ENDOMETRIAL ADENOGENESIS AND UTERINE IMMUNE REGULATION IN SHEEP By MARIA B. PADUA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004
Copyright 2004 by Maria B. Padua
I dedicate this thesis to my mo ther, uncle, nanny, and siblings.
iv ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Pete r J. Hansen, for giving me the opportunity to join his program. I am grateful to him for challenging me academically and encouraging me as a scientist during these ye ars. Also, I thank my committee members Dr. William Buhi and Dr. Nasser Chegini for their advice and suggestions for improving my research projects and academic training. I also thank my old lab mates, Dr. aban Tekin, Dr. Yaser Al -Katanani, Dr. Fabiola Paula-Lopes, Dr. RocÃo Rivera , Dr. Zvi Roth, Dr. Joel Hern Ã¡ndez, Paolete Soto, Heather Rosson, and Olga Ocon, and my current lab mates, Jeremy Block, Dean Jousan, MoisÃ©s Franco, Luiz Augusto de Castro, Amber Brad and Katherine Hendricks, for their great help through many different ways. I especially want to recognize Dr. aban Tekin for helping me to improve my lab skills. Thanks are also extended to all personnel in the Department of Animal Sciences, especially to Dean Glicco a nd the staff of the Meats Laborat ory for their assistance with sheep management and slaughter. I also am grateful to the University of Florida Diagnostic Referral Laboratories and th e Hybridoma Core Facility of the Interdisciplinary Center for Bi otechnology Research at the Univ ersity of Florida for their valuable help in my experiments. Special thanks go to Luis Felipe Deppe for all his support and patience and Dr. AndrÃ©s Kowalski, MÃ³nica Prado, JosÃ© Crist bal Nieto, and MarÃlia a nd JosÃ© Trujillo for their good advice and sincere friendship for many years. I really appreciate it. Likewise,
v thanks go to Dr. GermÃ¡n Portillo, Carlos Lu cena, Carlos RodrÃgu ez, Dervin Dean, Dr. TomÃ¡s and Gabriela Belloso, and Luis and Nadia ChÃ¡vez for helping me many times during this period of my life.
vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF FIGURES.........................................................................................................viii LIST OF ABBREVIATIONS..............................................................................................x ABSTRACT......................................................................................................................x ii CHAPTER 1 REVIEW OF LITERATURE.......................................................................................1 Mechanisms Proposed to Allow Fetal Evas ion of the Maternal Immune Response....1 Placenta as an Immunological Barrier...................................................................1 Antigenic Immaturity............................................................................................2 Immune Tolerance.................................................................................................4 Pregnancy in Sheep.......................................................................................................8 Placentation...........................................................................................................9 Major Histocompatibility Complex (MHC) Expression.....................................10 Uterine Leukocyte Populations in Sheep....................................................................11 Progesterone as Immunosuppressive Molecule in Sheep...........................................14 Characteristics of Ovine Uterine Serpin.....................................................................16 Relationship to Serpins........................................................................................16 Biochemical Properties........................................................................................17 Endometrial Secretion.........................................................................................18 Immunosuppressive Properties............................................................................19 Ovine Uterine Morphogenesis....................................................................................21 Uterine Gland Knockout Model.................................................................................23 Synopsis and Hypothesis............................................................................................25 2 ACTIONS OF PROGESTERONE ON UTERINE IMMUNOSUPPRESSION AND ENDOMETRIAL GLAND DEVELOPMENT IN THE UTERINE GLAND KNOCKOUT (UGKO) EWE.....................................................................................27 Introduction.................................................................................................................27 Materials and Methods...............................................................................................30 Materials..............................................................................................................30 Experimental Design...........................................................................................30
vii Collection of Tissues and Uterine Fluids............................................................31 Skin Graft and Uterine Histology........................................................................32 Progesterone Radioimmunoassay........................................................................32 Determination of Protein Concentra tion in Uterine Fluid and Flushing.............33 Detection of OvUS in Uterin e Fluid by Western Blotting..................................33 Immunohistochemistry for OvUS and CD45R+ Lymphocytes...........................34 Statistical Analysis..............................................................................................35 Results........................................................................................................................ .36 Progesterone Concentration in Plasma................................................................36 Gross Uterine Morphology..................................................................................36 Histological Analysis of Endometrium...............................................................39 Survival of Skin Grafts........................................................................................39 Total Protein Content in the Uterine Lumen.......................................................42 Presence of OvUS in Uterine Fluid.....................................................................45 Immunochemical Localization of OvUS.............................................................45 Immunolocalization of CD45R+ Lymphocytes...................................................45 Discussion...................................................................................................................51 3 GENERAL DISCUSSION.........................................................................................56 LIST OF REFERENCES...................................................................................................62 BIOGRAPHICAL SKETCH.............................................................................................75
viii LIST OF FIGURES Figure page 1-1 Schematic illustration of the hypothesis...................................................................26 2-1 Progesterone concentration in plasma ( ng/ml) 2, 8 and 24 hours after injection in ewes treated with corn oil vehi cle (open circles) or progesterone (closed circles) over a 45 days period. ...................................................................37 2-2 Gross appearance of the uteri. ................................................................................38 2-3 Endometrial histology for control an d uterine gland knockout (UGKO) ewes treated with corn oil (CO) or pr ogesterone (P4) for 60 days. ................................41 2-4 Gross appearance of surviving autografts and allografts placed into the uterus of control (A, D, E) and uterine glan d knockout (UGKO) ewes (B, C, F) 30 days after surgery. ..................................................................................................43 2-5 Histology of autografts (A -C) and allografts (D-F) 30 days after grafting into the uterus of control and uterine gland knockout (UGKO) ewes. ................................44 2-6 Total protein recovered from uterine fl uid of control and uterine gland knockout (UGKO) ewes treated with corn oil vehicl e (CO) or progesterone (P4) for a 60 days period. ............................................................................................................46 2-7 Representative western blot for detect ion of ovine uterine serpin (OvUS) in uterine fluid or flushings collected from control and uterine gland knockout (UGKO) ewes treated with corn oil (CO) or progesterone (P4) after 60 days. ......47 2-8 Immunolocalization of OvUS (A-F) and CD45R+ cells (G-L) in endometrium from control and uterine gland knockout (UGKO) ewes treated with corn oil vehicle (CO) or progesterone (P4) for 60 days. .....................................................49 2-9 Density of CD45R+ cells in different areas of the uterine endometrium according to the presence of autografts (b lack bars) and allografts (grey bars) for control and uterine gland knockout (UGKO) ewes treated with corn oil vehicle (CO) or progesterone (P4) for 60 days. ..................................................................50 3-1 Proposed model for the process of endo metrial gland formation in the uterus of the adult ewe. .....................................................................................................57
ix 3-2 Proposed model for homing lymphocyte s in the luminal (LE) and glandular epithelium (GE) in the uterine endometrium of adult ewes. ...................................60
x LIST OF ABBREVIATIONS CD Cluster of Differentiation CO Corn Oil Vehicle Con A Concanavalin A G-CSF Granulocyte Macropha ge-Colony Stimulating Factor HLA Human Leukocyte Antigen IDO Indoleamine 2,3-dioxygenase IFN Interferon IGF Insulin Growth Factor IL Interleukin IL-2R Interleukin 2 Receptor iNOS Inducible Nitric Oxide Synthase MCP Monocyte Chemotactic Protein MHC Major Histocompatibility Complex NK Natural Killer OvUS Ovine Uterine Serpin OVA Ovalbumin PHA Phytohemagglutinin PND Post Natal Day PolyIÂ•PolyC Polyinosinic-Polycytidylic Acid P4 Progesterone
xi PHA Phytohemagglutinin RCL Reactive Center Loop TCR T Cell Receptor TNF Tumor Necrosis Factor TGF Transforming Growth Factor UGKO Uterine Gland Knockout
xii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ENDOMETRIAL ADENOGENESIS AND UTERINE IMMUNE REGULATION IN SHEEP By Maria B. Padua December 2004 Chair: Peter J. Hansen Major Department: Animal Sciences Progesterone is the hormone of pregnanc y in mammals. Among its roles, it is involved in regulation of the maternal immune system to prevent rejection of the semiallograft conceptus. In ewes, progesterone induces secretion of a 55-57 kDa member of the serine proteinase inhibitor superfamily called ovine uterine serpin (OvUS) by the endometrial glands in the uterus. OvUS exhibits a variety of immunosuppressive properties towards lymphoid cells and has been proposed to be a mediator of progesterone actions during pregnancy. To de termine whether OvUS plays this role, an experiment was conducted to determine wh ether the immunosuppressive effects of progesterone occur in ewes that are treate d hormonally to prevent the development of uterine glands. These so-called uterine gland knockout (UGKO) ewes are produced by exposure of the neonate to synthetic progestin for at least 8 weeks. For this experiment, ovariectomized cont rol and UGKO ewes were treated with 100 mg/day progesterone for 30 days. An aut ograft and allograft of skin were then
xiii placed in each uterine lumen and treatments were continued for an additional 30 days before grafts were examined for survival. All autografts surviv ed and had a healthy appearance after histological analysis. Allograf ts were generally rejected in ewes treated with the vehicle but were protected in hor mone-treated ewes, regardless of uterine phenotype. Analysis of the hist oarchitecture and protein synthe tic capacity of the uterus revealed that progesterone induced differen tiation of endometrial glands and synthesis and secretion of OvUS in UGKO ewes. In part icular, endometrial gla nds were absent or greatly reduced in the UGKO ewes treated w ith the vehicle, but were present in UGKO ewes treated with progesterone . Indeed, OvUS was presen t in uterine fluid of UGKO ewes treated with progester one and was localized by i mmunohistochemistry on both luminal and glandular epithelium. Progest erone also reduced the difference between control and UGKO ewes in the density of CD45R+ lymphocytes residents in the uterine endometrium. Taken together, the results confirm that proge sterone delays graft response in uteri. The development of endometrial glands a nd induction of OvUS synthesis caused by progesterone treatment in the UGKO model limit conclusions regarding the role of endometrial glands in mediating the immunosuppressive action of progesterone. Nonetheless, responses of UGKO ewes to progesterone indicate that the hormone can induce de novo development and differentiation of endometrial glands and lymphocyte homing to the uterus. Use of the UGKO model should prove useful in further elucidating the molecular basis for progesterone-driven changes in uterine morphogenesis and immune function.
1 CHAPTER 1 REVIEW OF LITERATURE Mechanisms Proposed to Allow Fetal Evas ion of the Maternal Immune Response The maternal immune system has long been recognized as a threat to the conceptus because the expression of paternal genes makes the conceptus an allograft to the mother. In 1953, Medawar proposed three possible mech anisms to explain how the conceptus survives this threat: (i) the presence of an anatomical barrier which separates the fetus and the mother, (ii) the antigenic immatu rity of the conceptus, and (iii) the immunological tolerance of the mother. Meda warÂ’s landmark paper highlighting what he termed the Â“immunological paradox of pregnanc yÂ” has led to over 50 years of research into pregnancy immunology, mostly with mouse models, and the formulation of additional hypotheses to explain the survival of the fetus. Nonetheless, most current concepts can still be organized around MedawarÂ’s original three mechanisms. Placenta as an Immunological Barrier MedawarÂ’s idea that the placenta forms an immunological barrier between the mother and the conceptus is supported by obser vations that the placenta does inhibit the movement of antibodies and T cells from th e mother to the conceptus (Chaouat et al., 1983). In species where there is transfer of antibodies across the placenta (for example, mouse), there is some evidence that antibodies against the fetal antige ns are preferentially absorbed by placenta (Raghupathy et al., 1981). Recent data suggest that binding of antibody to the placenta does not lead to cell lysis because of local downre gulation of the complement sy stem. In mouse, the gene
2 product Crry, which controls the deposition of activated complement proteins C3 and C4 on the surface of fetal cells, plays a key role in regulation of complement activation at the placenta because developing Crry-/embryos had surface-deposited C3 in the trophoectoderm and the ectoplacental cone with an invasion of polymorphonuclear inflammatory granulocytes (Xu et al., 2000) . Further studies demonstrated that the alternative pathway of the complement system is the primary contributor to the fetal loss, specifically the maternal complement component C3 (Mao et al., 2003). T cell trafficking across the placenta is also limited. In mouse, the placenta acts as a barrier for passage of cells from the moth er to the fetus and vice versa, which could otherwise lead to graft vs. host disease (Hunz iker et al., 1984). In addition, it has been proposed that FasL of maternal and fetal orig in could protect the pl acenta from maternal and fetal cell trafficking across the placenta (Hunt et al., 1997; Makrigiannakis et al., 2001). However, at least in humans, some fe tal cells can pass th rough the placenta in normal pregnancies and these cells or their descendants can persist into the motherÂ’s circulation for decades (Lo et al., 2000; Bi anchi and Lo, 2001). This process, known as microchimerism, could affect the maternal immune system and have been recently implicated in the development of auto immune diseases in women (Nelson, 2002). Antigenic Immaturity MedawarÂ’s idea that the fetus is antigenic ally immature is wrong although he was correct in hypothesizing that the placenta would be of reduced antigenicity. In mouse, there is expression of paternal MHC antigens on the embryo as early as the 2-cell stage to the blastocyst stage (Searle et al., 1976; Webb et al., 1977; Fernandez et al., 1999). The definitive placenta also expresses MHC class I antigens. It is true, however, that MHC class I antigen expression is dow nregulated in those parts of the placenta in contact with
3 the maternal endometrium. In mice for exam ple, MHC class I expr ession occurs on the spongiotrophoblast, but not on the labyrinthine trophobl ast in contact with the maternal system (Billington and Bell, 1983). In hum an placenta, cytotrophoblast expresses MHC class I antigen, but the syncy tiotrophoblast which is in contact with endometrium is MHC class I negative (Sunderland et al., 1981). Development of an immune response to fetal MHC class I antigens, which has been docum ented in mouse pregnancy (Kiger et al., 1985), could be due to shedding of MHC class I positive trophoblast or fetal tissue into the maternal circulation. One implication of reduced antigenicity of the outer layers of trophoblast is that these tissues are potentially at risk for lysis by maternal natural killer (NK) cells. Natural killer cells recognize targets that do not express MHC class I antigen, which ordinarily interacts with an inhibitor receptor on the NK cell to bl ock lysis (Ljunggreen and Karre, 1990). However, human and mouse trophoblasts are resistant to NK cell lysis despite the lack of MHC class I expression (Zuckerm ann and Head, 1987; King et al, 1990). In humans at least, partial protection agai nst NK cell lysis is afforded by placental expression of a non-classical MHC class I called HLA-G (produced by HLA-class Ib genes) which has only limited antigenic vari ation between individuals, and can inhibit recognition of target cells by NK cells (Le Bouteiller and Mallet, 1997). Mice deficient in NK cells by homologous recombination form anomalies at implantation site and decidual spiral arteries (Croy et al ., 2002; 2003). It has been hypothesized that cytokine producti on by NK cells, especially IFN, is involved in uterine vascular remodeling during pre gnancy (Croy et al., 2002).
