Co-stimulation of human decidual natural killer cells by interleukin-2 and stromal cells

Increased angiogenic factor secretion by decidual natural killer cells from pregnancies with high uterine artery resistance alters trophoblast function

STUDY QUESTION

Are the concentrations of factors secreted by decidual natural killer (dNK) cells from pregnancies at high risk of poor spiral artery remodelling different to those secreted from pregnancies at low risk?

SUMMARY ANSWER

Expression levels of PLGF, sIL-2R, endostatin and angiogenin were significantly increased by dNK cells from high-risk pregnancies, and angiogenin and endostatin were found to alter trophoblast function.

WHAT IS KNOWN ALREADY

During early pregnancy, maternal uterine spiral arteries are remodelled from small diameter, low-flow, high-resistance vessels into larger diameter, higher flow vessels, with low-resistance. This change is essential for the developing fetus to obtain sufficient oxygen and nutrients. dNK cells have been implicated in this process.

STUDY DESIGN, SIZE, DURATION

dNK cells were isolated from first trimester terminations of pregnancies (obtained with local ethical approval) screened for normal- or high-resistance index, indicative of cases least (<1%) and most (>21%) likely to have developed pre-eclampsia had the pregnancy not been terminated (n = 18 each group). Secreted factors and the effects of these on the trophoblast cell line, SGHPL-4, were assessed in vitro.

PARTICIPANTS/MATERIALS, SETTING, METHODS

A multiplex assay was used to assess dNK cell-secreted factors. SGHPL-4 cell functions were assessed using time-lapse microscopy, 3D invasion assays, endothelial-like tube formation ability and western blot analysis.

MAIN RESULTS AND THE ROLE OF CHANCE

The expression levels of PLGF (P < 0.01), sIL-2R (P < 0.01), endostatin (P < 0.05) and angiogenin (P < 0.05) were significantly increased by dNK cells from high-risk pregnancies. Endostatin significantly decreased SGHPL-4 invasion (P < 0.05), SGHPL-4 tube formation (P < 0.05) and SGHPL-4 Akt^ser473 phosphorylation (P < 0.05). Angiogenin significantly decreased SGHPL-4 invasion (P < 0.05), but increased SGHPL-4 tube formation (P < 0.01) and decreased SGHPL-4 Akt^ser473 phosphorylation (P < 0.05).

LIMITATIONS, REASONS FOR CAUTION

The culture of dNK cells and protein concentrations in vitro may not fully represent the in vivo situation. Although SGHPL-4 cells are extravillous trophoblast derived, further studies would be needed to confirm the roles of angiogenin and endostatin in vivo.

WIDER IMPLICATIONS OF THE FINDINGS

The altered expression of secreted factors of dNK cells may contribute to pregnancy disorders associated with poor spiral artery remodelling.

STUDY FUNDING/COMPETING INTEREST(S)

This study was supported by the Wellcome Trust (project reference 091550). R.F. was a recipient of a PhD studentship from the Division of Biomedical Sciences, St. George's, University of London. The authors have no conflict of interests.

Expansion of decidual CD45RO⁺ T cells with high expression of CEACAM1 in the early stage of pregnancy

The aim of the present study was to investigate the mechanism involved in the expansion of CD45RO+ T cells in the decidual microenvironment, and in the expression of the inhibitory carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) on the surface of decidual CD45RO+ T cells. Twenty-one healthy nonpregnant females and seventeen healthy pregnant females in the first trimester were included in the study. Peripheral blood samples from nonpregnant and pregnant females, and decidual tissues from pregnant females following elective abortion, were obtained and analyzed by flow cytometry. The percentages of CD45RO+ T cells and CEACAM1-expressing CD45RO+ T cells were significantly higher in first trimester human decidua than in the peripheral blood. Conditioned medium from the coculture of monocytes and the human trophoblast HTR8/SVneo cell line (MHM) was added to the model for the generation of CD45RO+ T cells in vitro. MHM caused an increase in the percentage of CD45RO+ T cells in a monocyte chemoattractant protein-1 (MCP-1)‑dependent manner and an increase in the percentage of CEACAM1-expressing CD4+CD45RO+ T cells in the model. In conclusion, our results implied that trophoblast cells and monocytes may be involved in the increase of decidual CD45RO+ T cells and the high expression of CEACAM1 on their surfaces.

