Summary Embryos were recovered from diabetic rats on day 5 of pregnancy and incubated in vitro for up to 72 h. Compared to control embryos, blastocysts from diabetic rats showed a marked impairment in growth that resulted at 48 h in a higher rate of degeneration and a lower morphological score in the developing population. After 72 h in vitro, fewer developing blastocysts from diabetic rats formed trophoblastic outgrowths and fewer of those implanted developed an inner cell mass when compared with the control group. When assessed for their cell content, blastocysts from diabetic rats contained fewer cells than control embryos at the start of the culture. This difference persisted, and even worsened, during the ensuing incubation period. The increasing cellular deficiency in blastocysts from diabetic rats was primarily located to their inner cell mass lineage but trophoblast growth was also affected. When trophoblast outgrowths were compared for their surface area and number of nuclei, those collected from diabetic rats were smaller, contained fewer nuclei and had a higher proportion of giant nuclei than control outgrowths. Our data thus demonstrate that despite their removal from the abnormal intra-uterine environment, blastocysts from diabetic rats remain functionally affected by their early exposure and fare less well than control embryos cultured under the same standard conditions. [Diabetologia (1994) 37: 855-
862]Key words Blastocyst, trophoblast outgrowth, inner cell mass, carry-over effect, rat.Extensive clinical studies have confirmed the need to start maternal metabolic treatment before or at the time of conception if the rate of fetal malformations is to be significantly reduced [1][2][3]. The absolute necessity for periconceptional care is now well accepted, however, the mechanism by which an early exposure to uncontrolled maternal diabetes could be carried-over to the organogenic phase remains to be elucidated [4]. Interesting data pertaining to this issue have been obtained in experiments showing that preimplantation embryos from either drug-treated or spontaneous diabetic mice have a decreased ability to develop in vitro when maintained under the same culture conditions as control embryos [5,6]. These observations were made at the morphological level, however, and it is not known whether such a discrepancy in growth potential extends to more subtle differences in the rate of embryonic cell proliferation and differentiation.In a previous study on the impact of the maternal diabetic condition on preimplantation growth in vivo, we demonstrated that blastocysts collected from diabetic rats contained fewer cells than control embryos [7] and that cells from the lineage responsible for forming the fetus were more affected in their proliferation than the cell population that gives rise to the placenta [8]. This deficiency in preimplantation development was corrected by treating the dia-