You are on page 1of 7



Structural Aspects of the Zona Pellucida of In Vitro-Produced Bovine Embryos:
A Scanning Electron and Confocal Laser Scanning Microscopic Study
Geert Vanroose,
Hans Nauwynck,
Ann Van Soom,
Marie-There`se Ysebaert,
Gerard Charlier,
Patrick Van Oostveldt,
and Aart de Kruif
Department of Reproduction, Obstetrics and Herd Health,
Laboratory of Virology,
and Department of Physiology,
Biochemistry and Biometrics,
Faculty of Veterinary Medicine, University of Ghent, B-9820 Merelbeke, Belgium
Veterinary and Agrochemical Research Center,
1180 Brussels, Belgium
Laboratory of Biochemistry and Molecular Cytology,
Faculty of Agricultural and Applied Biological Sciences,
University of Ghent, B-9000 Gent, Belgium
Structural aspects of the bovine zona pellucida (ZP) of in
vitro-matured (IVM) oocytes and in vitro-produced (IVP) em-
bryos were studied in two experiments to nd a tentative expla-
nation for the zonas barrier function against viral infection.
In Experiment 1, the ultrastructure of the outer ZP surface
was studied. The diameter (nm) and the number of the outer
pores within an area of 5000 m
of 10 IVM oocytes, 10 zy-
gotes, 10 8-cell-stage embryos, and 10 morulae were evaluated
by scanning electron microscopy. In oocytes and morulae, the
ZP surface showed a rough and spongy appearance with nu-
merous pores. In zygotes, the ZP surface was found to have a
smooth, melted appearance with only a few pores. In 8-cell-
stage embryos, both surface patterns were found. The mean
number (per 5000 m
) and the mean diameter of the outer
pores were different between the four stages of development (P
0.001): 1511 pores in oocytes, 1187 in zygotes, 1658 in 8-
cell-stage embryos, and 3259 in morulae, with mean diameters
of 182, 223, 203, and 155 nm, respectively.
In Experiment 2, the continuity of the meshes (network of
pores) towards the embryonic cells was examined by confocal
laser scanning microscopy. Therefore, the passage through and
the location in the ZP of uorescent microspheres, with similar
dimensions as bovine viral diarrhea virus (BVDV, 4050 nm) and
bovine herpesvirus-1 (BHV-1; 180200 nm), were evaluated. For
all stages, the smallest beads were detected halfway through the
thickness of the ZP, whereas the beads with a size of 200 nm
were found only within the outer-fourth part of the ZP. It can
be concluded that the intact ZP of bovine IVM oocytes and IVP
embryos are constructed in such a way that BVDV and BHV-1
should not be able to traverse the ZP and reach the embryonic
cells. However, the risk exists that viral particles can be trapped
in the outer layers of the ZP.
The mammalian zona pellucida (ZP) is an extracellular
coat that surrounds growing oocytes, ovulated oocytes, and
early embryos. During the initial oocyte growth in second-
ary follicles, the zona pellucida is secreted by the devel-
oping oocyte and by surrounding follicular cells as extra-
This work was supported by grant no. 941437 from IWT Brussels.
Correspondence and reprint requests: Geert Vanroose, Department of Re-
production, Obstetrics and Herd Health, Faculty of Veterinary Medicine,
University of Ghent, Salisburylaan 133, B-9000 Gent, Belgium. Fax: 32
9 264 77 97; e-mail:
Received: 25 June 1999.
First decision: 16 August 1999.
Accepted: 24 September 1999.
2000 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363.
cellular patches that further coalesce into a uniform layer
[13]. The ZP has several functions such as binding sper-
matozoa in a species-specic manner, blocking polyspermy,
preventing the dispersion of blastomeres during preimplan-
tation development, facilitating the passage of the embryo
through the oviduct, and protecting the embryo during early
stages of development [3,4]. The ZP is composed of only
a few highly modied glycoproteins: bovine (b) ZP1, bZP2,
bZP3-, bZP3-, and bZP4 [5]. Recently, it was revealed
that bZP2 and bZP4 are fragments of bZP1 [5]. Each of
these glycoproteins has specic functions. In cattle, sperm
receptor activity is attributed to ZP3-. ZP2 serves as a
secondary receptor for acrosome-reacted sperm. After
sperm-egg fusion, cortical granule-associated enzymes con-
vert ZP2 and ZP3 to ZP2
and ZP3
, respectively. These
egg-induced modications play a role in the block to poly-
spermy [6]. The glycoproteins determine the specic struc-
ture of the ZP [4,5,7]. When visualized by scanning elec-
tron microscopy (SEM), the mammalian ZP is found to be
composed of a network dispersed with numerous pores [8]
and with morphologically dissimilar internal and external
surfaces. The external surface displays a fenestrated lattice-
like appearance and the internal surface shows a regular
rough appearance [9,10].
