Adherence of bovine viral diarrhea virus to bovine oocytes and embryos with a hardened zona pellucida cultured in vitro

Abstract

The purpose of this study was to investigate the adherence of bovine viral diarrhea virus (BVDV) to bovine mature, or immature, cumulus-free oocytes and to in vitro fertilized embryos, maintained in vitro in a ligated bovine oviduct to allow for the hardening of the zona pellucida. Incubation of the oocytes and embryos in the oviduct for 5 h caused hardening of the zona pellucida as measured by resistance to pronase digestion (which increased from approximately 3 min to 7 h; P >0.001). However, there was no difference between the number of infected oocytes and embryos (n = 965 in 193 samples) following experimental exposure to BVDV regardless of whether or not they were previously incubated in the oviduct (P > 0.05). It was concluded that the modification of the proteolytic resistance properties of the zona pellucida during in vitro oviductal incubation did not influence the adherence of BVDV to zona pellucida of oocytes or in vitro fertilized embryos.

Mammalian oocytes and preimplantation-stage embryos are surrounded by a zona pellucida (ZP), a glycoprotein acellular matrix. The ZP plays an important role in many physiological functions, including fertilization, block to polyspermy, transport of embryos through the oviduct, and containment of blastomeres. Furthermore, the ZP plays a major role as a protective shell against infection of embryonic cells and as a carrier of infectious agents in the spread of livestock diseases through embryo transfer (ET) practices (1).

Extensive experimentation has shown that an intact ZP, both in vivo and in vitro fertilized embryos, is an effective barrier against penetration by several animal pathogens, although some viruses and bacteria can bind strongly to it (2). The mechanism of pathogens binding to ZP is unknown. However, it has been demonstrated that pathogenic agents are more likely to adhere to the surface of ZP of in vitro fertilized (IVF) embryos than to that of in vivo fertilized embryos (2). Furthermore, rendering IVF embryos free from contaminants using the sequential washing procedure recommended by the International Embryo Transfer Society (IETS) is not effective (1). It has been suggested that divergent conditions for the production of these 2 types of embryos may lead to changes in their morphology and to differences in the interaction of the ZP with pathogens. After in vivo fertilization and the passage of the oocytes through the oviduct to the uterus, ZP gradually accumulates oviductally-derived, mucin-like glycoproteins on its surface and in the perivitelline space (3,4). In contrast, IVF involves oocytes being harvested from the ovaries, and then matured, fertilized, and cultured in vitro (outside of the utero-tubal environment) for approximately 10 d. Consequently, IVF embryos are not exposed to oviductal secretions. Besides the many morphological and physiological differences between in vivo and IVF embryos, the ZP of in vivo fertilized embryos is characterized by increased resistance to enzymatic and chemical dissolution. Recent reports (5,6) indicate that following incubation of bovine cumulus-free oocytes in oviducts maintained either in vitro or in vivo, similar alterations in ZP response to enzymatic treatment occur. This physical modification was termed hardening of ZP (HZP) (7).

The objective of this study was to investigate whether or not experimentally induced changes in HZP affect adherence of bovine viral diarrhea virus (BVDV) to the ZP of oocytes and embryos produced in vitro. In these experiments, oocytes and embryos were produced using standard methodology described previously, with some modifications (8,9). Briefly, cumulus-oocyte complexes were recovered by aspiration of 3 to 6 mm follicles from ovaries of slaughtered cattle, and subsequently matured in TCM-199 medium with Earle's salt (Sigma Chemical Company, St. Louis, Missouri, USA) supplemented with 35 μg/mL follicle stimulating hormone (FSH) (Folltropin; Vetrepharm, Belleville, Ontario), 10 IU/mL human gonadotropin (hCG) (Chorionic Gonadotropin; Ayerst Veterinary Laboratory, Guelph, Ontario) and 20% estrous cow serum at 38°C and 5% CO₂, in air for 24 h. After maturation, oocytes were exposed to approximately 1 × 10⁶/mL motile spermatozoa and then incubated for 18 h in modified Tyrode's solution (containing 10 μg/mL heparin). Oocytes were denuded of cumulus cells by vortexing them in Dulbecco's phosphate-buffered saline (PBS) with 1 mg/mL of polyvinyl alcohol (PVA) (Sigma) followed by repeated pipeting. Subsequently cumulus-free presumptive zygotes were placed in culture wells (Nunclon, Roskilde, Denmark) with modified synthetic oviductal fluid (SOF) (9) and incubated for 8 d under silicone oil at 38°C in 5% CO₂, 5% O₂, and 90% N.

