however, the PLA 2-CG labeling of both membrane domains was significantly higher than that of the other four membrane domains. Also, no statistical difference in the number of gold particles was
detected among these four membrane domains. After in vitro capacitation, the PLA 2-CG labeling was significantly reduced in the plasma membrane over the acrosomal region, whereas the concentration of phospholipids in the plasma membrane over the postacrosomal region, the middle
piece, and the principal piece was significantly elevated as determined
by Student's t-test with p values < 0.05.
the density of FSCs was significantly lower in the PMp
overlying the postacrosomal region (Fig. 13a). The PMp of
the middle piece and the principal piece carried a moderate
density of FSCs (Fig. 14a). The annulus was devoid of
FSCs in both noncapacitated and capacitated sperm. In addition, the zipper structure in the principal piece was not
labeled by filipin (Fig. 14a). When sperm were incubated
in KRB-BSA to induce capacitation before filipin treatment, the number of FSCs was significantly reduced in the
PMp over the anterior acrosome region and the equatorial
segment of the head (Figs. 13b and 15). In addition, a significant decrease in the number of FSCs was also seen in
the postacrosomal region and the middle piece as well as
the principal piece as compared with their counterparts in
noncapacitated sperm (Figs. 14b and 15). Therefore, in
vitro capacitation caused a decrease in the concentration of
cholesterol over various freeze-fractured plasma membrane
domains except for the annulus separating the middle piece
from the principal piece. Freeze-fracture replicas of control
samples showed that FSCs were absent over various fractured membrane structures of the sperm head (Fig. 13c),
confirming the specificity of labeling.
DISCUSSION
On the basis
immotile sperm,
(45% and 90%)
arate these two
of the differential densities of motile and
a two-step discontinuous Percoll gradient
[18] was used in the present study to sepcell populations. The motile sperm were
FIG. 9. Surface replica of a noncapacitated sperm showing weak
PLA 2-CG labeling over the three plasma membrane domains of the tail.
The zipper (small arrows) is unlabeled. MP: middle piece; A: annulus; PP:
principal piece. Bar = 0.5 pim.
denser and were sedimented as a dense cell layer between
the 45% and 90% Percoll fractions, whereas the immotile
sperm remained in the 45% Percoll and in the first interphase
layer (between the initial layer of sperm suspension and the
45% Percoll). More than 90% of the sperm collected from
the second interphase (between the 45% and 90% Percoll)
were motile as verified on the phase-contrast microscope and
normal in their ultrastructure as revealed in Epon-embedded
sections. When incubated in KRB-BSA medium for 1-1/2
h, a majority of the sperm possessed a hyperactivated motility pattern typical of that of capacitated sperm; such sperm
have been shown previously to bind to the egg surface [10].
Sperm incubated in a similar medium for a period of more
than 1-1/2 h showed a marked decrease in motility, suggesting that they might be going through senescence.
SPERM MEMBRANE PHOSPHOLIPIDS AND CHOLESTEROL
1141
FIG. 10. Surface replica of acapacitated sperm with coating materials (small arrows), showing a decrease of PLA 2-CG labeling density over the anterior
acrosome region (Aa) and the equatorial segment (Eq). However, a high concentration of PLA 2-CG labeling is detected in the upper portion of the
postacrosomal region (Pa). Bar = 1 RIm. Inset: High magnification of the region delimited by the box in Figure 10, showing the strongly labeled upper
portion of the postacrosomal region (Pa) separated from the weakly labeled equatorial segment by the subacrosomal ring (arrowheads). Bar = 0.5 apm.
In Vitro Capacitation Causes a Decrease of Phospholipid
Concentration in the Plasma Membrane of the
Mouse Sperm
Previous biochemical studies have shown no significant
changes in total phospholipids of the plasma membrane
during in vitro acrosome-like reaction, but changes have
been found to occur in specific classes of plasma membrane
lipids [3, 8]. In the present study, using fracture-labeling in
combination with PLA 2-CG, we demonstrated that in vitro
capacitation resulted in a significant decrease in PLA 2-CG
labeling in the PMp of the sperm head, especially over the
acrosomal region, and in the middle piece of the tail. No
changes were detected in the PMp of the annulus and the
principal piece. Physiological changes that occur in the acrosomal region of capacitated sperm have been previously
studied by FRAP with use of a fluorescent lipid analogue
[10]. It is possible that during capacitation, the decrease in
1142
LIN AND KAN
e
u
-
.E
C
-
0
O
C
EE
.)Ou
2E
oU
oE
_C
co-
<E
0.
