Hormonal modulation of human natural killer cell activity in vitro

This study investigates the effect of steroid and polypeptide hormones on human NK cell activity in vitro. Solutions of hormones were incubated with normal peripheral blood cells in a 4 h cytotoxic assay against the NK-sensitive erythromyeloid cell line K562. The steroid hormones oestradiol, progesterone and testosterone did not exhibit a significant effect on K562 lysis at concentrations within the normal physiological range. HCG and LH, however, had a marked effect on NK activity, the latter at concentrations within the range observed in normal menstruating females.     

Does fetal gender affect cytotrophoblast cell activity in the human term placenta? Correlation with maternal hCG levels

BACKGROUND: Pregnant women with female fetuses have higher maternal serum human chorionic gonadotropin (hCG) levels than pregnant women with male fetuses. Ki-67, a cell proliferation and activity marker, is confined mostly in the nuclei of villous cytotrophoblasts of the human placenta. In this study, we examined the effect of fetal gender on the cytotrophoblast cell activity in human term placenta, with special regard to maternal serum and cord blood hCG levels. METHODS: Thirty-four uncomplicated, singleton, term pregnancies (17 male and 17 female fetuses) were recruited in the study. hCG was measured in maternal peripheral serum and umbilical cord blood. Placental samples were collected in each patient during the cesarean section. Cytotrophoblast cell activity was measured by using immunohistochemistry for Ki-67 antigen. Ki-67 staining index values of the cytotrophoblasts were compared between the female and male placentas. RESULTS: Maternal serum and cord blood hCG levels were higher in pregnant women with female fetuses than in those carrying male fetuses. There was no sex difference in Ki-67 immunostaining rates of the cytotrophoblast cells. There was no correlation between maternal serum and cord blood hCG levels and Ki-67 staining index values of the cytotrophoblast cells. CONCLUSIONS: The difference in maternal serum and cord blood hCG levels in correlation with the fetal gender is not associated with cytotrophoblast cell activity in the human term placenta. The gender of the fetus does not seem to affect the regulation of cytotrophoblast cell proliferation.     

NK cell activity and estrogen hormone levels during normal human pregnancy

NK cell activity was determined in peripheral blood of 24 women during pregnancy, and compared to NK activity of 40 healthy nonpregnant women in generative age. An increase in the first trimester was followed by a significant decline of NK activity in the second trimester, and a further fall in the third trimester of pregnancy. The initial rise of NK activity was predominantly due to primigravidas, whereas the fall in the second trimester was mainly due to multigravidas. There was a significant negative correlation between NK activity and the increasing levels of estrogen hormones (beta-estradiol, estriol and estrone) in the sera of pregnant women. However, when analyzed for each trimester of pregnancy separately, a significant (p less than 0.02) negative correlation was only found with beta-estradiol, suggesting that high doses of this hormone could contribute to pregnancy-associated NK suppression.     

Suppression of murine allogeneic cell interactions by sex hormones

Investigations have been carried out on the action of several steroid hormones on lymphocyte functions in inbred strains of mice. The recognitive, proliferative and effector phases of allogeneic cell interactions in vitro were assessed using a mixed lymphocyte reaction (MLR) and cell-mediated lympholysis (CML). In MLR containing Balb/c (responder) and C57BL/6 (stimulator) splenocytes DNA synthesis was markedly reduced in the presence of progesterone, cortisol or estradiol. In CML, progesterone and estradiol (1–5 μg/ml) blocked in vitro generation of cytotoxic lymphocytes, while cultures with cortisol were partially inhibited. None of these hormones suppressed the cytotoxic activity of previously sensitized effector cells generated in vitro. Cultures containing testosterone expressed both normal DNA synthesis in MLR and cytotoxic activity in the CML test. These findings suggest a selective pattern of immunosuppression by sex hormones which may be important in preventing graft rejection or graft-versus-host interactions which may arise as a consequence of fetal engraftment during pregnancy.     

The balance between cytotoxic NK cells and regulatory NK cells in human pregnancy

NK cells kill tumor cells and virus-infected cells as well as secrete a variety of cytokines. These effector functions are regulated by the balance between activating receptor signals and inhibitory receptor signals which are triggered by specific major histocompatibility complex (MHC) or non-MHC ligands. It is thought currently that the balance between immunostimulation and immunoregulation in T cell immunity is achieved by a Th1/Th2/Th3/Tr1 and CD4(+)CD25(+) regulatory T (Treg) cell paradigm. Here, we discuss the cytokine paradigm of NK cells in human pregnancy. During normal, intact pregnancy, peripheral blood NKr1 cells and decidual NK3 cells increase, while these NK cell populations decrease significantly in miscarriage cases, suggesting that an imbalance in NK1/NK2/NK3/NKr1 is correlated with miscarriage. Recent investigations have shown that not only Treg cells, but also regulatory NK (NK reg) cells, play very important roles in the maintenance of pregnancy. We summarize the progress in studying NK reg cells and focus on how NK reg cells and cytotoxic NK cells affect the reproductive immune response.     

