The concentration of leukaemia inhibitory factor (LIF) was measured in uterine flushings obtained from normal fertile women, from women with unexplained infertility and from women who suffered recurrent miscarriage. In normal fertile women, LIF was not detected in flushings obtained on days luteinizing hormone (LH)+0 to LH+6 of the cycle, but concentrations gradually increased from day LH+7 to a maximum at day LH+12. The amount of LIF in flushings obtained from women with unexplained infertility was significantly lower than in those from normal fertile women on day LH+10 (P < 0.05). The production of LIF by cultured human epithelial and stromal cells was also investigated. LIF was not detectable in the supernatants of cultured stromal cells. Basal LIF production by epithelial cells varied according to the stage in the cycle at which the biopsy was taken. Significantly more LIF was produced by epithelial cells from late proliferative and early secretory endometrium compared with amounts produced by cells from early proliferative (P < 0.001) and late secretory (P < 0.01) endometrium. High doses of progesterone and oestradiol caused a small decrease in epithelial cell LIF production: the combined effect of progesterone and oestradiol (P < 0.01) was greater than the effect of either steroid alone (P < 0.05). The results show, for the first time, the capability of human endometrium to produce LIF in vivo. The fact that maximum LIF concentrations are present at implantation and that decreased concentrations occur in women with unexplained infertility suggest the importance of this cytokine in embryo implantation.

Preimplantation embryos from a variety of mammalian species contrast markedly in their response to culture in vitro. Murine preimplantation embryos display a wider tolerance than other mammalian species to culture environments, and this has contributed to the development of several effective defined culture media. Embryo coculture on somatic cells remains the most effective method of supporting reasonable rates of bovine preimplantation development in vitro. The patterns of gene expression for several antioxidant enzymes during preimplantation murine and bovine development were examined by use of the reverse transcription-polymerase chain reaction technique to determine whether the differential developmental capacity of mammalian preimplantation embryos in culture may reflect variations in the patterns of expression for a series of antioxidant enzymes. Transcripts for catalase, CuZn-containing superoxide dismutase (CuZn-SOD), Mn-SOD, glutathione peroxidase (GPX), and glutamylcysteine synthetase (GCS) were detected in mouse embryos at all stages of development regardless of in vivo or in vitro development. Preimplantation cow embryos produced by in vitro procedures expressed mRNAs for catalase, CuZn-SOD and GPX, whereas transcripts for Mn-SOD were not detected at any stage. GCS transcripts, although present in stages up to the morula, were not detected in cow blastocysts. Analysis of antioxidant gene expression in both bovine primary oviductal cell monolayer cultures and nonattached, ciliated oviductal cell vesicle cultures revealed a constitutive pattern of expression of all five enzymes for the 8-day culture interval. These experiments suggest that differences in gene expression may contribute to the variation in the ability of embryos to develop in vitro with respect to levels of oxygen and dependence on coculture.

In the mouse, estrogen and progesterone are required to prime the uterus for decidual cell reaction (DCR) in response to an intraluminal stimulus and, once DCR is induced, progesterone is required to maintain DCR. However, some evidence indicates that certain nonprogestational steroid hormones may also be involved in regulating DCR. The present study determined whether androgen plays any role in DCR. Adult CD1 mice were ovariectomized and treated with a regimen of estradiol and progesterone to prime the uterus for DCR and to maintain DCR. Sesame oil was injected into the uterine lumen to induce DCR on Day 5 of the treatment. DCR was determined by deciduomal weight-the difference between the wet weights of oil-injected and noninjected uterine horns. Testosterone, given at 1 mg/day during Days 3-5, could not replace progesterone in priming the uterus for DCR. However, the same dose of testosterone given during Days 6-8 maintained DCR. Alkaline phosphatase activity, a bio-marker for DCR, was present in the deciduoma maintained by either progesterone or testosterone, although the distribution of this enzyme activity was more intense in the antimesometrial pole in progesterone-maintained deciduoma. A nonaromatizable androgen, 5 alpha-dihydrotestosterone (DHT), was also effective in maintaining DCR, and this action of DHT was blocked by an androgen receptor antagonist, hydroxyflutamide. The relative potency of DHT in maintaining DCR was similar to that of progesterone. However, the regression of the deciduoma appeared to be advanced in DHT-treated mice. Ovariectomy on Day 6 of pregnancy resulted in resorption of the conceptus and regression of the decidua within 48 h. Treatment with DHT at the time of ovariectomy could not prevent fetal resorption, but it delayed the regression of decidua, as indicated, in part, by the presence of granulated metrial gland cells. In summary, androgen cannot prime the uterus for DCR, but it can maintain DCR once it is induced. The physiological significance of this finding remains to be determined.

