René De Hertogh

Decreased Inner Cell Mass Proportion in Blastocysts From Diabetic Rats

Late morulae and blastocysts were recovered from streptozocin-induced diabetic pregnant rats and individually examined for numbers of inner cell mass (ICM) cells and trophectoderm (TE) cells. Compared with embryos collected from control rats, exposure to maternal diabetes significantly decreased mean ICM cell number of blastocysts recovered on day 5 of gestation, but the TE population of these embryos remained unaffected. The mean ICM proportion was therefore significantly lower than that of control embryos. These differences were not observed between the two groups of morulae collected on day 5, suggesting that the distinctive susceptibility of the ICM was expressed after blastocyst formation. On day 6, a significant inhibitory effect of diabetes was observed on the growth of both ICM and TE cells, but because the reduction was more severe in the ICM than in the TE, the mean ICM proportion of these blastocysts was again significantly lower than in control embryos. A linear quadratic relationship was obtained between the numbers of ICM cells of individual blastocysts and their respective numbers of TE cells in each of the two experimental groups. However, the slope of the curve was slower in the diabetic group than the control group. The disturbed ICM cell growth in the blastocysts from diabetic rats was found to be associated with a significantly increased incidence of cell death predominantly located in the ICM. Because it is known that excessive reduction of the ICM is incompatible with normal embryogenesis after implantation, our results suggest that the differential sensitivity of ICM and TE cells in preimplantation blastocysts may contribute to the pattern of postimplantation defects described in diabetic pregnancies.

Stimulatory and inhibitory effects of glucose and insulin on rat blastocyst development in vitro.

The effect of glucose and insulin on the in vitro development of the rat preimplantation embryo was studied by incubating rat blastocysts recovered on days 5 or 6 of pregnancy in the absence or presence of increasing levels of glucose and/or insulin for 24 or 48 h. A differential cell-staining method allowed the separate counting of inner cell mass (ICM) and trophectoderm (TE) cells at the end of the incubation period. In a high-glucose medium (17 mM), ICM and, to a lesser extent, TE developments were significantly and irreversibly inhibited. Low insulin concentrations (3 pM) stimulated ICM and TE development in the presence of 1.1 or 6 mM glucose. Higher insulin levels (30–600 pM) in a 6-mM glucose medium, resulted in a dose-dependent inhibition of ICM and, to a lesser extent, TE development after both 24 and 48 h. This insulin-induced inhibition was reversible if insulin was removed from the medium after 24 h. In the absence of glucose in the medium, insulin was neither stimulatory nor inhibitory on ICM growth. Dead-cell occurrence in ICM after a 48-h incubation increased with increasing glucose concentration in the medium. Insulin alone did not increase dead-cell number but enhanced the effect of glucose. These results show that, in the presence of glucose, insulin might be stimulatory (at low concentrations) or inhibitory (at higher concentrations) on ICM development. A high glucose level was also inhibitory and increased dead-cell occurrence. The data suggest that insulin and glucose might interact and modulate blastocyst development as a function of their respective concentrations.

Maternal insulin treatment improves pre-implantation embryo development in diabetic rats

SummaryPre-implantation embryos were recovered from control, diabetic and insulin-treated diabetic rats on day 5 of pregnancy. Compared to control animals, diabetic rats had a 20 % reduction in the number of embryos per rat and blastocysts recovered from diabetic rats showed a 19 % decrease in total cell number. The cellular decrease observed in blastocysts was mainly at the expense of the inner cell mass. Insulin replacement therapy was started on day 1 of pregnancy and normalized the glycaemia of diabetic rats but failed to raise the number of embryos per rat toward the control value. Insulin treatment, however, fully restored the normal cell number in both the inner cell mass and trophectoderm of blastocysts. The dead cell index, which was significantly elevated in the inner cell mass of blastocysts from diabetic rats, also returned to the control value following insulin treatment. Our data suggest that diabetes-induced impairment of pre-implantation development can be partly prevented by insulin treatment started shortly after conception.

Possible role for TNF-alpha in early embryopathy associated with maternal diabetes in the rat.