4 Immune Tolerance Of all of MedawarÂ’s original hypotheses, mo st research has focused on the idea that there is either antigen-speci fic or nonspecific inhibition of immune responses against fetal antigens. There is evidence that both types of inhibition o ccur during pregnancy. Tafuri et al. (1995) proposed that matern al T cells undergo a transient tolerance to paternal alloantigens during pregnancy. This hypothesi s was tested by producing mice that were transgenic for TCR specifi c for the paternal alloantigen H-2Kb. In this way, the numbers of T cells against paternal antigen could be monitored by flow cytometry. During midpregnancy, transgenic mice had re duced numbers of T cells of the same clonotype when conceptuses were Kb positive, but not when conceptuses were of other haplotypes. In addition, Kb tumor grafts were not rejected when they were placed in H2kxd TCR transgenic mice bearing a Kb positive conceptus, but rejection occurred for syngeneic and thirdÂ–party allogeneic pregna ncies (Tafuri et al., 1995). The ability for graft rejection was restored after parturition. In another model, it was shown that T cells specific for fetal H-Y antigens were decrea sed during pregnancy in transgenic mice expressing TCR specific for H-Y antigen (J iang and Vacchio, 1998). Zhou and Mellor (1998) also showed that expr ession of paternally inherite d MHC class I molecules (H2Kb) by the trophoblast produced a reduction of CD8 expressed on the surface of maternal CD8+ T cells. Taken together, these experi ments indicate specif ic inhibition of T cell populations that rec ognizes paternal antigens. A host of molecules produced by the placenta and endometrium have also been proposed to act in a TCR-nonspecific manner to inhibit lymphocyte response . Among these molecules proposed for this ro le are transforming growth factor(TGF) (Arck et al., 1995), interleukin (IL)-10 (Cha ouat et al., 1995), leukemia inhibitor factor and
5 macrophage-stimulating factor (M-CSF) (C lark et al., 1994; Pi ccini et al., 2001), prostaglandin E2 (Low and Hansen, 1988; Parhar et al., 1989), indoleamine 2,3dioxygenase (IDO) (Munn et al., 1998) and prog esterone (Szekeres-Bartho et al., 1985; Hansen, 1986) among others. In mouse, much emphasis recently has focused on the enzyme IDO which catabolizes tryptophan and is expressed in macr ophages, specific subset of dendritic cells and in the human placenta (Mellor and Munn, 2001; Mellor et al., 2003). The expression of IDO is increased by IFNand, in dendritic cells by IL -10 (Grohmann et al., 2003). Exposure of pregnant mice to 1-methyl-t ryptophan, an inhibitor of IDO, induced rejection of allogeneic conceptuses but not syngeneic conceptuses (Munn et al., 1998). This result was interp reted to mean that IDO inhibits lymphocyte responses at the maternal-fetal interface by starving the lym phocytes of tryptophan while prevention of this depletion with an IDO inhibitor allows development of a maternal immune response against the conceptus (Munn et al., 1998). A si milar role for IDO has been demonstrated in vitro when T cell proliferation was i nhibited by monocytes differentiated into macrophages using M-CSF (Munn et al., 1999). Moreover, IDO is also involved in the regulation of the complement system si nce inflammation caused by C3 complement component deposition was detected at the matern al-fetal interface before fetal rejection of allogeneic pregnant mice exposed to th e IDO inhibitor (Mellor et al., 2001). T regulatory cells have also been implicated in the downregulation of the immune response at the maternal-fetal interface. Thes e naturally occurring cells are characterized by the surface coexpression of CD4+ and CD25+. Although the inhibitory effects of these regulatory cells seems to be mediated by th e production of immunosuppressive cytokines
6 such as IL-10 and TGF, recent experiments have shown that CD4+ CD25+ T regulatory cells can induce IDO production in dendritic ce lls via the cell surface marker cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) (F allarino et al., 2003; Wood and Sakaguchi, 2003). In human peripheral blood, T regulatory cells ar e found throughout pregnancy, peaking during the second trimester and decrea sing at the postpartum period (Somerset et al., 2004). The best evidence that T regulator y cells regulate maternal immune system against the fetus is that T cell-deficient BALB/c nude ( nu/nu ) mice receiving a lymphocyte preparation depleted of CD25+ lymphocytes did not sustain allogeneic pregnancies; fetal resorption was character ized by abnormal fetuses, hemorrhage and infiltration of CD3+ T cells at the maternal-fetal interface (Aluvihare et al., 2004). Transfer of lymphocyte depleted of CD25+ cells into females pregnant from syngeneic mating resulted in 50% of pregnancy success (Aluvihare et al., 2004). T-cell derived cytokines seems to be an im portant contributor to the regulation of both fetal survival and fetal rejection. Activated CD4+ T cells can be classified according to their cytokine production as type 1 CD4+ T cells (Th1), which produce IL-2, tumor necrosis factor(TNF), and IFNand which typically enhance cell-mediated immunity, and type 2 CD4+ T cells (Th2), which produce IL-4, IL-5, IL-6, IL-10 and IL13 and which typically promote B cell f unction (Saito, 2000). The production of cytokines by Th1 cells inhi bits the Th2 subset and vi ce versa (Raghupathy, 2001). During murine normal pregnancy, the Th2 cyt okines are constitutively present at the maternal-fetal interface, whereas IFNis transient (Lin et al., 1993). Disruption of normal pregnancy in mouse was caused by inj ection of proinflammatory cytokines IFN, TNFand IL-2 (Chaouat et al., 1990). In c ontrast, fetal resorption was prevented by
7 administration of IL-10 in abortion-prone mice (Chaouat et al., 1995). Defective cytokine production by Th2 cells subset, especi ally LIF, IL-4 and IL-10 was observed for decidual T cells of women with recurrent unexplained abortions and women undergoing abortions during the first trimester of pregnancy (Picci ni et al., 2001). Other studies have demonstrated that pr ogesterone, the hormone most directly responsible for pregnancy, provide s a transient state in which th e uterus is able to support the allograft conceptus. Moriyama and S ugawa (1972) shown that progesterone allowed implantation and proliferation of different xe nografts cells placed into the uterus of golden hamsters, but this effect was not achie ved in those animals treated with estradiol alone (Moriyama and Sugawa, 1972). Progester one also allows the survival of skin grafts and mouse hybridoma cells placed into uterine lumen of ovariectomized ewes (Hansen et al., 1986; Maje wski and Hansen, 2002). It is unclear whether the e ffect of progesterone on lym phocyte function is mediated directly through actions on the lymphocyte ac ting via progesterone receptor activation. Several authors have failed to identify proge sterone receptors on lymphocytes (Mansour et al., 1994; Schuts et al., 1996; King et al., 1996) or inhibit effects of progesterone on lymphocyte function with progesterone recepto r antagonists (Van Voorhis et al., 1989). In humans, Szekeres-Bartho et al. (1985) reported that progesterone receptors were induced in peripheral blood lymphocytes dur ing pregnancy suggesti ng that progesterone has direct immunomodulatory effects on lymphocytes. In cows, however, there is no difference in lymphocyte sensitivity to proge sterone between pregna nt and non-pregnant cows (Monterroso and Hansen, 1993). Other studies have shown that the immunosuppressive actions of progesterone on lymphocytes are exer ted by inhibition of
8 K+ channels to depolarize the pl asma membrane and inhibit Ca+2 signaling and subsequent IL-2 production (Ehring et al., 1998). Perhaps progesterone exerts its eff ects by inducing the production of other immunossupresive molecules. It has been shown that Th2 cells are upregulated by progesterone (Piccini et al., 2001). Expression of IDO in mo st human cells are possibly regulated by progesterone (Mel lor and Munn, unpublished observat ions cited in Mellor et al., 2001). Szekeres-Bartho et al. (2001) ha ve proposed that human lymphocytes from pregnant individuals and pe ripheral blood lymphocytes from non-pregnant individuals exposed to progesterone produce a 34 kD a protein known as progesterone-induce blocking factor, which has inhi bitory effects on NK cell activ ity and lymphocytes, actions on the cytokine balance (Th2 over Th1) and also reverse high resorption rates in the murine abortion system. In summary, many mechanisms have been proposed to help the fetus to evade the maternal immune system during pregnancy. In the remainder of this chapter, relevant aspects of pregnancy in the ewe and possi ble mechanisms by which the allogeneic conceptus survive in the uterus of that species will be reviewed. Pregnancy in Sheep Pregnancy lasts approximately 147 days in ewes and is maintained by progesterone. During the first 50 days of gestation plasma concentration of progest erone range from 2-3 ng/ml at values similar to those found during th e luteal phase of the estrous cycle (Bassett et al., 1969; Stabenfeldt et al ., 1971). After day 50 of pregnancy, removal of the corpus luteum does not cause abortion (Casida and Warwick, 1945), suggesting that the placenta is a sufficient source of progest erone during the latter part of pregnancy. Indeed, there is a steady increase in progesterone concentration to concentrati ons about five times greater
9 (10-15 ng/ml) than values found during early pr egnancy; concentratio ns start to decline again around two weeks before parturition (Bas sett et al., 1969; Stabenfeldt et al., 1971). Placentation The placenta in sheep is of the epitheliochoria l type, but it is generally referred to as synepitheliochorial because the uterine epithe lium is modified by invasion and fusion of fetal trophectoderm binucleate cells to fo rm a syncytium which lasts throughout pregnancy (Wooding 1982). The migration of binucleate cells towards the maternal tissue seems to be involved with the deliver y of protein molecules such as placenta lactogen (Wooding 1982) and pregnancy-associat ed glycoprotein (Xie et al., 1991) to the maternal circulation. The most intimate a ttachment between placenta and endometrium, and the major site of gaseous and nutrie nt exchange, occurs at structure called placentomes (Davies and Wimsatt, 1966; Perry, 1981). These are formed by the combination of knob-like structures on the placenta called cotyledons and aglandular cup-like structures on the endometrium ca lled caruncules (Davies and Wimsatt, 1966; Perry, 1981). The number of placentomes is around 60 to 100; they increase in size and number until day 90 of pregnancy when they start to shrink and possibly decrease in number (Davies and Wimsatt, 1966). Within the placentomes, at the base of the chorionic villi, maternal blood is leaked in to spaces between the maternal and fetal tissues (Perry, 1981). Unlike for placentomes, glandular epitheliu m is present in the interplacentomal endometrium (Perry, 1981). In this region, feta l and maternal epithelia are in contact and the luminal epithelium of the endometrium is infiltrated by fetal binucleate cells (Davies and Wimsatt 1966; Perry, 1981).
10 Major Histocompatibility Complex (MHC) Expression MHC class I molecules were detected in most of the stromal cells and in the epithelium of the placentomes before day 19 of pregnancy when formation of the syncytium became apparent (Gogolin-Ewens et al., 1989). However, MHC class I molecules were not detected in the tro phoblast or in the syncytial layer of the placentomes during latter stages of pregnanc y (Gogolin-Ewens et al., 1989). Similarly, MHC class I mRNA or protein were not detected in the co nceptus trophectoderm, even though 2-microglobulin mRNA (but not protein) was found in the trophectoderm at day 20 of pregnancy (Choi et al., 2003). Thus, the absence of MHC class I expression may provide the trophoblast protection against T-cel l cytotoxic activity. Similarly, placental tissues were negative for expression of MH C class II antigen expression (Gogolin-Ewens et al., 1989). In contrast, endometrial tissu e does express MHC class I molecules on both placentomal and interplacentomal uterine epithelial cells (Go golin-Ewens et al., 1989). In the interplacentomal regions, MHC class I was detected in the glandular epithelium and connective tissues at all later stages of pregnancy (Gogolin-Ewens et al., 1989). During days 10 and 12 of pregnancy, MHC class I and 2-microglobulin mRNA and protein were detected in the luminal epithe lium only (Choi et al., 2003). At days 14 to 20, these mRNAs and proteins were localized instead in the middle and deep glandular epithelium and stroma and there was a 3-fold increase in intensity of expression as compared to tissues from days 14 or 16 of the estrous cycle (Choi et al., 2003). Intrauterine infusi on of interferon(IFN) increased the amount of mRNA for MHC class I and 2-microglobulin in the glandular epit helium and stroma in ovariectomized, progesterone-treated ewes (Choi et al., 2003).
11 MHC class II molecules in the endometrium were found in the placentomal area throughout pregnancy with more positive cells during early stages (Gogolin-Ewens et al., 1989). In the interplacentomal regions, MHC class II molecules were localized in epithelium and stroma; numbers of positive cells decreased in the subepithelial stroma as pregnancy advanced (Gogolin-Ewens et al., 1989). In ovariectomized ewes, MHC class II positive cells were localized throu ghout the caruncular and intercaruncular endometrium with more intense staining in the luminal epithelium and subepithelial layers (Gottshall and Hansen, 1992). Treatment with progesterone for 60 days reduced the number of MHC class II positive cells (Gottshall and Hansen, 1992). Uterine Leukocyte Populations in Sheep In the uterus of cyclic ewes, lympho cytes are found in the caruncular and intercaruncular epithe lium (Lee et al., 1988) but, during pregnancy, lymphocytes become nearly absent in the placentomes (Gogolin-Ewe ns et al., 1989). In the intercaruncular endometrium, lymphocytes are mainly localized in the luminal and glandular epithelium and in some areas of the stroma immediatel y beneath these epithe lia (Lee et al., 1988; Gotshall and Hansen, 1992; Majews ki et al., 2001). In the non-pregnant sheep uterus, around 50% of the intraepith elial lymphocyte population of the endometrium is composed of CD8+ CD45R TCR cells and the remaining popul ation consists of equal numbers of CD8+ CD45R+ TCR+ cells and CD8+ CD45R+ TCR cells (Meeusen et al., 1993). During the later st ages of pregnancy, in c ontrast, the majority of intraepithelial lymphocytes are large granulated CD8+ CD45R+ TCR+ cells (Meeusen et al., 1993). These cells c ontain the cytolytic molecule perforin (Fox and Meeusen, 1999). The proportion of CD8+ CD45R+ TCR+ cells increase in number in the luminal epithelium of the interplacento mal areas during mid and late pregnancy (Lee et al., 1992;
12 Meeusen et al., 1993; Nasar et al., 2002). In contrast, this cell population remains nearly constant in numbers in the glandular epith elium throughout pregnancy (Lee et al., 1992; Nasar et al., 2002). During partur ition, a significant decrease o ccurs in the pe rcentage of both large granulated and non-granulated ly mphocytes in the luminal and glandular epithelium (Nasar et al., 2002). Both lymphoc yte populations remain constant at day 1 postpartum in both epithelia, but at postpart um day 3, there was an increase of large granulated lymphocytes in the luminal epithelium and in both types of lymphocytes in the glandular epithelium (Nasar et al., 2002). While CD8+ cells are present in the endome trium, there are very few CD4+ T helper cells, mast cells, or B cells in the endometriu m of cyclic or pregnant ewes (Lee et al., 1988; 1992, Gogolin-Ewens et al., 1989). Cells involved in innate immunity have been recently described in the ovine endometrium. Macrophages are mainly localized in the intercaruncular stroma, between th e luminal and glandular epithelium, and their numbers increase greatly during pregnancy (Tekin and Hansen, 2004). Inducible nitric oxide synthase (iNOS), an enzyme required for activ ated macrophages to produce nitric oxide, is found in the luminal and glandular epitheliu m, and stromal cells of the intercaruncular endometrium, placentomes and intercotyledonary placenta of pregnant ewes (Zheng et al., 2000; Kwon et al., 2004). In the interc otyledonary placenta and intercaruncular endometrium, iNOS levels were high at day 60 of gestation, declin e at day 80 and then increase again on days 100 and 120, respectivel y (Kwon et al., 2004). The highest levels of activity of iNOS were detected in the placentomes between days 100-120 of pregnancy (Kwon et al., 2004).