The effects of sex, menstrual cycle, and oral contraceptives on the number and activity of natural killer cells

OBJECTIVES: The aim of this study was to study the impact of sex, the menstrual cycle, and the use of oral contraceptives (OC) on the number and activity of natural killer (NK) cells. METHODS: Both the number and the activity of NK cells were assessed per milliliter of blood, and NK activity (NKA) per NK cell and per lymphocyte was calculated. NKA was measured in each subject using a whole blood assay, which preserves the plasma and all blood cells, and using a washed blood assay, in which plasma is replaced with an artificial medium. The subjects were young (20-29 years old) women with a regular menstrual cycle (n = 39; 26 tested on both the follicular and the luteal phases), age-matched women who use OC (n = 26), and age-matched men (n = 20). RESULTS: Men showed markedly and significantly higher NKA than women with regular menstrual cycles or women using OC, who had the lowest levels of NKA. No significant differences in blood concentration of NK cell were found. Differences in NKA were of similar magnitude in the whole and washed blood assays per milliliter of blood, per NK cell, or per lymphocyte. The menstrual cycle had no significant effect on activity levels of NK cells, but during the periovulatory phase, the number of NK cells per milliliter of blood increased significantly. CONCLUSIONS: The observed differences are independent of the presence of serum factors during the in vitro assessment of NKA, but may be related to chronic exposure to sex steroids and to fluctuation in the NK cell expression of beta-adrenoceptors.     

Uterine NK cell development, migration and function

Uterine natural killer (uNK) cells represent the predominant lymphocytes in the uterus during early pregnancy and in the secretory phase of the menstrual cycle. They are CD56(high)CD16(-) and have low cytotoxicity, but constitutively secrete a number of cytokines, chemokines and angiogenic molecules. uNK cells differ from CD56(high) blood NK cells in several ways, including the killer cell immunoglobulin-like receptor repertoire and expression of some genes induced by hormone environment. uNK cells may arise by in-utero proliferation and differentiation of NK cell progenitors under the control of the sex steroid hormones and/or cytokines, such as interleukin-15, and/or be recruited from CD56(+) blood NK cells that would undergo tissue-specific differentiation in the uterine microenvironment. There is evidence showing that uNK cells display a different pattern of chemokine receptors and adhesion molecules, thus leading to a different migratory response. It has not yet been fully defined which uNK cell function(s) are critical for successful pregnancy. The close encirclement of spiral arteries by NK cells, together with their ability to produce angiogenic factors, suggests that they might influence mucosal vascularization. Their proximity to the extravillous trophoblast supports the idea that uNK cells could recognize these cells as fetal, and regulate their invasion during placentation.     

Suppression of natural killer cell activity by sera from patients with endometriosis.

OBJECTIVE: We determined the effect of sera from patients who have endometriosis on natural killer cell activity. STUDY DESIGN: The natural killer cell activity of lymphocytes from healthy volunteers was examined after incubation with sera from patients who had endometriosis or from controls, with K562 cells used as targets. RESULTS: Lymphocytes treated with sera from patients who had endometriosis expressed significantly lower levels of cytotoxicity compared with lymphocytes treated with control sera. This suppression of cytotoxicity was dose dependent, and the degree of suppression was proportional to the incubation time of the effector cells with the sera. Decreased cytotoxicity after serum treatment was also observed with sera from patients who had been treated with danazol. CONCLUSIONS: These findings show that humoral factors that can inhibit natural killer cell activity in vitro are present in the peripheral blood of patients who have endometriosis; moreover, they suggest that the suppressed natural killer cell activity may allow the development of endometrial cells at ectopic sites.     

Decreased natural killer cell activity in women with endometriosis.