Embryo implantation in the mouse is an invasive process and requires the action of proteinases, including plasminogen activator (PA) and metalloproteinases. After the implanting embryo establishes close contact with the endometrium, the invasion process begins, at least in part, through interactions of the embryo with the extracellular matrix in the endometrium. This study determined whether embryo interaction with extracellular matrix components would affect the secretion of PA in vitro. PA in vitro. Mouse embryos were collected from the uterus on Day 3.5 of development, just before implantation, and were cultured dishes precoated with bovine serum, plasma fibronectin, or BSA (control). Embryos cultured on serum- or fibronectin-coated dishes secretes significantly more PA than those cultured on BSA. The effect of fibronectin was inhibited by hexapeptides that contained the integrin-recognizing Arg-Gly-Asp sequence. This indicates that the action of fibronectin in enhancing PA secretion is mediated through its receptor (integrins) in the embryo. Fibronectin fragments reproduced the effect of the whole fibronectin molecule, suggesting that the clustering of integrins by specific ligands is responsible, at least in part, for the increase PA secretion. The increase in PA secretion was a specific response to fibronectin rather than a reflection of increased total protein secretion, and was at least partially a result of the increased steady-state level of PA mRNA in the cultured embryos. Laminin was as effective as fibronectin in promoting PA secretion. Epidermal growth factor increased PA secretion, probably by promoting the interaction of the embryos with the extracellular matrix. In summary, our findings indicate that the interactions of the implanting embryos with their extracellular matrix may regulate trophoblast invasion by controlling PA secretion.

The rat endometrium undergoes decidualization, a tissue remodeling process, during embryo implantation. Plasminogen activator (PA), particularly the urokinase-type PA (uPA), has been implicated in tissue remodeling. The present study determined whether rat endometrial stromal cells secrete uPA during decidualization in vitro and, if so, whether the secretion is regulated by prostaglandins that are required in decidualization. Endometrial stromal cells were obtained from rats that had been treated with estrogen and progesterone to sensitize their uteri for decidualization, and the cells were cultured for up to 72 h in a serum-free medium. PA activity in the conditioned medium, as determined by a chromogenic assay, increased steadily during the 72-h culture period. PA secretion decreased when endogenous prostaglandin synthesis was inhibited by the addition of indomethacin to the cell cultures. The inhibitory effect of indomethacin on PA secretion was reversed by prostaglandin E2, and much less effectively by prostaglandin F2 alpha. PA activity in the medium was due primarily to uPA because 1) PA activity was inhibited by a uPA-specific inhibitor-amiloride-and by an anti-mouse uPA antibody, and 2) the predominant PA activity in the medium, as identified in zymography, had a molecular mass of approximately 40 kDa, similar to that reported for uPA. Northern blot analyses of RNA from the cultured cells demonstrated that the steady-state levels of mRNA for uPA, but not for tPA and plasminogen activator inhibitor-1, were decreased by indomethacin; this decrease was reversed by prostaglandin E2. Taken together, the data indicate that rat endometrial stromal cells secrete uPA during decidualization in vitro, and that prostaglandin E2 regulates uPA secretion by the decidualizing cells by directly increasing uPA gene transcription and/or stabilizing its transcripts. These findings may help to partially elucidate the mechanism of action of prostaglandin E2 in decidualization.