M'C) O
._s
~.0
E
c5
C
c)
C
SPERM MEMBRANE PHOSPHOLIPIDS AND CHOLESTEROL
1143
the PLA2 -CG labeling in the plasma membrane over these
specific regions was due to the hydrolysis of phospholipids
by sperm endogenous phospholipases. The membrane-associated phospholipase A2 has been found to be distributed
in the mouse sperm plasma membrane, and its activity is
increased during capacitation and acrosome reaction [25].
Activation of the membrane phospholipase [26] and phospholipase D [27] also occurs during capacitation and acrosome reaction. The activation of these membrane phospholipases will hydrolyze the phospholipids into lysophospholipids and cis-unsaturated free fatty acids, which are
thought to be endogenous fusogens [25]. Those hydrolytic
products are believed to cause local perturbation in membrane, which may facilitate membrane fusion [28]. Therefore, during capacitation, changes in the PLA 2-CG labeling
of the PMp over the anterior acrosome region and the equatorial segment indicate that a decrease in phospholipid content may facilitate the occurrence of the acrosome reaction
and the initial binding of sperm to the egg; the change in
the membrane phospholipid concentration of the postacrosomal region may be involved in the preparation of this
region for sperm-egg fusion. Since the middle piece is the
energy source for motility, the change of phospholipid concentration in the middle piece may be involved in alterations of other plasma membrane components of the sperm
tail, resulting an increase in the intracellular Ca 2+ and
cAMP levels that support hyperactivated motility for penetrating an egg [27].
Reorganization of Phospholipids Also Appears on the
Surface of the Plasma Membrane
Morphological alterations resulting from capacitation
have been extensively studied, although the biochemical basis for these changes is still uncertain [29]. Signs of changes
during capacitation occur in each of the distinct membrane
domains as each prepares to participate in the fertilization
process. Some of these changes include removal of the coating materials from the acrosomal region [30], relocation of
sperm surface antigens from one plasma membrane domain
to another, and an increase in intramembranous particles in
the postacrosomal region [24]. In the present study, a striking
morphological change, which was characterized by the presence of small aggregates of elevated particles on the surface
of the plasma membrane immediately below the subacrosomal ring, was found in surface replicas of capacitated
sperm. Because the external surface of the plasma membrane
is directly exposed to the capacitating environment, it is not
surprising that significant changes take place on the membrane surface during capacitation. Our results also reveal that
in vitro capacitation induces removal of the coating materials
mainly from the equatorial segment, constituting an important step in capacitation [27]. When noncapacitated and capacitated sperm were directly exposed to PLA 2-CG, a noticeable redistribution of phospholipids was found on the
external surface of the plasma membrane. In noncapacitated
sperm, a uniform distribution of PLA2 -CG was confined
mainly to the acrosomal region. After capacitation, our results indicate not only that PLA2-CG labeling is reduced in
the acrosomal region, but also that gold particles are redistributed and accumulated in the upper portion of the postacrosomal region immediately below the subacrosomal ring.
The redistributed PLA 2-CG labeling and the aggregates of
elevated membrane particles were colocalized below the subacrosomal ring. An increase of PLA 2-CG labeling was also
detected over the middle piece and the principal piece of
FIG. 12. In vitro capacitation results in an increase of PLA 2-CG labeling
on the external surface of the plasma membrane over the middle piece
(MP) and the principal piece (PP) of capacitated sperm tail. A: annulus.
Bar = 0.5 tFm.
capacitated sperm tail. It is noteworthy that no significant
changes were found over either the PMp or the external
surface of the plasma membrane of the annulus during capacitation, suggesting that capacitation does not change the
organization of phospholipids in the plasma membrane of
this particular domain. However, changes in the phospholipid
concentration of the surface plasma membrane over the anterior acrosome region and the equatorial segment may play
an important role in acrosome reaction and initial binding of
the sperm to the egg by these two membrane domains. The
striking changes resulting from capacitation in the postacrosomal region, characterized by reorganization of the membrane structure and redistribution of phospholipids, could influence reorganization of other plasma membrane molecules
in this region and may play a role in the preparation of this
region for sperm-egg fusion.