A cross-talk of decidual stromal cells, trophoblast, and immune cells: a prerequisite for the success of pregnancy

Embryo implantation and formation of a functional placenta are complex processes that require a plethora of regulatory mechanisms involving both mother and embryo cells. Recently, an important role in this complicated cells and factors network was assigned to the decidual stromal cells (DSC) and trophoblast cells. Decidualization includes biochemical changes that trigger DSC to produce a number of factors required for the implantation and induction of immunotolerance in maternal immune system. Immunotolerance is achieved by a cascade of strictly controlled events starting with selective homing of immune cells to the feto-maternal site, regulated proliferation, and predominant differentiation into a regulatory type of immune cells. Furthermore, cytotoxic effector functions are reduced owing to the influence of steroid hormones, factors, cytokines, and inhibitory receptors. Altogether the entire immune system of the mother is switched to tolerogenic functional state which is a prerequisite for the successful maintenance of pregnancy.

Progestin regulates chemokine (C-X-C motif) ligand 14 transcript level in human endometrium

Leukocyte populations change profoundly in the human endometrium during the menstrual cycle. However the predominant cell, the uterine natural killer (uNK) cell does not contain steroid receptors. From gene array analysis we identified a transcript encoding chemokine (C-X-C motif) ligand 14 (CXCL14) which is markedly up-regulated in the secretory phase of the cycle. We confirm this data by northern blotting and quantitative PCR. Using in situ hybridization we localized CXCL14 mRNA to the glandular epithelial cells where it was detected only in the secretory phase of the cycle. Candidate progesterone response elements were identified at positions -2028/-2007 and -722/-697 (PRE1 and PRE2, respectively) relative to the translation start site. These were functionally tested using luciferase reporter deletion constructs, electrophoretic mobility shift assays and site-directed mutagenesis. The deletion/mutation of these sites reduced progesterone induction by 40 and 20%, respectively. Finally, we demonstrated that recombinant CXCL14 stimulated uNK cell chemotaxis in vitro. We therefore conclude that CXCL14 is likely to be regulated by progesterone in human endometrium and that it may exert a chemoattractive effect on uNK cells and in part be responsible for their clustering around the epithelial glands.

Prevention of embryo loss in non-obese diabetic mice using adoptive ITGA2(+)ISG20(+) natural killer-cell transfer

Both regulatory T cells and regulatory natural killer (NK) cells may play essential roles in the maintenance of pregnancy. In this study, we show that a significantly high percentage of spontaneous embryo loss was observed in both allogeneic and syngeneic pregnant non-obese diabetic (NOD) mice. The percentage of embryo loss in allogeneic pregnant mice was further increased by the administration of anti-asialo ganglio-N-tetraosylceramide to deplete NK cells, but was decreased by the adoptive transfer of ITGA2(+)ISG20(+) (CD49b(+) CD25(+)) NK cells from normal mice. No such trend was observed in syngeneic pregnant NOD mice. The pattern of CXCR4 (specific receptor for CXCL12) expression on NK cells was analyzed and NK-cell migration was confirmed by in vitro and in vivo migratory assays. Since CXCL12 production by murine trophoblast cells was confirmed previously, our findings suggest that the recruitment of peripheral CXCR4-expressing ITGA2(+)ISG20(+) NK cells into pregnant uteri may be important in the regulation of feto-maternal tolerance.