Tumor necrosis factor (TNF) bioactivity was assessed in culture media conditioned with uterine cells collected from control or diabetic rats on days 5 and 8 of pregnancy. On both days, diabetic uterine cells released significantly more biologically active TNF than did control cells, and this activity was significantly decreased by the addition of anti-TNF-α antibodies but not by the addition of normal IgG when WEHI 164 cells were used as a target. When uterine tissues from day 5 or day 8 pregnant diabetic rats were tested by Northern blot analysis, TNF-α mRNAs were twofold more abundant than in control samples, but the difference was not statistically significant (P = 0.086 and 0.100, respectively). Immunohistochemical analysis of diabetic day 5 uterine sections revealed that most of the TNF-α synthesis occurs in the epithelium lining the uterine lumen. Finally, the growth of day-5 embryos in culture medium conditioned with day-5 diabetic uterine cells was significantly reduced when compared with that of embryos in medium conditioned with control cells. Embryonic development was markedly improved when anti-TNF-α antibodies were added to the diabetic-cell conditioned medium. Our data support the hypothesis that TNF-α may be implicated in the developmental deficiencies observed in preimplantation embryos from pregnant diabetic rats.

Identification of Caspase-3 and Caspase-Activated Deoxyribonuclease in Rat Blastocysts and Their Implication in the Induction of Chromatin Degradation (but Not Nuclear Fragmentation) by High Glucose

Abstract Previous investigations have shown that maternal diabetes impairs rodent embryo development during the earliest phase of gestation. Exposure to high concentrations of glucose before implantation results in a decrease in the number of cells per embryo and in a concomitant increase in two nuclear markers of apoptosis, chromatin degradation and nuclear fragmentation. In the present study, we show that two intracellular effectors of apoptosis, caspase-3 and caspase-activated deoxyribonuclease (CAD), are involved in the embryotoxicity of high glucose. Using reverse transcription-polymerase chain reaction and immunocytochemistry, we first demonstrated that these two effectors were expressed in rat blastocysts. The two effectors were detected in all the cells of the blastocysts and the immuno-signals were excluded from the nuclei. Rat blastocysts were incubated for 24 h in either 6 mM or 28 mM glucose in the presence or absence of specific inhibitors (DEVD-CHO [10 μM] against caspase-3 and aurin [1 μM] against CAD). After incubation, blastocysts were examined for the proportion of nuclei showing signs of chromatin degradation or nuclear fragmentation. Addition of DEVD-CHO or aurin was found to inhibit the increase in chromatin degradation induced by high glucose. None of these two inhibitors prevented the increase in nuclear fragmentation triggered by excess glucose. Our data indicate that chromatin degradation and nuclear fragmentation are two nuclear damages that are induced separately by high glucose in rat blastocysts. Chromatin degradation is apparently mediated by the activation of caspase-3 and CAD.

Increased synthesis of tumor necrosis factor-alpha in uterine explants from pregnant diabetic rats and in primary cultures of uterine cells in high glucose.

The production of tumor necrosis factor-α (TNF-α) was investigated in uterine explants from normal, diabetic, or insulin-treated diabetic pregnant rats. Explants from diabetic rats released more soluble TNF-α than did those in the other groups. The extent of this secretion was correlated with blood glucose concentration at the time of explantation. The concentration of cell membrane–associated TNF-α in the explants was not altered by diabetes. Daily insulin administration failed to normalize uterine TNF-α secretion despite correction of glycemia in the diabetic rats. Explants from normal pregnant rats cultured in vitro with increasing concentrations of D-glucose showed a dose-dependent increase in TNF-α secretion. The production of TNF-α in high glucose was also tested in primary cultures of uterine cells isolated from either immature or adult rats. TNF-α secretion was increased in high D-glucose but not in iso-osmolar concentrations of L-glucose, D-raffinose, D-galactose, or mannitol. Cell membraneassociated TNF-α was not influenced by high D-glucose. Semiquantitative reverse transcription–amplification of RNA extracted from primary cultures of uterine cells showed that the steady-state level of TNF-α transcripts was increased by high D-glucose but not by high L-glucose. The results are consistent with the hypothesis that hyperglycemia is instrumental in the overexpression of TNF-α in the diabetic uterus. Because TNF-α has a demonstrated negative impact on embryonic growth, enhanced TNF-α synthesis in the pregnant uterus may contribute to the embryopathy associated with maternal diabetes.

Decreased expression of fibroblast growth factor-4 and associated dysregulation of trophoblast differentiation in mouse blastocysts exposed to high D-glucose in vitro.