1144
LIN AND KAN
FIG. 13. a) Freeze-fracture preparation of noncapacitated sperm showing a high density of FSCs (arrows) in the PMp overlying the anterior acrosome
region and the equatorial segment. However, the number of FSCs is lower in the postacrosomal region. b) Freeze-fracture replica of a capacitated
mouse sperm treated with filipin. The density of FSCs (arrows) is significantly reduced in the PMp over the anterior acrosome region, equatorial segment,
and the postacrosomal region. c) Freeze-fracture replica of a control sample showing the absence of FSCs over the fractured plasma membrane when
sperm were incubated in 0.01 mM PBS for 1 h prior to freeze-fracture. Aa: anterior acrosome region; Eq: equatorial segment; Pa: postacrosomal region.
Bar = 1 m.
SPERM MEMBRANE PHOSPHOLIPIDS AND CHOLESTEROL
1145
FIG. 14. a) Freeze-fracture replica of a noncapacitated sperm showing a moderate labeling density of FSCs (arrows) in the PMp of the middle piece
(MP) and the principal piece (PP) of the tail. The annulus (A)and the zipper (arrowheads) are devoid of FSCs. b) After capacitation, the labeling density
of FSCs (arrows) is significantly reduced in the middle piece and the principal piece. The annulus is devoid of FSCs. Bar = 0.5 m.
Cholesterol Depletion of the Plasma Membrane Is an
Important Step of In Vitro Capacitation
Filipin, a polyene antibiotic, has been widely used in
freeze-fracture studies to visualize the distribution of sterol
in biological membrane [31-34]. Cholesterol is the major
sterol in the sperm plasma membrane. With the use of routine freeze-fracture technique in conjunction with filipin as
probe, cholesterol was revealed to be heterogeneously distributed in the protoplasmic fracture-face of the plasma
membrane of the mouse sperm. In vitro capacitation did
not change the distribution pattern of cholesterol in various
fractured membrane domains. However, a quantitative comparison of the number of FSCs between noncapacitated
sperm and capacitated sperm indicated a significant decrease of FSCs in the acrosomal region, the postacrosomal
region, the middle piece, and the principal piece. Cholesterol is believed to limit protein insertion into the phospholipid bilayer to restrict lateral mobility of membrane components and to modulate the activity of membrane proteins
by changing their conformation [35, 36]. It is also an antifusigenic component [24]. Membrane-mediated events
such as capacitation, acrosome reaction, and sperm-egg fusion require that the fusigenic areas in the plasma and acrosomal membranes participate in the steps leading to fertilization. Thus, a decrease in cholesterol level probably
destabilizes plasma membrane that can readily fuse with
the underlying outer acrosomal membrane [37]. It has been
suggested that sperm cholesterol can be transferred to capacitating medium during incubation [6] and to uterine fluid
during capacitation [37, 38]. Our results are in accord with
the notion that cholesterol depletion in the acrosomal region
and postacrosomal region may be a requirement for the
initiation of the acrosome reaction and sperm-egg binding.
Similarly, the changes in cholesterol level in the plasma
membrane over the middle piece and principal piece may
increase the lateral mobility of membrane components, supporting hyperactivated motility for penetrating an egg.