Improvement of fertility with adoptive CD25+ natural killer cell transfer in subfertile non-obese diabetic mice

To investigate the role that CD25(+) natural killer (NK) cells may play in the establishment of pregnancy, the effect of adoptive CD25(+) NK cell transfer on pregnancy outcome in subfertile non-obese diabetic (NOD) mice was investigated. Phenotypic analysis of NK cells was performed by flow cytometry before and after the transfer. The proportion of subfertile NOD female mice that failed to become pregnant when co-caged with C57Bl/6 males for 16 weeks was significantly higher in female NOD mice than in normal female BALB/c controls (53.1% versus 15.1%; P < 0.01). The subfertile NOD mice were divided into three groups receiving CD25(+) NK cells (group 1), CD25(-) cells (group 2) or RPMI 1640 medium only (group 3). Group 1 had significantly more pregnancies than those receiving CD25(-) NK cells (77.8% versus 11.1%; P < 0.01) and controls injected with RPMI 1640 medium (0.0%; P < 0.01). Improved fertility was concomitant with an increase in placental CD49b(+) NK cells expressing Foxp3. Foxp3 expression was confirmed in the CD25(+) NK cells before the transfer. These results indicate that subfertility in NOD mice may be partially attributed to the insufficiency of CD25(+) and Foxp3(+) NK cells recruited into the pregnant uterus.

Human uterine natural killer cells: a reappraisal

The presence of granulated cells within the uterus of many species has been recognised for many years but only recently have these been recognised to be a type of NK cell. Various terms have been applied to the cells, including endometrial granulocyte, K cell and, in mouse and rat, granulated metrial gland cell. Although early studies are often based on histology and electron microscopy, these often include important information for current studies. In vitro studies of purified cells have focused particularly on cytotoxicity and cytokine production and roles in the control of trophoblast invasion and spiral artery remodelling in human pregnancy have been proposed. Evidence in mouse has implicated uNK cell production of IFN-gamma in vascular remodelling but evidence for such a role for human uNK cells remains to be established. Investigation of uNK cells in human pregnancy is hampered by the lack of availability of tissues from the first half of the second trimester of pregnancy when vascular remodelling occurs and also by possible differences between cells from different regions of decidua. The presence of similar cells in species with no trophoblast invasion into the uterus and epitheliochorial placentation raises the question of whether control of trophoblast invasion by human uNK cells is important in vivo and raises the possibility of another function which is conserved between species.

Apoptosis of extravillous trophoblast cells limits the trophoblast invasion in uterine but not in tubal pregnancy during first trimester

During the first trimester of pregnancy extravillous trophoblast cells (EVT) invade the maternal decidua. Invasion normally is reduced from the second trimester onwards and stops in the inner third of the myometrium. By contrast, in extrauterine tubal pregnancy, trophoblast invasion may even penetrate the tubal wall, which ultimately leads to the rupture of the fallopian tube. Induction of apoptosis of EVT cells, by maternal immune competent cells, may be an important mechanism to limit EVT invasion in uterine pregnancy. Tissue specimens from first and second trimester uterine pregnancy and first trimester tubal pregnancy were analyzed for apoptosis by TUNEL- and M30-staining. By immunohistochemical double labelling, maternal leukocyte subtypes were co-localized to apoptotic cells and in this context, the number of CD56(+)NK cells was analyzed. Our data show that apoptosis is confined to the decidua basalis. Most apoptotic cells are single cytokeratin-positive epithelial cells residing in the stromal compartment. Consequently these cells can only be EVT cells. Maternal leukocytes are not apoptotic. They are located in close contact to apoptotic cells. The number of apoptotic cells in the second trimester (1.8+/-0.7 per cent) is reduced compared to first trimester (5.6+/-0.7 per cent) of uterine pregnancy. In parallel, the number of NK cells declines from first (24.4+/-2.9) to second (12.4+/-1.8) trimester. Furthermore, apoptosis is significantly reduced in ectopic (0.9+/-0.3 per cent) compared to eutopic first trimester pregnancies. Consequently, we suggest that in first trimester uterine pregnancy, induction of EVT cell apoptosis by the maternal immune system is one mechanism to limit EVT invasion. During the second trimester, in parallel to declining numbers of NK cells, the mechanism changes. However, in tubal pregnancy due to differing immunological microenvironments at the ectopic implantation site, apoptosis induction fails, which deleteriously may result in uncontrolled invasion and penetration of the tubal wall.