Natural killer (NK) cell activity is characterized by the spontaneous capacity of lymphoid cells derived from nonimmunized hosts to recognize and lyse certain tumor cell lines, virus-infected cells and transplanted tumor cell lines. Endometriosis is characterized by implantation and proliferation of autologous ectopic tissue in the pelvic cavity. Therefore, we focused on examining NK cell activity in women with endometriosis. NK cell activity in peripheral blood from women with endometriosis was lower than in women without endometriosis. On the basis of this finding, we analyzed also the effect of peripheral sera of women with endometriosis on NK cell activity. In the presence of peripheral sera of women with endometriosis, NK cell activity was significantly suppressed as compared with the sera of women without endometriosis. The suppressive effect of sera of women with endometriosis on NK cell activity showed dose-dependent curves. These studies provide the speculation that natural immunity mediated by NK cells may modulate the development of endometrial implants.     

Human seminal plasma suppresses lymphocyte responses in vitro in serum-free medium

The in vitro immunosuppressive properties of human seminal plasma have been re-investigated in serum-free medium in view of recent suggestions that the previously observed effects might be dependent on the presence of exogenous serum co-factors present in the culture media. The present studies reveal that low concentrations of seminal plasma can inhibit the ability of peripheral blood leukocytes to lyse K562 target cells in the absence of fetal calf or new-born calf serum. These inhibitory effects could be achieved by pre-incubating the effector cells in seminal plasma at 37 degrees C prior to use in the natural killer cell assay or by incorporating it into the assay system. Additional studies revealed that human seminal plasma could also inhibit the proliferative response of peripheral blood lymphocytes to phytohaemagglutinin in serum-free HB103 medium. These effects were most marked and consistent if the seminal plasma was present throughout the period of culture. Overall, these studies indicate that the previously reported suppressive effects of human seminal plasma in these systems cannot be entirely attributable to cytotoxic factors generated by exogenous serum components.     

Differential actions of progesterone and cortisol on lymphocyte and monocyte interaction during lymphocyte activation--relevance to immunosuppression in pregnancy

Progesterone and glucocorticoids share important anti-inflammatory and immunosuppressive properties. Both hormones have potent anti-proliferative effects in MLR, mitogen activation and cytotoxic T-cell generation. We investigated the cellular target of this in vitro anti-proliferative activity by comparing the effects of progesterone and cortisol on lymphocyte-monocyte interaction in concanavalin (Con A) induced human T-cell activation. Three different in vitro systems for assessing monocyte dependent T-cell activation by Con A were used: (1) limiting concentration of monocyte, (2) preincubation of isolated populations of monocytes and T cells with steroids and (3) role of steroid on action of Interleukin-1 (IL-1) activity. Monocytes separated from human peripheral blood leukocytes by flotation gradients and adherence to plastic were cultured at concentrations of 0.5-10% with constant numbers of isolated autologous T cells. Inhibition of Con A activation in cortisol (0.1-10 micrograms/ml) treated cultures was inversely proportional to percent monocytes, whereas in progesterone (2.0-20 micrograms/ml) treated cultures, inhibition was independent of monocyte concentration. Separated monocytes preincubated with progesterone and cultured with fresh T cells supported normal (108 +/- 7% control) levels of activation, but progesterone treated T cells and fresh monocytes responded at about 60% control levels. Similar experiments with cortisol (1 or 10 micrograms/ml) revealed significantly reduced responses when either cell population was preincubated with steroid. IL-1 induced by LPS stimulation of monocytes was blocked in its ability to stimulate Con A induced T cell proliferation with either steroid present during the assay of IL-1. These data provide additional support for local immunosuppression by steroids in the placenta during pregnancy. They suggest that progesterone selectively blocks T cell activation by a direct effect on T cells, whereas cortisol interferes with both monocytes and T cells.     

Effects of human serum on rat cumulus-oocyte complex functions in vitro: Oocyte activation and endocrine secretions

Secretion of hormones and oocyte meiotic events were assessed following in vitro culture of ovulated rat cumulus—oocyte complexes (COCs) in media containing different types of human serum. Both toxic and nontoxic (determined by mouse embryo test) samples of fetal cord or adult female serum were utilized for these experiments. After short-term culture (4.5 hr), with or without COCs, medium containing 10% serum was collected and analyzed for its content of estradiol, progesterone, and prostaglandin E (PGE), and oocytes were cytologically evaluated for spontaneous activation (second polar-body extrusion). Activation of oocytes occurred in all media tested. Steroids (progesterone and estradiol) levels were markedly elevated in culture medium containing cord serum as compared to medium containing adult female serum. The progesterone content of culture medium decreased after incubation of COCs with cord serum and increased when incubated with adult female serum. Little or no prostaglandin was detected in any control media. However, COCs secreted prostaglandin during culture in all media. COCs secreted estradiol when cultured in medium containing cord but not adult female serum. Results demonstrate that two types of serum utilized for in vitro culture of COCs varied markedly in their hormone content and differentially affected the secretion of hormones by COCs during culture. The results are discussed in relation to the success of IVF procedures.     