Conclusion
Using fracture-label, surface-replica, and routine freezefracture techniques in combination with PLA 2-CG and filipin
as probes, this study presents three major findings: 1) there
is a decrease of phospholipid concentration in the protoplasmic fracture-face of the plasma membrane overlying the
sperm head and the middle piece of the tail of capacitated
sperm; 2) in vitro capacitation induces morphological
changes and redistribution of phospholipids on the external
surface of the plasma membrane over the equatorial segment
and the postacrosomal region; 3) in vitro capacitation also
results in a decrease of cholesterol concentration in various
plasma membrane domains. Taken together, our results suggest that during in vitro capacitation, changes in the concentration of phospholipids of the plasma membrane over the
anterior acrosome region and the equatorial segment may be
involved in the acrosome reaction and initial binding of
sperm to the egg plasma membrane by providing more fusigenic areas in these two domains; reorganization and redistribution of phospholipids in the equatorial segment and
the postacrosomal region during capacitation may play an
important role in the preparation of these two regions for
1146
LIN AND KAN
350
300
E
e
n
X 250
+1
W
0 200
- 150
to
0
, 100
V
.0
Z
50
0I
1
I1
I
anterior equatorial
acroeome egment
region
I I\
m ~xl
i\N
poetmiddle annulus
acrosomal piece
region
,
principal
piece
FIG. 15. Comparative analysis of the labeling density of FSCs over various freeze-fractured plasma membrane domains of mouse sperm before
and after in vitro capacitation. In noncapacitated sperm, there was a statistically significant difference in the mean labeling density of FSCs among
the six different plasma membrane domains as determined by KruskalWallis one-way ANOVA with p values < 0.001. Pairwise multiple comparison analysis with p values < 0.05 showed that the labeling densities
of FSCs in the plasma membrane over the anterior acrosome region and
the equatorial segment were not significantly different from each other;
however, the labeling intensity of FSCs of both membrane domains was
significantly higher than that of the other four membrane domains. In
addition, no statistical difference in the number of FSCs was found among
the postacrosomal region, the middle piece, and the principal piece. After
capacitation, the labeling density of FSCs was significantly reduced in all
freeze-fractured membrane domains (except the annulus) as evaluated by
Student's t-test with p values < 0.05.
sperm-egg interaction (fusion); and removal of antifusigenic
sterol from various plasma membrane domains may facilitate
the occurrence of the acrosome reaction and support hyperactivated motility for penetrating an egg.
REFERENCES
1. Drobins EZ. Capacitation and the acrosome reaction. In: Scialli AT,
Zinaman MJ (eds.), Reproductive Toxicology and Infertility. New York:
McGraw-Hill; 1993: 77-132.
2. Bearer EL, Friend DS. Modifications of anionic lipid domains preceding
membrane fusion in guinea pig sperm. J Cell Biol 1982; 92:604-615.
3. Evans RW, Weaver DE, Clegg ED. Diacyl, alkenyl and alkyl ether
phospholipids in ejaculated, in utero-, and in vitro-incubated porcine
spermatozoa. J Lipid Res 1980; 21:223-228.
4. Koehler JK. The mammalian sperm surface: studies with specific labeling techniques. Int Rev Cytol 1978; 54:73-108.
5. Schwarz MA, Koehler JK. Alterations in lectin binding to guinea pig
spermatozoa accompanying in vitro capacitation and the acrosome reaction. Biol Reprod 1979; 21:1295-1307.
6. Davis BK, Byrne R, Hungund B. Studies on the mechanism of capacitation. II. Evidence for lipid transfer between plasma membrane of rat
sperm and serum albumin during capacitation in vitro. Biochim Biophys
Acta 1979; 558:257-266.
7. Langlais J, Zollinger M, Plante L, Chapdelaine A, Bleau G, Roberts
KD. Localization of cholesteryl sulfate in human spermatozoa in support of a hypothesis for the mechanism of capacitation. Proc Natl Acad
Sci USA 1981; 78:7266-7270.
8. Nikolopoulou M, Soucek DA, Vary JC. Modulation of the lipid composition of boar sperm plasma membranes during an acrosome reaction
in vitro. Arch Biochem Biophys 1986; 250:30-37.
9. Tesarik J, Flechon JE. Distribution of sterol and anionic lipids in human
sperm plasma membrane: effects of in vitro capacitation. J Ultrastruct
Mol Struct Res 1986; 97:227-237.
10. Wolf DE, Hagopian S, Ishijima S. Changes in sperm plasma membrane
lipid diffusibility after hyperactivation during in vitro capacitation. J
Cell Biol 1986; 102:1372-1377.
11. Pinto da Silva P Kachar B, Torrisi MR, Brown C, Parkison C. Freezefracture cytochemistry: replicas of critical point-dried cells and tissues
after fracture-label. Science 1981; 213:232-233.
12. Christopher WS, David D. Membrane organization in mouse spermatozoa revealed by freeze-etching. Ultrastruct Res 1974; 49:167-187.
13. Severs NJ. Freeze-fracture cytochemistry: review of method. J Electron
Microsc 1989; 13:175-203.
14. Kan FWK. Cytochemical evidence for the presence of phospholipids in
epithelial tight junction strands. J Histochem Cytochem 1993; 41:649656.
15. Yanagimachi R. Mechanisms of fertilization in mammals. In: Mastroianni L, Biggers JD (eds.), Fertilization and Embryonic Development In
Vitro. New York: Plenum Press; 1981: 81-182.
