Cells of the metrial gland: their relation to trophoblast cells and reproductive characteristics

Data on the origin, morphology and function of metrial gland cells are reviewed. Characteristic features of metrial gland cells are the availability of numerous eosinophilic granules lying near two round or oval nuclei and peripheral zone of the cytoplasm, generally devoid of organelles. This zone can generate pseudopodia-like projections. The notable peculiarity of metrial gland cells involves their ability to penetrate into blood vessels, to migrate towards the embryo, and to achieve the ectoplacental cone. The majority of metrial gland cells is accumulated in the decidua basalis zone where the tertiary trophoblast cells usually migrate. The metrial gland cells seem to constitute a cell population analogous to that of decidual cells. Data on the protective role of metrial gland cells are discussed. The metrial gland cells are proven to be polyploid. Polyploid nuclei are found both in mononucleate and binucleate cells. Acytokinetic mitosis is presumably a way leading to polyploidization of metrial gland cells.     

The structure and differentiation of granulated metrial gland cells of the pregnant mouse uterus

Summary A study was made with the light and electron microscopes of the granulated metrial gland cells of the decidua basalis of the pregnant mouse uterus, up to day 11 of pregnancy. The granulated metrial gland cells are large, up to 50 in diameter, mono- or binucleate and the glycogen rich cytoplasm typically contains many large glycoprotein granules which may be up to 5 in diameter. Morphological evidence is described in support of a lymphocyte-like cell being the precursor to the granulated metrial gland cell. This differentiation sequence is similar to that already proposed in the rat but differences between the ultrastructure of the mature metrial gland cells of rats and mice were noted.     

Glycogen-containing cells in the maternal and embryonic portions of the placenta in the rat and the common vole Microtus subarvalis

Differentiation sequences and further transfiguration of glycogen-rich cells during placenta development were investigated for the rat and field vole Microtus subarvalis (11-20 day gestation). The presence of glycogen is a characteristic feature of decidual cells located in the region of lateral sinusoids, as well as of metrial gland cells, secondary giant trophoblast cells and trophoblast cells in the connective zone of placenta. Glycogen-containing metrial gland cells and trophoblast cells of connective zone of placenta are found to underlie the layer of tertiary giant trophoblast cells that cover the wall of the central arteria. Thus, both maternal and embryo-derived glycogen-containing cells always accompany the tertiary giant trophoblast cells that penetrate deeply into the maternal part of placenta but do not contain glycogen. In the field vole placenta the cells of peripheral trophoblast subpopulation of the connective zone of placenta attaching to the decidua basalis are stained by PAS-reaction more intensely than deeply situated ones. These data, as well as other phenomena revealed here, show that maternal and trophoblastic cells attaching to each other in placenta contain, as a rule glycogen. Glycogen cells in rat placenta and trophoblast cells of peripheral subpopulation of connective zone of placenta are similar in many respects. In this connection, a possible protective role of glycogen-containing cells, that probably favour the co-existence of maternal and embryo-derived cells in placenta, is discussed.     

Comparative enzyme histochemistry of the early and term rat decidua with special attention to decidual regression

Summary As the early rat decidua is believed to fulfil functions other than the late or basal decidua, the question as to whether this difference is reflected in decidual cell metabolism was investigated. Using cryosections of pregnant rat uteri of the 10th, 15th and 21st gestational day, activities of oxyradical-forming enzymes and hydrolases were analysed histochemically. The enzyme activities of decidual stromal cells and fibroblasts of the metrial gland exhibited three main fluctuations. One group of enzyme activities did not change during gestation, a second group decreased or disappeared, and a third group increased or was expressed in the late decidua only. Enzymes of the purine and polyamine pathway, including oxyradical-forming oxidases, were absent from early mesometrial decidual cells, but were highly active in the late regressing decidua and metrial gland. Some acid hydrolases and neutral proteases became active in the mature decidua. The possibility that purine-degrading and oxyradical-forming enzymes support decidual as well as metrial gland regression, and thus placental separation, by direct tissue damage and/or by indirect rupture of lysosomal membranes, inducing the release of acid hydrolases, is considered. Dedicated to Professor Zdeněk Lojda on the occasion of his 65th birthday.     