Reports concerning the properties of the ZP of bovine
in vitro-produced (IVP) embryos are limited. Riddell et al.
[11] demonstrated by SEM that the external surfaces of the
ZP of immature oocytes are very irregular, with uneven
distribution of numerous pores, crevices, and projections.
Recently, the surface of the bovine ZP of immature oocytes,
in vitro-maturing and in vitro-matured (IVM) oocytes, and
in vitro-inseminated oocytes was studied by Suzuki and co-
workers [1214]. They found that the ne structure of the
ZP of immature oocytes is characterized by a network with
numerous wide meshes and deep holes. After maturation,
this network becomes ner, and the meshes and holes seem
to be less deep. No meshes have been found on the ZP of
inseminated oocytes, probably due to fusion of several lay-
ers of the network.
Variations in the architecture of the ZP exist between
species. When the porous surface is compared among spe-
cies, the largest pores are observed in the ZP of the rabbit
and the cat, the smallest in the cow. The pores are largest
at the outer surface of the ZP but decrease in size centrip-
etally [8,15]. In general, pore shape varies within and be-
tween the zonae of the species observed, but many zonae
exhibit a more elliptical than circular shape. The arrange-
ment of zona pores appears random in all species except
the cat, where a denite trend toward concentric pore ar-
rangement exists. Variation in network thickness has also
been observed, with the broadest plate-like reticulum ex-
hibited by the hamster zona. In the rabbit and the cat, this
network is more complex and distinct, with a reduction of
ber width. The smallest size of the bers composing the
zona network is found in the cow and the opossum [8].
Since embryo transfer has become an expanding busi-
ness, questions have arisen concerning the risk of virus
transmission through embryos and the epidemiological
complications of embryo transfer [1618]. Two of the most
important viral pathogens in cattle, herpesvirus-1 (BHV-1)
and bovine viral diarrhea virus (BVDV), are associated
with reproductive failure, such as abortion and infertility;
thus far, a limited number of investigations have been re-
ported on the interactions of these viruses with IVP embry-
os. Recent studies suggest that the sanitary risks for ex-
posure to BHV-1 and BVDV may be more important for
embryos produced in vitro than in vivo, due to differences
in the ZP of those produced in vitro, enabling easier ad-
sorption of viruses [17,19]. In previous studies, we dem-
onstrated that no virus replication nor signs of embryonic
degeneration were detected in ZP-intact, IVP bovine em-
bryos incubated with BHV-1 and BVDV [20,21]. From
these ndings, it was concluded that neither virus is able
to cross the ZP. In other words, an intact zona of IVP em-
bryos acts perfectly well as a protective barrier against
BHV-1 and BVDV. Therefore, the aim of the present study
was to characterize structural aspects of the ZP of IVP em-
bryos at different stages of development to nd an expla-
nation for its barrier function against viral infection. The
diameter of the outer pores and the continuity of the meshes
towards the embryonic cells were examined by SEM and
confocal laser scanning microscopy (CLSM), respectively.
General Procedures
In vitro embryo production. Bovine embryos were pro-
duced by standard in vitro fertilization (IVF) procedures.
Ovaries from healthy animals with normal reproductive
tracts were collected at slaughter and transported to the lab-
oratory in warm physiological saline containing 25 g/ml
kanamycin sulfate at 30C, within 4 h after collection. Ova-
ries were processed according to methods described pre-
viously [22]. Briey, oocytes were aspirated from 26-mm
follicles and matured for 24 h in modied bicarbonate-buff-
ered TCM 199 medium (Gibco BRL, Merelbeke, Belgium)
supplemented with 20% heat-inactivated estrous cow se-
rum, 0.2 mM sodium pyruvate (Sigma, Bornem, Belgium),
50 g/ml gentamicin sulphate medium (Gibco BRL), and
0.04 mM glutamine (Sigma). Then, the cumulus-oocytes
complexes were matured at 38.5C in 5% CO
in air at
100% humidity for 24 h. A fourth of the matured oocytes
were vortexed to remove the cumulus cells. Residual cu-
mulus cells were removed by pipetting. The remaining ma-
ture oocytes were inseminated with a nal sperm concen-
tration of 10
sperm/ml in 500 l IVF-TALP (Tyrodes so-
lution supplemented with albumin, lactate, and pyruvate)
supplemented with 6 mg/ml fatty acid-free BSA, 20 g/ml
D-penicillamine, 10 M hypotaurine, 1 M epinephrine,
and 25 g/ml heparin (all from Sigma) for 24 h at 38.5C
in 5% CO
in air with maximal humidity. For each exper-
iment, three 0.25-ml straws lled with semen from a bull
free of BHV-1 and BVDV were used. Frozen-thawed sperm
was separated over a Percoll gradient (4590%; Pharmacia,
Uppsala, Sweden). Maturation and fertilization media were
not covered with parafn oil. After insemination, oocytes
were vortexed to remove excess sperm and cumulus cells.