Marrow Stromal Cells for Interleukin-2 Delivery in Cancer Immunotherapy

Marrow stromal cells (MSCs) can be easily gene-modified and clonally expanded making them ideal candidates for transgenic cell therapy. However, recent reports suggest that MSCs possess immunosuppressive effects, which may limit their clinical applications. We investigated whether interleukin (IL)-2 gene-modified MSCs can be used to mount an effective immune response against the poorly immunogenic B16 melanoma model. We first show that primary MSCs mixed with B16 cells and injected subcutaneously in syngeneic recipients do not affect tumor growth. On the other hand, IL-2-producing MSCs mixed with B16 cells significantly delayed tumor growth in an IL-2 dose-dependent manner. Furthermore, we observed that matrix-embedded IL-2-producing MSCs injected in the vicinity of preestablished B16 tumors led to absence of tumor growth in 90% of treated mice (p < 0.001). We demonstrated that tumor-bearing mice treated with IL-2-producing MSCs developed CD8-mediated tumor-specific immunity and significantly delayed tumor growth of a B16 cell challenge (p < 0.05). In addition, treatment of cd8-/-, cd4-/- and beige mice revealed that CD8+ and natural killer (NK) cells, but not CD4+ cells, were required to achieve antitumor effect. In conclusion, MSCs can be exploited to deliver IL-2 and generate effective immune responses against melanoma in mice with normal immune systems.

Osteopontin is up-regulated in human decidual stromal cells

OBJECTIVE

To analyze expression and regulation of osteopontin in human stromal endometrial and decidual cells.

DESIGN

Study by immunohistochemistry, RNase protection assay, and in vitro study.

SETTING

Academic research unit.

PATIENT(S)

Twenty-five fertile women.

INTERVENTION(S)

Expression of osteopontin mRNA and protein was analyzed by RNase protection assay (RPA) and immunohistochemistry in secretory endometrium and decidua. Regulation of osteopontin expression in endometrial epithelial cell and stromal cells was analyzed by incubation with 17beta-estrogen, progesterone, IL-1beta, IL-6, and LIF and by incubation with cell culture supernatants of decidual natural killer cells. Furthermore, osteopontin expression was studied in stromal cells, decidualized in vitro.

MAIN OUTCOME MEASURE(S)

mRNA and protein expression and regulation of endometrial stromal osteopontin.

RESULT(S)

Osteopontin mRNA and protein are expressed at increasing concentrations in human secretory endometrium and in decidua. High concentrations in total decidua were due to high osteopontin concentrations in decidualized stromal cells. Osteopontin mRNA and protein expression was not regulated by incubation of endometrial epithelial and stromal cells with steroids, cytokines, and natural killer cell supernatant for 6 and 24 hours. In contrast, decidualization of stromal cells for 7, 14, and 21 days resulted in significant up-regulation of osteopontin mRNA and protein.

CONCLUSION(S)

Osteopontin is expressed at high concentrations in secretory endometrium in epithelial cells and in decidua in stromal cells. Decidual stromal expression of osteopontin is up-regulated by decidualization.