Expression of mRNA encoding insulin‐like growth factors I and II by uterine tissues and placenta during pregnancy in the rat

The uterus and the placenta synthesize insulin‐like growth factors (IGFs) and insulin‐like binding proteins (IGFBPs). These growth factors are implicated in processes of proliferation and differentiation that occur in the uterus. To determine the patterns of expression of IGFs during rat pregnancy we used in situ hybridization with digoxigenin labeled probes on uterus from day 7 to day 16 of pregnancy. In early gestation days (7–8) both IGF mRNAs showed similar tissue distribution with relative abundance in the stroma and circular muscle layer. On days 11 and 12 expression for IGF‐I mRNA was found in the mesometrial decidua and metrial gland and in the ectoplacental cone while clear expression of IGF‐II mRNA could only be found in the latter. On days 13 and 14, expression for IGF‐I mRNA could be detected in the mesometrial decidua and metrial gland but no expression was observed for IGF‐II mRNA. A gradient of IGF‐I mRNA expression could be observed in the placenta on day 16, with the trophoblastic cells of the basal zone expressing the signal with stronger intensity than in the labyrinthine zone. For IGF‐II mRNA the highest expression was associated with the labyrinthine zone. Endovascular trophoblast was positive for both mRNAs. The spatial and temporal patterns of expression suggests a role for IGFs in the process of decidualization as well as in the establishment, growth and differentiation of the various trophoblast cells of the placenta. Mol. Reprod. Dev. 53:294–305, 1999. © 1999 Wiley‐Liss, Inc.     

Expression of vascular endothelial growth factor by granulated metrial gland cells in pregnant murine uteri

Granulated metrial gland (GMG) cells are a characteristic uterine component belonging to a natural killer cell lineage. This study is aimed at revealing their kinetic and spatial relationship with vascular growth during pregnancy and the expression of vascular endothelial growth factor (VEGF). GMG cells and blood vessels were identified by periodic-acid-Schiff-reagent (PAS)-stained granules and positive staining for factor-VIII-related antigen, respectively. GMG cells were widely distributed in the decidua and metrial gland and showed a numerical increase with a peak at day 13 in parallel with the increase of vascular density. Preceding the maximal vascular development at day 13, microvessels with a narrow lumen representative of neovascularization prevailed at days 7-9, and the VEGF content in the decidua/metrial gland was significantly elevated at days 7-13 concurrently with mRNA expression. By immunolight microscopy combined with PAS staining, GMG cells with PAS-stained granules were positive for VEGF. Immunoelectron microscopy demonstrated that immunoreactions were diffuse in the cytoplasm but not localized in the granules. In contrast, fibroblast-like stromal cells were negative. These data indicate that GMG cells express VEGF and may play inducing roles in uterine neovascularization during pregnancy.     

Synthesis of 'decidualization-associated protein' in tissues of the rat uterus and placenta during pregnancy.

The synthesis of a soluble protein (referred to as 'decidualization-associated protein', DAP), has been examined in uterine and placental tissues of rats during pregnancy by means of polyacrylamide gel electrophoretic analysis of [3H]leucine-labelled soluble proteins. No synthesis of the protein was detected in non-implantation regions of the uterus. In implantation site tissue, no synthesis was detected on Days 6 or 7 of pregnancy. Only slight synthesis was present in the endometrium on Day 8, but synthesis rose rapidly from Days 9 to 12 in both the endometrium and myometrium although differences in the rates of increase were observed. Synthesis fell from Day 12 to 14 in both tissues. Synthesis by the myometrium was entirely localized in the mesometrial region, which contains the metrial gland. After Day 12, when the endometrium is represented by the chorioallantoic placenta, synthesis was examined in the labyrinthine and the decidua basalis/basal zone placenta tissues. No synthesis of 'DAP' was detected in the labyrinthine placenta from Day 16 of pregnancy. Synthesis observed in the decidua basalis/basal zone placenta fell dramatically from Day 14 to 20. The pattern of synthesis of 'DAP' during pregnancy suggests a role in the establishment of the chorioallantoic placenta and metrial gland in the rat.     

Localization of a pore-forming protein (perforin) in granulated metrial gland cells

Several studies have suggested that the granulated metrial gland (GMG) cells of the metrial gland (MG) may be natural killer (NK)-like cells. The cytotoxicity of NK cells involves secretion of a pore-forming protein termed perforin, which can polymerize on the target cell membrane to form transmembrane pores that are thought to be involved in target cell death. In the present study, we used an antiserum against perforin to determine whether this protein can be detected immunohistochemically in GMG cells. Mouse uteri were fixed by vascular perfusion with several fixatives on Day 14 of pregnancy, and tissue sections were labeled by an indirect immunofluorescence method. Specific perforin labeling was detected in GMG cells throughout the MG, in the decidua basalis, and in the labyrinthine placenta. The presence of perforin in GMG cells supports the suggestion that they may be NK-like cells.     