Groups of 25 inseminated oocytes were transferred into 50-
l droplets of oviduct coculture overlaid with parafn oil
(Merck-Belgolabo, Overijse, Belgium). The culture medi-
um was Menezo-B2 (INRA, Paris, France) supplemented
with 2.5 g/ml fungizone (Gibco), 10% FCS, and bovine
oviduct epithelial cells (BOEC). Coincubation took place
at 38.5C in 5% CO2 in air with maximal humidity.
Preparation of oocyte/embryos for SEM. Mature oocytes
without their cumulus investments, zygotes (1 day post fer-
tilization), 8-cell stage embryos (3 days post fertilization),
and morulae (6 days post fertilization) were xed for SEM
as follows: the ZP-intact oocytes/embryos were washed 3
times in distilled water. The oocytes/embryos were then
xed for 1 h in 2.5% glutaraldehyde in 0.1 M cacodylate
buffer (Agar Scientic Ltd., Stansted, UK), pH 7.2, at 4C.
After xation, the oocytes/embryos were washed twice for
5 min in 2% cacodylate buffer (Agar Scientic) at 4C;
after washing, they were postxed in 1% osmium tetroxide
in distilled water for 2 h at 20C. Subsequently, the oocytes/
embryos were rinsed for 10 min in distilled water and then
were dehydrated in a series of increasing concentrations of
ethanol and acetone. Following dehydration, the oocytes/
embryos were dried in a Bal-Tec CPD 03 critical-point dry-
er using liquid CO
as the transitional uid. After drying,
they were coated with 20-nm gold by a Balzers sputtering
device (Balzer, Liechtenstein). SEM observations were made
with a Philips (Eindhoven, The Netherlands) 501 scanning
electron microscope at an accelerating voltage of 30 kV.
Since the preparation of oocytes and embryos might result
in some xation artifacts, a cautious interpretation of the
results is desirable.
Preparation of oocytes/embryos for CLSM. The conti-
nuity of the meshes towards the embryonic cells was ex-
amined by CLSM. For this purpose, uorescent micro-
spheres (Fluospheres, Molecular Probes Europe, Leiden,
The Netherlands) were used: one type (crimson red uo-
rescence with excitation/emission maxima of 365/415) with
a size of 200 nm, and another type (blue and yellow-green
uorescence with excitation/emission maxima of 625/645)
with a diameter of 40 nm. These beads have the same phys-
ical dimensions as BVDV (4055 nm) and BHV-1 (180
200 nm), respectively. The permeability of the ZP to these
microspheres was investigated by the following experimen-
tal design: ZP-intact cumulus-free embryos were put in a
drop of medium on a microscopic slide and evaluated with
an inverted microscope equipped with two positioning ma-
nipulators and with two 3-dimensional remote-control mi-
cromanipulators. Two holding pipettes with an outer di-
ameter of 110 m and an opening (inner diameter) of 25
m were used. One pipette held the embryo; the other pi-
pette was placed on the surface of the ZP and a mixture of
the microspheres was deposited there with positive pressure.
The localization of the microspheres in the ZP was visually
demonstrated by a Bio-Rad MRC 1024 Laser Scanning
Confocal Microscope (Bio-Rad House, Hertfordshire, UK)
linked to a Nikon Diaphot 300 microscope (Nikon Corp.,
Tokyo, Japan) and interfaced to a Compaq Prosignia 300
(Compaq Computer Corp., Houston, TX). Krypton-argon
laser light was used to excite crimson red uorescence
(625-nm line) and yellow-green uorescence (567-nm line)
of the microspheres. Extended focus images were obtained
with Bio-Rad COSMOS and Lasersharp software (Bio-Rad
House). Some caution with interpretation of the results is
desirable since the possibility exists that the pressure of the
FIG. 1. SEM images of the ZP of a zygote (A) and an 8-cell-stage embryo
(B). Arrows indicate that the diameter of pores appear to decrease towards
the inner layers of the ZP (centripetal narrowing architecture). Scale bar
1000 nm.