13 Activity characteristic of natu ral killer-like cells was found in endometrial epithelial cells (Tekin and Hansen, 2002). Moreover, th ere was intense immunol ocalization of the cytolytic protein perforin in the glandular epithelium, less intense staining in the luminal epithelium and a few areas of staining in the endometrial stroma of pregnant and nonpregnant uterine horns of unilaterally-p regnant ewes (Tekin and Hansen, 2003). Also, eosinophils were localized in the st ratum compactum stroma after day 11 in the pregnant ewe and cells expressed mRNA le vels for the chemoattractants monocyte chemotactic protein (MCP)-1 and -2 (Asselin et al., 2001). Intrauterine infusion of IFNin ovariectomized progesterone-treated ewes increased the abunda nce of endometrial MCP-1 and MCP-2 mRNAs (Asselin et al., 2001). Messenger RNA for a variety of proinflammat ory cytokines such as tumor necrosis factor(TNF), IL-1 , IL-1 , IL-6, IL-8, IL-10, IFNand TGFwas found in the uterine epithelium of nonpregna nt and pregnant ewes, alt hough there was no differences in expression between the groups (Fox et al ., 1998). In contrast, there was very low expression of IL-3 and granulocyte macrophage -colony stimulating fact or (G-CSF) in the endometrium and no expression of IL-2 or IL -4 in the endometrium from nonpregnant or pregnant sheep (Fox et al., 1998). There is evidence for local ac tivation of endometrial lym phocytes by the conceptus. Expression of activation markers such as CD25, CD44, CD29 and L-selectin on TCR+ intraepithelial lymphocytes recovered from th e pregnant horn of unilaterally-pregnant was higher than expression for lymphocytes fr om the non-pregnant horn or cyclic ewes (Liu et al., 1997). There is systemic regul ation of endometrial leukocytes numbers as well, however. In particular, the increase in the number of TCR+ cells in the luminal
14 epithelium during mid and late pregnancy (L ee et al., 1992) is likely due to endocrine changes associated with pregnancy and not due to a local signal from the conceptus because numbers of TCR+ were higher in the luminal ep ithelium of both pregnant and nonpregnant uterine horns of unilaterally-pregna nt ewes as compared to ovariectomized ewes (Majewski et al., 2001). Moreover, th e difference in numbers between pregnant and non-pregnant horns was small and non-signi ficant. Macrophage levels seem to be regulated by both systemic and local signals because the accumula tion of macrophages in both horns of unilaterally-pregnant ewes wa s higher in the non-pregnant ewes and accumulation was greater in the pregnant ut erine horn than in the non-pregnant horn (Tekin and Hansen, 2004). Progesterone as Immunosuppressive Molecule in Sheep The identity of pregnancy-a ssociated systemic and loca l regulators of endometrial lymphocyte function is largely unknown. One molecule that has been well documented as a regulatory molecule for uterine immune function is progesterone. Progesterone can inhibit two aspects of the immune function Â– clearance of bacter ia and rejection of allografts and xenografts. In cyclic ewes, intrauterine inoculations with Actinomyces pyogenes and Escherichia coli caused the establishment of uterine infections when inoculations were performed on day 7 of the estrous cycle, but not when inoculations were performed at day 0 of the cycle (R amadan et al., 1997; Seals et al., 2003). Injections of progesterone for 20 days to ovariectomized-postpartum ewes resulted in uterine infections after intrauterine infusions of A. pyogenes and E. coli (Seals et al., 2002; Lewis 2003). Such uterine infections did not occur in animals treated with the vehicle (Ramadan et al., 1997; Seals et al., 2002; 2003; Lewis 2003). Daily injections of progesterone did not prevent reje ction of skin grafts placed into the uterus when the
15 hormone was given three days before the placem ent of the grafts into the uterus (Reimers and Dziuk, 1974). However, long-term exposur e to progesterone in ovariectomized ewes beginning 30 days before grafting prolonged surv ival of skin grafts (Hansen et al., 1986) and mouse hybridoma cells (xenografts) placed into uterine lumen of sheep (Majewski and Hansen, 2002). Progesterone can also lead to change s in the population of endometrial lymphocytes. Treatment with progesterone for 60 days reduced the number of MHC class II+ and CD45+ cells in the intercaruncular en dometrium of ovariectomized ewes (Gottshall and Hansen, 1992). In contrast, th e number of macrophages was not affected by progesterone treatment in ovariectomized ewes (Tekin and Hansen, 2004) and expression of MCP-1 and MCP-2 mRNA in en dometrial eosinophils in ovariectomized ewes was increased by progesterone treatment (Asselin et al., 2001). The mechanisms by which progesterone inhibit uterine immune function are incompletely understood. It is unlikely that progesterone acts directly on uterine lymphocytes. High concentrations of progesterone (10-6 to 10-5 M) are required to suppress lymphocyte prolifera tion induced by mitogens (Staples et al., 1983; Low and Hansen, 1988; Monterroso and Hansen, 1993). These concentrations are higher than the KD (10-10 M) of the progesterone receptor (Olea-Serrano et al., 1985) and higher than concentrations in circulations (~10-8 M). Probably, progesterone inhibits lymphocytes through a receptor-independent mechanism, because inhibition was not affected by the presence of the progesterone receptor an tagonist RU 38486 (Monterroso and Hansen, 1993). It is possible that i nhibitory actions of progeste rone are exerted through the induction of synthesis and secretion of ot her molecules in the uterus that have
16 immunosuppressive properties. Treatment of ovariectomized ewes with progesterone induced the appearance of lymphocyte-inhibitory activity in uterine flushings or uterine fluid (Stephenson and Hansen, 1990; Hansen and Skopets, 1992). Moreover, uterine fluids from unilaterally-pregnant ewes contain a factor that can inhibit mitogen-induced lymphocyte proliferation (Stephenson et al., 1989a ). This molecule has been identified as ovine uterine serpin (Skopets and Hansen, 1993). Characteristics of Ovine Uterine Serpin Relationship to Serpins Ovine uterine serpin (OvUS) belongs to the serpin superfamily of serine proteinase inhibitors (Ing and Roberts, 1989), which also includes -1-antitrypsin, angiotensinogen, and ovalbumin among others. Uterine serpins are also found in ut erine secretions from pregnant cows, sows and goats (Leslie et al., 1990; Malathy et al., 1990, Tekin et al., 2004). Ovine uterine serpin shows about 96% amino acid sequence identity to caprine uterine serpin (CaUS) (Tekin et al., 2004), 72% identity to bovine uterine serpin (BoUS) but only about 50% and 56% identity, respectivel y, to two distinct porcine uterine serpins (PoUS-1 and PoUS-2) (Mathialagan and Hansen, 1996). The structure of serpins is characterized by three -sheets, nine -helices and the presence of a reactive center loop (RCL) that is exposed for interaction with the proteinase (Irving et al., 2000; Silverman et al., 2001; van Ge nt et al., 2003). Binding of proteinase and serpin leads to a cleavag e and inactivation of the serpin and a conformational change that makes the protein more thermodynamically stable (Irving et al., 2000; Silverman et al., 2001; van Gent et al., 2003). Not all the members of the serpin superfamily function as proteinase in hibitors. Among the non-inhibitory serpins are the molecular chaperone heat shock pr otein 47, the hormone transport proteins
17 corticosteroid binding globulin and thyroxine binding globulin, and proteins without a well-understood function like ovalbumin (Irving et al., 2000; van Gent et al., 2003). Ovine uterine serpin also appears to be an inactive proteinase . The protein does have some inhibitory activity to the as partic proteinases pepsin A and pepsin C (Mathialagan and Hansen, 1996; Peltier et al ., 2000a), but the concentration required to inhibit pepsin is too high us ing a serpin-like inhibitory m echanism. Also, OvUS did not inhibit a wide range of serine proteinases (I ng and Roberts, 1989). The tertiary structure of OvUS appears to be different from a protot ypical serpin since, unlike a typical serpin, limited proteolysis with trypsin did not cleave its RCL or affect its secondary structure, thermal stability or biological act ivity (Peltier et al., 2000a). Biochemical Properties Ovine uterine serpin exists in uterine flui d as a pair of basic glycoproteins with molecular weights of 55,000 and 57,000, derive d from a single 54,000 precursor (Moffatt et al., 1987; Hansen et al., 1987). The isoel ectric point is 9.2 (Hansen et al., 1987). Amino acid sequence of OvUS indicates th e presence of two N-linked glycosylation sites, indicating that the two major forms of OvUS ma y differ in the number of carbohydrate chains they possess after posttr anslational modification (Hansen et al., 1987; Ing and Roberts, 1989). The carbohydr ate content of OvUS consists of 2.8% neutral sugars, 2.5% amino sugars, and 0.3% sialic acid (Hansen et al., 1987). Ovine uterine serpin can bind the pregna ncy-associated glycoproteins which are inactive members of the aspartic proteina se family produced by the ovine trophoblast (Mathialagan and Hansen, 1996). It also bi nds activin A present in allantoic fluids (McFarlane et al., 1999) a nd the immunoglobulins IgA and IgM, but not IgG (Hansen and Newton, 1988). Thus, OvUS may act as a carrier serpin for other proteins.
18 Similarly, PoUS was found to mediated iron tr ansfer across the pig placenta by binding to the endometrial iron-binding protei n uteroferrin (Renegar et al., 1982) Endometrial Secretion Ovine uterine serpin is th e major protein produced by the sheep uterus during most of pregnancy (Bazer et al., 1979; Moffatt et al., 1987). It is also transported across the placenta and can be found in amniotic and al lantoic fluids (Newton et al., 1989). Large amounts of the protein are present in the uterus Â– at day 140 for example, the total protein present in the uterine fluid recovered from unilaterally-pr egnant ewes is around 15 g and most of this, as identified by sodium dodecy l sulfate polyacrylamide gel electrophoresis (SDS-PAGE), is OvUS (Bazer et al., 1979; Moffat et al., 1987). Ovine uterine serpin mRNA can first be detected around Days 13-16 of the estrous cycle and at Days 13-15 of pr egnancy (Ing et al., 1989; Stew art et al., 2000). In the intercaruncular endometrium, a 3-fold increa se of steady-state le vels of OvUS mRNA occurs between Days 20 and 60, another 3-fold between Days 60 and 80, and a decline at Day 120 of pregnancy (Stewart et al., 2000). Between days 20 and 50 of gestation, the expression of OvUS mRNA is lower in the deep endometrial glandular epithelium than in the upper glandular epithelium of the stratum spongiosum (Ste wart et al., 2000). There was however, no difference in mRNA expressi on between the upper and lower glandular epithelium of the stratum spongiosum between days 50 and 60 of pregnancy (Stewart et al., 2000). OvUS mRNA was expressed at high levels in all glandul ar epithelium in the stratum spongiosum between days 60 and 120 of gestation (Stewart et al., 2000). On postpartum day 1, OvUS mRNA was still dete cted in the stratum spongiosum of the glandular epithelium, but not in the superficial glandular epithelium (stratum compactum)
19 (Gray et al., 2003). At postpartum days 7 a nd 28, OvUS mRNA was not detected in the endometrial glands (Gray et al., 2003). The secretion of OvUS is under influence of progesterone (Moffatt et al., 1987; Ing et al., 1989; Leslie and Hansen, 1991). In ova riectomized ewes, ovine uterine serpin can be detected after 6 days of progesterone therapy (Ing et al., 1989). A large increase in the secretion of OvUS is observed by 14-30 days of progesterone treatm ent (Ing et al., 1989; Leslie and Hansen, 1991). Levels of OvUS mRNA were not affected by uterine infusion of placental lactogen and grow th hormone in progesterone-treated ovariectomized ewes (Spencer et al., 1999a), but was increased in the glandular epithe lium by uterine infusion of placental lactogen and/or grow th hormone combined with IFNinfusions (Spencer et al., 1999a; Noel et al., 2003). A decrease of OvUS mRNA in the glandular epithelium of ovariectomized ewes was induced by co-admin istration of estradio l with progesterone, which up-regulates the expression of progester one receptors in the endometrium (Spencer et al., 1999a). It appears that OvUS is in itially produced only by the uterine glands and then its expression spreads to the luminal epitheliu m. Although OvUS protein was found only in glandular epithelium at day 60 of pregnancy, it was immunolocalized in the luminal and glandular epithelium of the intercarun cular endometrium by days 120 to 140 of pregnancy (Moffatt et al., 1987; Stephenson et al., 1989b). Immunosuppressive Properties Immunosuppressive actions of OvUS have been broadly tested. Purified OvUS inhibited lymphocyte pro liferation produced by exposure to the antigen Candida albicans , phytohemagglutin (PHA), concanaval in A (Con A), and in the mixed lymphocyte reaction (Segerson et al., 1984; Stephenson et al., 1989a; Skopets and
20 Hansen, 1993; Skopets et al., 1995). A lthough OvUS inhibited T cell-dependent lymphocyte proliferation, it did not caus e any immunosuppressive activity against lymphocytes activated by the T and B cell mitogen pokeweed (PWM) (Skopets and Hansen, 1993). OvUS also reduced the anti body titer in ewes imm unized against the Tcell dependent antigen ovalbumin (OVA) (Skopets et al., 1995). Ovine uterine serpin had no effect on the reduction of skin-f old thickness caused by Mycobacterium tuberculosis in sheep (Skopets et al., 1995) and failed to inhibit expression of CD25 on -T + cells induced by Con A (Peltier et al., 2000b). Perhaps, the failure of OvUS to suppress proliferation of -T + cells reflects the increase in numbers of these cells in the endometrial epithelium during mid and late pregnancy (Lee et al., 1992) when OvUS is produced at high concentrations in the uterus. Ovine uterine serpin also inhibits NK cells . This was first demonstrated by Liu and Hansen (1993) who found OvUS inhibited NK-like activity in sheep lymphocytes and mouse splenocytes against K562 and YAC-1 ta rget cells (Liu and Hansen, 1993). OvUS also inhibited lytic activity of NK-like cel ls in peripheral blood lymphocytes and endometrial epithelium against D-17 cells infect ed with bovine herpes virus-1 (Tekin and Hansen, 2002). In vivo, OvUS blocked aborti on induced by poly(I)Â•poly(C) in pregnant mice (an NK-cell mediated phenomenon; Kinsky et al., 1990), and reduced basal splenocyte NK cell activity (Liu and Hansen, 1993). Binding of OvUS to lymphocytes is specific, dose dependent and saturable (Liu et al., 1999). OvUS contains seve ral phosphorylation sites such as tyrosine kinase, protein kinase C, and cyclic adenosine monophosphate (Peltier et al., 2000c ). Although the exact mechanism by which OvUS inhibits lym phocyte proliferation remains unknown, the
21 protein does block IL-2 indu ced proliferation and reduced expression of CD25 (IL-2R chain) (Peltier et al., 2000b). OvUS does not inhibit the cost imulatory effect of CD26 on PHA-stimulated lymphocytes (Liu and Hansen, 1995) and does not block the increase in IL-2 mRNA caused by Con A (Peltier et al ., 2000b). The immunosuppressive effect of OvUS is not blocked by addition of neutraliz ing antibody to transforming growth factor(Skopets and Hansen, 1993). Ovine Uterine Morphogenesis In sheep, uterine morphogenesis occurs dur ing fetal and neonatal life. Uterine horns are already fused and slightly cu rved on gestational days 55 Â– 60 and the mesenchyme is already differentiated into endometrium and myometrium (Wiley et al., 1987). Fetal uteri at day 90 Â– 100 have curved uterine horns, characte ristic of the adult uterus, with clearly define d nodular (aglandular) and inte rnodular (glandular) areas (Wiley et al., 1987). Although uterine glands are not present during fetal development, small invaginations in the mucosal epithelial are observed in the in ternodular areas on days 135 to 150 of fetal life (Wiley et al., 1987). Endometrial glands are still absent in the neonate at postnatal Day 0 Â– 1 (PND 0 1) (Bartol et al., 1988ab; Taylor et al., 2000). Adenogenesis starts between PND 0 Â– 7 when shallow invaginations appear in the luminal epithelium; tubular structures that branch and coil into the stroma are seen on PND 7 Â– 14 (Wiley, 1987; Bartol et al., 1988ab; Taylor et al ., 2000). Extensive uterine gland development that advances to the myometrium is reached on PND 21 Â– 28 (Wiley et al., 1987; Bartol et al., 1988b; Taylor et al., 2000). Complex, coiled and branched tubular glands are present throughout the stro ma on PND 42 Â– 56 that appear very similar to the adult uterus (Taylor et al., 2000).
