Accepted for publication in Therio
Detection of ovine lentivirus in the cumulus cells, but not in the oocytes or
follicular fluid, of naturally-infected sheep
C. Cortez Romeroa, F. Fienia*, C. Rouxa, P. Russoc, J.M. Guibertc, F. Guiguenb, Y. Cheblouneb , M. Pépind and 4
a UPSP 5301 DGER , National Veterinary School, Nantes France
b UMR 754 INRA/ENVL/UCBL, Lyon Cedex 07 France
c AFSSA- Sophia Antipolis, France
d AFSSA-LERPAZ Maisons-Alfort, France
The aim of this study was to examine the infection status of oocytes, cumulus cells, and
follicular fluid taken from 140 naturally-infected ewes with Maedi-Visna virus (MVV). MVV
proviral DNA and MVV RNA were detected using nested-PCR or RT-PCR MVV gene
amplification respectively in the gag gene. Nested-PCR analysis for MVV proviral-DNA was
positive in peripheral blood mononuclear cells in 37.1% (52/140) of ewes and in 44.6%
(125/280) of ovarian cortex samples.
The examination of samples taken from ovarian follicles demonstrated that 8/280 batches of
cumulus cells contained MVV proviral-DNA, whereas none of the 280 batches of oocytes
taken from the same ovaries and whose cumulus cells has been removed, was found to be
PCR positive. This was confirmed by RT-PCR analysis showing no MVV-viral RNA
detection in all batches of oocytes without cumulus cells (0/280) and follicular fluid samples
taken from the last 88 ovaries (0/88).
The purity of the oocyte fraction and the efficacy of cumulus cell removal from oocytes was
proved by absence of granulosa cell-specific mRNA in all batches of oocytes lacking the
cumulus cells, using RT-PCR.
This is the first demonstration that ewe cumulus cells harbor MVV genome and despite being
in contact with these infected cumulus cells, the oocytes and follicular fluid remain free from
infection. In addition, the enzymatic and mechanical procedures we used to remove infected-
cumulus cells surrounding the oocytes, are effective to generate MVV free-oocytes from
Key Words: Maedi-Visna, oocytes, cumulus cells, lentivirus
* Corresponding author: Department of Research into the Health Risk and Biotechnology of Reproduction, 41
National Veterinary School, BP 40706, 44307 Nantes CEDEX 03, France. Tel.: +33-2-40687710; fax: +33-2-42
40687748. E-mail address: firstname.lastname@example.org (F. Fieni).
Maedi Visna (MV) is a disease of sheep  caused by a RNA virus from the Lentivirus genus
of the Retroviridae family [2,3]. This family of viruses also includes the caprine arthritis-
encephalitis virus (CAEV), which usually affects goats. These two viruses are considered as
In sheep, Maedi Visna virus causes progressive degenerative inflammatory disease in multiple
organs including the lungs (pneumonia - maedi), mammary gland, joints, and nervous system
(meningoencephalomyelitis - visna). Except in Australia and New Zealand which are the only
countries where lentiviral infections have been observed in goats but not in sheep , MV has
been identified worldwide. The prevalence of infection in many countries ranges from 15% to
Nowadays, reproductive biotechnologies, such as multiple ovulation embryo transfer (MOET)
or in vitro embryo production (IVP), are essential to ensure the improvement and the
dissemination of genetic material. In small ruminants, these techniques provide a source of
low-cost embryos for basic research in developmental biology and physiology, and for
commercial use . Nevertheless, these techniques have been held responsible for spreading
infectious diseases. To prevent MVV transmission though reproductive biotechnologies, the
use of non contaminated semen from certified MVV-free rams, bred under controlled
conditions, is reliable and simple. However, it does not appear to be quite so easy to certify
embryos or oocytes from donor ewes, as some MVV-infected ewes may be latently infected
and are seronegative. Multiple ovulation embryo transfer uses cumulus-oocyte complexes,
released from the ovary or taken from ovarian follicles, for the in vitro production of
In cows, bovine viral diarrhea virus (BVDV) has been found in follicular epithelial cells,
follicle fluid , and cumulus cells  taken from persistently infected heifers. If cumulus
cells are not removed during IVF, BVDV was found in association with morulae and
blastocysts . However, if these cells are washed and the embryos cultured in a disease-free
medium, BVDV-free embryos can be obtained after 10 days of culture . In goats, we
recently demonstrated that CAEV can infect granulosa cells in vitro , and CAEV-proviral
DNA was identified in cumulus cells using nested-PCR amplification from naturally infected
goats in vivo . However, this last study demonstrated that despite being surrounded by
infected cumulus cells, the goat oocytes were not infected .