FIG. 2. SEM images of the ZP of IVM bovine oocytes and IVP bovine
embryos. A) A matured oocyte with a rough surface, remnants of probably
cellular material (arrows) and granular as well as knobby processes (as-
terisks). B) A zygote with a smooth and compact surface. C) An 8-cell-
stage embryo with granular deposits (asterisks) and a stratied upbuilding
process of the zona (arrows). D) Morula with a rough surface, granular
deposits (asterisks), and depositions of biological material in layers (ar-
rows). Scale bar 1000 nm (all panels). Images on the left and right are
of the same embryonic stage and of the same oocyte or embryo, but of
another place on the ZP surface.
deposition may cause some beads to progress further into
the ZP than a nonmotile virus might.
In an additional experiment, ZP-intact cumulus-free em-
bryos were incubated for 24 h in medium containing a mix-
ture of the two types of microspheres. The localization of
the microspheres in the ZP was visually demonstrated as
described above.
Experimental Design
Experiment 1. The ZP of 10 IVM oocytes, 10 zygotes,
10 8-cell stage embryos, and 10 morulae were studied. All
these embryonic stages were collected in a single sequence
of in vitro maturation, fertilization, and culture. The outer
pores present within ve surface areas (arbitrarily selected)
of 100 m
each per embryo were examined by SEM. The
actual percentage of ZP surface area surveyed was about
2.5% of the total ZP surface. For each pore, the smallest
diameter (nm) was measured by an image analysis system
(NIH Image 1.60). The percentage of pores that were large
enough to allow the entry of BHV-1 or BVDV within the
examined surface of the ZP was calculated. The ultrastruc-
ture of the ZP surface areas was evaluated and compared
within and between the stages of development.
Experiment 2. The ZP of 10 IVM oocytes, 10 zygotes,
10 8-cell stage embryos, and 10 morulae were studied by
CLSM as described above. All these embryonic stages were
collected in the same sequence of IVM/IVF/IVC as in Ex-
periment 1. The passage and the localization of uorescent
microspheres of different size, respectively through and in
the ZP, were evaluated and compared within and between
the stages of development.
Statistical Analysis
The results of this completely hierarchical design were
analyzed following a general linear procedure (SPSS for
Windows, 7.0, Chicago, IL) with stage of embryo (S) as a
xed main factor, embryo per stage (E), surface area (A)
per embryo, and pores (P) per surface area as consecutive
random subfactors. The model was: Y S E/S A/E/
S P/A/E/S, with Y the observed diameter of the pores,
or the number of pores per area. In the latter case, the last
term in the model is not considered. Equality of variances
within groups was checked before proceeding with analy-
sis. In the analysis of the observations of the diameter of
pores, Type I Sums of Squares were chosen. This is rec-
ommended as the unequal number of pores per surface area,
resulting in an unequal number of observations, and it is a
reection of the population size of pores in an embryo [23].
Variance components among groups were calculated for the
random factors. Scheffes procedure was followed for com-
parison of means of the xed factor.
Experiment 1: Scanning Electron Microscopic Study
General observations. The outer ZP-surface of the in-
vestigated embryos showed two principal patterns: 1) a net-
work with a spongy appearance containing numerous pores,
and 2) a more compact structure with fewer pores, which
were mostly larger in comparison with the pores of the
spongy surface. In all four stages of development, the deep-
er lying pores were smaller in comparison with the super-
cial pores (the diameters of the pores decreased the nearer
the pores were situated to the inner ZP; Fig. 1). The pore
shape was in general circular or elliptical, and the pores
were arbitrarily distributed.
Embryonic stage-dependent observations. In observa-
tions of matured oocytes (Fig. 2A), SEM analysis showed
that the ZP surface was mainly characterized by a spongy
network with a rough surface. Remnants of presumably cel-
lular material ( 1 m in diameter) were regularly per-
ceived. Globular as well as knobby processes ( 250500
nm in diameter) were observed in all investigated surface
areas and were seen to be touching the ZP.
Examination of zygotes (Fig. 2B) revealed that the
spongy network had become more compact and the pores
were less numerous in comparison with the pores in the ZP
FIG. 3. SEM images of the zona pellucida of an 8-cell-stage embryo (A)
and a morula (B). Arrows indicate bridges of biological material between
neighboring pores. Scale bar 1000 nm.
TABLE 1. SEM analysis of the diameter of the outer pores of the ZP of
bovine oocytes and IVP bovine embryos.
Stage of
No. of
oocytes or
No. of
5000 m
Mean dia-
meter of pores
(nm) SD
% Pores
50 nm
% Pores
200 nm
8-cell stage
182 69
223 95
203 79
155 71
Values with different superscripts within the same column differ sig-
nicantly (P 0.001, nested ANOVA, Scheffes pairwise comparisons of
means; standard error of the difference between number of pores
TABLE 2. Partitioning of the variation between observations.