Granulated metrial gland cells in the murine uterus: localization, kinetics, and the functional role in angiogenesis during pregnancy

Granulated metrial gland (GMG) cells are a major immune cell population in the murine pregnant uterus, and contribute to the maintenance of pregnancy by functioning as uterus-specific natural killer (NK) cells. In order to reveal their kinetics, activation, and functional roles in pregnancy, we conducted quantitative and immunohistochemical analyses in normal and immuno-modulator-treated mice. Under a light microscope, GMG cells were identified by red cytoplasmic granules in periodic-acid-Schiff (PAS)-stained sections. They progressively increased in number and size with the peak at day 12-14 of pregnancy in the decidua and metrial gland. New vessel formation was most prominent around day 8, and the total vascular area reached the peak at day 13. GMG cells were often located near the blood vessels, and expressed vascular endothelial growth factor (VEGF), suggesting their possible inducing role in angiogenesis during the development of decidua/metrial gland. While blood vessels in the non-pregnant uterus were negative for vascular cell adhesion molecule (VCAM)-1, those in the pregnant one were positive. Treatment with neutralizing antibody against VCAM-1, however, did not decrease the number of GMG cells. On the other hand, mitosis of GMG cells was frequently observed. These data suggest that the increment of GMG cells during pregnancy may largely result from local proliferation in the uterus rather than an increased influx of precursor cells. Although we attempted to induce in vivo activation of GMG cells by administration of interleukin-12 (IL-12) or alpha-galactosylceramide, a potent activator for natural killer-T (NK-T) cells, the number of GMG cells did not appreciably increase. The present study has demonstrated that GMG cells locally proliferate in the pregnant uterus, not being related to VCAM-1 expression by the uterine vasculature or systemic activation of NK cells and NK-T cells, and seem to be involved in angiogenesis in the pregnant uterus through VEGF production.

Leukocyte networks and human endometrial remodelling

Remodelling of the human endometrium occurs during the normal menstrual cycle. This process involves the disintegration of the superficial or functionalis layer of the endometrium following the fall in progesterone resulting from the demise of the corpus luteum and the reconstruction of a new layer without scarring. The degradative properties of matrix metalloproteinases (MMP) and their presence in the endometrium during remodelling events suggests that they are effector molecules in this process. The features of menstruation parallel those of an inflammatory response and the abundance of leukocytes in the endometrium prior to the onset of menstruation indicates a role for these cells in the remodelling process. This review examines the relationship between leukocytes and the local production and activation of MMP within the endometrium.

Conceptus influences the distribution of uterine leukocytes during early porcine pregnancy

Pregnancy in humans and rodents is associated with dramatic changes in leukocyte populations within the uterus. In these species, recruitment of leukocytes, mostly natural killer (NK) lymphocytes, accompanies decidualization of endometrial stroma even in the absence of pregnancy. In the pig, a nondecidualizing species, the predominant lymphocytes in the pregnant uterus are T and/or NK cells, but their distribution relative to embryonic attachment sites has not been reported. The objective of this study was to compare the abundance of leukocytes in porcine endometrium in contact with trophoblast with that between attachment sites during the early postattachment period. Uteri were recovered on Days 15-17 (n = 4), 18 and 19 (n = 4), 21 and 22 (n = 5), and 25-27 (n = 2) of gestation and from cycling pigs during the luteal phase (Day 15; n = 3). Leukocytes were identified in uterus obtained at versus between attachment sites using an antibody reactive with all leukocytes (CD44). In all pregnant animals, leukocytes were diffusely scattered throughout the endometrial stroma but were rare or absent in the luminal epithelium. Leukocyte density was approximately 3-fold greater in endometrium in contact with conceptuses than in endometrium between attachment sites throughout the early postattachment period. Leukocyte density during the luteal phase was similar to that between attachment sites, suggesting that leukocyte recruitment was a localized response to the embryo. The ability of an individual porcine conceptus to recruit maternal leukocytes to the adjacent stroma may be a vital step in early placental development and embryo survival.