Bovine oviducts in the follicular phase of the estrous cycle (based on the presence of preovulatory follicle on the ovary) were collected from slaughtered cows. Subsequently, the oviducts were trimmed free of fat and connective tissue and rinsed twice in PBS. Oocytes or embryos in groups of 30 to 50 were aspirated into 0.25 mL artificial insemination plastic straws in approximately 10 μL of PBS. A loaded straw connected to the 1 mL syringe was introduced through the infundibulum into the ampullar part of the oviduct where the oocytes or embryos were deposited. After ligation of both ends, the oviduct was placed in TCM-199 medium (Sigma) and incubated at 38°C in 5% CO₂, 5% O₂, and 90% N for 5 h. Following incubation, the oocytes and embryos were flushed from the oviduct with PBS containing 1 mg/mL PVA, retrieved, washed, and divided into treatment groups. A proportion of oocytes and embryos were exposed to 1.0% pronase (Sigma) in PBS to determine HZP, while the remaining oocytes and embryos were incubated with BVDV. The time of the ZP lysis was defined as the interval between the last washing and the complete visual digestion of the ZP as observed under (×25) a stereomicroscope.

Groups of cumulus-free immature (with compact cumulus cells after their retrieval from the follicular fluid) and mature oocytes (with expanded cumulus cells after in vitro maturation) and embryos, were incubated with 10⁶ TCID₅₀/mL of either noncytopathic (NY-1) or cytopathic (NADL) strains of BVDV at 38°C for 3 h. Subsequently, oocytes and embryos were washed according to the method recommended by IETS (1); 10 successive washes of 10 or fewer embryos in fresh aliquots of 2 mL of PBS medium supplemented with 1 mg/mL PVA. To transfer oocytes and embryos between washes, a semi-automatic micropipette with a 5 μL disposable microtip was used. For each transfer, a fresh aliquot of medium (2 mL) and a new microtip were used (dilution 1:400).

In the present report the term of the sample refers to groups of oocytes, embryos, oviductal cells, and follicular fluids, which were tested for the presence of BVDV. Samples of sonicated oocytes and embryos (5 per sample) were inoculated on monolayers of Madin-Darby bovine kidney (MDBK) cells in 75 mL culture flasks for 5 d. The cultures were fixed and stained by an immunoperoxidase according to a method previously described for BVDV isolation (10).

Samples of oocytes and embryos found to be negative at this stage underwent 2 further passages and immunoperoxidase testing before being declared negative for BVDV. In addition, all negative samples were tested for the presence of BVDV RNA using the reverse transcription nested polymerase chain reaction (RT-n PCR) procedure. The RNA was extracted from the samples using a commercial kit as recommended by the manufacturer (QIAamp; Qiagen, Valencia, California, USA). The 3 primers, BVD 100, HCV 368, and BVD 180, were synthesised and used for the standard RT-PCR and the hemi-nested PCR as described in detail by Givens et al (11).

Samples of oviductal cells, pools of follicular fluids, and oocytes from each day of ovary and oviduct collection were also tested for the presence of BVDV. Corresponding groups of oocytes and embryos that were not incubated in an oviduct, but exposed to the same concentrations of BVDV, were tested and served as controls. The difference between the ZP dissolution times was analyzed using the nonparametric Kruskal-Wallis test performed with software (Instat software; GrapfPad, San Diego, California, USA). The association of oocytes and embryos with BVDV was compared using a Chi-square test with the same software.