Number of pores
of total
Diameter of pores
of total
Surface area
P 0.001 for all factors.
FIG. 4. The distributions of the pores in relation to their smallest di-
of the oocytes. This compact pattern, which was found in
all the embryos, displayed a smooth appearance. Some-
times the surface of the ZP had a melted look with only a
few pores, which were mostly not as deep as the outer pores
visible in oocytes. No remnants of cellular material were
detected, and only a few processes as seen on the ZP of
matured oocytes were observed. Some crevices with the
print of a sperm head were seen. Occasionally, a sperm cell
was detected near such a ssure, which was at least 1 m
deep and 12 m 46 m in surface area. Penetrating
and penetrated sperm cells were regularly noticed; of the
latter, only the tails were visible. Completely melted ZP
surfaces, without pores, were always observed in the im-
mediate environment of the sperm cells.
In 8-cell-stage embryos (Fig. 2C), the two principal pat-
terns were found to the same extent. The largest pores were
found in surface areas that displayed a more compact ap-
pearance. In certain areas, granular depositions were ob-
served on the ZP surface. There were no traces of pene-
trated or penetrating sperm cells. Sometimes an adhering
sperm head or sperm cell (with tail) was seen. Stratied
processes of the ZP were visible.
In morulae (Fig. 2D), a more rough surface of the ZP
was demonstrated. An occasional single sperm cell was de-
tected on the surface of some morulae. There were many
small pores, which were accompanied by deposits of layers
of biological material. Granular depositions were noticed,
which were sometimes closely associated with the ZP. In
8-cell-stage embryos and morulae, narrow bridges of bio-
logical material were present between 2 pores, as if these
bridges were splitting up one large pore (Fig 3).
Statistical Analysis
The observed number of outer pores in 5 surfaces of 10
oocytes or embryos (a total area of 5000 m
) and the
mean diameter of these pores are summarized in Table 1.
In matured oocytes, 1511 pores/5000 m
were detected.
After fertilization, the number of pores remained more or
less stable at rst; but from the third day onwards, a dra-
matic increase was seen, with the largest number of pores
being present in morulae (3259 pores/5000 m
) (P
0.001). The mean diameter of the outer pores was signi-
cantly different between the 4 stages of development (F
14.14, df 3 and 36, P 0.001), with the largest diameter
in zygotes (223 nm). The diameter decreased signicantly
in more developed stages (P 0.001).
A signicantly negative correlation was observed be-
tween the number of pores per area and the mean diameter
of these pores (r 0.56 ; P 0.001). This correlation
was clearly demonstrated when the percentage of pores
with a diameter of 200 nm were studied. In morulae (the
stage with the highest number of pores per area) only 19%
of these pores was large enough to allow entry of BHV-1,
whereas in the other 3 stages these percentage were more
than 34%. At least 97% of the pores (in all stages) were
large enough to allow the entry of BVDV. The distributions
of the pores in relation to their smallest diameter are shown
in Figure 4.
In order to establish the source of the observed differ-
ences (number and diameter of pores), the variation in the
observations was partitioned for each possible source of the
variation: xed (stage) or random (embryo, surface area,
pores) (Table 2). The percent variation in the number of
pores between stages (55%) far exceeds the percent random
variation between embryos (25%) and between different
areas (20%). As for the variation in diameter, the main con-
tribution (64%) comes from differences between pores per
observed area; the remaining variation is due mainly to
differences of the mean diameter between the stages (25%).
Experiment 2: CLSM Study
After CLSM analysis, a homogeneous dispersion of the
microspheres was detected in the outer layers of the ZP.
FIG. 5. CLSM images of uorescent mi-
crospheres in the outer layers of the ZP of
a zygote. The beads were deposited to the
surface of the zona with positive pressure
(small arrows). A) Crimson red uoresence
of microspheres with a diameter of 200
nm. B) Yellow-green uorescence of mi-
crospheres with a diameter of 40 nm.
Scale bar 10.5 m diameter of the
zona. Large arrows indicate the inner side
of the zona.
FIG. 6. Light microscopy (A) and CLSM
images (B,C) of uorescent microspheres
in the outer layers of the ZP of a zygote,
after 24-h incubation in a suspension of
the beads. Arrows indicate the inner side
of the zona. B) Crimson red uorescence
of microspheres with a diameter of 200
nm. C) Yellow-green uorescence of mi-
crospheres with a diameter of 40 nm.