22 The process of adenogenesis requires an in crease in cell prolifer ation of epithelial cells of the luminal and glandular epithelium (Bartol et al., 1988b; Ta ylor et al., 2000). The initiation of endometrial gland formati on is an ovary-independent event since gland development on PND 14 was not affected by ova riectomy at birth (B artol et al 1988a; Carpenter et al., 2003a). Although ovariectomy on PND 7 did not affect the number of superficial ductal invaginations of the gl andular epithelium from the luminal epithelium or the density of endometrial glands in the stratum compactum area of the stroma on PND 56, it reduced the total number of endometr ial glands and the density of endometrial glands in the stratum spongiosum area of the stroma, suggesting that some ovarian derived factors regulate in pa rt the process of coiling and branching between PND 14 and 56 (Carpenter et al., 2003a). In addition, reduction in mRNAs expression for follistatin, activin subunit A, activin receptor types IA and II and an increase in activin subunit B expression were detected by in situ hybridization on PND 56 in ewes ovariectomized on PND 7 (Carpenter et al., 2003a; Hayashi et al., 2003). Taylor et al. (2001) suggests that uterine gland prolif eration may be promoted by the action of stromal insulin-like growth f actor-I (IGF-I) and IGF -II acting through IGF-I epithelial receptor. IGF-I and IGF-II mRNAs expressi on were localized abundantly in the intercaruncular stroma underlying prolifer ating and differentiating endometrial glands on PND 7 to 56 and PND 21 to 42, respectively (T aylor et al., 2001). In contrast, IGF-I receptor mRNA expression was particularly a bundant in the luminal epithelium on PND 1 and also in the nascent and proliferating glands on PND 21 to 56 (Taylor et al., 2001). In addition, fibroblast growth f actor-7 and hepatocyte growth factor may be involved in the process of coiling and branching of gland morphogenesis because there was an
23 increase in expression of their mRNA for th ese growth factors in the intercaruncular endometrium after PND 21 (T aylor et al., 2001). Activation of the estrogen receptor (ER) seems to regulate in part the process of gland formation in the intercaruncular endom etrium between PND 14 and 56. The initial stages of the adenogenesis process between birth and PND 14, wh ich involves budding differentiation and formation of tubules, were not affected by treatm ent of neonatal ewes with an antagonist of ER and ER , but the antagonist retarded the coiling and branching processes of the gland formation that occu r between PND 14 and 56 (Carpenter et al., 2003c). In recent studies, endometrial adenogenesis was down-regulated by hypoprolactinemia and up-regulated by hyperprol actinemia, demonstrating that prolactin is implicated in uterine gland formation (C arpenter et al., 2003b). In addition, signal transducers and activators of transcriptions (STAT) 1, 3 and 5 were expressed in the nascent glandular epithelium and prolactin increased phosphorylation of STATs 1 and 5 in uterine explants (Car penter et al., 2003b). Uterine Gland Knockout Model In ewes, administration of a potent synthetic progestin to the neonate inhibits endometrial adenogenesis to generate an adult uterine gland knockout (UGKO) phenotype (Spencer et al., 1999b; Gray et al., 2000ab; 2001ab). Depending upon the animal, the intercaruncular endometrial ar eas contain ruffled luminal epithelium and compact stroma with either complete absence of glands, slight gla ndular invaginations or infrequent cyst and gland-like structures in the stroma (Gray et al., 2000ab; 2001a). Development of the UGKO phenotype requires a bout eight weeks of progestin exposure; neonatal administration for 13 days did not i nduce a complete ablati on of uterine glands
24 (Bartol et al., 1988b; Gray et al., 2000a). E xposure of lambs to estradiol valerate also generates an UGKO adult phenotype (Carpent er et al., 2003c; Haya shi et al., 2004). Uterine morphology and function is disrupted by long term-exposure of neonatal pigs to estradiol valerate (Spencer et al., 1993; Tarlenton et al., 2003). Treatments with progesterone and estradiol produced partia l to complete uterine gland knockout phenotype in cows (Bartol et al., 1995). The mechanism by which neonatal exposure to progestin blocks development of endometrial glands is not known, but the inhibition of adenogenesis is not produced by direct inhibi tion of endometrial ce ll proliferation (Gray et al., 2000b). Although UGKO ewes exhibit variability in th e length of the estrous cycle, they have normal progesterone concentrations in plasma and respond to prostaglandin F2 by undergoing luteinization (Gray et al., 2000a). Histoarchitectur al evaluation of UGKO ewes showed normal ovaries, oviductal ampulla r and isthmus, cervix, vagina and uterine myometrium (Gray et al., 2000b; 2001a). Ewes with the UGKO phenotype also have normal expression of endometria l progesterone, estrogen and oxytocin receptors (Gray et al., 2000a). UGKO ewes are not able to support pre gnancy to day 25 (Gray et al., 2000a; 2001ab; 2002). Immunoreactive IFNin uterine flushes from UGKO ewes was present in very low amounts or was undetectable and ewes either had no conceptus, degenerating tubular conceptus or fragmenting filamentous conceptus (Gray et al., 2001b; 2002). In addition, other molecules i nvolved in cell-cell adhesions such as osteopontin and GlyCAM-1 were undetectable or present in lo w levels in uterine flushes of UGKO ewes at 14 days after mating (Gray et al., 2002) . In contrast, normal patterns of
25 immunoreactive Muc-1 and integrins were pres ent in the luminal epithelium of the endometrium of UGKO ew es (Gray et al., 2002) and expression of IFNdependent genes in the endometrium could be induced by intra-uterine injections of recombinant ovine IFN(Gray et al., 2002). Synopsis and Hypothesis Based on this literature, there is evidence to support the idea th at OvUS inhibits lymphocyte proliferation and is a mediator of immunosuppressive e ffects of progesterone in the ewe. The definitive test to determ ine whether OvUS plays such a role is to determine whether removal of OvUS (usual ly performed homologous recombination) blocks the effect of progesterone on uterine im mune function. Since it is not feasible to use the transgenic model in sheep, the UGKO ew e is an alternative model to test the effects of OvUS on uterine graft rejection because the absence of uterine endometrial glands should result in a progesterone-tr eated ewe without th e capacity for OvUS synthesis. Therefore, the objective presente d in this dissertati on is to use the UGKO model to evaluate the ro le of endometrial glands and by in ference, OvUS in regulation of uterine immune function by pr ogesterone (Figure 1-1).
26 Figure 1-1. Schematic illustration of the hypothe sis. For control ewes, progesterone (P4) induces the secretion of OvUS from the uterine endometrial glands, and OvUS in turn decreases activation of lymphocytes in response to the skin allograft placed within the uterus. As a result, the allograft survives within the uterus. For the uterine gland kn ockout ewes (UGKO) ewes, progesterone will not induce the synthesis of OvUS b ecause uterine endometrial glands are absent. As a result, lymphocytes against the allograft will become Control ewes UGKO ewes P4 OvUS OvUS Graft survival Graft rejection P4
27 CHAPTER 2 ACTIONS OF PROGESTERONE ON UT ERINE IMMUNOSUPPRESSION AND ENDOMETRIAL GLAND DEVELOPMENT IN THE UTERINE GLAND KNOCKOUT (UGKO) EWE Introduction Among its many actions to maintain pregnanc y, progesterone acts to inhibit uterine immune function which may prevent immunologi cal rejection of th e conceptus (Hansen, 1998). In sheep, for example, progesterone reduces numbers of sp ecific populations of lymphocytes in the uterine endometrium (Gottshall and Hansen, 1992; Majewski and Hansen, 2002) and delays rejection or promotes survival of skin allografts (Hansen et al., 1986) and hybridoma xenografts (Majewski and Hansen, 2002) placed within the uterine lumen. The inhibitory effects of progesterone on uterine graft rejection are be lieved to be indirect because concentrations of progesterone required to directly inhibit lymphocytes are much higher than achieved in studies where progesterone inhi bited uterine immune response (Low and Hansen, 1988; Monterroso and Hansen, 1993). Rather, it has been hypothesized that immunosuppressive effects of progesterone in the uterus are mediated by secretion of a lymphocyte-i nhibitory molecule produced by the uterus in response to progesterone. Indeed, treatment of ovariectomized ewes with progesterone results in the appearance of lymphocyte-inhibi tory activity in uterine fl uid (Stephenson et al., 1989a; Hansen and Skopets, 1992).
28 A likely candidate for the progesteroneinduced immunosuppressive molecule in sheep is OvUS, also known an ovine uterine m ilk protein. This protein is a member of the serine proteinase inhibitor superfamily (Ing and Roberts, 1989) and produced by uterine endometrium under the influence of pr ogesterone (Moffat et al., 1987; Leslie and Hansen, 1991; Spencer et al., 1999a). Related members of this family have been also reported in pregnant cows (L eslie et al, 1990; Mathialaga n and Hansen, 1996) and pigs (Malathy et al., 1989). The major site of synthesis of OvUS is the glandular epithelium on the endometrium. Expression of OvUS mRNA was limited to glandular epithelium between days 17 and 120 of pregnancy (Ste wart et al., 2000) and OvUS protein was found in glandular epithelium but not lumi nal epithelium at day 60 of pregnancy, (Stephenson et al., 1989b). By days 120 to 140 of pregnancy, however, immunoreactive OvUS was also detected in luminal epithe lium (Moffatt et al., 1987; Stephenson et al., 1989b). A role of OvUS in inhibition of uterine immune responses is indicated by several observations. In vitro, OvUS inhibi ts lymphocyte activation and proliferation induced by T cell mitogens and interleukin-2 (Segerson et al., 1984; Skopets and Hansen, 1993; Skopets et al., 1995; Pelt ier et al., 2000b) and natural ki ller cell activity (Liu and Hansen, 1993; Tekin and Hansen, 2002). In vivo, OvUS inhibits T-cell dependent antibody production in sheep (Skopets et al., 199 5) and fetal loss induc ed by natural killer cell activation mediated by injection of poly (I) Â• poly(C) (Liu and Hansen, 1993). Conclusive evidence that OvUS mediates the effects of progesterone on uterine immune function will be dependent upon dem onstrating that progesterone is unable to regulate uterine immune function in sheep incap able of OvUS synthesis. While it is not practical to use homologous recombination to generate sheep without a functional OvUS
29 gene, it is possible to produce ep igenetic changes in ewes to l ead to an animal without the presence of endometrial gla nds or the ability to produ ce glandular-derived OvUS. Changes in uterine morphology and functi on caused by the action of sex steroid hormones have been reported in many livestock animals (Spencer et al., 1993; Bartol et al., 1995; Tarlenton et al., 2003; Carpenter et al., 2003c). Long-term exposure of lambs to norgestomet generates an adult that has e ither an absence of glands, slight glandular invaginations into the stroma, or limited numbe rs of cystor gland-like structures (Gray et al., 2000ab; 2001a), without apparent effect s on development of other extrauterine reproductive tract structures or the ovary (Gray et al., 2000b; 2001a). Ewes with the uterine gland knockout (UGKO) phenotype do no t show disturbances in circulating concentrations of progesterone and retain the ability to respond to prostaglandin F2 (Gray et al., 2000a). The uteri of cyclic UGKO ewes displays normal expression patterns of progesterone, estrogen, and oxytocin recep tors and several adhesion molecules on the uterine luminal epithelium (Gray et al., 2000a; 2002), and retains a normal response to interferon(Gray et al., 2002). Ho wever, UGKO ewes exhibit a recurrent pregnancy loss that involves loss of the elongation c onceptus between days 9 and 14 of pregnancy (Gray et al., 2000a; 2001ab; 2002). Available evidence supports the hypothesis that one or more adhesion proteins are deficient in the secretions of the uterus that are required to support early conceptus survival and development (Gray et al., 2002). In the present experiment, we tested the hypothesis that progesterone is unable to prolong survival of skin allografts placed w ithin the uterine lumen of UGKO ewes. An unexpected finding, that prolonge d progesterone treatment induced the development and differentiation of endometrial glands in UGKO ewes, prevented testing of the role of
30 OvUS but also provide evidence that one of th e actions of progesterone in adult animals is to stimulate histogenesi s of endometrial glands. Materials and Methods Materials Progesterone was obtained from Sigma-Al drich (St. Louis, MO). Hybond ECL nitrocellulose membranes and ECL chemiluminescence Western blot kit were purchased from Amersham Bioscience (Piscataway, NJ). Precast ready gels, kaleidoscope protein standard, 2-mercaptoethanol and gelatin were obtained from BioÂ–Rad (Hercules, CA). Hybridoma cells producing m onoclonal antibody to CD45R+ (clone 73B) were purchased from the European Collection of Cell Cu ltures (Salisbury, UK). Ascites fluid for CD45R+ was produced by the Hybridoma Core Facil ity of the Interdisciplinary Center for Biotechnology Research at the University of Florida. Monoclona l antibodies against OvUS (HLÂ–218 and HLÂ–708) were made as desc ribed previously (Leslie et al., 1990) and were prepared as hybridoma supernatants . Ovine uterine serpin was purified from crude uterine fluid of unilatera lly-pregnant ewes as describe d elsewhere (Liu and Hansen, 1995). Experimental Design A total of 23 Rambouillet crossbred ewes, 12 controls and 11 UGKO ewes, were used in the experiment. UGKO ewes were prod uced as described previously (Spencer et al., 1999b; Gray et al., 2000a) by implanting crossbred Rambouillet ewe lambs with a single Synchromate BÂ® (Sanofi, Overland Park, KS) impl ant within 12 hours of birth and every two weeks thereafter for a total of 8 week s. Implants were inserted subcutaneously in the periscapular area a nd released approximately 6 mg of norgestomet (17 -acetoxy11 -methyl-19-norpreg-4-ene-3,20-dione), a po tent synthetic 19-nor progestin, over a 14
31 day period (Bartol et al., 1988b). Normal c ontrol ewes did not receive implants. The normal control and UGKO ewes used in the present study were approximately three years of age. All ewes were bilaterally ovariectomized via midventral laparatomy 30 days before the initiation of the experiment. Treatments we re arranged according to a 2 x 2 factorial design with main effect of type (control vs UGKO) and hormone treatment (vehicle or progesterone). Ewes were randomly assigned w ithin type to hormonal treatment so that 8 control ewes and 8 UGKO ewes received dail y subcutaneous injections of 5 ml of 20 mg/ml progesterone (i.e., 100 mg/day) dissolved in a corn oil vehicle whereas 4 control ewes and 3 UGKO ewes received daily injections of 5 ml corn oil. On day 30 after the first injection, two skin grafts were placed in the uterus a ccording to procedure described by Hansen et al., 1986. An autograft (a piece of skin of the abdominal area from the same ewe) was placed in one randomly-chosen uterine horn while an al lograft (a piece of skin of the abdominal area from a different ew e) was placed into the other uterine horn. Daily injections were continued for an additional 30 days. On day 15 after graft placement, 10 ml blood samples were collected via jugular venipuncture at 2, 8 and 24 hours after injection to determine plasma concen trations of progesterone. On day 30 after graft placement, ewes were slaughtered by captive bolt stunning and exsanguination and reproductive tracts were recovered fo r examination of graft survival. Collection of Tissues and Uterine Fluids Visible uterine fluid was collected via aspi ration using an 18 ga needle and syringe. The total amount of uterine fluid collected wa s recorded and the flui d centrifuged twice at 3600 x g at 4oC for 20 minutes and the supernatant fraction stored at Â– 20oC for further analysis. When visible uterine fluid was not present, the uter us was flushed with 20 ml of
32 DulbeccoÂ’s phosphate buffered saline (DPBS) pH 7.3. After collection of fluid, the uterus was opened longitudinally and the surv ival of the skin graf ts and their general appearance recorded. Pieces of surviving gr afts were immediately preserved in a 10 % (w/v) neutral buffered formalin solution. Afte r graft collection, thr ee tissue samples (3-4 mm3) of the intercaruncular endometrium were collected at random from each uterine horn, near the area where the graft was place d, and also preserved in neutral buffered formalin solution. Skin Graft and Uterine Histology Uterine and skin graft tissue sections were dehydrated, em bedded in paraffin blocks, and 5Âµm sections prepared and mounted on slides. Histological appearance was determined after staining with hematoxylin and eosin and examination under bright field with a Zeiss Axioplan microscope (C arl Zeiss, Inc., GÃ¶ttingen, Germany). Photomicrographs were prepared using a Sony CD Mavica 400 digital camera (San Diego, CA, USA). Progesterone Radioimmunoassay Blood samples collected via jugular venipunctu re into heparinized tubes were placed on ice until they could be centrifuged at 2000 x g for 20 minutes, and the plasma harvested and stored at -20oC until the day of the assay. Progesterone was measured using a solid-phase 125I radioimmunoassay kit (Coat-ACountÂ® Progesterone Diagnostic Products Laboratory, Los Angeles). Sensitivity of the assay (90% Bo) was 0.1 ng/ml and the intrassay and interassay CV were 11.42% and 4.13% respectively. For statistical analysis, plasma samples with concentrations below the sensitivity of the assay were assigned a concentration equal to the sensitivity of the assay.
