A.8 Placenta (1996), Vol. 15

Interferon Tau: A Novel Pregnancy Recognition Signal. F.W. Bm T.E. Spencer and T.L. Ott, Department of Animal Scie\ce, Institute of Biosciences and Technology, Texas A&M University, College Station, Texas 77843

Trophectoderm of ruminant conceptuses secrete tau interferons (IFNz) as the pregnancy recognition signal. IFNz acts on uterine epithellum to suppress transcription of the estrogen receptor (ER) and oxytocin receptor (OTR) genes which blocks development of the uterine luteolytic mechanism and, therefore, release of luteolytic pulses of prostaglandin F,,. Continued secretion of progesterone by the corpus luteum ensures establishment of pregnancy. Secretion of IFNT on Days 12-13 for sheep and Days 14-17 for cows and goats is essential for pregnancy recognition. IFNt is hypothesized to affect endometrial gene expression by activation of the JaklStat pathway resulting in formation of the ISGF3a transcription factor complex. ISGF3cl binds to interferon stimulated response elements and activates transcription of interferon-responsive genes such as interferon regulatory factor-l (IRF-1) which, in turn, activates expression of the negative-acting transcription factor IRF-2. Pregnancy results in a transient increase in endometrial IRF-1 expression followed 24-48 h later by a sustained elevation in IRF-2. We propose that IRF-2 negatively regulates ER gene expression and blocks expression of the OTR gene to abrogate the luteolytic mechanisms and ensure the establishment of pregnancy.

NK cells in the uterus during human pregnancy. Ashlev Kinq, Tanya D. Burrows, Sue Hiby, Lucy Gardner and Y.W. Loke, Research Group in Human Reproductive Immunobiology, Department of Pathology, University of Cambridge, Cambridge, U.K.

The uterine mucosa is populated by abundant CD56brrShf CD16- mCD3- cells with LGL morphology. Interestingly, many characteristics of decidual NK cells (low cytotoxicity against the K562 cell line; lack of expression of CD16 and CD57; expression of CD69 and cytoplasmic CD3; response to low dose IL-2) are similar to those described for fetal NK ceils. This raises the possibility that adult uterine NK cells are a distinct subpopulation, perhaps arising from fetal NK stem cells. An alternative explanation is that adult NK ceils undergo tissue-specific homing and differentiation. The function(s) of these distinctive tissue NK cells in relation to human reproduction are unknown. Their presence correlates with the period of placentation when trophoblast migrates into the decidua. These invading extravillous trophoblast cells express HLA-G and HLA-C, the latter mainly in the form of free heavy chains. Decidual NK cells do possess Killer Inhibitory Receptors (KIR) which are known to recognise HLA class I antigens, This raises the question whether these NK cells exert control over the semi- allogeneic implanting placenta by interacting with trophoblast HLA-G and HLA-C.

THI- and TH-2 Type Immunity to Trophohlast in Women with Pregnancy Success and Failure. 1.A Hill, D I Anderson, Fearing Laboratory. Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA.

Following antigen activation CD4 and CD56 lymphocytes produce one of two distinctive cytokine profiles leading to differences in immune function These cells, termed THI and TH2 cells, secrete primarily E-2, IN&y and TNF and IL-4, IL-5 and IL-IO respectively. THI-type cytokines, predominantly n”F-y and TNF have been demonstrated to impair embryo development and trophoblast growth in vitro and to mediate abortion in mice. while, the THZ-type cytokines, IL-4 and IL-IO have been demonstrated in the decidua of normal murine pregnancy. As the human adorn&urn and decidua contain CD4 and CDS6 IvmDhocvtes caoahle of cvtokine 1 . secretion, we have hypothesized that lymphocyte activationby trophoblast or other antigens leading to the release of THI or TH2-type cyiokines may be a mechanism underlying recunent pregnancy loss or pregnancy success, respectively. Lymphocytes from many women with unexplained recurrent spontaneous abortion (URA) repond to trophoblast extracts by both proliferating and releasing soluble factors that adversely affect in-vitro embryo and trophoblast viability. while, lymphocytes from parous women with a history of normal pregnancies andwomen with a history of preynancy loss due to cbramosomal or maternal anatomic causes do not generally iho& this reponse In a recent series, embryotoxic activity was detected in woernatants from 125 of 244 URA oatients and was siwificantlv associated with embryotoxicity. Of20 supematants f&m these patients &en f&further study. all were QOSitiVe for TNF-a. 17 for TNF-P. two for IL-IO, and one for IL-4 No cytokines were detected in ~upernatanf~ from unstimulated or RBC-membrane activated (control) cells of women with URA In contrast. tropboblart-activatzd l&mphocye supematants from control women and men neither were embryotoxic nor contained THI-type cytokines, but contained the TH2-type cytokine IL-IO and three contained IL-4. In separate studies in 39 women with URA progesterone (IV’ M) was able to block THI-type cytokine secretion by trophoblast-activated lymphocytes but did not cause the secretion ofTHY&type cytokines. In other experiments involving 5 wanen with w in vitro administration of IL-IO to trophoblast-activated lymphocytes was also able to inhibit INF-y secretion. These data provide evidence that THI-type immunity to rrophoblast is associated with URA and may play a role in human reoroductive failure: while. TH?-tvoe immunitv to troohoblast mav be a natural respome contributing to successful pr&ancy. Th&pies reducing THi-type cpokine and increasing THt-type cytokine secretion to trophoblast in women with recurrent abortion need to be studied in appropriately designed clinical trials

Assessment of Uterine Natural Killer Cell Functions from Studies of Trans enic and Mutant Mice. B.A. Croy, .!.A. Luross, M-4. 2. urmond, Department of Biomedrcal ;~Gn~ceiUnrversrty of Guelph, Guelph, Ontario, Canada

Uterine Natural Killer (uNK) cells, a cell population formerly called Granulated Metrial Gland (GMG) cells, are lymphocytes which are localised within the murine uterus during pregnancy. Although several functions have been proposed for uNK cells during pregnancy, a critical role for these cells has not been documented. The use of various transgenic and mutant strains of mice provides an opportunity to study the development of the implantation site throughout gestation within the context of a deleted or significantly reduced population of uNK cells. Reproductive studies involving the T- and NK-cell defi@r&transgenic mouse TgE26 and the double mutant ~56 ‘.IL-2R67lL-2Rp mouse have demonstrated irregularities in placental development and reproductive performance that may be attributable to the lack of uNK cells at the implantation sites. The actions of interleukin-2 (IL-2) on uNK cell function has not been documented in viva. To examine the role of IL-2 on uNK cell growth and differentiation during gestption, histological analysis of jmplantation sites from IL-2‘ mutant mice and IL-2--@,m‘ - double mutant mice was examined at multiple time- points during gestation. These tissues revealed an implantation site morphology and uNK cell frequency similar to that of wild-type mice. This data suggests that IL-2 is not a significant signal for uNK cell differentiation or precursor viability, and that the placental pathology in mice deficient in T- and NK-cells is independent of IL-2. Supported by NSERC and OMAFRA.