Relaxin in sera during the luteal phase of in-vitro fertilization cycles.

To identify the time when relaxin can first be detected in peripheral sera after in-vitro fertilization (IVF) and embryo transfer, blood samples were collected from 20 women up to 14 days after oocyte retrieval. Sixteen women did not become pregnant and in eight of them relaxin (but not beta-human chorionic gonadotrophin, beta-hCG) was measurable for the first time at days 6 to 12. Concentrations of other hormones measured were also different in these eight women compared with the remaining eight non-pregnant women; their serum concentrations of 17 alpha-OH progesterone, progesterone and oestradiol were higher but concentrations of luteinizing hormone and follicle-stimulating hormone were lower. Three women became pregnant; relaxin and beta-hCG were first detected on the same day (10 to 12). The remaining woman had increased beta-hCG levels but did not develop a clinical pregnancy. Measurement of serum relaxin during IVF cycles may allow assessment of corpora luteal function before its identification by levels of steroid hormones.     

Inhibition of natural killer cell activity by retroplacental sera as compared to peripheral sera. Lack of influence of immune complexes

A decrease of natural killer cell activity (NKCA) during human pregnancy might contribute to the acceptance of the allogeneic fetus by the maternal host. The inhibition of NKCA might be due to serum factors derived from the trophoblast. We focused especially on the role of immune complexes, as it has already been described that these complexes depress NKCA and as they are found frequently in retroplacental serum. We have compared the influence of 19 paired retroplacental and peripheral blood sera on NKCA of normal donors. One peripheral and eight retroplacental sera contained immune complexes. Normal donor mononuclear cells were incubated with carboxyfluorescein-labeled K562 cells in the presence of retroplacental serum or peripheral serum. NKCA was measured on a FACS Analyzer. Ten of 19 retroplacental sera inhibited NKCA significantly in comparison to the corresponding peripheral serum (P = 0.003). There was no correlation between NKCA and the immune complex level. We conclude that, as compared to peripheral serum taken at delivery, there is a retroplacental serum-induced inhibition of NKCA, which is not correlated with the presence of immune complexes.     

Sex differences in correlation coefficient among the cord serum levels of LH-hCG, beta-hCG, FSH, estradiol cortisol and testosterone (author's transl)

Lutenizing hormone (LH-hCG), follicle stimulating hormone (FSH), beta-human chorionic gonadotropin (beta-hCG), estradiol, cortisol and testosterone were determined and correlated with each hormones in 62 cord sera (32 male and 35 female infants). Mean (+/- S.E.)male cord sera cortisol concentrations (36.2 +/- 3.9 ug/dl) were significantly (p less than 0.05) higher than that of female (24.1 +/- 3.7 ug/dl). Another hormones were not found significant sex difference. Regression analysis showed significant positive correlations between LH-hCG (r = 0.525, p less than 0.001) or beta-hCG (r = 0.461, p less than 0.005) levels and FSH levels in all cord sera (N = 67). In male cord sera, there were significant positive correlations between LH-hCG (r = 0.493, p less than 0.01) or beta-hCG (r = 0.485, p less than 0.01) levels and testosterone levels. There was a significant positive correlation (p less than 0.01) between estradiol levels and testosterone levels in female cord sera. These data suggest (1) the sex difference of cortisol levels which indicate the response for the stress during labor and delivery, (2) there was a significant sex difference in the maturation of the feed back mechanism in pituitary-gonadal axis.     

Changes in NK cell activity during the estrous cycle and pregnancy in mice

Changes in NK cell activity during the estrous cycle and pregnancy in mice were studied by a 4 h 51Cr-release assay using YAC-1 cells as a target. NK activity both in the spleen and peripheral blood mononuclear cells showed a single fluctuation with a peak at met-estrous-2 during the estrous cycle. Splenic NK activity was suppressed in the early to mid-stages of pregnancy but increased sharply in the late stage. The activity declined thereafter, reaching the estrous level post-partum. NK activity in the peripheral blood showed a decrease throughout the entire pregnancy, but increased on Day 1 post-partum, returning to the estrous level thereafter. To elucidate the mechanisms involved in these variations of NK activity, NK-enriched and non-adherent cells were prepared from the spleens of mice at estrous and met-estrous-2 and those at early and late stages of pregnancy and then examined for NK activity. The results showed that there was no difference in the cytotoxic activity among these purified NK cells. Adherent cells purified from the spleens of mice in the early stage of pregnancy when co-cultured with the non-adherent NK cell fractions suppressed the NK cytotoxicity. These results strongly suggest the possibility that phagocytic and/or adherent cells may be involved in the regulation of splenic NK activity during the estrous cycle and pregnancy.     