Scale bar 10.5 m diameter of the
The smallest microspheres (40 nm) were shown to be de-
posited obviously deeper in the ZP in comparison with the
beads of a size of 200 nm (Fig. 5). Microspheres were not
detected in the inner layers of the ZP or close to the oocyte
After incubation of the embryos in a suspension of the
microspheres for 24 h, the smallest beads (40 nm) were
detected halfway through the thickness of the zona. The
beads with a size of 200 nm were found only within the
rst one-fourth of the thickness of the ZP (Fig. 6).
The present study is the rst attempt to analyze the ZP
of bovine IVM oocytes and IVP embryos by means of SEM
and CLSM, in order to nd a tentative explanation for its
barrier function against viral infection with BHV-1 or
Our results clearly demonstrated that the surface of oo-
cytes and embryos show a variety of morphological struc-
tures. We have distinguished two principal patterns: 1) a
network with a rough and spongy appearance containing
numerous pores, and 2) a more compact and melted struc-
ture with fewer pores, which are mostly larger in compar-
ison with the pores of the spongy surface. The same pat-
terns have also been demonstrated in IVM human oocytes
[24]. In our study, all oocytes and embryos were found to
display the two patterns. However, the proportion of these
two patterns was dependent upon the stage of embryonic
development. In matured oocytes, the surface was rougher,
probably due to remnants of cumulus cells and corona ra-
diata cells. Suzuki et al. [1214] reported similar observa-
tions in immature, maturing, and matured bovine oocytes.
They inferred that the rough surface was composed of re-
siduals of the corona radiata cells. The latter cells form
cytoplasmic processes that pass through the zona and ter-
minate with gap junctions on the vitelline membrane of the
oocyte [25]. It is known also that after maturation, most of
these cell processes retract or disintegrate and that the ma-
trix of the ZP closes the channels left by the cytoplasmic
processes [26]. The pores that we detected on the outer
surface of mature oocytes probably marked the entries of
cell processes. Whereas in pig and rat the diameter of these
processes is about 50100 nm [26], we found in bovine
oocytes a mean diameter of 182 nm. The surface of the
zygotes was found to have a smooth, melted appearance
with only a few pores. On some of the surface areas in-
vestigated, we detected fewer than 5 pores. The melted look
of the ZP, without pores, in the direct proximity of the pen-
etrating sperm cells is probably caused by the lytic action
of acrosomal enzymes such as acrosin and hyaluronidase.
This acrosome reaction is necessary for spermatozoa to tra-
verse the ZP in order to fuse with the oolemma. The lytic
action of the enzymes probably explains the general melted
view of the ZP shortly after fertilization. The same obser-
vation was made by Suzuki and coworkers [13], who de-
scribed a dramatic change of the ZP network after fertil-
ization. This structural change, accompanied by modica-
tion of ZP glycoproteins and by changed permeability of
macromolecules (dextrans and lectins), is known as hard-
ening, which plays a critical role in preventing polysper-
my [27,28].
Three days after IVF, the two principal patterns were
found to the same extent on the ZP of 8-cell-stage embryos.
The rst signs of granular deposits on the surface of the
ZP were noted. At 6 days postinsemination, the rough sur-
face was the dominant pattern on the ZP surface of moru-
lae. It is known that oviductal cells secrete proteins that
attach to the ZP [29,30]. Therefore, we speculate that the
rough aspect of post fertilization embryonic stages (8-cell
stage and morula) could be a result of the depositions of
biological material secreted by the oviductal cells used in
our culture system. The outer pores were the smallest in
morulae, and bridges were found in the middle of some
large pores, features that may also have been caused by
these depositions in and around the pores. All the afore-
mentioned observations clearly demonstrate that the ZP is
a dynamic structure that can be changed by its environment
[29] and that is able to repair damage [24].
Concerning the risk of infection of embryonic cells with
BHV-1 or BVDV, we studied the outer pores, which could
be ports of entry for viruses. We have shown that pores at
the surface of the ZP are large enough to allow entry of
BHV-1 (180200 nm) and BVDV (4555 nm). In all em-
bryonic stages investigated, more than 96% of the outer
pores were large enough for the entry of BVDV. The per-
centage of pores large enough for the entry of BHV-1 was
dependent on the embryonic stage, with a minimum of 19%
in morulae and a maximum of 51% in zygotes.
By means of SEM, it was observed that the diameter of
the channels in the ZP decreased centripetally, which is in
agreement with reports in humans [15] and in other animals
[8]. Our observation was conrmed by means of CLSM
images. From these data, it may be assumed that intact ZP
of IVP bovine oocytes and zygotes are constructed in such
a way that BHV-1 and BVDV may enter the channels in
the outer layers of the ZP but do not pass through inner
layers of the ZP towards the embryonic cells. This centrip-
etal narrowing architecture of the meshes of the ZP has
been demonstrated in the hamster [9,10]. At present, only
one virus has been demonstrated to be able to penetrate
through the ZP of farm animals. This is the porcine par-
vovirus, which is a very small virus of only 20 nm diameter.