Abstract.Aims/hypothesis: Paracrine interactions are thought to operate between the inner cell mass and trophectoderm cell lineages to co-ordinate their expansion and differentiation during embryo implantation. We aimed to determine whether hyperglycaemia could interfere with this regulatory process. Methods: Mouse blastocysts were pre-exposed to either control or high concentrations of d-glucose for 24 h in vitro and tested for their ability to attach and spread over a fibronectin-coated culture substrate. Quantitative immunocytochemistry was done on blastocysts to assess the protein expression of Fibroblast Growth Factor-4, a growth factor preferentially produced by the inner cell mass and thought to restrict trophectoderm differentiation into giant trophoblasts. Experiments were then done combining the pre-exposure to high d-glucose with the addition of recombinant fibroblast growth factor-4. Results: Compared with control blastocysts, high d-glucose pre-treated embryos were found to form 18 % larger trophoblast outgrowths. These blastocysts also showed a 30 % reduction in the expression of fibroblast growth factor-4 protein by inner cell mass cells as they were outgrowing. The trophoblast surface area per outgrowth and the trophoblast nuclear area were corrected when the addition of recombinant fibroblast growth factor-4 was combined with the pre-exposure to high d-glucose. Conclusion/interpretation: The data suggests that trophectoderm differentiation is impaired by high d-glucose and that this effect is secondary to a deficiency in fibroblast growth factor-4 protein in the inner cell mass. These observations add a novel perspective to the study of the peri-implantation embryopathy associated with maternal diabetes. [Diabetologia (2001) 44: 1318–1325]

Cell allocation to the inner cell mass and the trophectoderm in rat embryos during in vivo preimplantation development

SummaryThe number of trophectoderm (TE) and inner cell mass (ICM) cells was determined by complementmediated lysis and differential staining in rat embryos collected at different times during in vivo preimplantation development. At 90 h after fertilization, two groups of morulae were discriminated according to the presence or absence of detectable ICM cells, and the analysis of their total cell number indicated that acquisition of a permeability seal between TE cells begins at the 14-cell stage. On the other hand, our data confirmed that blastocoele formation occurs after the fourth cleavage division in the rat. The total cell number increased exponentially with time in blastocysts recovered between 90 h and 127 h but the cell kinetics of TE and ICM cells were different. The proportion of ICM cells consequently varied throughout blastocyst development, with a peak value for expanded blastocysts at 103 h. Finally, a linear-quadratic relationship was found between the numbers of TE and ICM cells when all the embryos with a detectable ICM were analysed together.

Estrogen-receptors and Estrogen-induced Protein-synthesis in the Uterus of Diabetic Rats

SummaryThe present study was undertaken to examine whether the diabetic state influenced the early stages of oestrogen action in the uterus. Ovariectomized streptozotocin diabetic rats were given intravenous infusions of oestradiol-17 B one day to 4–7 months after the initiation of diabetes. Oestradiol was taken up to a similar extent (on a wet weight basis) by the uteri of diabetic and control animals. Oestradiol receptor levels were similar in the cytosol of control and diabetic rats (1.06±0.19 and 1.01±0.27 pmol/uterus respectively). Nuclear translocation of the receptor complex under the influence of oestradiol infusion occurred similarly in both groups. The stimulation of a specific oestrogeninduced protein in the uteri occurred at a similar rate in both groups from one day to 4–7 months after the onset of diabetes. Thus streptozotocin diabetes of short to long duration does not interfere with the early stages of oestrogen action.

Oestrogen and progestogen receptors in endometrium and myometrium at the time of blastocyst implantation in pregnant diabetic rats.

SummaryA suitable hormonal environment is a prerequisite for blastocyst implantation. Experimental diabetes was previously shown to modify the hormonal milieu and produce alterations in oestrogen receptor kinetics in the uterine tissue. In the present work, oestrogen and progestogen receptor levels were measured on the morning of day 6 of pregnancy in normal and in streptozotocin-induced diabetic rats, both in implantation sites and in interembryonic segments of endometrium and myometrium. Receptor levels were different in the implantation sites compared to the interembryonic segments of endometrium, both in the control and in the diabetic animals. Indeed, implantation sites were characterized by lower oestrogen receptor levels in cytosol and higher progestogen receptor levels in cytosol and nuclei. However, compared to the control rats, the diabetic rats had lower oestrogen receptor levels in implantation sites, both in cytosol and nuclei. In the myometrium, the differences between sites or between types of rats were minimal. Plasma levels of oestradiol were lower in diabetic rats than in control animals, whereas progesterone levels were similar. A 20% lower implantation rate was found in diabetic rats, compared to normal rats. These results show that the specific distribution of oestrogen and progestogen receptors between implantation sites and interembryonic segments was preserved in the diabetic rats; however the absolute level of oestrogen receptor was lower. This abnormal endocrine milieu might arise from a lower oestradiol level and a decreased oestradiol/progesterone ratio in the circulating blood. Whether the lower implantation rate in diabetic rats might be a consequence of the overall disturbed hormonal status remains to be elucidated.