Immunochemical investigations on the protein of the "pregnancy zone". X. Demonstration of immunosuppressive activities of the pregnancy-associated alpha2-glycoprotein (pregnancy zone protein) by means of the macrophage electrophoretic mobility test (author's transl)]

By means of the macrophage electrophoretic mobility (MEM) test the supposed immunosuppressive activity of the pregnancy-associated alpha2-glycoprotein (pregnancy zone protein) could be demonstrated significantly. Pregnancy sera and sera of nonpregnant women with hormonal contraception containing this pregnancy protein showed an immunosuppresive activity, too. On the contrary, serum of male donors with the blood group AB and human serum albumin had no inhibitory effects. The results suggest that the pregnancy-associated alpha2-glycoprotein rather influences the afferent limb of the immune response and does not interfere with the activity of the mediator (migration inhibiting factor).     

Progesterone as an immunologic blocking factor in human pregnancy serum

The influence of estriol and progesterone on lymphocyte reactivity was examined by testing the cytotoxic effect on human embryonic fibroblast cells of non-pregnant women's lymphocytes incubated with different concentrations of estriol and progesterone and with complete and progesterone-depleted third trimester pregnancy sera. Progesterone, but not estriol, had a significant inhibitory effect on lymphocyte cytotoxicity at concentrations comparable to those present in pregnancy serum. 95% depletion of progesterone from pregnancy sera caused an 80% loss of inhibitory activity on lymphocyte cytotoxicity. These data suggest that the blood level of progesterone in pregnancy is sufficient to depress lymphocyte reactivity and that progesterone is responsible for the greater part of the serum inhibitory activity in at least the later stages of pregnancy.     

Level of S100B protein expression in the amnion at various gestational ages in the third trimester of normal pregnancies

BACKGROUND: S100B protein is a unique calcium-binding protein. Its biological role within the cell populations is not completely defined. Some pathological conditions that develop during pregnancy could affect S100B concentrations in the amniotic fluid, cord blood, and maternal serum. The aim of our study was to assess the correlation between S100B protein expression in the amnion, amniotic fluid and gestational age in the third trimester of uncomplicated pregnancies. METHODS: Amnion, amniotic fluid, maternal peripheral and umbilical cord blood samples were collected from healthy women who delivered at 31-36 weeks (n=17), 37-40 weeks (n=22), and 41-42 weeks (n=21). The expression of S100B in the amnion was assessed by immunohistochemistry and real-time (RT)-PCR, and its concentrations in amniotic fluid, maternal and cord blood sera were determined by ELISA. RESULTS: The S100B protein expression in the amnion and its concentrations in amniotic fluid, maternal and cord blood sera of patients in the third trimester were not significantly different at various gestational ages. CONCLUSIONS: The S100B protein expression in the amnion and the S100B protein concentrations in amniotic fluid, maternal and cord blood do not vary significantly in the third trimester of uncomplicated pregnancies.     

Cytotoxic activity and phenotypic analysis of natural killer cells in early normal human pregnancy

Peripheral blood lymphocytes taken from healthy women planning a pregnancy and then at various intervals up to the 16th week of pregnancy were assayed for natural killer (NK) cell activity against K562 target cells both in a 51Cr-release assay and in a single cell cytotoxicity assay. Results indicated a depression in NK activity from the earliest stages of pregnancy. The target binding capacity of the effector cells remained unimpaired up to 16 weeks, but a significant reduction in the post-binding lytic potential was observed, which parallelled the drop in cytotoxicity as assayed by the 51Cr-release method. The ability of individual effector cells to recycle and kill multiple targets remained essentially unimpaired. Analysis of lymphocyte subpopulations using the monoclonal antibodies anti-Leu-7 and anti-Leu-11b, which recognize NK cell-associated antigens, showed a significant reduction in the proportion of the mature, lytically active Leu-7-11+ cells capable of both binding and lysing K562 target cells. The suggestion that the depression in cytotoxicity may be associated with the reduction in the Leu-7-11+ subpopulation is supported by the high correlation levels observed between the proportion of Leu-7-11+ cells and target cell lysis.