Uterine natural killer cells in a three-dimensional tissue culture model to study trophoblast invasion

The high numbers of CD56(+) cells with natural killer (NK) functions present in the uterine mucosa during the late secretory phase of the menstrual cycle and during early pregnancy have been considered to be implicated in implantation and in the regulation of trophoblast invasion. A three-dimensional organ culture model was used to study the interactions of these uterine NK cells with Jeg-3 and BeWo choriocarcinoma cells as a model of the invasive trophoblast. For this purpose, fragments of endometrial and decidual tissue were put in close contact with multicellular spheroids of choriocarcinoma cells in small silicon funnels. After the formation of stable contacts, the confrontation cultures were transferred to spinner flasks, cultivated for up to 6 days, and prepared for immunohistochemistry. During 2 days of cocultivation, the first cells started to move forward into the stromal component of the confrontation culture as demonstrated by staining of the choriocarcinoma cells using anti-human cytokeratin. Invasion advanced until, after a total of 6 days, some choriocarcinoma cells had already penetrated deeply into the host tissue. After a cultivation period of 1 week, both the endometrial and decidual tissue fragments still contained several CD56(+) uterine NK cells, and some of them expressed the proliferation-associated marker Ki-67 without any exogenous activation. A few CD56(+) cells were found directly at the invasion front, as well as between the choriocarcinoma cells. These cells also contained the cytolytic granule protein perforin indicating a migration of NK cells with cytolytic potential toward the potentially invasive cells. In conclusion, this human system closely resembles the in vivo conditions during trophoblast invasion and provides an appropriate in vitro model for studying dynamic processes involving various cell types during trophoblast invasion at the experimental level. Moreover, it enables us to study the effects of cytokines and growth factors that possibly regulate trophoblast invasion.

Effects of trophoblast invasion on the distribution of leukocytes in uterine and tubal implantation sites

OBJECTIVE

To distinguish endocrine and paracrine influences on leukocyte subpopulations at uterine and tubal implantation sites.

DESIGN

Retrospective immunohistochemical study.

SETTING

Departments of Anatomy, and Obstetrics and Gynecology, School of Medicine, RWTH University of Aachen, Aachen, Germany.

PATIENT(S)

Ten women with a viable ectopic pregnancy (EP), 25 women who had undergone elective first-trimester termination of pregnancy, and 4 women who had undergone hysterectomy with adnexectomy.

INTERVENTION(S)

None.

MAIN OUTCOME MEASURE(S)

Quantitative analysis of leukocyte subpopulations at the implantation sites and their corresponding noninvaded tissues, decidual tissue from patients with EP, and tubal mucosa from normal menstrual cycle.

RESULT(S)

Similar numbers and characteristic distribution patterns of macrophages, T cells, and B cells were found at both normal intrauterine and tubal implantation sites. Natural killer (NK) cells were always absent from tubal mucosa. The number and distribution of leukocytes within decidual tissue from women with EP corresponded to those in the noninvaded decidual compartment in intrauterine pregnancy (IUP).

CONCLUSION(S)

Leukocyte populations present in the tubal and uterine mucosa are an intrinsic characteristic of these tissues. The distinct leukocyte distribution pattern at the implantation sites suggests that the invading trophoblast exerts a paracrine influence on endometrial and endosalpingeal leukocytes. The absence of natural killer cells from the tubal wall may be one reason for the higher degree of invasiveness of the trophoblast at the tubal implantation site.

Angiogenic growth factor messenger ribonucleic acids in uterine natural killer cells