Lectin histochemical studies of mouse granulated metrial gland cells

A lectin histochemical study has been carried out on mouse granulated metrial gland cells, the major leucocyte population that differentiates in the uterine wall in pregnancy. The binding characteristics of 26 lectins were examined using light microscopical methods. Fourteen of the lectins, with affinities ranging through N-acetylgalactosamine, galactose, N-acetylglucosamine, mannose and sialic acid residues, bound to the cytoplasmic granules of granulated metrial gland cells, and each appeared to bind to the limiting membrane of the granules. The binding characteristics of three of these lectins (Wheat germ agglutinin, Concanavalin A and Helix pomatia agglutinin) were examined using electron microscopical methods. These showed a different binding pattern to the cytoplasmic granules of granulated metrial gland cells compared with that found using light microscopical methods, as they appeared to bind evenly across the granule's matrix. This binding pattern corresponds to the reactivity of the granule matrix in the periodic acid--Schiff technique. Six lectins bound to the cell membranes of granulated metrial gland cells. These included the E and L isoforms of Phaseolus vulgaris agglutinin, with affinities for complex carbohydrates, whose binding differences were related to the stage of differentiation of the granulated metrial gland cells. The lectin binding described presents additional markers of granulated metrial gland cells and tools for investigating carbohydrate moieties in the functional activities of granulated metrial gland cells This revised version was published online in November 2006 with corrections to the Cover Date.     

Lectin histochemical studies of mouse granulated metrial gland cells

A lectin histochemical study has been carried out on mouse granulated metrial gland cells, the major leucocyte population that differentiates in the uterine wall in pregnancy. The binding characteristics of 26 lectins were examined using light microscopical methods. Fourteen of the lectins, with affinities ranging through N-acetylgalactosamine, galactose, N-acetylglucosamine, mannose and sialic acid residues, bound to the cytoplasmic granules of granulated metrial gland cells, and each appeared to bind to the limiting membrane of the granules. The binding characteristics of three of these lectins (Wheat germ agglutinin, Concanavalin A and Helix pomatia agglutinin) were examined using electron microscopical methods. These showed a different binding pattern to the cytoplasmic granules of granulated metrial gland cells compared with that found using light microscopical methods, as they appeared to bind evenly across the granule's matrix. This binding pattern corresponds to the reactivity of the granule matrix in the periodic acid--Schiff technique. Six lectins bound to the cell membranes of granulated metrial gland cells. These included the E and L isoforms of Phaseolus vulgaris agglutinin, with affinities for complex carbohydrates, whose binding differences were related to the stage of differentiation of the granulated metrial gland cells. The lectin binding described presents additional markers of granulated metrial gland cells and tools for investigating carbohydrate moieties in the functional activities of granulated metrial gland cells     

Measurement of natural killer activity and target cell binding by mouse metrial gland cells isolated by enzymic or mechanical methods

Cells of the metrial glands of mice were isolated by enzymic or mechanical dissociation procedures. Morphological observations indicated that up to half of the enzymically dissociated cells and nearly all of the mechanically dissociated cells were granulated metrial gland cells, but the presence of some fibroblast-like stromal cells among the latter population was not ruled out. Moreover, the granulated metrial gland cells had lost a substantial part of their granule content during isolation. Both cell preparations had little or no natural killer (NK) activity, indicating either that granulated metrial gland cells are not NK-like or that their NK activity was impaired by loss of granule-associated lytic substances or by other factors. Enzymically dissociated metrial gland cells did not bind significantly to the NK target cell YAC-1, nor did they develop granules, NK activity, or the ability to bind YAC-1 cells during culture in vitro, either in normal medium or with the addition of indomethacin or lymphokines. Mechanically dissociated metrial gland cells bound avidly to YAC-1 cells but not to P815 cells or adult thymus cells, which are not NK target cells. Since many if not most of the mechanically dissociated metrial gland cells appeared morphologically to be granulated metrial gland cells, their selective binding to an NK target cell suggests that granulated metrial gland cells may be related in some way to NK cells.     