For immature and matured oocytes and embryos not exposed to the oviduct the ZP dissolution times were 3.6, s_¯x= 0.24 (n = 20); 3.8, s_¯x = 0.24; and 4.0, s_¯x = 1.24 min, respectively (P > 0.05). Corresponding times for those incubated in the oviduct were 393, s_¯x = 47; 431.0, s_¯x = 0.5; and 467, s_¯x = 61 min, respectively (P > 0.05). These results showed that the oviductal environment significantly delayed proteolytic digestion of ZP, both of oocytes and IVF embryos, as compared to the control groups, indicating some structural modifications within the ZP (P < 0.001). These findings were consistent with those reported by Smorag et al (5), Katska et al (12), and Duby et al (13) who investigated the resistance of bovine ZP to pronase in more detail. These investigators demonstrated that in contrast to mice and humans, spontaneous hardening of the bovine ZP does not occur during in vitro culture, but it can be induced by the incubation of cumulus-free preovulatory oocytes in homologous and heterologous oviducts under in vitro or in vivo conditions. In addition, it was noted that the ZP hardening was not related to the follicular phase of the estrous cycle. Also Pollard and Leibo (14), when investigating cooling and freezing sensitivity of embryos, reported that the ZP of in vivo fertilized embryos collected from superovulated cows were more resistant to pronase digestion than the ZP of in vivo fertilized embryos cultured in TCM-199 medium.

In total, 193 samples containing 965 oocytes and embryos were used during 4 experimental replications in the present study. Fifty samples were declared free from BVDV after the 3rd passage and immunoperoxidase staining. Lower, but not significant, differences in percentages of samples associated with the infectious cytopathic strain of BVDV as compared to a noncytopathic strain were detected (P > 0.05) (Table I). However, the application of the PCR in addition to the virus isolation test revealed that almost all negative samples (46 out of 50) were associated with BVDV nucleic acid (92%) regardless of whether or not they were incubated in the oviduct. This is consistent with findings of other authors who reported difficulties in rendering IVF embryos free from several strains of BVDV when washed according to the recommended method by IETS (15,16,17).

Table I.

In the present experiments, BVDV was chosen as an infectious agent model which can be removed from in vivo fertilized embryos by a simple washing procedure, as opposed to IVF embryos to which the BVDV adheres more firmly. Based on the BVDV isolation results presented here (Table I), it appears that in spite of induced changes in proteolytic resistance of the ZP of oocytes and embryos, the oviductal environment did not prevent the adherence of BVDV to the ZP. This finding may suggest that the binding of BVDV to the ZP is not related to the developmental stage of the oocytes and embryos in vitro. However, as the quantitative viral assay was not used in this study, the concentration of BVDV associated with the ZP remains unknown.

In these studies, IVF embryos were produced in the presence of isolated bovine oviductal epithelial cells (BOEC), which may corroborate our data indicating that secretions of BOEC cells in vitro are not able to protect embryos against adherence or embedding of viral particles in the matrix of ZP. It is conceivable that other tubal fluid components; such as, transudate of blood serum proteins, electrolytes, amino acids, and other metabolites (in combination with secretions by the oviductal ciliated and nonciliated secretory cells), are required for embryo protection against the binding of BVDV to the surface of ZP under in vivo conditions (18). However, this concept remains elusive and speculative as there are no data related to the identity of such a possible oviductal factor. To our knowledge, the present study is the first attempt to investigate the physico-chemical modifications of the ZP in relation to its interaction with a viral agent. We also realize that other pathogenic agents may interact differently with ZP of embryos exposed in vitro to oviductal secretions.

In conclusion, it appears that an oviductal factor, which may protect fertilized in vivo embryos from the firm attachment of BVDV, is either not related to HZP or is not secreted by the oviduct under the conditions used in this study. A further detailed investigation is warranted as essential for the development of a method for disinfection of IVF embryos to render them free of BVDV.