Depression of blast transformation of peripheral leukocytes by plasma from pregnant women

Pregnancy is associated with suppression of the nuclear factor kappaB/IkappaB activation pathway in peripheral blood mononuclear cells

Modulations of maternal immune cell function are critical for successful growth and development of an antigenically distinct fetus. It has been proposed that pregnancy is associated both with suppression of the adaptive immune system and a generalised maternal inflammatory response with changes in immune function resembling those associated with septicemia, and these changes are more exaggerated when pregnancies are complicated with pre-eclampsia. The nuclear factor (NF)-kappaB family of transcription factors play a significant role in immune regulation. We hypothesised therefore that if pregnancy is associated with activation of the maternal immune system, this would be supported by the activation of NF-kappaB and degradation of IkappaBalpha and beta in peripheral blood mononuclear cells (PBMCs). We demonstrate the contrary: NF-kappaB activity is suppressed in PBMCs from pregnant females and more in pre-eclampsia. The inhibition of NF-kappaB activation in pregnancy is not attributed to over-expression of IkappaBalpha or beta. In contrast, levels of IkappaBalpha and beta in cytoplasmic extracts from PBMCs in pregnancy are decreased compared with non-pregnant controls, and IkappaBalpha levels are decreased more so in pre-eclampsia. We have shown that activation of NF-kappaB in PBMCs from patients with septicemia follows the classical pathway. This pathway is differentially regulated in pregnancy. Alterations in NF-kappaB nuclear binding and IkappaBalpha levels were reproducible by culturing PBMCs in pooled pregnant serum. Taken together, these data indicate that pregnancy-specific factors exist to regulate expression of NF-kappaB/IkappaB in a pregnancy-specific manner, and may underlie one mechanism by which the fetus avoids maternal rejection throughout pregnancy.

Immunology of pregnancy: towards a unifying hypothesis

The variable findings of hormonal-immunoregulation and the variable cellular and humoral immune responses in pregnancy have been considered in relationship to the physiological response. From such considerations it appears that the peripheral blood lymphocyte/leukocyte response in pregnancy is not important, but rather the local uterine immune response at implantation and throughout pregnancy. It is proposed, and evidence is presented, that a normal allogeneic immune response is initiated at the time of implantation of the blastocyst. This immune response regulates the invasive nature of the trophoblast and initiates the first stage of parturition. The initiation and maintenance of this immune response is based on an interplay between maternal and paternal HLA and trophoblast antigens. In the case of HLA-incompatible donor-recipient blastocyst transplants, a more pivotal role for immunoregulation by trophoblast antigens is proposed. This is because it is considered that the local uterine immune response suppresses the expression of allogeneic HLA. This concept is further developed in terms of haploid HLA suppression on maternal and fetal lymphocytes that cross the placenta. This is considered to allow the interaction of these lymphocytes with each other and explains maternal transfer of cell-mediated immunity.

Glycoprotein glycosylation and the immunosuppressive effects of human pregnancy serum

Pregnancy serum contains a factor or factors which suppress T lymphocyte proliferation, although the identity of the factor(s) is still unclear. We have demonstrated that the immunosuppressive activity of pregnancy sera can be destroyed by treatment with periodate which oxidises protein-linked oligosaccharides. Similar effects have been noted with uromodulin, a potent immunosuppressive glycoprotein initially isolated from pregnancy urine. We find, however, that uromodulin is present in both pregnancy and non-pregnancy sera, and that removal of uromodulin from pregnancy serum by lectin affinity chromatography is not associated with loss of activity, ruling out this glycoprotein as the immunosuppressive factor. The possible role of protein-linked oligosaccharides of other serum glycoproteins in causing the pregnancy-related immunosuppression is discussed.

Detection of immunoregulatory factors in retroplacental serum in human pregnancy

Retroplacental serum (RPS) obtained from pregnant women at term deliveries was studied for regulatory effects on T-lymphocyte proliferation and for pregnancy-associated substances and compared with peripheral serum (PS) of the same donor. Proliferative response to phytohemagglutinin and alloantigens in RPS was lower than that in PS. RPS contained higher levels of human placental lactogen, progesterone, estradiol, and prostaglandin E2 than of PS. However, there were no differences in concentrations of pregnancy-associated alpha 2-glycoprotein, pregnancy-specific beta 1-glycoprotein, prostaglandin F2 alpha, alpha-fetoprotein, human chorionic gonadotropin, cortisol, carcinoembryonic antigen, and immunoglobulins between RPS and PS. The amounts of human placental lactogen, progesterone, or prostaglandin E2 seen in RPS did not inhibit T-cell proliferation. Mixtures of various doses of these three substances were still not inhibitory. Thus, the suppressive activity of RPS could not be explained by these pregnancy-associated substances, but a possible involvement of unknown immunoregulatory factors at fetomaternal interface might be suggested.