Evaluation of the pregnancy immunotrophism hypothesis by assessment of the reproductive performance of young adult mice of genotype scid/scid.bg/bg

The hypothesis that enhancement of pregnancy success results from immune recognition of the conceptus was evaluated by studying reproductive performance in a new line of mice deficient in NK cells and lacking B cells and T cells. Doubly mutant mice of genotype scid/scid.bg/bg are both viable and fertile. The numbers of offspring born to pairs of this genotype were not different from numbers born to heterozygous pairs. Differences in prenatal loss could not be found between genotypes by counts of either fetal resorption sites or corpora lutea. The timing of developmental stages and the differentiation of trophoblast, placenta, decidua and metrial gland in scid/scid.bg/bg mice appeared normal. These results suggest either that lymphokine influences on trophoblast cells in vivo do not contribute, in a major way, to pregnancy success or that the important cytokines are derived from uterine cell populations that are not classical, mature B cells, T cells or NK cells.     

Immunosuppressor activity of rat endometrial granulated cells and their differentiation

Natural killer and natural suppressor activities of the rat endometrial granulated cells were assayed on day 13 of pregnancy or pseudopregnancy. Metrial gland granulated cells were used as endometrial granulated cells. The natural killer activities of metrial gland granulated cells and other cells were determined by means of Hashimoto-Sudo test with K562 cells as targets. The estimation of natural killer activity included removal of the cells sticking to glass from a suspension of material gland granulated cells. Cytochemically, metrial gland granulated cells were identified by the presence of PAS-positive granules in the cytoplasm after treatment of the cells with diastase and identification of a specific antigen with the help of specific antisera. The natural killer activity of metrial gland granulated cells was twice weaker than that of splenocytes from the same pregnant or pseudopregnant females. The level of natural killer activity was proportional to the content of metrial gland granulated cells in a cell system. These data suggest that the natural killer activity of metrial gland granulated cells is realized via their contact with cell targets. Natural killer and suppressor activities were determined simultaneously for metrial gland granulated cells and splenocytes of the same rat with common cell targets. When estimating the nuclear suppressor activity of metrial gland granulated cells, the splenocytes of the same rat were used as an effector in a natural killer test. Various amounts of metrial gland granulated cells were added to the effector : target system at a ratio of 50:1. The natural suppressor activity of metrial gland granulated cells did not depend on the amount of metrial gland granulated cells present in a natural killer system. After fractionation in a Percoll gradient, the highest natural killer activity was recorded in a 30% Percoll fraction. The highest and lowest natural suppressor activities were recorded in 30% and 60% Percoll fractions, respectively. The culture medium was characterized by natural suppressor activity as well. The differences in mean areas of metrial gland granulated cells in 30 and 60% Percoll fractions between the pregnant (144.7 +/- 13.4 and 75.0 +/- 12.5 microm2, respectively) and pseudopregnant (97.5 +/- 4.9 and 69.2 +/- 3.5 microm2) females were reliable. The natural killer activity was estimated in all studied 23 samples of metrial gland granulated cells, among which 18 (79.6 +/- 7.8%) displayed the natural suppressor activity as well. The absence of natural suppressor activity in five samples was combined with the absence of this activity in their culture medium and with a reduction in the mean area of metrial gland granulated cells in 30% Percoll fraction to 109.1 +/- 5.2 microm2. The data obtained confirm the known data on a low activity of metrial gland granulated cells and demonstrated for the first time the natural suppressor activity of these cells. It was concluded that the natural suppressor activity of metrial gland granulated cells is due to their differentiation from metrial gland granulated cells with natural killer activity.     

Immunosuppressor activity of rat endometrial granulated cells and their differentiation