33 Determination of Protein Concentrat ion in Uterine Fluid and Flushing Protein concentration of uter ine fluid was determined using the Bradford procedure (Bradford, 1976) with bovine serum albumin as standard. Total protein content of the uterine lumen was calculated as the protein concentration times either the volume of uterine fluid recovered, or for ewes in which flushing was performed, the volume of DPBS used for flushing. Detection of OvUS in Uterine Fluid by Western Blotting Aliquants of 1 Âµg uterine protein diluted in a total volume of 20 Âµl with DPBS were mixed 1:1 (v/v) with loading buffer [0.125 M Tris HCl pH 6.8 containing 20% (v/v) sucrose, 10% (w/v) SDS, a trace amount of bromephenol blue and 5% (v/v) 2mercaptoethanol] and boiled for 3 minutes. Sa mples of 0.5 Âµg of uterine protein, as well as samples of purified OvUS and OVA were then separated according to molecular weight using one-dimensional discontinuous sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with 4 Â– 15% (w/v) gradient polyacrylamide gels and Tris HCl buffer. Proteins were transferre d electrophoretically to Hybond ECL 0.2 Âµm nitrocellulose membranes. Conditions for transfer were 200 mA for 1 hour at room temperature using a degassed buffer of 25 mM Tris, 193 mM glycine and 20% (v/v) methanol. Membranes were blocked overnight in TBS-T [10 mM Tris pH 7.6, 0.9% (w/v) NaCl and 0.3% (v/v) Tween -20] that also contained 1% (w/v) gelatin (TBS-TG). Membranes were rinsed four times with TBST and then incubated for 1 hour at room temperature with a mouse monoclona l antibody recognizing OvUS (HLÂ–218, 1:32,000 dilution of hybridoma supernatant in TBS-TG) and then washed as described before. Membranes were incubated for 1 hour at room temperature with horseradish peroxidaseconjugated sheep anti-mouse IgG (1:8000 dilution in TBS-TG) and washed as previously
34 described. Blots were deve loped using the ECL Western blotting chemiluminescence substrate kit for 1 minute. Sp ecificity of labeling was eval uated by including a negative control in which primary antibody was repl aced with hybridoma cell culture medium. Immunohistochemistry for OvUS and CD45R+ Lymphocytes Immunohistochemistry was performed usi ng the HistoScan Un iversal Monoclonal Detector kit (Biomeda, Foster City, CA) that utilizes strepavidi n-biotin peroxidase complex for detection. The procedure was performed on 5 Âµm-thick formalin-fixed, paraffin-embedded sections placed on poly-L-lysine-coated slides. After deparaffinization and rehydrati on slides were microwaved while immersed with 10 mM citrate pH 6.0. The procedure was perfor med three times for 2 minutes each and specimens were allowed to cool between pro cedures and finally for 20 minutes. Slides were then washed twice in deionized water (5 minutes each), once in phosphate buffered saline [PBS; 0.1 M sodium phosphate, pH 7.4 containing 0.9% (w/v) sodium chloride] containing 2% (v/v) hydrogen peroxide for 5 mi nutes and once in PBS for 5 minutes. All further steps were performed in a humidity ch amber at room temperature and slides were washed for 3 minutes between each step with PBS-GS [PBS containing 2% (v/v) donor goat serum]. After application of the tissue conditioner provided in the kit for 5 minutes, sections were incubated for 30 minutes w ith the primary antibody (for OvUS, HL-708, 1:800 dilution of hybridoma supern atant in PBS-GS; for CD45R+, clone 73B, 1:800 dilution of ascites fluid in PBS-GS). Ne gative controls were incubated with mouse ascites fluid, clone NS-1 (Sigma-Aldrich, St. L ouis, MO), at the same dilution as used for primary antibodies. Reagents provided in the kit were used for the other steps as recommended by the manufacturer. Incubati on with biotin-conjugated goat anti-mouse IgG and streptavidin-peroxidase were fo r 30 minutes each, incubation with the
35 chromogen reagent, 3-amino-9-ethylcarbazole was for 15 minutes and counterstaining with hematoxylin was for 3 minutes. Slides were then rinsed with deionized water for several minutes and blotted before applyi ng Crystal/Mount medium (Biomeda, Foster City, CA) and coverslips. Slides were examined for staining usi ng bright field with a Zeiss Axioplan microscope (Carl Zeiss, Inc., GÃ¶ttingen, Ge rmany). Photomicrographs were prepared using a Sony CD Mavica 400 digital camera (S an Diego, CA, USA). Presence of OvUS was evaluated qualitatively. CD45R+ cells were evaluated by scoring the relative abundance in the luminal epithelium, glandular epithelium and stroma on a scale from 0 (no positive cells) to 4 (very dense accumulati on of positive cells). One section per horn was evaluated for each sheep. Statistical Analysis Data were analyzed by least-square analys is of variance using the General Linear Models procedure of the Statistical Analys is System (SAS for Windows, Release 8.02, SAS Institute, Cary, NC, USA). For repeated-measures data (progesterone concentrations and numbers of CD45R+ cells), ewe was considered a random effect and other main effects were considered fixed. Tests of significance were determined using error terms determined after calculation of th e expected means squares. In general, the mathematical model considered main effects and all interactions. The one exception was for numbers of CD45R+ cells in the glandular epithelium where the absence of glands in all but one UGKO ewe treated with corn oil required analyses with several models. For these data, various tests of subsets of data were performed to determine differences between control and UGKO ewes treated with corn oil and effects of progesterone and type of graft on control ewes.
36 Data on total protein in the uterus exhibited heterogeneity of variance. Therefore, data were log-transformed be fore analysis and are presented as means Â± individual SEM for each group. For other variables, heterogene ity was not apparent and data are reported using a pooled estimate of error. Results Progesterone Concentration in Plasma For both control and UGKO ewes on day 15 of treatment, concentrations of progesterone were higher (p<0.001) for ewes treated with 100 mg/day of the hormone than for ewes receiving corn oil vehicle (Fig ure 2-1). Concentrations in the progesteronetreated ewes peaked at a c oncentration of 17.9 ng/ml at 8 hours after injection and then decline to a nadir of 13.1 ng/ml at 24 hours af ter injection (i.e., immediately before the subsequent progesterone injec tion). For ewes treated with the vehicle, values were generally below the limit of detection of the assay and in no case greater than 0.53 ng/ml. Gross Uterine Morphology While uterine weights were not recorded, treatment with progesterone caused an increase in uterine size in control ewes (c ompare Figure 2-2B showing a uterus from control ewe treated with corn oil vehicle with the uterus on the right of Fi gure 2-2C that represents a uterus from a control ewe treated with progesterone). In contrast, there was no obvious increase in size of the uterus of UGKO ewes treated with progesterone as compared to UGKO ewes treated with the co rn oil vehicle (compare Figures 2-2A and the left-hand uterus in Figure 2-2C). More over, the uterus of UGK O ewes treated with progesterone was much smaller than those of control ewes treated with the hormone (Figure 2-2C).
37 Figure 2-1. Progesterone concentration in plasma (ng/ml) 2, 8 and 24 hours after injection in ewes treated with corn oil vehicle (open circles) or progesterone (closed circles) over a 45 days period. Data represent least square means Â± SEM. Progesterone concentrations differed between the two groups (p<0.001). Time after injection (hours) 051015202530 Progesterone concentration (ng/ml) -5 0 5 10 15 20 25
38 Figure 2-2. Gross appearance of the uteri. Pane ls A and B illustrate uteri from a control (A) and uterine gland knockout (UGKO) ew e (B) that were treated with corn oil for 60 days. Panel C shows uteri of a UGKO (left side) and control ewe (right side) treated with pr ogesterone for 60 days. Note the difference in size between the two uteri. 2.6 cm 2.6 cm 2.6 cmA B C 2.6 cm 2.6 cm 2.6 cmA B C
39 There were other characteristic s of the uterus of the UGKO ew es that differed from the typical appearance of the sheep uterus. Of tentimes, the uterine wall was thin and appeared friable. The uterine lumen also usua lly contained a dark br own fluid Â– this was true for both vehicle-treated and progesterone -treated ewes. Fina lly, caruncules were almost totally absent on the endometrium of the vehicle-treated ewes. There were, however, prominent caruncules in most of the progesterone-treated ewes. Histological Analysis of Endometrium The presence of uterine endometrial glands is summarized in Ta ble 2-1. For control ewes both luminal and glandular epithelia we re present in all an imals regardless of hormonal treatment. The chief difference betwee n groups was the larger size of glands in the progesterone-treated animals (compare Figures 2-3A and 2-3B). For UGKO ewes treated with corn oil vehicle, luminal epithelium was present in all cases, but glandular epithelium was absent or greatly reduced in 2 of 3 ewes. A few scattered cyst-like or primitive glands could be identified but othe rwise uterine endometrium was composed of luminal epithelium and stroma (Figure 2-3C). In contrast to this pattern, well-defined glandular epithelium was presen t in the remaining corn oiltreated ewe (Figure 2-3D) and for all UGKO ewes treated with progesterone. Fo r the latter case, gla nds were present in either one uterine horn (n=4; 2 on the autograft side and 2 on th e allograft side) or in both uterine horns (n=4) (Figure 23E and 2-3F respectively). Survival of Skin Grafts Results of skin graft survival are summarized in Table 2-1. All autografts survived regardless of treatment. Grossly, the grafts a ppeared healthy and most were attached to
40 Table 2-1. Survival of skin grafts and presen ce of uterine glands and ovine uterine serpin (OvUS) in control and uterine-gla nd knockout ewes (UGKO) treated with corn oil vehicle (CO) or progesterone (P4).1 1 Results are fraction of graft surviving (graft survival), the fraction of ewes with uterine glands (Histology), the proportion of ewes in which OvUS was detected in uterine fluid or flushings (Western blot) and th e proportion of ewes in which OvUS was immunolocalized to endometrium by immunohistochemistry (IHC). Graft survival Histology Western blot IHC Ewe type Treatment Autograft Allograft Uterine glands OvUS OvUS Control CO 4/4 0/4 4/4 0/4 0/4 UGKO CO 3/3 1/3 1/3 0/3 0/3 Control P4 8/8 8/8 8/8 8/8 8/8 UGKO P4 8/8 8/8 8/8 8/8 8/8
41 Figure 2-3. Endometrial histol ogy for control and uterine gland knockout (UGKO) ewes treated with corn oil (CO) or progester one (P4) for 60 days. Sections were stained with hematoxylin and eosin. Pa nels A and B show uteri from control ewes with well defined luminal and glandul ar epithelium. Panels C, D, E and F show uteri from UGKO ewes treated with corn oil vehicle (C and D) and progesterone (E and F). Note the lack of glandular epithelium in C and E. Two of three UGKO corn-oil treated ewes had no glands (C), while one ewe possessed glands (D). All UGKO ewes tr eated with progesterone had glands in one uterine horn (n=4) or both (n =4) uterine horns. LE, luminal epithelium; GE, glandular epithelium; S, stroma. (A) Control-CO (B) Control-P4 200 mLE GE GE S S LE LE LE LE GE LE GE(C) UGKO-CO (F) UGKO-P4 (E) UGKO-P4 (D) UGKO-COS S S S (A) Control-CO (B) Control-P4 200 m 200 mLE GE GE S S LE LE LE LE GE LE GE(C) UGKO-CO (F) UGKO-P4 (E) UGKO-P4 (D) UGKO-COS S S S
42 the uterine endometrium (see Figu re 2-4 for examples). Histol ogical analysis of the skin grafts confirmed the visual observations. Au tografts were well organized and viable, with the presence of well defined keratin, ep idermis and dermis (Figure 2-5A to C). Allograft survival, in contrast, depended upon treatment and, when present, allograft displayed signs of necrosis. For ewes receivi ng corn oil vehicle allografts from 4 of 4 control ewes and 2 of 3 UGKO ewes had been adsorbed when examined 30 days after grafting; traces of wool were still in the uterus but the skin tissue was gone (Figure 2-4A and 2-4B). In the third UGKO ewe treated with vehicle, the allograf t was present (Figure 2-4C). The pattern of graft su rvival was altered by progesterone treatment. In this case allografts were present in 8 of 8 control ew es and, 8 of 8 UGKO ewes (Figures 2-4D, E and F). The gross appearance of surviving a llografts was often n ecrotic, however, with graft appearing brown and having a soft c onsistency. Histological examination of surviving grafts demonstrated that grafts we re disorganized, lacked identifiable keratin and epidermis and were characterized by abunda nt infiltration of leukocytes (Figures 25D, E and F). Total Protein Content in the Uterine Lumen Total uterine protein cont ent of the uterine lumen was greater for UGKO ewes than control ewes (p<0.001). For both gr oups, total protein content was higher for progesterone-treated ewes than for ewes trea ted with corn oil vehicle (p<0.01), but there was an interaction between ewe type and treatment (p < 0.01) that reflects the fact that total protein content was incr eased by progesterone treatment to a greater extent for UGKO ewes than control ewes (Figure 2-6).
43 Figure 2-4. Gross appearance of surviving autogr afts and allografts placed into the uterus of control (A, D, E) and uterine gl and knockout (UGKO) ewes (B, C, F) 30 days after surgery. Ewes were treat ed with corn oil (CO) (A-C) or progesterone (P4) (D-F). Note that au tografts were presen t in all ewes and that allografts had been completely reab sorbed in all ewes treated with corn oil [note the wool (w) in panels A and B; the tissue had been completely reabsorbed] except for one UGKO ewe (C). In contrast to the situation in ewes treated with corn oil, all allograf ts were present in ewes treated with progesterone; this was true for contro l (D and E) and UGKO ewes (F). A F D C Ew w A F D C Ew w
44 Figure 2-5. Histology of autograf ts (A-C) and allografts (D-F ) 30 days after grafting into the uterus of control and uterine gl and knockout (UGKO) ewes. Ewes were treated with corn oil (CO) or progester one (P4). Sections were stained with hematoxylin and eosin. Panels AC show well organized skin tissue (autografts) with presence of keratin (K), epidermis (E) and dermis (D). Panels D-F show allografts that were undergoing degeneration; note the lack of keratin and epidermis (A) Control-CO (E) Control-P4 (D) UGKO-CO (C) UGKO-P4 (B) UGKO-CO (F) UGKO-P4 200 mD D D D D D E E E KK K (A) Control-CO (E) Control-P4 (D) UGKO-CO (C) UGKO-P4 (B) UGKO-CO (F) UGKO-P4 200 m 200 mD D D D D D E E E KK K
45 Presence of OvUS in Uterine Fluid A representative immunoblot fo r the presence of OvUS in uterine fluid is shown in Figure 2-7 while a summary of the incidence of OvUS in uterine fluid is presented in Table 1. Immunoreactive OvUS was not detected in uterine fluids or flushings of control or UGKO ewes treated with corn oil vehicle. However, a single immunoreactive band at a molecular weight of 55,000 -57,000 was detected in uterine fluids from all control and UGKO ewes treated with progesterone. The immunoreactive bands seen using antiOvUS were not visible for ne gative control reactions in which culture medium replaced primary antibody (data not shown). Immunochemical Localization of OvUS A summary of the presence of OvUS in ut erine tissue sections is presented in Table 2-1. Immunoreactive OvUS was not de tected in any sections of uterine endometrium from corn oil-treated control (F igure 2-8B) or UGKO ewes (Figure 2-8C). Immunoreactive OvUS was observed, however, in all endometrial sections from progesterone-treated ewes, whether from cont rol (Figure 2-8D) or UGKO (Figures 2-8E and 2-8F). Immunoreactive OvUS was not detect ed in sections of the endometrium used as negative controls (Figure 2-8A). The protein was immu nolocalized to the glandular epithelium (Figures 2-8D and 2-8E) and in so me areas of the luminal epithelium (Figure 2-8F). No positive reaction was detected in areas of the stroma. Immunolocalization of CD45R+ Lymphocytes Regardless of treatment, CD45R+ cells in sections of the endometrium where autografts were present were mainly located in the luminal and glandular epithelium of control and UGKO ewes treated either with corn oil vehicle or progesterone. Some CD45R+ cells were localized in the stroma area (F igures 2-8G and 2-8H respectively).