It has been detected by means of a transmission electron
microscope in embryonic cells of ZP-intact in-vivo-derived
porcine embryos after incubation with the virus [31]. In
mice, Mengo encephalitis virus (2728 nm in diameter) has
also been shown to pass through the ZP of 2-cell mouse
embryos and mouse morulae [32,33]. It has been suggested
that viral penetration presumably occurs along channels left
when the follicle cell processes are withdrawn.
Although we have found no evidence in previous reports
that viruses are able to cross the ZP [20,21], one particular
observation in this study deserves further attention. In some
fertilized zygotes, excavation or ssures with the exact di-
mensions of a sperm head (which was probably caused by
enzymatic digestion) were observed. These holes could
provide a point of entry for the virus at fertilization, besides
serving as a potential hiding place for the virus. Virus res-
ident in such a cavity might be protected against eventual
sanitary measures, e.g. washings and trypsin treatment, and
pose a danger for infecting susceptible cows after transfer-
ence of such treated embryos. Possible virus entry in the
ZP during fertilization has been previously reported for por-
cine parvovirus, pseudorabies virus, and porcine enterovi-
ruses, which were found in sperm tracks in pores in the ZP
[34]. However, these tracks tend to close up quickly after
fertilization, thus it would be difcult for viruses, which are
nonmotile, to traverse these channels in the ZP up to the
embryonic cells [26]. Porcine parvovirus, pseudorabies vi-
rus, and porcine enteroviruses have also been associated
with porcine sperm cells at the outer surface of the ZP of
porcine embryos.
In conclusion, this study provides evidence that the in-
tact ZP of IVM bovine oocytes and IVP bovine embryos
are constructed in such a way that BHV-1 and BVDV
should not be able to reach the embryonic cells. However,
even though the structure of the ZP serves as a barrier, the
risk exists that viral particles could be embedded in the
outer layers of the ZP. Therefore, additional research is
needed to determine if viral particles remain in the ZP after
washing and/or after trypsin treatment, and to establish ef-
fective embryo treatments which can ensure that virus-free
IVP embryos are produced.
The authors wish to thank H. Favoreel, A. Geldhof, J. Mestach, G.
Spaepen, and N. Vanbruaene for excellent technical assistance.
1. Dunbar BS, Prasad SV, Timmons TM. Comparative structure and
function of mammalian zonae pellucidae. In: Dunbar BS, ORand MG
(eds.), A Comparative Overview of Mammalian Fertilization. New
York: Plenum Press; 1991:97114.
2. Dunbar BS, Avery S, Lee V, Prasad S, Schwahn D, Schoebel E, Skin-
ner S, Wilkins B. The mammalian zona pellucida: its biochemistry,
immunochemistry, molecular biology, and development expression.
Reprod Fertil Dev 1994; 6:331347.
3. Epifano O, Dean J. Biology and structure of the zona pellucida: a
target for immunocontraception. Reprod Fertil Dev 1994; 6:319330.
4. Wassarman PM. Regulation of mammalian fertilization by zona pel-
lucida glycoproteins. J Reprod Fertil Suppl 1990; 42:7987.
5. Topper E, Kruijt L, Calvete J, Mann K, Topfer-Petersen E, Woelders
H. Identication of bovine zona pellucida glycoproteins. Mol Reprod
Dev 1997; 46:344350.
6. Kopf GS. Zona pellucida-mediated signal transduction in mammalian
spermatozoa. J Reprod Fertil Suppl 1990; 42:3349.
7. Green D. Three-dimensional structure of the zona pellucida. Rev Re-
prod 1997; 2:147156.
8. Dudkiewicz A, Williams W. Fine structural observations of the mam-
malian zona pellucidae by scanning electron microscopy. Scanning
Electron Microsc 1977; Vol II:317324.
9. Phillips D, Shalgi R. Surface properties of the zona pellucida. J Exp
Zool 1980; 213:18.
10. Keefe D, Tran P, Pellegrini C, Oldenbourg R. Polarized light micros-
copy and digital image processing identify a multilaminar structure of
the hamster zona pellucida. Hum Reprod 1997; 12:12501252.
11. Riddell KP, Stringfellow DA, Gray BW, Riddell MG, Wright JC, Galik
PK. Structural and viral association comparisons of bovine zonae pel-
lucidae from follicular oocytes, day-7 embryos and day-7 degenerated
ova. Theriogenology 1993; 40:12811291.