16. Bedford JM. Sperm capacitation and fertilization in mammals. Biol Reprod 1970; 2:128-158.
17. Johnson MH. The macromolecular organization of membranes and its
bearing on events leading up to fertilization. J Reprod Fertil 1975; 44:
167-184.
18. Tanphaichitr N, Smith J, Kates M. Levels of sulfogalactosylglycerolipid
in capacitated motile and immotile mouse spermatozoa. Biochem Cell
Biol 1990; 68:528-535.
19. Aguas AP, Pinto da Silva P. The acrosome membrane of boar sperm: a
Golgi-derived membrane poor in glycoconjugates. J Cell Biol 1985;
100:528-534.
20. Kan FWK. Use of Peldri 11as a sublimation dehydrant in place of critical
point drying in fracture label cytochemistry and in backscattered electron
imaging fracture label. J Electron Microsc Tech 1990; 14:21-31.
21. Friend DS, Fawcett DW Membrane differentiations in freeze-fractured
mammalian sperm. J Cell Biol 1974; 63:641-664.
22. Fawcett DW The mammalian spermatozoa. Dev Biol 1975; 44:394-436.
23. Bradley MP, Rayns DG, Forrester IT. Effects of filipin, digitonin, and
polymyxin B on plasma membrane of ram spermatozoa-an EM study.
Arch Androl 1980; 4:195-204.
24. Toshimori K, Higashi R, Oure C. Distribution of intramembranous particles and filipin-sterol complexes in mouse sperm membranes: polyene
antibiotic filipin treatment. Am J Anat 1985; 174:455-470.
25. Thakkar JK, East J, Seyler D, Franson RC. Surface-active phospholipase
A2 in mouse spermatozoa. Biochem Biophys Acta 1983; 754:44-50.
26. Roldan ERS, Murase T, Shi QX. Exocytosis in spermatozoa in response
to progesterone and zona pellucida. Science 1994; 266:1578-1581.
27. Yanagimachi R. Mammalian fertilization. In: The Physiology of Reproduction, Vol. 1, 2nd ed. New York: Raven Press; 1994: 189-317.
28. Meizel S, Turner K. Stimulation of an exocytotic event, the hamster
sperm acrosome reaction, by cis-unsaturated fatty acids. FEBS Lett
1983; 161:315-318.
29. Peterson RN, Gillott M, Hunt M, Russell LD. Organization of the boar
spermatozoa plasma membrane: evidence for separate domains (subdomains) of integral membrane proteins in the plasma membrane overlying the principal segment of the acrosome reaction. J Cell Sci 1987;
88:343-349.
30. Friend DS, Orici L, Perrelet A, Yanagimachi R. Membrane particle
changes attending the acrosome reaction in guinea pig spermatozoa. J
Cell Biol 1977; 74:561-577.
31. Elias PM, Friend DS, Goerke J. Membrane sterol heterogeneity: freezefracture detection with saponins and filipin. J Histochem Cytochem
1979; 27:1247-1260.
32. Friend DS. Plasma membrane diversity in a highly polarized cell. J Cell
Biol 1982; 93:243-249.
33. Pelletier RM, Vitale ML. Filipin vs enzymatic localization of cholesterol
on guinea pig, mink, and mallard duck testicular cells. J Histochem
Cytochem 1994; 42:1539-1554.
34. Robinson JM, Karnovsky MJ. Evaluation of the polyene antibiotic filipin as a cytochemical probe for membrane cholesterol. J Histochem
Cytochem 1980; 28:161-168.
35. Muller CP, Kruger GRE Modulation of membrane proteins by vertical
phase separation and membrane lipid fluidity. Basis for a new approach
to tumor immunotherapy. Anti-Cancer Res 1986; 6:1181-1194.
36. Yeagle PL. Lipid regulation of cell membrane structure and function.
FASEB J 1989; 3:1833-1842.
37. Davis BK. Uterine fluid binds sperm cholesterol during capacitation in
the rabbit. Biol Reprod 1981; 24(suppl 1): 149A (abstract 254).
38. Davis BK. Timing of fertilization in mammals: sperm cholesterol/phospholipid ratio as a determinant of the capacitation interval. Proc Natl
Acad Sci USA 1981; 78:7560-7564.