46 Figure 2-6. Total protein recovered from ut erine fluid of control and uterine gland knockout (UGKO) ewes treated with corn oil vehicle (CO) or progesterone (P4) for a 60 days period. Data represents means Â± SEM. Total protein was affected by ewe type (UGKO vs control; p<0.001), progesterone treatment (p<0.01) and the interaction of ewe type with progesterone treatment (p<0.01). Control-COControl-P4UGKO-COUGKO-P4 Total uterine protein recovered (mg) 0 200 400 600 800 1000 1200
47 Figure 2-7. Representative west ern blot for detection of ovine uterine serpin (OvUS) in uterine fluid or flushings collected from control and uterine gland knockout (UGKO) ewes treated with corn oil (CO) or progesterone (P4) after 60 days. Purified OvUS and ovalbumin (OVA) were used as positive and negative control proteins respectively. OvUS OvUS OVAL OVAL Control CO Control CO UGKO CO UGKO CO UGKO P4 UGKO P4 UGKO P4 UGKO P4 Control P4 Control P4 Control P4 Control P4210 kDa 131 kDa 89 kDa 41.3 kDa 31.8 kDa 18.1 kDa 7.1 kDa 55-57 kDa 210 kDa 131 kDa 89 kDa 41.3 kDa 31.8 kDa 18.1 kDa 7.1 kDa 55-57 kDa
48 Positive cells were also detected in the lu minal and glandular epithelium of the uterine endometrium where allografts were present of corn oil-treated ewes (Figure 2-8I) and in hormone-treated ewes but abundant infiltration of cells was observed into the stroma was also apparent (Figures 2-8K and L respectively). Few immunoreactive CD45R+ cells were detected in the luminal epithelium and stroma of UGKO corn oil-treated ewes (Figure 2-8J). The density of CD45R+ cells was estimated by subjective scoring of each section of endometrium examined Â– results are shown in Figure 2-9. For luminal epithelium, density of CD45R+ cells was lower for UGKO ewes than for control ewes (p<0.01). In both types of ewes, the presence of allograf ts in the uterus produced an increase of CD45R+ cells in the luminal epithelium (p<0.01). There was a type x treatment x graft interaction (p=0.08). In particular, the pres ence of an allograft caused an increase in numbers of CD45R+ cells for control ewes treated with corn oil. Progesterone blocked this increase. In the UGKO ewes in cont rast, the increase in numbers of CD45R+ cells caused by the allograft was small and progest erone caused an increase in numbers of CD45R+ cells in both uterine horns. Among control ewes, numbers of CD45R+ cells in the glandular epithelium were higher in the uterine horn with the allograf t than for the horn bearing the autograft (p<0.05) and the difference between horns contai ning allografts and au tografts tended to be reduced in the progesterone-treated ewes (treatment x graft; p=0.07). For glandular epithelium, density of CD45R+ cells in the progesterone-treated groups was lower for UGKO ewes than for control ewes (p=0.06) . For the UGKO ewes treated with
49 Figure 2-8. Immunolocalizati on of OvUS (A-F) and CD45R+ cells (G-L) in endometrium from control and uterin e gland knockout (UGKO) ewes treated with corn oil vehicle (CO) or proge sterone (P4) for 60 days. Panel A represents a negative control for Ov US. Panels B and C the lack of immunoreactive OvUS in endometrium from control (B) and UGKO ewes treated with corn oil. Panels D, E and F illustrate detection of immunoreactive OvUS in the glandular (D and E) and luminal (F) epithelium of endometrium from cont rol (D) and UGKO (E and F) ewes treated with progesterone. Panels G-H represent immunoreactive CD45R+ cells in the endometrium on the side of the autogr aft for control (G) and UGKO (H) ewes treated with progesterone. Pane ls I-L represent immunoreactive CD45R+ cells in the endometrium on the side of the allograft of control (I) and UGKO (J) ewes treated with th e vehicle; and for UGKO (K) and control (L) ewes treated with progesterone. AB CD F G J H I L K 200 m 200 m E
50 Figure 2-9. Density of CD45R+ cells in different areas of the uterine endometrium according to the presence of autografts (black bars) and allografts (grey bars) for control and uterine gland knockout (UGKO) ewes treated with corn oil vehicle (CO) or progesterone (P4) for 60 days. Data represents least square means Â± SEM. In the luminal epithelium, the population of CD45R+ cells was affected by type (UGKO vs control; p<0.001), graft (allograft vs autograft; p<0.01) and the interactions of type by treatment (p=0.07) and type x treatment x graft (p=0.08). Am ong control ewes, numbers of CD45R+ cells in the glandular epithelium was affect ed by graft (p<0.05); treatment x graft was p=0.07. For progesterone-treated groups, density of CD45R+ cells in the glandular epithelium was affected by type (p=0.06). There were no significant effects on numbers of CD45R+ cells in stroma. Control-COControl-P4UGKO-COUGKO-P4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Density of CD45R + cells 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Luminal Epithelium Glandular Epithelium Stroma
51 progesterone, moreover, there was no difference in density of CD45R+ cells between horns with autografts or allografts. In the stroma, there was an increase of CD45R+ cells regardless of the type, but it was not significant. Discussion Results from this experiment confirmed th at progesterone delaye d the rejection of allogeneic tissues placed into the uter ine lumen. An unexpected finding was that prolonged progesterone treatment was cap able of inducing development and differentiation of endometrial glands into f unctional cells capable of OvUS secretion. This effect of progesterone may have abrogated the UGKO phenotype and allowed progesterone to maintain skin graft surviv al through induction of OvUS synthesis or some other mechanism. The induction of functional endometrial glands made it impossible to answer the question posed at the beginning of the study, i.e. whether progesterone inhibits uterin e immune function through m echanisms independent of induction of OvUS synthesis. Nonetheless, these results provide some novel insights into uterine biology including the c onclusion that the adult uterus retains the ability to form endometrial glands and that the developmen t processes causing differentiation of these cells into endometrial glands is under control of progesterone. Results also suggest the involvement of progesterone and, possibly endom etrial glands in homing of lymphocytes to the endometrium. Other studies have shown that administrati on of progesterone at doses ranging from 50 to 200 mg/day maintain the presence of allo geneic and xenogeneic tissues in the sheep uterus (Hansen et al., 1986; Ma jewski and Hansen, 2002). It is clear that progesterone is not preventing gr aft rejection per se , but rather delaying reje ction because surviving allografts were undergoing tissue disorganiz ation and neutrophil invasion. Similar
52 findings were reported by Hansen et al. (1986). This delay in tissue graft rejection is likely to reflect a decrease in function of eff ector lymphocytes or ot her leukocytes in the uterus. The role of T cells in rejection is illustrated in the present study by the observation that the endometrium in the uterine horn containing the allograft had increased accumulation of CD45R+ cells, which in the sheep uterus represent mostly T cells (Meussen et al., 1993). Indeed, proge sterone has been re ported to decrease lymphocyte numbers in the endometrium (G ottshall and Hansen, 1992) and progesterone blocked the increase in numbers of CD45R+ cells in luminal and glandular epithelium caused by the local presence of the allograft in control ew es from the present study. The concentrations of progesterone causing a delay in graft rejection (in this case, a peak of 17.9 ng/ml) are below the concentra tions of progesterone required to inhibit lymphocyte proliferation (Low and Hanse n, 1988; Monterroso and Hansen, 1993). The hypothesis that has been put forward to explain the progesterone-induced delay in rejection of tissue grafts in the uterus is that the immunosuppre ssive protein OvUS mediates the effects of proge sterone. A test of this hypothesis using the UGKO ewe was not possible, however, because progesterone induced appearance of endometrial glands in the UGKO ewe and these glands produced and s ecreted OvUS as indicated by results of immunohistochemistry and west ern blotting. Thus, the newl y-differentiated glandular epithelium in the UGKO ewe induced by pr ogesterone treatment was functional to respect to OvUS secretion. The actions of progesterone to induce new endometrial gland development and to cause these glands to diffe rentiate into functiona l glands capable of secretion of the prototypi cal progesterone-induced prot ein in the sheep was an unexpected finding that casts light on the developmental processes controlling uterine
53 differentiation and function. In ewes, the deve lopment of the glandular epithelium in the uterus is a postnatal event, starting between days 0 and 7, with bud formation from the luminal epithelium and proliferation into the stroma (Bartol et al., 1988ab; Taylor et al., 2000). Tubular structures that start to branch and coil are found by day 21 after birth (Taylor et al., 2000) and the endometrial ade nogenic process seems to be completed by day 56 of life when the histoarchitecture of th e uterus resembles the adult ewe (Taylor et al., 2000). The processes involved in gla nd formation, which include invasion of endometrial cells into the stroma and their proliferation and organization into branched and coiled glands, requires remodeling of the stroma extracellular matrix, epithelialepithelial and epithelial-stromal cell inter actions, and other cellular and biochemical events. Undoubtedly, the pro cess is under control of endocri ne and paracrine regulators (Gray et al., 2001c). Clearly, the existe nce of the UGKO phenotype indicates that neonatal exposure disrupts one or more of these regulatory systems to intercept the normal course of adenogenesis. What the pres ent results indicate is that the endometrium retains the ability to initiate and comple te glandular formation and that prolonged treatment with a high dose of progesterone rest ores one or more of the components of the adenogenesis pathway that was disrupted by ne onatal progestin treatment. There is evidence for the existence of stem cells for epithelial and stromal cells in endometrium from adult women (Chan et al., 2004; Cho et al., 2004). Present results suggest cells (probably luminal epith elial cells) with the capacity fo r differentiating into glandular epithelium persist in the UGKO ewe and can be activat ed by prolonged progestin treatment.
54 A previously undescribed characteristic of the UGKO phenotype in the absence of progesterone is the large reduc tion in the population of CD45R+ cells. Perhaps, it is low numbers of these cells that le d to one allograft being pres ent in a UGKO ewe treated with corn oil. Reduction in the number of CD45R+ in the UGKO ewe suggests that the lymphocyte homing mechanism is altered in th is phenotype. One molecule involved in leukocytes extravasation, glycosylation-depe ndent cell adhesion molecule 1 is expressed in the endometrial epithelium of the sheep (Spencer et al., 1999c) and there are undoubtedly others. Perhaps, neonatal progestin treatment disrupts th e normal pattern of expression of lymphocyte homing receptors. Alternately, progestin treatment may change endometrial function so that ly mphocyte egress from the endometrium is hastened. It can also not be excluded that changes in lymphocytes numbers may not represent a disruption in ly mphocyte trafficking in the UGKO ewe but rather in situ differentiation of lymphocytes in the glandul ar epithelium. Recombinase genes (RAG-1 and RAG-2) have been found expresse d in human decidual mononuclear cells (Hayakawa et al., 1994). The fact that progesterone treatment of UGKO ewes increased numbers of endometrial CD45R+ cells means that, as for its effects on endometrial gland morphogenesis, progesterone treatment of th e adult can reverse actions of neonatal progestin exposure. It is possible that the two effects of progesterone in the adult UGKO ewe, inducing gland formation and increas ing numbers of epithelial lymphocytes induction are unrelated. Altern atively, however, it is possible that the induction of new gland development is func tionally related to restor ation of homing of CD45R+ lymphocytes to the uterus, i.e., that migrati on of lymphocytes to the endometrium or their
55 retention in the endometrium depends upon the glandular epithelium. One possibility is that some CD45R+ lymphocytes enter the endometrium through homing to the glandular epithelium and then traverse to the luminal epithelium . In conclusion, results confirm that proge sterone delayed reject ion of allogeneic tissue placed into the uterine lumen and show ed that progesterone can reverse the effects of neonatal progestin expos ure on endometrial gland morphogenesis to induce appearance of functional glands capable of OvUS synthesi s. Among the differentiation events in the endometrium that are di srupted by neonatal progestin exposure are formation of the pool of CD45R+ lymphocytes resident in th e endometrial epithelium. This effect of progestin treatment, too, could be reversed by progesterone exposure during adulthood and it is possible than the i nduction of glandular development induced by progesterone is functionally related to the increase in lymphocyte numbers in the endometrium. Nonetheless, these results provide some novel insights into uterine biology including the conclusion that the adu lt uterus retains the ability to form endometrial glands and that the developmen t processes causing differentiation of these cells into endometrial glands are under influence of progest erone. Results also suggest the involvement of progesterone and, po ssibly endometrial glands, in homing of lymphocytes to the endometrium. Overall, results of the presen t study demonstrate the potential of the UGKO ewe as a tool to study uterine morphogenesis and immune function.
56 CHAPTER 3 GENERAL DISCUSSION The main goal of this dissertation was to understand the possible role of OvUS in the survival of the allogeneic conceptus during pregnancy in ewes. As stated before, the specific hypothesis for the experiment was to te st the role of OvUS in the rejection of allografts placed into the uterine lumen usi ng the UGKO ewe as a model. Results of the research described in this thesis could not adequately test this hypothesis because of the unexpected result that progesterone induced th e appearance of endometrial glands in the UGKO ewes and because this newly-different iated glandular epithelium was functionally active as indicated by immunolocalization of OvUS in uterine fluids and endometrial tissues of ewes treated with progesterone. Th is result is revealing because it implicates progesterone in the process of endometrial adenogenesis in adult ewes and the possible link between endometrial glands and homi ng of lymphocytes into the uterus. An unanswered question from this thesis a nd one that could be subject to additional research is how progesterone can initiate endometrial gland fo rmation in the adult uterus. Possible mechanisms for such an action of pr ogesterone on the adult ewe are illustrated in Figure 3-1. As a result of exposure to proge stin of lambs at bi rth, the adult uterine endometrium of UGKO ewes can have a ruffl ed luminal epithelium and compact stroma with no glands, small glandular invagina tions, or occasional cyst and gland-like structures in the stoma (Gray et al., 2000a b; 2001a). Progesterone could induce the reinitiation of the glandular epithelium by either inducing invagination of the luminal epithelium or by causing further differentiation of the cyst-like structures in the stroma.
57 Figure 3-1. Proposed model for the process of endometrial gland formation in the uterus of the adult ewe. In the neonate, ad enogenesis is initiated with shallow invaginations in the luminal epithe lium (LE). It is proposed that progesterone (P4) can initiat e this process in adult ewes (1). In addition, progesterone may act to induce gland formation causing differentiation of cyst and gland-like structures (CG-LS) pr esent in the stroma (S) (2) or from stem cells (SC) resident in the luminal epithelium and stroma (3). GE LE S (1) (1) (2)P4 (3) SC CG-LS ? ?
58 Of the two possibilities, the firs t is more likely because of previous data suggesting that glands form by invaginations of the lumi nal epithelium (Wiley, 1987; Bartol et al., 1988ab; Taylor et al., 2000). Th ere is also a possibility that the actions of progesterone on the adenogenesis process are the result of actions on stem cells present in the endometrium. Such cells have been detected in epithelial and stroma l cells of the human endometrium (Chan et al., 2004; Cho et al., 20 04). Progesterone may also have induced the release of growth factors from keratinocytes present in the skin graft placed into the uterus which could have promoted the devel opment of endometrial glands. An important question to address is whether the ability of progesterone to induce the formation of new glands is a phenomenon limited to UGKO ewes , which have abnormal gland formation, or whether if it also occurs in normal ewes. In addition to the effects of progester one on the induction of endometrial gland morphogenesis, results presented in this diss ertation also shown that progesterone could restore the population of lymphocytes present in the uterine endometrium and that this effect may be related with the developm ent of the glandular epithelium caused by progesterone. In the sheep ut erus, lymphocytes are localized in the luminal and glandular epithelium of the intercaruncular endometriu m (Lee et al., 1998; Gotshall and Hansen, 1992; Majewski et al., 2001). The lymphocyt e population that resides in the pregnant uterus is mainly composed of CD8+ CD45R+ TCR+ cells and these increase in number in the luminal epithelium during mid and late pregnancy (Lee et al., 1992; Meussen et al., 1993; Nasar et al., 2002). Results presente d in this disserta tion shown that CD45R+ cells were greatly reduced in the luminal epithelium of UGKO ewes treated with the vehicle and this population of cells seems to be restored in the luminal epithelium by
59 progesterone. Perhaps the restoration of lymphocyte numbers in the epithelium and development of endometrial glands are func tionally linked. The increase in the numbers of lymphocytes in the endometrium may have beneficial effects in the formation of new uterine glands due to the increase in cytoki nes released by these ce lls. A proposed model that illustrates the possible relationship between gland morphogenesis and homing of lymphocytes in the uterine endometrium in the adult ewe is shown in Figure 3-2. Initially, CD45R+ cells could migrate from the cap illary network to the luminal and glandular epithelium of the intercaruncu lar endometrium and from the glandular epithelium some cells migrate to the luminal epithelium. In the absence of endometrial glands, CD45R+ cells can only migrate from the capillary network to the luminal epithelium and the traffic of these cells fr om the glandular to the luminal epithelium would be eliminated. If so, progest erone could restore numbers of CD45R+ cells in the luminal epithelium by replacing the glandular ep ithelium. Another possibility is that UGKO ewes have a deficiency in chemokine s that attract lymphocytes because the glandular source for these molecules is eliminated. Given the inadequacy of the UGKO model, an important question is how the hypothesis about the effect of OvUS on allograf t rejection could be te sted in the future using the sheep as a model of experimenta tion. One possibility could be to produce UGKO ewes in which elimination of glands was irreversible, for example by extending the period of neonatal treatment with progest in. Another approach could be to test whether infusion of purified OvUS (purified fr om pregnant uterine fluid or recombinant OvUS) into the uterine horn would block allo graft rejection. The concentration of OvUS to be infused should be at the same concentrat ion used to inhibit lymp hocyte proliferation
60 Figure 3-2. Proposed model for homing lympho cytes in the luminal (LE) and glandular epithelium (GE) in the uterine endometr ium of adult ewes. In the initial process, CD45R+ lymphocytes (red circles) coul d migrate from the capillaries to the luminal (1) and glandular epithe lium (2). Perhaps lymphocytes resident in endometrial glands traffic to the lu minal epithelium. In the absence of endometrial glands, this r oute of delivery of CD45R+ lymphocytes to the luminal epithelium is blocked (3). GE LE CD45R+ (1) (2) ? (3) Cytokines ? (-) Cytokines ? (+)
61 in vitro (1 mg/ml) (Liu and Hansen, 1993). Th ere is also the possibility of using small interfering RNA molecules to inhibit or silen ce the OvUS gene and block the synthesis of the protein in the uterine endometrium. Although this novel t echnique has had good results in in-vitro models, it is still a challeng e for in vivo studies because of lack of a good delivery system that mediate efficient thei r cellular uptake and release (Wang et al., 2003). In summary, progesterone delayed skin graft rejection placed into the uterine lumen of UGKO ewes and this process was in conjunction with the development of functional endometrial glands th at were able to synthesize OvUS. Results also indicate a possible connection between pr ogesterone and gland forma tion in regulation of the lymphocyte population of the uterus.