In the present study our main aim was to determine the presence or absence of MVV proviral-
DNA and/or MVV viral RNA in oocytes, cumulus cells, and follicular fluid aspirated from
ovaries taken from naturally infected ewes.
2. MATERIALS AND METHODS
2.1. Sample collection
The blood samples and ovaries were collected at the slaughterhouse (winter 2002-2003,
spring 2003 and 2005) from 140 Lacaune and crossed-breed ewes, originating from breeding
flocks in the south of France.
Whole blood samples (10 ml) were taken by jugular venipuncture into anticoagulant (lithium
heparin), from each ewe just prior to removal of the ovaries.
The ovaries (n=280) were removed immediately after slaughter and were carried to the
laboratory in Dulbecco’s phosphate buffer saline (PBS, Eurobio®, France) in a thermos flask
(+4ºC). At the laboratory, each ovary was washed in sterile saline solution and processed
individually in a Petri dish. A new scalpel blade and sterile disposable material were used at
the slaughterhouse and in the laboratory, for each ewe and each ovary.
From the first 192 ovaries, cumulus-oocyte complexes (COCs) were harvested using the
slicing technique [11,12]. Each ovary was rinsed in 1 x Minimal Essential Medium (MEM,
Gibco-BRL, Paris, France) supplemented with 10% fetal calf serum (FCS, F-7524, Sigma,
Saint Quentin Fallavier, France). The rinsing solution was recovered in a Petri dish. The
oocytes, surrounded by their cumulus cells, were identified using a binocular
stereomicroscope, washed 10 times by successive passages in 1 x MEM medium and then
divided into 2 batches per ovary. The first batch consisted of oocytes with their cumulus cells.
The second batch contains oocytes depleted from their cumulus cells following incubation for
3 minutes in 1 x PBS containing 300µg/ml hyaluronidase (Sigma, H-4272: Saint Quentin
Fallavier, France). They were pipetted several times during this period, under binocular
stereomicroscope control, to ensure that all cumulus cells had been eliminated [13,14]. Zona
Pellucida (ZP)-intact oocytes with no cumulus cells were then washed 10 times over in MEM
medium with 10% FCS.
In the last 88 ovaries, follicular fluid, containing cumulus-oocyte complexes, was aspirated
from ovarian follicles with a diameter of more than 3mm, using an 18 gauge needle connected
to a syringe. This follicular fluid was centrifuged (2000 x g for 5 min). The supernatant was
recovered and samples of 300 to 350 µl of follicular fluid were stored at - 80 ºC. The tube was
then rinsed with MEM medium and the rinsing solution was collected in a Petri dish to
recover the cumulus-oocyte complexes. Cumulus-oocyte complexes present in the small
ovarian follicles were then harvested from each ovary using the slicing technique. All
cumulus-oocyte complexes recovered from the same ovary were washed 10 times over in
MEM medium. The cumulus cells were then removed from each individual batch of oocytes,
and the latter grouped into one batch. The fluids used for enzymatic washing and the ten
washing cycles were centrifuged (2000 x g for 5 min) to recover the cumulus cells; the
supernatant was carefully removed and the cumulus cells were then washed twice with 1 ml
of sterile PBS.
After the cumulus-oocyte complexes had been collected, fragments of 5g of outer ovarian
cortex were taken from each ovary tissue.
The samples were stored at -80°C for subsequent nucleic acid extraction to search for MVV
proviral DNA using specific nested-PCR, and for MVV viral RNA and cumulus cell mRNA
using RT-nested PCR.