12. Suzuki H, Yang X, Foote RH. Surface characteristics and size changes
of immature, in vitro matured and in vitro fertilized bovine oocytes.
Theriogenology 1994; 41:abstract 307.
13. Suzuki H, Yang X, Foote RH. Surface alterations of the bovine oocyte
and its investments during and after maturation and fertilization in
vitro. Mol Reprod Dev 1994; 38:421430.
14. Suzuki H, Presicce GA, Yang X. Analysis of bovine oocytes matured
in vitro vs in vivo by scanning electron microscopy. Theriogenology
1996; Abstract 271.
15. Nikas G, Paraschos T, Psychoyos A, Handyside A. The zona reaction
in human oocytes as seen with scanning microscopy. Hum Reprod
1994; 9:21352138.
16. Stringfellow DA, Riddell KP, Zurovac O. The potential of embryo
transfer for infectious disease control in livestock. N Z Vet J 1991;
17. Stringfellow DA, Wrathall AE. Epidemiological implications of the
production and transfer of IVF embryos. Theriogenology 1995; 43:
18. Wrathall A. Embryo transfer and disease transmission in livestock: a
review of recent research. Theriogenology 1995; 43:8188.
19. Guerin B, Nibart M, Marquant-Le Quienne B, Humblot P. Sanitary
risks related to embryo transfer in domestic species. Theriogenology
1997; 47:3342.
20. Vanroose G, Nauwynck H, Van Soom A, Vanopdenbosch E, de Kruif
A. Susceptibility of zona-intact and zona-free in vitro-produced bo-
vine embryos at different stages of development to infection with
bovine herpesvirus-1. Theriogenology 1997; 47:13891402.
21. Vanroose G, Nauwynck H, Van Soom A, Vanopdenbosch E, de Kruif
A. Replication of cytopathic and noncytopathic bovine viral diarrhea
virus in zona-free and zona-intact in vitro-produced bovine embryos
and the effect on embryo quality. Biol Reprod 1998; 58:857866.
22. Mahmoudzadeh A, Van Soom A, Bols P, Ysebaert M, de Kruif A.
Optimalization of a simple vitrication procedure for bovine embryos
produced in vitro: effect of developmental stage, two-step addition of
cryoprotectant and sucrose dilution on embryonic survival. J Reprod
Fertil 1995; 103:3339.
23. Maxwell S, Delaney H. Designing Experiments and Analyzing Data.
Pacic Grove, CA:Broks/Cole Publishing Company; 1990.
24. Schwartz P, Magerkurth C, Wichelmann H. Scanning electron mi-
croscopy of the zona pellucida of human oocytes during intracyto-
plasmic sperm injection (ICSI). Hum Reprod 1996; 11:26932696.
25. Allworth A, Albertini D. Meiotic maturation in cultured bovine oo-
cytes is accompanied by remodeling of the cumulus cell cytoskeleton.
Dev Biol 1993; 158:101121.
26. Anonymous. Recommendations for the sanitary handling of embryos.
In: Stringfellow DA, Seidel SM (eds.). Manual of the International
Embryo Transfer Society, 3th ed. Champaign, IL: International Em-
bryo Transfer Society; 1998: 26.
27. Hatanaka Y, Nagai T, Tobita T, Nakano M. Changes in the properties
and composition of zona pellucida pigs during fertilization in vitro. J
Reprod Fertil 1992; 95:431440.
28. Legge M. Oocyte and zygote zona pellucida permeability to macro-
molecules. J Exp Zool 1995; 271:145150.
29. Wegner C, Killian G. In vitro and in vivo association of an oviduct
estrus-associated protein with bovine zona pellucida. Mol Reprod Dev
1991; 29:7784.
30. Verini Supplizi A, Monaci M, Stradaioli G, Greve T, parillo F. Iden-
tication of glycoconjugates in the zona pellucida of in vitro matured
and tubal unfertilized bovine oocytes by lectin histochemistry. Anim
Reprod Sci 1996; 43:99111.
31. Bane D, James J, Gradil C, Molitor T. In vitro exposure of preim-
plantation porcine embryos to porcine parvovirus. Theriogenology
1990; 33:553561.
32. Gwatkin R. Effect of viruses on early mammalian development. I.
Action of Mengo encephalitis virus on mouse ova in vitro. Proc Nat
Acad Sci USA 1966; 50:576.
33. Gwatkin R. Passage of Mengovirus through the zona pellucida of the
mouse morula. J Reprod Fertil 1967; 13:577578.
34. Bolin S, Turek J, Runnels L, Gustafson D. Pseudorabies virus, porcine
parvovirus, and porcine enterovirus interactions with the zona pellu-
cida. Am J Vet Res 1983; 44:10361039.