62 LIST OF REFERENCES Aluvihare VR, Kallikourdis M, Betz AG. Regul atory T cells mediate maternal tolerance to the fetus. Nat Immunol 2004; 5: 266-271. Arck PC, Merali FS, Manuel J, Chaouat G, Clark DA. Stress-triggered abortion: inhibition of protective s uppression and promotion of tumor necrosis factor(TNF) release as mechanis m triggering resorptions in mi ce. Am J Reprod Immunol 1995; 33: 74-80. Asselin E, Johnson GA, Spencer TE, Bazer FW . Monocyte chemotactic protein-1 and -2 messenger ribonucleic acid in the ovine ut erus: regulation by pr egnancy, progesterone and interferon. Biol Reprod 2001; 64: 992-1000. Bartol FF, Johnson LL, Floyd JG, Wiley AA , Spencer TE, Buxton DF, Coleman DA. Neonatal exposure to progesterone and es tradiol alters ut erine morphology and luminal protein content in adult beef heifers. Therioge nology 1995; 43: 835-844. Bartol FF, Wiley AA, Coleman DA, Wo lfe DF, Riddell MG. Ovine uterine morphogenesis: effects of age and proge stin administration and withdrawal on neonatal endometrial development and DNA synthesis. J Anim Sci 1988b; 66: 30003009. Bartol FF, Wiley AA, Goodlett. Ovine uter ine morphogenesis: histochemical aspects of endometrial development in the fetus and neonate. J Anim Sci 1988a; 66: 1303-1313. Basset JM, OxborrowTJ, Smith ID, Thorburn GD. The concentration of progesterone in the peripheral plasma of the pregnant ewe. J Endocrinol 1969; 45:449-457. Bazer FW, Roberts RM, Basha SM, Zavy MT, Caton D, Barron DH. Method for obtaining ovine uterine secreti ons from unilaterally pregna nt ewes. J Anim Sci 1979; 49: 1522-1527. Bianchi DW, Lo DMY. Fetomaternal cellula r and plasma DNA trafficking: the yin and the yang. Ann N Y Acad Sci 2001; 995: 119-131. Billington WD, Bell SC. Fetal histocompatib ility antigens and maternal immune responses. Ciba Found Symp 1983; 96: 69-88.
63 Bradford M. A rapid and sensitive method fo r the quantification of microgram quantities of protein utilizing the prin ciple of protein-dye binding. Anal Biochem 1976; 72: 248-254. Carpenter KD, Gray CA, Bryan TM, Welsh Jr TH, Spencer TE. Estrogen and antiestrogen effects on neonatal uterine de velopment. Biol Reprod 2003c; 69: 708717. Carpenter KD, Gray CA, Noel S, Gertler A, Bazer FW, Spencer TE. Prolactin regulation of neonatal ovine uterine gland mor phogenesis. Endocrinology 2003b; 144: 110-120. Carpenter KD, Hayashi K, Spencer TE. Ovarian regulation of endometrial gland morphogenesis and activin-folli statin system in the neona tal ovine uterus. Biol Reprod 2003a; 69: 851-860. Casida LE, Warwick EJ. The necessity of the corpus luteum for maintenance of pregnancy in the ewe. J Anim Sci 1945; 4: 34-36. Chan RWS, Schwab KE, Gargett CE. Clonogeni city of human endomet rial epithelial and stromal cells. Biol Reprod 2004; 70: 1738-1750. Chaouat G, Assal Meliani A, Martal J, Raghupathy R, Elliot J, Mossmann T, Wegmann TG. IL-10 prevents naturally occurri ng fetal loss in the CBA x DBA/2 mating combination, and local defect in IL-10 produc tion in this aborti on-prone combination is corrected by in vi vo injection of IFN. J Immunol 1995; 152: 2411-2420. Chaouat G, Kolb JP, Wegmann TG, The muri ne placenta as an immunological barrier between the mother and the fetus. Immunol Rev 1983; 75: 31-60. Chaouat G, Menu E, Clark DA, Dy M, Minkow ski M, Wegmann TG. Control of fetal survival in CBA x DBA/2 mice by lymphokine therapy. J Reprod Fert 1990; 89: 447458. Cho NH, Park YK, Kim YT, Yang H, Kim SK. Lifetime expression of stem cell markers in the uterine endometrium. Fertil Steril 2004; 81: 403-407. Choi Youngsok, Johnson GA, Spencer TE, Ba zer FW. Pregnancy and interferon tau regulate major histocompatib ility complex class I and 2-microglobulin expression in the ovine uterus. Biol Reprod 2003; 68: 1703-1710. Clark DA, Chaouat G, Mogil R, Wegmann TG. Prevention of spontaneous abortion in DBA/2-mated CBA/J mice by GM-CSF involves CD8+ T cell-dependent suppression of natural effector cell cyto toxicity against trophoblast target cells. Cell Immunol 1994; 154: 143-152.
64 Croy BA, Chantakru S, Esadeg S, Ashkar AA, We i Q. Decidual natural killer cells: key regulators of placental development (a review). J Reprod Immunol 2002; 57: 151168. Croy BA, Esadeg S, Chantakru S, van den Heuvel M, Paffaro VA, He H, Black GP, Ashkar AA, Kiso Y, Zhang J. Update on pathways regulatin g the activation of uterine natural killer cells, their interactio ns with decidual spiral arteries and homing of their precursors to the uterus . J Reprod Immunol 2003; 59: 175-191. Davies J, Wimsatt W. Observation on the fine structure of the sheep placenta. Acta Anat 1966; 65: 182-223. Ehring GR, Kerschbaum HH, Eder C, Neben AL, Fanger CM, Khoury RM, Negulescu PA, Cahalan MD. A nongenomic mech anism for progesterone-mediated immunosuppression: inhibition of K+ channels, Ca+2 signaling, and gene expression in T lymphocytes. J Exp Med 1998; 1593-1602. Fallarino F, Grohmann U, Kwa ng WH, Orabona C, Calcinar o, Vacca C, Bianchi R, Belladonna ML, Fioretti MC, Alegre ML, Puccetti P. Modulation of tryptophan catabolism by regulatory T cells. Nat Immunol 2003, 4: 1206-1212. Fernandez N, Cooper J, Sprinks M, Mohamed A, Fiszer D, Kurpis z M, Dealtry G. A critical review of the role of the major histocompatibility complex in fertilization, preimplantation development and feto-mater nal interactions. Hum Reprod Update 1999; 5: 234-248. Fox A, Lee CS, Brandon MR, Meeusen ENT. E ffects of pregnancy within sheep uterine interplacentomal epithelium. Am J Reprod Immunol 1998; 40: 295-302. Fox A, Meeusen E. Sheep perfor in: identification and expression by T cells from pregnant sheep uterine epithelium. Vet Immunol Pathol 1999; 68:293-296. Gogolin-Ewens K, Lee CS, Mercer WR, Bra ndon MR. Site-directed differences in the immune response to the fetu s. Immunology 1989; 66: 312-37. Gotshall SL, Hansen PJ. Regulation of leukocyte subpopulations in the sheep endometrium by progesterone. Immunology 1992; 76: 636-641. Gray CA, Bartol FF, Tarlenton BJ, Wile y AA, Johnson GA, Bazer FW, Spencer TE. Developmental biology of uterine gl ands. Biol Reprod 2001c; 65: 1311-1323. Gray CA, Bartol FF, Taylor KM, Wiley AA, Ramsey WS, Ott TL, Bazer FW, Spencer TE. Ovine uterine gland knock-out model: effects of gland ablation on the estrous cycle. Biol Reprod 2000a; 62: 448-456.
65 Gray CA, Bazer FW, Spencer TE. Effects of neonatal progestin exposure on female reproductive tract structure and function in the adult ewe. Biol Reprod 2001a; 64: 797-804. Gray CA, Burghardt RC, Johnson GA, Bazer FW, Spencer TE. Evidence that absence of endometrial gland secretions in uterine gland knockout ewes compromises conceptus survival and elongation. Reproduction 2002; 124: 289-300. Gray CA, Stewart MD, Johnson GA, Spencer TE. Postpartum uterine involution in sheep in endometrial gene expre ssion. Reproduction 2003; 185-198. Gray CA, Taylor KM, Bazer FW, Spencer TE. Mechanisms regulating norgestomet inhibition of endometrial gland morphogene sis in the neonatal ovine uterus. Mol Reprod Dev 2000b; 57: 67-78. Gray CA, Taylor KM, Ramsey WS, Hill JR , Bazer FW, Bartol FF, Spencer TE. Endometrial glands are required for pr eimplantation conceptus elongation and survival. Biol Reprod 2001b; 64: 1608-1613. Grohmann U, Fallarino F, Pu ccetti P. Tolerance, DCs and tryptophan much ado about IDO. Trends Immunol 2003; 24: 242-248. Hansen PJ. Regulation of uterine immune function by progesterone-lessons from the sheep. J Reprod Immunol 1998; 40:63-79. Hansen PJ, Bazer FW, Segerson EC. Skin graf t survival in the uterine lumen of ewes treated with progesterone. Am J Re prod Immunol Microbiol 1986; 12: 48-54. Hansen PJ, Ing NH, Moffatt RJ, Baumbach GA, Saunders PTK, Bazer FW, Roberts RM. Biochemical characterization and biosynthesi s of the uterine milk proteins of the pregnant sheep uterus. Biol Reprod 1987; 36: 405-418. Hansen PJ, Newton GR. Binding of immunogl obulins to the major progesterone-induced proteins secreted by the sheep uterus . Arch Biochem Bi ophys 1988; 260: 208-217. Hansen PJ, Skopets B. Temporal relati onship between progesterone and uterine lymphocyte-inhibitory activity in ewes. Vet Rec 1992; 131: 371-372. Hayakawa S, Saito S, Nemoto N, Chishima F, Akiyama K, Shiraishi H, Hayakawa F, Karasaki-Suzuki M, Fujii KT, Ichijo M, Sakurai I, Satoh K. Expression of recombinase-activating genes (RAG-1 and 2) in human decidual m ononuclear cells. J Immunol 1994; 153 : 4934-4039. Hayashi K, Carpenter KD, Gray CA, Spencer TE. The activin-follistatin system in the neonatal ovine uterus. Biol Reprod 2003; 69: 843-850.
66 Hayashi K, Carpenter KD, Spencer TE. Neonatal estrogen exposure disrupts uterine development in the postnatal sh eep. Endocrinology 2004; 145: 3247-3257. Hunt JS, Vassmer D, Ferguson TA, Miller L. Fas ligand is positioned in mouse uterus and placenta to prevent trafficking of ac tivated leukocytes between the mother and the conceptus. J Im munol 1997; 158: 4122-4128. Hunziker RD, Gambel P, Wegmann TG. Placenta as a selective barrier to cellular traffic. J Immunol 1984; 133: 667-671. Ing NH, Francis H, McDonnell JJ, Amann JF, Roberts RM. Progesterone induction of the uterine milk proteins: major secretor y proteins of sheep endometrium. Biol Reprod 1989; 41: 643-654. Ing NH, Roberts RM. The major progesteron e-modulated proteins secreted into the sheep uterus are members of the serpin supe rfamily of serine pr otease inhibitors. J Biol Chem 1989; 264: 3372-3379. Irving JA, Pike RN, Lesk AM, Whisstockn JC . Phylogeny of the serpin superfamily: implications of patterns of amino acid conservation for structure and function. Genome Research 2000; 10: 1845-1864. Jiang SP, Vacchio MS. Multiple mechanism of peripheral T cell tolerance to the fetal Â“allograft.Â” J Immunol 1998; 160: 3086-3090. Kiger N, Chaouat G, Kolb JP, Wegmann TG, Guenet JL. Immunogenetic studies of spontaneous abortion in mice. Preimmunizati on of females with allogeneic cells. J Immunol 1985; 134: 2966-2970. Kinsky R, Delage G, Rosin M, Thang N, Ho ffman M, Chaouat G. A murine model of NK cell mediated resorption. Am J Reprod Immunol 1990; 24: 195-205. King A, Gardner L, Loke YW. Evaluati on of oestrogen and progesterone receptor expression in uterine mucosal lympho cytes. Hum Reprod 1996; 11: 1079-1082. King A, Kalra P, Loke YW. Human trophoblas t cell resistance to decidual NK lysis is due to lack of NK target struct ure. Cell Immunol 1990; 127: 230-237. Kwon Hyukjung, Wu Guoyao, Meininger CJ, Ba zer FW, Spencer TE. Developmental changes in nitric oxide synthesis in th e ovine placenta. Biol Reprod 2004; 70: 679686. Le Boutellier P, Mallet ValÃ©rie. HL A-G and pregnancy. Rev Reprod 1997; 2: 7-13. Lee CS, Gogolin-Ewens K, Brandon MR. Identification of a unique lymphocyte subpopulation in the sheep uterus . Immunology 1988; 63: 157-164.
67 Lee CS, Meeusen E, Gogolin-Ewens K, Brandon MR. Quantitative and qualitative changes in the intraepithelial lymphocyte popul ation in the uterus of nonpregnant and pregnant sheep. Am J Reprod Immunol 1992; 28: 90-96. Leslie MV, Hansen PJ. Progest erone-regulated secre tion of the serpin-like proteins of the ovine and bovine uterus. Steroids 1991; 56: 589-597. Leslie MV, Hansen PJ, Newton GR. Uterine se cretions of the cow contain proteins that are immunochemically related to the ma jor progesterone-induced proteins of the sheep uterus. Domest Anim Endocrinol 1990; 7:517-526. Lewis GS. Role of ovarian progesterone and potential role of prostaglandin F2 and prostaglandin E2 in modulating the uterine response to infectious bacteria in postpartum ewes. J Anim Sci 2003; 81: 285-293. Lin H, Mosmann TR, Guilbert L, Tuntipopipat S, Wegmann TG. Synt hesis of T helper 2-type cytokines at the maternal-fetal interface. J Immunol 1993; 151: 4562-4573. Liu WJ, Gottshall SL, Hansen PJ. Incr eased expression of cell surface markers on endometrial T-cell receptor+ intraepithelial lymphocytes by the local presence of the sheep conceptus. Am J Reprod Immunol 1997; 37: 199-205. Liu WJ, Hansen PJ. Effect of progesterone -induced serpin-like proteins of the sheep endometrium on natural-killer cell activity in sheep and mice. Biol Reprod 1993; 49: 1008-1014. Liu WJ, Hansen PJ. Progesterone-induced secr etion of dipeptidyl peptidase-IV (cluster differentiation antigen-26) by the uterine endometrium of the ewe and cow that costimulates lymphocyte prolifera tion. Endocrinology 1995; 136: 779-787. Liu WJ, Peltier MR, Hansen PJ. Binding of ovi ne uterine serpin to lymphocytes. Am J Reprod Immunol 1999; 41: 428-432. Ljunggreen HG, Karre K. In search of Â“missing selfÂ”: MHC molecules and NK cell recognition. Immunol Today 1990; 11: 237-244. Lo YMD, Lau TK, Chan LYS, Leung TN, Chang AMZ. Quantitative analysis of the bidirectional fetomaternal transfer of nucle ated cells and plasma DNA. Clin Chem 2000; 46: 1301-1309. Low BG, Hansen PJ. Actions of steroids and prostaglandins secreted by the placenta and uterus of the cow and ewe on lymphocyte proliferation in vitro. Am J Reprod Immunol Microbiol 1988; 18: 71-75. Majewski AC, Hansen PJ. Progesterone inhib its rejection of xenogeneic transplants in the sheep uterus. Horm Res 2002; 58:128-135.