Natural killer and natural suppressor activities of the rat endometrial granulated cells were assayed on days 13 and 14 of pregnancy or pseudopregnancy. Metrial gland granulated cells were used as endometrial granulated cells. The natural killer activities of metrial gland granulated cells and other cells were determined by means of Hashimoto-Sudo test with K562 cells as targets. The estimation of natural killer activity included removal of the cells sticking to glass from a suspension of material gland granulated cells. Cytochemically, metrial gland granulated cells were identified by the presence of PAS-positive granules in the cytoplasm after treatment of the cells with diastase and identification of a specific antigen with the help of specific antisera. The natural killer activity of metrial gland granulated cells was twice weaker than that of splenocytes from the same pregnant or pseudopregnant females. The level of natural killer activity was proportional to the content of metrial gland granulated cells in a cell system. These data suggest that the natural killer activity of metrial gland granulated cells is realized via their contact with cell targets. Natural killer and suppressor activities were determined simultaneously for metrial gland granulated cells and splenocytes of the same rat with common cell targets. When estimating the natural suppressor activity of metrial gland granulated cells, the splenocytes of the same rat were used as an effector in a natural killer test. Various amounts of metrial gland granulated cells were added to the effector: target system at a ratio of 50 : 1. The natural suppressor activity of metrial gland granulated cells did not depend on the amount of metrial gland granulated cells present in a natural killer system. After fractionation in a Percoll gradient, the highest natural killer activity was recorded in a 60% Percoll fraction. The highest and lowest natural suppressor activities were recorded in 30% and 60% Percoll fractions, respectively. The culture medium was characterized by natural suppressor activity as well. The differences in mean areas of metrial gland granulated cells in 30 and 60% Percoll fractions between the pregnant (144.7 ± 13.4 and 75.0 ± 12.5 µm², respectively) and pseudopregnant (97.5 ± 4.9 and 69.2 ± 3.5 µm², respectively) females were reliable. The natural killer activity was estimated in all studied 23 samples of metrial gland granulated cells, among which 18 (79.6 ± 7.8%) displayed the natural suppressor activity as well. The absence of natural suppressor activity in five samples was combined with the absence of this activity in their culture medium and with a reduction in the mean area of metrial gland granulated cells in 30% Percoll fraction to 109 ± 5 µm². The data obtained confirm the known data on a low killer activity of metrial gland granulated cells and demonstrated for the first time the natural suppressor activity of these cells. It was concluded that the natural suppressor activity of metrial gland granulated cells is due to their differentiation from metrial gland granulated cells with natural killer activity.Translated from Ontogenez, Vol. 36, No. 1, 2005, pp. 26–34.Original Russian Text Copyright © 2005 by Podporina, Mikhailov.     

The metrial gland

The information available about the metrial gland of the pregnant rodent uterus with its content of granulated metrial gland (GMG) cells is reviewed. Recent research shows that GMG cells differentiate from bone marrow cells and supports the suggestion that GMG cells are involved in the immunological relationship between mother and foetus. There is probably a complex association between GMG cells and stromal cells of the metrial gland, and it is suggested that the association between GMG cells and the placental labyrinthine cells represents a functional interaction.     

Studies of implantation traces in rats. III. Histological examination

We carried out a histological examination of the implantation traces in delivered rats. The implantation traces could be identified more than 500 days after delivery on the mesometrial side as black and brown spots. The implantation traces were recognizable as a cicatrix remaining in the parametrium, mesometrial triangle, which was formed by repair of injury caused by placental desquamation. In this area, metrial gland cells which were laid down through pregnancy were recognized for about two months after delivery. The implantation traces consisted of cicatrix tissue associated with collagen production and hemosiderin. It was possible to distinguish old and new traces by the size of the siderophile cells and by the degree of hemosiderin present. It was also possible to discriminate new traces as yellowish-brown areas, covered with a yellowish-white mass of degenerated metrial gland, in cleared uteri stained with 2% NaOH solution. Siderophile cells on the implantation traces were derived from giant cells which persisted around the peripheral region of the placental desquamation site, and these giant cells were considered to be identifiable with metrial gland cells. It was considered that formation of the cicatrix is essentially the same in abortion, stillbirth and normal delivery. However, it was found that the implantation traces had different histological appearances depending on the degree of injury to the endometrium and myometrium and time of placental desquamation. The iron content of the implantation traces corresponded quantitatively with the hemosiderin observed in the histological investigations. The iron content decreased rapidly up to 21 days after delivery, decreasing gradually thereafter. The iron in the implantation traces could, however, be analyzed quantitatively by atomic absorption spectroscopy until day 365 after delivery.     

In situ localization and characterization of bone marrow-derived cells in the decidua of normal murine pregnancy.