2.2. Preparation of samples for nested-PCR and RT-PCR
2.2.1. Peripheral Blood Mononuclear cells (PBMC)
Whole blood samples collected in heparin tubes (10 ml) were centrifuged at 2000 x g for 10
minutes at room temperature. The buffy coat containing the cells was recovered and PBMC
were obtained by Ficoll density-gradient centrifugation (Histopaque®-1077, Sigma
Diagnostics, Saint Quentin Fallavier, France). At the end of this phase, mononuclear cells
were recovered from the gradient, washed in sterile PBS pH 7.2 and centrifuged at 700 x g for
5 min, three times over. The supernatant was removed and the cell pellet stored at – 80 ºC
prior to DNA extraction. DNA extraction from PBMC: 5 X 105 to 106 cells, were used to
isolate DNA with a Qiagen® kit (QIAamp kit®, blood protocol; Qiagen, Courtaboeuf, France),
in accordance with the manufacturer's instructions. One hundred and forty PBMC samples
were processed and stored at -80 ºC, until PCR processing.
After thawing, DNA was extracted from the ovarian samples, the batches of oocytes with
cumulus cells from the 192 first ovaries, and the batches of isolated cumulus cells from the
last 88 ovaries, using a QIAamp Tissue kit® (Qiagen, Courtabœuf, France) in accordance with
the manufacturer’s instructions.
RNA was extracted from the 88 batches of follicular fluid, with 300 to 350 µl of fluid per
batch, using a Qiagen® kit (QIAamp kit®, viral RNA, large sample volume, Qiagen,
Courtaboeuf, France), in accordance with the manufacturer's instructions.
DNA and RNA were extracted together from the batches of oocytes without cumulus cells
using a Qiagen® - RNA/DNA kit, in accordance with the manufacturer’s instructions. In the
first 192 ovaries, the extraction was performed separately for each ovarian batch. In the
following 88 ovaries, extraction was performed from all of the oocytes pooled together into
one batch. This dual extraction made it possible to assay MVV proviral-DNA using PCR, and
granulosa cell-specific mRNA using RT-PCR.
Following extraction, the samples were stored at -80°C until processing using PCR and RT-
2.3. Procedure for nested-polymerase chain reaction (nested-PCR)
A nested-PCR was performed as described previously . Two rounds of PCR amplification
were used to detect the MVV gag sequence. Sequences from the gag region were used for the
first amplification using the external primers GEX5 (5’- GAA GTG TTG CTG CGA GAG
GTG TTG -3’) and GAG EX3 (5’- TGC CTG ATC CAT GTT AGC TTG TGC -3’),
corresponding to bases 393-416 and the complement of bases 1268-1291, respectively.
Internal primers were used for the second amplification: GAG IN5 (5’- GAT AGA GAC
ATG GCG AGG CAA GT -3’) and GAG IN3 (5’- GAG GCC ATG CTG CAT TGC TAC
TGT -3’), situated at positions 524-546 and 1013-1036, respectively, . The sequences
corresponding to these primers are well conserved in CAEV and MVV isolates. Sample DNA
integrity was controlled by amplifying the β-actin gene using primers, based on the human
sequence ; external - ES30 (5’- TCA TGT TTG AGA CCT TCA ACA CCC CAG –3’)
and ES32 (5’- CCA GGG GAA GGC TGG AAG AGT GCC –3’), and internal - ES31 (5’-
CCC CAG CCA TGT ACG TTG CTA TCC –3’) and ES33 (5’- GCC TCA GGG CAG CGG
AAC CGC TCA –3’).
PCR reactions were performed with 0.5–1µg of total DNA in 20 µl of distilled water added to
30 µl of an amplification solution containing: 5 µl of reaction buffer [10X, 650 mM of Tris-
HCl (pH 8,8), 160 mM of (NH4)2SO4, 0.1% of Tween-20], 5 µl of MgCl2 (50 mM), 1 µl of
dNTP (25 mM of each deoxynucleotide triphosphate: dATP, dGTP, dCTP, dTTP), 0.25 µl of
TAQ Polymerase (5 U/µl, EUROBLUETAQ® ADN Polymerase-Thermostable, EUROBIO,
Les Ulis, France), 0,6 µl of each primer GAG EX3 and GAG EX5 or ES30 and ES32 (33 µM,
GIBCO BRL Custom primers – Life Technologies, Grand Island, NY), and 18,15 µl diethyl
pyrocarbonate (DEPC)-treated water. For the second amplification, 5 µl of the product from
this first amplification were added to 45 µl of second amplification solution, containing the
same reagents as in the first solution with an extra 15 µl of DEPC-treated water and using the
internal primers : GAG IN5 and GAG IN3 or ES31 and ES33.