68 Majewski AC, Tekin , Hansen PJ. Local versus systemic control of numbers of endometrial T cells during pregnanc y in sheep. Immunology 2001; 102: 317-322. Makrigiannakis A, Zoumakis E, Kalantaridou S, Coutifaris C, Margioris AN, Coukos G, Rice KC, Gravanis A, Chrousos GP. Co rticotropin-releasing hormone promotes blastocyst implantation and early maternal tolerance. Nat Immunol 2001; 2: 10181024. Malathy PV, Imakawa K, Simmen RC, R oberts RM. Molecular cloning of the uteroferrin-associated protein, a major pr ogesterone-induced serpin secreted by the porcine uterus, and the expression of its mRNA during pregnancy. Mol Endocrinol 1990; 4: 428-440. Mansour I, Reznikoff-Etievant MF, Netter A. No evidence for the expression of the progesterone receptor on peripheral blood lymphocytes during pregnancy. Hum Reprod 1994; 9: 1546-1549. Mao D, Wu X, Deppong C, Friend LD, Doleck i G, Nelson DM, Molina H. Negligible role of antibodies and C5 in pregnancy lo ss associated exclusively with C3-dependent mechanisms through complement alterna tive pathway. Immunity 2003; 19: 813-822. Mathialagan N, Hansen TR. Pepsin-inhibitory activity of the uterine serpins. Proc Natl Acad Sci USA 1996; 93: 13653-13658. McFarlane JR, Foulds LM, OÂ’Connor AE, Ph illips DJ, Jenkin G, Hearn MTW, Kretser DM. Uterine milk protein, a novel activin g-binding protein, is present in ovine allantoic fluid. Endoc rinology 1999; 140: 4745-4752. Medawar PB. Some immunological and endoc rinological problems raised by evolution of viviparity in vertebrates. Symp Soc Exp Biol 1953; 7: 320-328. Mellor AL, Baban B, Chandler P, Marshall B, Jhaver K, Hansen A, Koni PA, Iwashima M, Munn DH. Induced indoleamine 2,3 dioxygenase expression in dendritic cells subsets suppress T cell clonal expans ion. J Immunol 2003; 171: 1652-1655. Mellor AL, Munn DH. Extingui shing maternal immune res ponses during pregnancy: implications for immunosuppression. Semin Immunol 2001; 13: 213-218. Mellor AL, Sivakumar J, Chandler P, Smith K, Molina H, Mao D, Munn DH. Prevention of T cell-driven complement activation and inflammation by tryptophan catabolism during pregnancy. Nat I mmunol 2001; 2: 64-68.
69 Meussen E, Fox Annette, Brandon M, Lee CS. Activation of uterine epithelial T cell receptor-positive lymphocyte during pr egnancy. Eur J Immunol 1993; 23: 11121117. Moffatt J, Bazer FW, Hansen PJ, Chun PW, Roberts RM. Purification, secretion and immunocytochemical localization of the ut erine milk proteins, major progesteroneinduced proteins in uterine secretions of the sheep. Biol Reprod 1987; 36: 419-430. Monterroso VH, Hansen PJ. Regulation of bovine and ovine lymphocyte proliferation by progesterone: modulation by steroid recepto r antagonists and physiological status. Acta Endocrinol 1993; 129: 532-535. Moriyama I, Sugawa T. Progesterone fac ilitates implantation of xenogeneic cultured cells in hamster uterus. Nat New Biol 1972; 236: 150-152. Munn DH, Shafizadeh E, Attwood JT, Bondarev I, Pashine A, Mellor AL. Inhibition of T cell proliferation by macrophage trypt ophan catabolism. J Exp Med 1999; 189: 1363-1372. Munn DH, Zhou M, Attwood JT, Bondarev I, C onway SJ, Marshall B, Brown C, Mellor AL. Prevention of allogeneic fetal rejec tion by tryptophan catabolism. Science 1998; 281: 1191-1193. Nasar A, Rahman A, Meeusen E, Lee CS. Peri-partum changes in the intraepithelial lymphocyte population of sheep interplacentomal endometrium. Am J Reprod Immunol 2002; 47: 132-141. Nelson JL. Microchimerism: incidental byprodu ct of pregnancy or active participant in human health?. Trends Mol Med 2002; 8: 109-113. Newton GR, Hansen PJ, Bazer FW, Leslie MV , Stephenson DC, Low BG. Presence of major progesterone-induced protei ns of the sheep endometrium in fetal fluids. Biol Reprod 1989; 40: 417-424. Noel S, Herman A, Johnson GA, Gray CA, Stewart MD, Bazer FW, Gertler A, Spencer TE. Ovine placental lactogen specifically bi nds to endometrial glands to the ovine uterus. Biol Reprod 2003; 68: 772-780. Olea-Serrano N, Devleeschouwer N, Leclercq G, Heuson JC. Assay for estrogen and progesterone receptors of breast cancer cell lines in monolayer culture. Eur J Cancer Clin Oncol 1985; 21: 965-973. Parhar RS, Yagel S, Lala PK. PGE2-m ediated immunosuppression by first trimester human decidual cells blocks activation of maternal leukocytes in the decidua with potential anti-trophoblast activit y. Cell Immunol 1989; 120: 61-74.
70 Peltier MR, Grant TR, Hansen PJ. Distinct phys ical and structural properties of the ovine uterine serpin. Biochim Bi ophys Acta 2000a; 1479: 37-51. Peltier MR, Liu WJ, Hansen PJ. Regulati on of lymphocyte proliferation by uterine serpin: interleukin-2 mRNA production, CD 25 expression and responsiveness to interleukin-2. Proc Natl Acad Sci USA 2000b; 223: 75-81. Peltier MR, Raley LC, Liberles DA, Benner SA, Hansen PJ. Evolutionary history of the uterine serpins. J Exp Zool 2000c; 288: 165-174. Perry JS. The mammalian fetal membrane s. J Reprod Fertil 1981; 62: 321-335. Piccini MP, Scaletti C, Vultaggi o A, Maggi E, Romagnani S. Defective production of LIF, M-CSF and Th2-type cytokines by T cells at fetomaternal interface is associated with pregnancy loss. J Reprod Immunol 2001; 52: 35-43. Raghupathy R. Pregnancy: success and failu re within the Th1/Th2/Th3 paradigm. Semin Immunol 2001; 13: 219-227. Raghupathy R, Singh B, Barrington-Leigh J, Wegmann TC. The ontogeny and turnover kinetics of paternal H-2K antigenic determ inants on the allogeneic murine placenta. J Immunol 1981; 127: 2074-2079. Ramadan AA, Johnson III GL, Lewis GS. Regul ation of uterine im mune function during the estrous cycle and in response to infec tious bacteria in sheep. J Anim Sci 1997; 75: 1621-1632. Reimers TJ, Dziuk PJ. The survival of intrau terine skin autografts and allografts in sheep. J Reprod Fertil 1974; 38: 465-467. Renegar RH, Bazer FW, Roberts MR. Placental transport and distribution of uteroferrin in the fetal pig. Biol Reprod 1982; 27: 1247-1260. Saito S. Cytokine network at the feto-m aternal interface. J Reprod Immunol 2000; 47: 87-103. Schust DJ, Anderson DJ, Hill JA. Progesterone-induced immunosuppression is not mediated through the progesterone receptor. Hum Reprod 1996; 11: 980-985. Seals RC, Wulster-Radcliffe MC, Lewis GS. Modulation of the uterine response to infectious bacteria in postpartum ew es. Am J Reprod Immunol 2002; 47: 57-63. Seals RC, Wulster-Radcliffe MC, Lewis GS. Ut erine response to infectious bacteria in estrous cycle ewes. Am J Re prod Immunol 2003; 49: 269-278.
71 Searle RF, Sellens MH, Elson J, Jenkinson EJ , Billington WD. Detection of alloantigens during preimplantation development and early trophoblast differentiation in the mouse by immunoperoxidase labeli ng. J Exp Med 1976; 143: 348-359. Segerson EC, Moffatt RJ, Bazer FW, Roberts RM. Suppression of phytohegglutininstimulated lymphocyte blastogenesis by ovi ne uterine milk protein. Biol Reprod 1984; 30: 1175-1186. Silverman GA, Bird PI, Carrell RW, Church FC, Coughlin PB, Gettins PGW, Irving JA, Lomas DA, Luke CJ, Moyer RW, Pemberton PA, Remold-OÂ’Donnell E, Salvesen GS, Travis J, Whisstock JC. The serpins ar e an expanding superfam ily of structurally similar but functionally diverse prot eins. J Biol Chem 2001; 276: 33293-33296. Szekeres-Bartho J, Barakonyi A, Par G, Polgar B, Palkovics T, Szereday L. Progesterone as immunom odulatory molecule. Int Immunopharmacol 2001; 1: 10371048. Szekeres-Bartho J, Hadnagy J, Pacsa AS. The suppressive effect of progesterone on lymphocyte cytotoxicity: unique progesterone sensitivity of pregnancy lymphocytes. J Reprod Immunol 1985; 7: 121-128. Skopets B, Hansen PJ. Identification of the predominant proteins in uterine fluids of unilaterally pregnant ewes th at inhibits lymphocyte pro liferation. Biol Reprod 1993; 49: 997-1007. Skopets B, Liu WJ, Hansen PJ. Effects of endometrial serpin-lik e proteins on immune responses in sheep. Am J Reprod Immunol 1995; 33: 86-93. Somerset DA, Zheng Y, Kilby MD, Sans om DM, Drayson MT. Normal human pregnancy is associated with an el evation in the immune suppressive CD25+ CD4+ regulatory T-cell subset. Immunol 2004; 112: 38-43. Spencer TE, Bartol FF, Bazer FW, Johns on GA, Joyce MM. Identification and characterization of glycosyl ation-dependent cell adhesi on molecule 1-like protein expression in the ovine uterus. Biol Reprod 1999c; 60: 241-250. Spencer TE, Gray A, Johnson GA, Taylor KM , Gertler A, Gootwine E, Ott TL, Bazer FW. Effects of recombinant ovine inte rferon tau, placental lactogen, and growth hormone on the ovine uterus. Biol Reprod 1999a; 61: 1409-1418. Spencer TE, Stagg AG, Joyce MM, Jenster G, Wood CG, Bazer FW, Wiley AA, Bartol FF. Discovery and characterization of endometrial epithelial messenger ribonucleic acids using the ovine uterine gland k nockout model. Endocrinology 1999b; 140: 4070-4080.
72 Spencer TE, Wiley AA, Bartol FF. Neonatal age and period of estrogen exposure affect porcine uterine growth, mor phogenesis and protein synthesi s. Biol Reprod 1993; 48: 741-751. Stabenfeldt GH, Drost M, Franti CE. Periphe ral plasma progesterone levels in the ewe during pregnancy and parturition. Endocrinology 1971; 90: 144-149. Staples LD, Binns RM, Heap RB. Influe nce of certain steroids on lymphocyte transformation in sheep and goats studied in vitro. J Endocrinol 1983; 98: 55-69. Stephenson DC, Hansen PJ. Induction by pr ogesterone of immunosuppressive activity in uterine secretions of ovariectomized ewes. Endocrinology 1990; 126: 3168-3178. Stephenson DC, Hansen PJ, Newton GR, Bazer FW, Low BG. Inhi bition of lymphocyte proliferation by uterine fluid from the pregnant ewe. Biol Reprod 1989a; 41: 10631075. Stephenson DC, Leslie MV, Low BG, Newton GR, Hansen PJ, Bazer FW. Secretion of the major progesterone-induced proteins of the sheep uterus by caruncular and intercaruncular endometrium of the pregna nt ewe from days 20-140 of gestation. Domest Anim Endoc rinol 1989b; 6: 349-362. Stewart MD, Johnson GA, Gray CA, Burghard t RC, Schuler LA, Joyce MM, Bazer FW, Spencer TE. Prolactin receptor and uterin e milk protein expression in the ovine endometrium during the estrous cycle a nd pregnancy. Biol Reprod 2000; 62: 17791789. Sunderland CA, Redmann WG, Stirrat GM. HL A A, B, C, antigens are expressed on the nonvillous trophoblast of the human placen ta. J Immunol 1981; 127: 2614-2615. Tafuri A, Alferink J, MÃ¶ller P, GÃ¼nter JH , Arnold B. T cell awareness of paternal alloantigens during pregnancy. Science 1995; 270: 630-633. Tarlenton BJ, Braden TD, Wiley AA, Bartol FF. Estrogen-induced disruption of neonatal porcine uterine development alters adult uterine function. Bi ol Reprod 2003; 68: 1387-1393. Taylor KM, Chen C, Gray CA, Bazer FW, Spencer TE. Expression of messenger ribonucleic acids for fibroblast growth factor s 7 and 10, hepatocyte growth factor, and insulin-like growth factors a nd their receptors in the ne onatal ovine uterus. Biol Reprod 2001; 64: 1236-1246. Taylor KM, Gray CA, Joyce MM, Stewart MD , Bazer FW, Spencer TE. Neonatal ovine uterine development involves alterations in expression of receptors for estrogen, progesterone and prolactin. Biol Reprod 2000; 63: 1192-1204.
73 Tekin , Hansen PJ. Natural killer-like cells in the sheep: functional characterization and regulation by pregnancy-associated pr oteins. Exp Biol Med 2002; 227: 803-811. Tekin , Hansen PJ. Lymphocyte-mediated lysi s of sheep chorion: susceptibility of chorionic cells to third-party and matern al cytotoxic lymphocytes and presence of cells in the endometrium exhibiting cytotoxic ity toward natural-killer cells targets. Theriogenology 2003; 59: 787-800. Tekin , Hansen PJ. Regulation of number of macrophages in the endometrium of the sheep by systemic effects of pregnancy, local presence of the conceptus, and progesterone. Am J Reprod Immunol 2004; 51: 56-62. Tekin , Padua MB, Newton GR, Hansen PJ. Identification and cloning of caprine uterine serpin. Mol Repr od Dev 2004; in press. Van Gent D, Sharp P, Morgan K, Kalsheke r N. Serpins: structure, function and molecular evolution. Int J Bioc hem Cell Biol 2003; 35: 1536-1547. Van Voorhis BJ, Anderson DJ, Hill JA. The effects of RU 486 on immune function and steroid-induced immunosuppr ession in vitro. J Clin Endocrinol Metab 1989; 69: 1195-1199. Wang QC, Nie QH, Feng ZH. RNA interferen ce: antiviral weapon and beyond. World J Gastroenterol 2003; 9: 1657-1661. Webb CG, Gall WE, Edelman GM. Synthesi s and distribution of H-2 antigens in preimplantation mouse embryos. J Exp Med 1977; 146: 923-932. Wiley AA, Bartol FF, Barron DH. Histogenesis of the ovine uterus. J Anim Sci 1987; 64: 1262-1269. Wood KJ, Sakaguchi S. Regul atory T cells in transplant ation tolerance. Nat Rev Immunol 2003; 199-210. Wooding FB. Structure and function of placenta l binucleate (Â‘giantÂ’) cells. Bibl Anat 1982; 22: 134-139. Xie S, Low BG, Nagel RJ, Kramer KK, Ant hony RV, Zoli AP, Beckers JF, Roberts RM. Identification of the major pregnancy-specifi c antigens of cattle and sheep as inactive members of the aspartic proteinase fa mily. Proc Natl Acad Sci USA 1991; 88: 10247-10251. Xu C, Mao D, Holers VM, Palanca B, Cheng AM, Molina H. A critical role for murine complement regulator Crry in fetomate rnal tolerance. Science 2000; 287: 498-501.
74 Zheng J, Li Y, Weiss AR, Bird IM, Ma gness RR. Expression of endothelial and inducible nitric oxide synt hases and nitric oxide produc tion in ovine placental and uterine tissue during late pre gnancy. Placenta 2000; 21: 516-524. Zhou M, Mellor AL. Expanded cohorts of maternal CD8+ T-cells specific for paternal MHC class I accumulate during pregnanc y. J Reprod Immunol 1998; 40: 47-62. Zuckermann FA, Head JR. Murine trophoblast resists cell-mediated lysis. J Immunol 1987; 139: 2856-2864.
75 BIOGRAPHICAL SKETCH Maria Beatriz Padua was born in 1968 in Ca racas, Venezuela. She graduated from El Carmelo High School in the same city in 1985 and enrolled the next year in the School of Agronomy at the University Centroccident al Lisandro Alvarado in Barquisimeto, Lara State, Venezuela, where she received her Bachelor of Science degree in agricultural engineering in 1994. In 1995, she worked as a sales representative for the veterinary health supply firm, Grupo Catalina C.A., in Ca racas, Venezuela. From 1996 to 2000, she was a pharmacist assistant for Celbefar C.A. Farmacia El Roble in the same city. She enrolled in an English language program at the University of Fl orida in 2001 and she enrolled in the Animal Molecular and Cell Biology Graduate Program of the Department of Animal Sciences at the Un iversity of Florida under the supervision of Dr. Peter J. Hansen in April, 2002. She is currently a Ma ster of Science candi date. Upon completion of her degree, she will pursue a Doctor of Philosophy degree at the University of Florida under Dr. Hansen.