163 FUNCTIONAL COMPARISON OF BOVINE TROPHOBLASTIC VESICLES DERIVED FROM FRESHLY COLLECTED CONCEPTUSES AND SERIALLY PASSAGED TROPHOBLAST CELLS

The co-transfer of trophoblastic vesicles (TVs), derived from in vivo-recovered conceptuses, is a well-known method for promoting the successful implantation of embryos through the action of bovine interferon-tau (bIFN-τ) on maternal-fetal recognition. However, the preparation of these TVs is a tedious process. Techniques have progressed for obtaining large numbers of TVs consistently from serially passaged trophoblast cells. The aim of the present study was to compare the function of TVs for co-transfer derived from in vitro production with those derived from in vivo-flushed embryos, and without co-transfer. Production of TVs from serially passaged trophoblast cells was carried out according to the method previously described by Takahashi et al. (2000, Cloned animals and placentation, 147–151, Yokendo: Tokyo, Japan). The in vitro TVs (IVP-TVs), 1–2 mm in diameter, were taken from cells passaged 48th to 52th and 96th. In vivo-derived TVs (vivo-TVs) were prepared from recovered elongating blastocysts 7 days after transfer of frozen–thawed embryos on Day 8 of the estrous cycle. Demi-embryos were produced from fresh embryos of flushed Japanese Black cows by bisection using a micromanipulator. Pairs of demi-embryos, without (control) or with 2 to 4 TVs, were transferred into the uterus ipsilateral of 65 Japanese Short Horn recipients to the functional corpus luteum. Pregnancy diagnoses were performed twice, from Day 30 to 70, by ultrasound scanning. BIFN-τ content of culture media was measured by RIA (Takahashi et al. 2005 Theriogenology 63, 1050–1060). Data were analyzed by chi-square test and Fisher's exact test. The pregnancy rate in the IVP-TVs group (20.0%, 5/25) had a tendency to be lower than in the vivo-TVs (46.2%, 6/13), although this did not reach statistical significance (P = 0.09). These pregnant animals were transferred TVs from passaged 48th-52nd (5/19). In the control group 33.3% (9/27) of recipients conceived. Twin pregnancy rate was also lower in the IVP-TVs (0/5) than in the other two groups (3/6 and 5/9, P = 0.09 and P = 0.04, respectively). Fetal losses occurred only in the single pregnancies of IVP-TVs (60.0%, 3/5) and vivo-TVs (16.7%, 1/6) up until Day 70. Reproductive efficiency, based on numbers of delivered offspring, was significantly lower in the IVP-TVs (8%, 2/25) compared with the vivo-TVs (61.5%, 8/13) and control (51.9%, 14/27) groups, respectively (P < 0.01). BIFN-τ levels secreted from the TVs derived from passages 46th and 48th were 0.157 ng/mL-1 and 0.113 ng/mL-1, respectively. In conclusion, compared with the in vivo TVs, those from serially passaged trophoblast cells had a negative effect on pregnancy, although the morphology of the two differently derived TVs was similar.