Samples were denatured at 94 ºC for 3 min, followed by 35 amplification cycles (denaturation
at 92 ºC for 1 min, annealing at 56 ºC for 1.5 min, and extension at 70 ºC for 3 min), and final
extension at 72 ºC for 10 min, in a Thermocycler (Eppendorf®, Mastercycler). At the end of
the second round, 15 µl of this reaction (10 µl of the amplified fraction were added to 5 µl of
dyed loading buffer in each gel well) were recovered and loaded onto 1.5% agarose gel
(ultrapure, electrophoresis grade, GIBCO BRL LIFE® Technologies, Paisley, Scotland),
containing ethidium bromide (GIBCO BRL LIFE® Technologies, Paisley, Scotland), in 1x
TAE buffer, and electrophoresis was carried out. The DNA-amplified bands were visualized
using transillumination with UV light (312 nm). A marker was used as a molecular weight
marker (5 µl of Smart-Ladder, GIBCO BRL LIFE Technologies), comprising 14 bands
calibrated between 10,000 and 200 bp. Two controls were used for each gel: a negative
control (distilled water) and a positive control, MVV proviral-DNA, from infected GSM (goat
The sensitivity of this technique enables the detection of one in vitro infected GSM (goat
synovial membrane) cell and a minimum of 10 in vitro CAEV-infected fibroblast cells .
2.4. Reverse transcriptase-polymerase chain reaction (RT-PCR) procedures
RT-PCR was first used to detect MVV-viral RNA in oocytes without cumulus cells and in
follicular fluid , and secondly to detect granulosa cell-specific mRNA in the oocytes
without cumulus cells in order to evaluate the efficacy of the technique used to remove these
For each sample, the cDNA was synthesized using 5 µl of total RNA (containing from 10 to
100 ng, corresponding to 2 x 104 to 105 cells) added to 15 µl of a solution containing: 1 µl of a
dNTP (25 mM of each deoxynucleotide triphosphate: dATP, dGTP, dCTP, dTTP; EUROBIO,
France), 1 µl (0.4 µg) of random hexamers (Biolabs® Inc, S1230S, New England), 4 µl of
buffer (5x, Kit-M-MLV Reverse Transcriptase, Promega®, Madison, USA; 3681), 2 µl (400
units) of Reverse Transcriptase enzyme (Kit-M-MLV RT RNase H minus, point mutant,
Promega®, Madison, USA; M3683), 2 µl (80 units) of RNasin® Ribonuclease Inhibitor,
Promega®, Madison, USA; N2111) and 5 µl of RNase-free water for PCR. The reaction was
performed for 30 min at 37ºC, followed by 5 min at 95 ºC. 233
After reverse transcription, 20 µl of the product of this reaction were amplified using a
specific nested-PCR protocol. For the detection of MVV DNA, the nested PCR protocol was
used as described above. For the detection of granulosa cell DNA, two FSH amplifications
were performed to induce CREM genes (cAMP responsive element binding modulator) and
ICER genes (inducible cAMP early repressor) coding for FSH-induced granulosa cells. For
the first amplification using nested-PCR, primer sequences were for CREM 5 (5’- TGG AAA
CAG TTG AAT CAC AG -3’) and for CREM 3 (5’- CTA CTA ATC TGT TTT GGG AG -
3’). This round was immediately followed by a second round using internal primers ICER 5
(5’- ACT CTG TAT GCA AAA GCC CA -3’) and ICER 3 (5’- CTA CTA ATC TGT TTT
GGG AG -3’), .
The products of each RT-PCR were separated by electrophoresis on 1.5% agarose gel (ultra
pure, electrophoresis grade, GIBCO BRL LIFE® Technologies, Paisley, Scotland), containing
ethidium bromide (GIBCO BRL LIFE® Technologies, Paisley, Scotland), in 1x TAE buffer.
The bands were visualized using transillumination with UV light (312 nm). A marker was
used as a molecular weight marker (5 µl of Smart-Ladder, GIBCO BRL LIFE Technologies),
comprising 14 bands calibrated between 10,000 and 200 bp. Two controls were performed for
each gel: a negative control (distilled water) and a positive control (ovarian granulosa cells
and/or RNA extracted from the supernatant of MVV-infected GSM culture).
2.5. Statistical analysis
A Chi2 test was used to compare the incidence rate of MVV proviral-DNA in oocytes, with or
without cumulus cells, between PCR blood-cell-positive and PCR blood-cell-negative ewes.