Angiogenesis is essential for endometrial growth and repair, and disruption of this process may lead to common disorders of women, including menorrhagia and endometriosis. In pregnancy, failure of the endometrial spiral arterioles to undergo remodeling leads to preeclampsia. Here we report that in addition to vascular endothelial growth factor A (VEGF-A), human endometrium expresses messenger ribonucleic acids (mRNAs) encoding VEGF-C, placenta growth factor (PlGF), the angiopoietins, angiopoietin 1 (Ang1) and Ang2, and the receptors VEGFR-3 (Flt-4), Tie 1, and Tie 2. Levels of VEGF-C, PlGF, and Tie 2 changed during the menstrual cycle. Intense hybridization for VEGF-C and PlGF mRNAs was found in uterine nature killer cells in secretory phase endometrium and for Ang2 mRNA in the same cells in the late secretory phase. Interleukin-2 (IL-2) and IL-15 up-regulated VEGF-C, but not PlGF or Ang2, mRNA levels in isolated NK cells. Conditioned medium from decidual NK cells did not induce human umbilical vein endothelial cell apoptosis. These results indicate that human endometrium expresses a wide range of angiogenic growth factors and that uterine nature killer cells may play an important role in the abnormal endometrial angiogenesis that underlies a range of disorders affecting women.

Mouse granulated metrial gland cells require contact with stromal cells to maintain viability

Granulated metrial gland (GMG) cells differentiate in the uterine wall in pregnancy in mice but the mechanisms which control their differentiation and maintenance are unknown. In vivo, GMG cells share an intimate association with fibroblast-like stromal cells. The importance of this association has been assessed by examining the effects of withdrawal of stromal cell contact on GMG cell maintenance in vitro. When single cell suspensions of cells were prepared from mouse metrial glands there was a steady decline in numbers with days of culture but usually some remained at 7 d of culture. The ability of metrial gland cells to kill Wehi 164 target cells in 51Cr-release cytotoxicity assays was retained by cells cultured for at least 3 d. When explants of metrial gland were maintained in culture to allow GMG cell migration onto the culture flask, the attached GMG cells were lost by 1 d later. Overall, these results suggest that a juxtacrine regulatory mechanism maintains GMG cells. The rapid loss of unsupported GMG cells in culture has major implications in the design of assays to examine GMG cell function.

IL-15 expression at human endometrium and decidua

A large number of natural killer (NK) cells appear in human uterine mucosa during the secretory phase and first trimester pregnancy. We investigated the expression of interleukin (IL)-15, a possible stimulator for these NK cells, in human endometrium and first trimester decidua. Semiquantitative reverse transcriptase-polymerase chain reaction revealed that IL-15 mRNA expression was stronger during the secretory phase and first trimester pregnancy than during the proliferative phase. Immunohistochemistry revealed that immunoreactivity for anti-IL-15 was higher during the secretory phase than it was during the proliferative phase. This was prominent in the perivascular stromal cells around invading spiral arteries during the mid- to late-secretory phase. In first trimester decidua, endothelial cells were also stained as strongly as stromal cells. A membrane-bound IL-15 molecule was detected on the surface of first trimester decidual cells by flow cytometry. Progesterone stimulated the release of soluble IL-15 in the supernatant of cultured decidual cells. These results suggest that IL-15 expression in human uterine mucosa corresponds to the fluctuation of uterine NK cells and that its production is hormonally controlled, especially by progesterone.

Human decidual natural killer cells express the receptor for and respond to the cytokine interleukin 15

The natural killer (NK) cells that are present in the uterine mucosa (decidua) during early pregnancy have a distinctive phenotype, CD56(bright) CD16(-). These cells have previously been shown to proliferate and be activated by interleukin (IL)-2. However, IL-2 is absent from the decidua and placenta, and we have therefore investigated whether IL-15 is present in the uterus and can act on decidual NK cells. Both IL-15 mRNA and protein were found in a variety of cells but particularly in decidual macrophages. IL-15 induced a proliferative response in decidual NK cells that was blocked by anti-IL-15 and was augmented by stem cell factor. The cytolytic activity of decidual NK cells against K562 was augmented. Interestingly, in contrast to IL-2, although activation with IL-15 resulted in some killing of JEG-3 choriocarcinoma cells, normal trophoblast cells remained resistant to lysis. These findings suggest that IL-15 is a candidate cytokine responsible for NK cell proliferation in vivo in the progesterone-dominated secretory endometrium and early decidua.