The study reported here was designed to examine the in situ distribution and characteristics of hemopoietically derived decidual cells during normal pregnancy in mice prenatally reconstituted with bone marrow cells carrying a transgenic marker. Bone marrow cells from a transgenic CD-1 strain (CD-1 beta; carrying 1000 copies of beta-globin genes in tandem) were injected into the yolk sac of Day 17 conventional CD-1 embryos. The pregnant females were allowed to deliver normally, and the female offspring raised to puberty were mated with CD-1 males and then killed on Day 12 of gestation. The extent of chimerism in sections of their spleens, uteri, and other organs was evaluated by in situ hybridization of the sections with a biotinylated cDNA probe specific for the beta-globin genes followed by avidin-biotin-peroxidase staining. Tissue controls were provided by CD-1 beta and CD-1 mice, respectively. Tissues were also processed without the application of the probe or with the application of biotinylated lambda DNA as specificity controls. Reconstituted mice exhibited variable degrees of hemopoietic chimerism as indicated by labeling of their splenic lymphocytes (18-54%; mean 42%) as well as hemopoietic cells in other organs. Variable cellular labeling was also noted in their decidua basalis and metrial glands. Labeled cells in these tissues were identified as typical decidual cells, macrophages, and granulated metrial gland (GMG) cells. Labeling of typical decidual cells varied extensively among implantation sites in the same chimera, the average labeling ranging from 17% to 33% (mean 24%) in various chimeras. Labeling was also noted in GMG cells, lymphocytes, and some decidual cells migrating out of metrial gland explants after 24-h culture. The non-pregnant uterus of a chimeric mouse revealed significant labeling of endometrial stromal cells indicative of their hemopoietic origin. These results revealed a hemopoietic origin of certain typical decidual cells and GMG cells identified in situ during normal murine pregnancy and a hemopoietic origin of certain endometrial stromal cells that may represent precursors of decidual cells. The precise timing of the predecidual stem cell migration from the bone marrow to the uterus remains to be defined.     

Migration of granulated metrial gland cells from cultured explants of mouse metrial gland tissue

Summary The migration of granulated metrial gland (GMG) cells from cultured explants of metrial gland tissue obtained from mice killed between days 10 and 16 of pregnancy has been studied. GMG cells migrated from all of the explants but more GMG cells were found around explants obtained from mice at day 10 of pregnancy than around explants obtained at later stages of pregnancy. The number of GMG cells found around each explant reached a peak at days 1, 2 or 3 of culture but only a few GMG cells were found around the explants by day 7 of culture.     

Uptake of immunoglobulin and albumin by granulated metrial gland cells in vitro

Single cell suspensions of metrial gland tissue from rats at Day 14 of pregnancy were prepared for maintenance in vitro. During the first 2 days of culture IgG was detected in glycoprotein granule-containing granulated metrial gland (GMG) cells. Albumin was also detected in GMG cells at the same stages. The IgG and albumin were not detected during the next 4 days in culture. When metrial gland cells, maintained in vitro for 5 days, were incubated with rat serum for a further 24 h, IgG and albumin were detected in GMG cells. When similar cultures were incubated for 24 h with purified rat IgG or purified rat albumin, GMG cells were positive for IgG and albumin respectively. Albumin was not detected in GMG cells in wax sections of metrial gland tissue, although IgG has previously been demonstrated. The uptake of serum proteins by GMG cells in vitro has been clearly shown but the difference in IgG and albumin content of these cells in paraffin-wax sections indicates that the means by which IgG accumulates intracellularly may be different in vitro and in vivo.     

Alterations in the expression of homing-associated molecules at the maternal/fetal interface during the course of pregnancy.

One of the most fascinating immunologic questions is how the genetically distinct fetus is able to survive and develop within the mother without provoking an immune rejection response. The pregnant uterus undergoes rapid morphological and functional changes, and these changes may influence the nature of local immune responses at the maternal/fetal interface at different stages of gestation. We hypothesized that specialized mechanisms exist to control access of maternal leukocyte subsets to the decidua and that these mechanisms are modulated during the course of pregnancy. At the critical period of initial placenta development, the maternal/fetal interface displays an unparalleled compartmentalization of microenvironmental domains associated with highly differentiated vessels expressing vascular addressins in nonoverlapping patterns and with recruitment of specialized leukocyte subsets (monocytes, granulated metrial gland cells, and granulocytes) thought to support, modulate, and regulate trophoblast invasion. One of the most striking observations at this time of gestation is the almost complete exclusion of lymphocytes from the maternal/fetal interface. The second half of pregnancy is characterized by a partial loss of microenvironmental specialization and different switches in vascular specificity within the decidua basalis, paralleling dramatic changes in the populations of recruited leukocytes (e.g., a striking influx of lymphocytes, especially T cells). In the term pregnant uterus, the expression of all vascular addressins decreased dramatically; only weakly staining maternal vascular segments remained. These segments may define sites of extremely low residual traffic in the term decidua, which contains remarkably few maternal leukocytes overall. Our results suggest that the maternal/fetal interface represents a situation in which leukocyte trafficking is exquisitely regulated to allow entry of specialized leukocyte subsets that may play a fundamental role in immune regulation during pregnancy.