Values of P<0.05 were considered to be significant.
To examine the infectivity status of blood and reproductive organs of ewes we used the
nested-PCR technique with the specific primers reported in the Materials and Methods
section. The detection of a specific band at 507 bp in the nested-PCR product with DNA
extracted from PBMC, ovaries, oocytes with cumulus cells, or cumulus cells demonstrated the
presence of MVV proviral-DNA. This band was also detected with DNA of the positive
control on agarose gel after electrophoresis and examination under UV light. Whereas the 393
bp bands, generated from the amplification of the endogenous β-actin gene, used as internal
control, was present in both non-infected and MVV-infected oocytes with or without cumulus
cells (Fig. 1).
To ensure that removal of cumulus cells from oocytes of the batch of cumulus-free oocytes
we used specific primers to amplify the mRNA of ICER and CREM FSH-induced that are
expressed in granulosa cells but not in the oocytes. Presence of RT-PCR specific product will
indicate the contamination of oocytes with cumulus cells and the absence will certify the
purity of oocytes. The mRNA analysis in cumulus cells using RT-nested PCR was considered
as being positive when two bands of 785 bp (ICER) and 1400 bp (CREM) molecular weight,
readily obtained with samples of the positive control, were amplified from the examined
sample and detected on agarose gel following electrophoresis and UV light illumination (Fig.
The examination of DNA isolated from peripheral blood mononuclear cells (PBMC) showed
that 52/140 (37.1%) ewes were harboring the provirus genome of MVV. The DNA-provirus
was also identified in 125 (44.6%) out of the 280 sampled ovaries. A total of 1,949 oocytes
were collected from 280 ovaries (1,210 and 739 oocytes from the first 192 and last 88 ovaries
respectively), giving an average of 7.0 ± 3.9 oocytes per ovary.
Nested-PCR analysis for MVV proviral-DNA was positive in 7 out of the 192 batches of
oocytes with cumulus cells taken from the first 192 ovaries, and in 1 of the 88 batches of
cumulus cell samples that had been removed by enzymatic washing from the oocytes taken
from the last 88 ovaries, giving a total of 8/280 positive samples (2.8%). However, none of
the batches of oocytes lacking the cumulus cells, removed using hyaluronidase treatment, was
found to be positive (0/280) (Table 1). RT-PCR analysis for MVV-viral RNA was also
negative for batches of oocytes without cumulus cells (0/280) and for batches of follicular
fluid removed from the last 88 ovaries (0/88).
To test the efficacy of the enzymatic and mechanical technique used to remove the cumulus
cells, an RT-nested PCR was used to detect granulosa cell mRNA. Results of this test were
negative for all 280 batches of oocytes without cumulus cells.
The main objective of our study was to determine whether sheep oocytes, cumulus cells and
follicular fluid are targets to Maedi-Visna virus infection. Samples taken from 140 Lacaune
ewes from the slaughterhouse were examined for MVV-proviral genome and this was
detected in PBMC of 37.1% (52/140) and in 44.6% (125/280) of ovarian cortex samples of
ewes. This infection rate ranging 40% is similar to that readily observed sheep flocks in the
South of France and from where these ewes were originated.
The data obtained by nested-PCR and RT-PCR amplification in the gag coding sequences of
MVV, with the samples taken from this ewe population, clearly demonstrated that cumulus
cells were infected by MVV. In contrast, neither oocyte cells nor follicular fluid samples were
found to be infected with MVV. In an other term, proviral-DNA was not detected in DNA
samples from the 280 batches of oocytes without cumulus cells whilst 7/192 batches of
oocytes with cumulus cells and 1/88 batches of cumulus cells, were found to be positive.