Maternal immune responses to the fetus in early pregnancy and recurrent miscarriage

Direct evidence that maternal immune rejection of the fetus causes some unexplained recurrent spontaneous abortions was sought in 18 women with this condition. Tests of maternal cell-mediated immunity to fetal (paternal) antigens were done before conception, in early pregnancy, and at miscarriage, and were compared with those in 10 controls in their first pregnancies. Maternal cytotoxic alloantibody production and the blocking effect of maternal sera on maternal lymphocyte activation were also evaluated. There was no evidence for maternal cell-mediated reactivity to paternal antigens in normal early pregnancy or in most women who aborted, but circulating cytotoxic cells were found at miscarriage in a third of affected women. There was no correlation between the production of cytotoxic antibodies and serum blocking activity, and the success of the pregnancy. These data provide evidence that cell-mediated immune reactivity may be changed in some women who abort recurrently but show that circulating immunological blocking factors are not relevant to the success of pregnancy. Their induction by maternal immunisation with paternal leucocytes does not explain why this procedure prevents recurrent spontaneous abortions.

The effect of human pregnancy serum on the synthesis and action of interleukin-1

Supernatants from mixed lymphocyte reactions performed in the presence of pregnancy serum (PS) or normal human serum (NHS) were tested for interleukin-1 (IL-1) activity in a thymocyte assay. A lower response was obtained with PS-containing supernatants, suggesting that PS exerts its suppressive effect by inhibiting IL-1 synthesis or by blocking its mode of action. IL-1 activity generated by human peripheral blood mononuclear cells (PBMC) stimulated with LPS and cultured in the presence of PS or NHS was similar. However, when PS was present, the proliferation of thymocytes co-mitogenically stimulated with PHA and exogenous IL-1 was diminished compared with the control (NHS), exhibiting its maximum inhibition when PS was added at the highest concentration and at the beginning of the incubation period. The suppressive effect of PS in the thymocyte assay disappeared with the combined action of exogenous IL-1 and IL-2. The response of T lymphocytes allogeneically stimulated was decreased in the presence of PS even when IL-1 was added; the recovery of proliferation being strictly dependent on the presence of the appropriate dose of exogenous IL-2. These results indicate that PS was interfering with IL-1 activity by rendering IL-2-producer cells unable to synthesize IL-2 and subsequently to proliferate.

Effect of pregnancy plasma and monoclonal antibodies to HLA and DR antigens on leukocyte migration

We studied the effect of pregnancy plasma and monoclonal antibodies to the histocompatibility antigens HLA and DR on migration of normal leukocytes from capillaries. It was observed that plasma from women in the first trimester of pregnancy significantly enhanced leukocyte migration at a concentration of 10% (mean area of migration = 50.7 +/- 9.2 cm2), as compared with plasma from nonpregnant women (22.8 +/- 7.0 cm2; p less than 0.05). This effect was less during the second trimester (39.5 +/- 3.5 cm2; p less than 0.05) and no enhancement was noted with plasma obtained from the third trimester of pregnancy. Similar results were obtained with a 20% concentration of plasma also. On the other hand, the monoclonal anti-HLA and anti-DR antibodies had an inhibitory effect on migration of leukocytes. These results are discussed in relation to the immunoregulatory role of pregnancy plasma in the nonrejection of the fetal allograft.

Induction of suppressor cell activity by human pregnancy serum

Third trimester pregnancy serum caused a dose-dependent inhibition of the one-way allogeneic mixed leukocyte reaction (MLR) through an effect that occurred during the first 48 hours of culture. Pregnancy serum also inhibited the mitogenic responsiveness of normal mononuclear leukocytes to concanavalin A (Con A) while the responses to phytohemagglutinin (PHA) and pokeweed mitogen (PWM) were less affected. Preincubation of normal peripheral blood mononuclear cells (PBMC) for 48 hours with 10% pregnancy serum enhanced a suppressor activity transferable with cells. These pregnancy serum-induced effector cells suppressed the MLR only when they were autologous to the responder population (p less than 0.05). The same suppressor cell preparation inhibited the proliferative responses of autologous PBMC to Con A (p less than 0.001) and PWM (p less than 0.05). These data suggest that one or more factors in pregnancy serum can induce or enhance suppressor cell activity in vitro. A comparable increase in suppressor cell activity in vivo may be responsible for blocking maternal rejection of the fetus and for the observed improvement in clinical activity of rheumatoid arthritis during pregnancy.

A longitudinal study of cell-mediated immunity in human pregnancy

Studies of cell-mediated immunity, comprising delayed hypersensitivity skin testing with four antigens, PHA and antigen-induced lymphocyte transformation and detection of the lymphokine, monocyte chemotactic factor, were undertaken in 14 women during pregnancy and in the post-natal period. Skin test reaction sizes to SKSD, PPD, Candida albicans extract and tetanus toxoid were not significantly different during pregnancy. In vitro lymphocyte transformation induced by PHA, SKSD and Candida albicans was enhanced during pregnancy. Monocyte chemotactic factor production was not significantly different during pregnancy. These in vitro tests were performed in the absence of autologous sera. These studies indicate that cell-mediated immune responses are normal during human pregnancy.

Association between pregnancy-associated alpha2-glycoprotein (alpha2-PAG) and mixed leucocyte reaction determinants on the leucocyte surface

alpha2-PAG is present on the surface on mononuclear blood leucocytes and can be demonstrated predominantly on B-lymphocytes and monocytes. Pretreatment of cells with antibody to alpha2-PAG leads to a marked reduction in Fc-rosette formation. Competitive blocking experiments with specific antisera reveal a particularly close association between alpha2-PAG and MLR (mixed leucocyte reaction) determinants on the cell surface. These findings suggest one mechanism whereby alpha2-PAG may modify cell-mediated immune responses.