This is the first study demonstrating the infection of oocyte cumulus cells by MVV. Despite
its preferential tropism for cells of the monocyte-macrophage lineage , MVV has been
now shown to infect epithelial cells from different organs as aortic smooth muscle cells ,
milk cells and mammary tissue , third eyelid  and intestinal cells . Caprine
arthritis encephalitis virus (CAEV), which is very similar to Maedi Visna virus, has been
shown to infect various types of epithelial cells in vitro, including milk epithelial cells ,
oviduct epithelial cells , luminal epithelial cells  and aortic endothelial cells
[28,29,30]. Furthermore, granulosa cells cultured in vitro have demonstrated susceptibility to
CAEV , whilst in a similar study, Ali Al Ahmad et al.  detected CAEV pro-viral DNA
in goat oocyte cumulus cells in vivo. However, we should point out that the proportion of
batches with CAEV infected-cumulus cells, 64/246 , was greater than that seen with
MVV in this study (8/280). This could be related to the fact that the proportion of infected
goats (75/123 i.e. 61%) in the study of Ali Al Ahmad et al.  was higher than that of
infected sheep in our study (52/140 i.e. 37.1%). Another explanation could be that CAEV has
a higher infectivity and tropism for epithelial cells than MVV.
The oocyte is ovulated with a covering of expanded cumulus cells and these cells are
dispersed in the ampullary region of the oviduct within a few hours . The cumulus cells
play a key role in acquisition of full developmental competence during oocyte maturation 
and are currently used for in vitro fertilization (IVF) during 24-hour in vitro maturation (IVM)
of the oocyte . If oocytes are denuded of their cumulus cells prior to in vitro maturation
(IVM), their developmental capacity is reduced . Denudation of preantral mouse oocytes
prior to in vitro culture inhibits their growth, alone or in co-culture with granulosa cells .
One of the most striking changes to occur during the growth phase of an oocyte is the
secretion of a glycoprotein membrane, the zona pellucida (ZP), which forms a protective coat
around the oocyte consisting of three glycoproteins (ZP1, ZP2 and ZP3), . The highly
specialized cumulus cells have trans-zonal cytoplasmic processes, which penetrate through
the zona pellucida and abut the oocyte membrane [37,38], forming the cumulus-oocyte
complex (COC). Gap junctions at the end of these processes allow the transfer of small
molecular weight molecules between oocyte and cumulus cell, and between cumulus cells,
whereas larger molecules are transported by receptor-mediated endocytosis [39,40,41,42].
These junctional contacts are formed by connexin proteins of which connexin-37 is expressed
by the oocytes at all stages of folliculogenesis [43,44]. These gap junctions facilitate bi-
directional communication and enable the transfer of nutrients, metabolic precursors (e.g.
amino-acids and nucleotides), cytokines and other hormones (e.g. neurotransmitters and
growth factors), and inhibitory and stimulatory meiotic signals [45,46], that are necessary for
oocyte growth, as well as small regulatory molecules that control oocyte development . In
this way, MVV may be able to come into direct contact with the oocytes via the cumulus cells
during oogenesis or oocyte maturation in vivo.
Nevertheless, despite having been in contact with infected cumulus cells none of the 1,949 Download full-text
oocytes without granulosa cells or of the 88 batches of follicular fluid were found to contain
MVV provirus-DNA or MVV RNA. This suggests that oocytes are able to resist MVV
infection. One possible mechanism of resistance could be the absence of functional receptors
for MVV at their surface as previously reported  or the very slow metabolism of the
oocytes imposed by granulosa cells, slowing development and preventing meiosis. However,
for small ruminant lentivirus there is no correlation between cell multiplication and
incorporation and/or replication of the virus into the cellular genome [48,49,50].
Nevertheless, the absence of viral-RNA in follicular fluid suggests that the activity of MVV-
viral replication in cumulus cells is weak, even hopeless. The most likely hypothesis is that
the resistance of the ovocytes could be due to the absence of oocyte membrane receptors,
which are required for the internalization of the MVV particle; the structure of these receptors
remains to be elucidated for MVV.
Therefore, even though, in vivo, the oocytes and follicular fluid were not found to be infected
by MVV, samples from cumulus cells that were found to be positive (8/280, i.e. 2.8%)
suggest that there is a risk of oocyte infection during in vitro embryo production (IVP). This
risk would appear to be significant, since in the last 15 years, IVP has shown an increasing
interest because it is the most readily available method and a cheaper source of mature
oocytes, zygotes and embryos for research use . In cattle, embryos produced in vitro are
potential vectors for the transmission of bovine viral diarrhea virus (BVDV) [52,53,54,55]. It
has been show that oviduct epithelial cells and granulosa cells of persistently infected cattle
contained both viral antigen and RNA  and that the BVDV was found in association with
morulae and blastocysts when these cells are not removed during IVF . However, if these
cells are washed and the embryos cultured in an uninfected medium, BVDV-free embryos can