Chapter 6.
The Fetal Immune System in
and Under Pathological Effects
6.1) The common immune system in human fetuses
In mammals, the immune system begins its maturation early in fetal life. In humans, for example, B lymphocytes have developed in the liver by 9 weeks of gestation and are present in the blood and spleen by 12 weeks (1). T lymphocytes start to leave the thymus from about 14 weeks of gestation and subsequently, cells with helper and suppressor phenotypes appear in the spleen. The relative lack of development of secondary lymphoid tissues in healthy fetuses most probably reflects a lack of antigenic stimulus.
Although there is considerable exchange of materials, the fetus remains largely separated from the mother's tissues. The fetus is inside the mother's uterus but is never in direct physical contact with the uterine walls. Two membranes—the trophoblast (chorion) and the amnion—surround the human fetus, and the inner space is filled with the amniotic fluid. Only a thin tube—the umbilical cord—penetrates these membranes to connect the fetus and the mother. Physiological exchange between mother and fetus occurs only at the interface where the umbilical cord fuses with the uterine walls. This interface, where fetal tissues and maternal tissues interact, is called the placenta.
Different genes are expressed in the trophoblast (fetal part of the placenta) and amniotic membrane. For example, HLA-A and B class I genes are downregulated in human trophoblast cells, whereas non-polymorphic class I molecules, e.g., HLA-G class Ib, are expressed in the extravillous cytotrophoblast and also in the endothelial cells of fetal vessels in the chorionic villi as well as in the amnion cells and amniotic fluid (2). HLA-G presents antigens for γ/δ T cells and at the same time defends the trophoblast from cytotoxic effector mechanisms. Since polymorphic major histocompatibility (MHC) is absent from the trophoblast, presentation of fetally derived antigens is unlikely to be MHC-restricted. Most γ/δ T cells recognize unprocessed foreign antigens without MHC.
In the decidua (maternal part of the placenta), γ/δ TCR-positive cells significantly increase in number, and most of these cells are in an activated form due to recognition of conserved mammalian molecules on the trophoblast (2). Following recognition of fetally derived antigens, the maternal immune system reacts with the setting in of a wide range of protective mechanisms.
Many observations suggest that a successful pregnancy is associated with an altered TH1/TH2 balance, an important prerequisite to the maternal immune system not rejecting the fetus (3-5). Maternal immune response is biased toward humoral immunity and away from the cell-mediated immunity which could be harmful to the fetus. Cytokines of maternal origin act on placental development. On the other hand, antigen expression on the placenta determines maternal cytokine pattern. Increased concentrations of proinflammatory cytokines in the amniotic fluid indicate the presence of intra-amniotic or placental inflammation and increase the risk of preterm birth, cerebral palsy, and bronchopulmonary dysplasia (6,7)
Normal human pregnancy is characterized by low peripheral natural killer (NK) activity, and increased NK activity seems to play a role in spontaneous abortions of unknown etiology (2). In early human pregnancy, most uterine lymphocytes are CD56 granulated NK cells, which do not express CD16 or CD3 (8). In early pregnancy, uterine NK cells become enriched at sites where the fetal trophoblast infiltrates the decidua (9). The dynamics of the appearance of uterine NK cells suggests that one of the functions of these cells is control of placentation (10,11).
Another protective mechanism operating in favor of pregnancy is progesterone-dependent immunomodulation (12). The biological effect of progesterone is mediated by a 34-kDa protein termed progesterone-induced blocking factor (PIBF). PIBF, synthesized by lymphocytes of healthy pregnant women in the presence of progesterone, inhibits the release of arachidonic acid and NK activity, and modifies the cytokine balance. PIBF supports TH2 cytokines, inhibits NK cells, and induces an increased production of non-cytotoxic blocking antibodies (13,14). The production of pro-inflammatory, cytotoxic cytokines, such as INFγ, TNFą, and IL-2, is reduced.
In the fully developed placenta, the fetal-derived trophoblast forms fingerlike villi that penetrate and intermingle with the surface layer (endometrium) of the uterus (15). The fetal circulation extends down the umbilical cord and branches into the capillaries inside these villi. The villi are surrounded by a network of intervillous spaces, and the mother's endometrial arteries fill these spaces with blood. Endometrial veins remove this blood. As a result, maternal blood flows continuously around the villi, and they are the sites for the exchange of materials between the fetal and maternal circulatory systems. The mother's circulatory system is not continuous with that of the fetus. Blood does not normally flow from the mother to the fetus and back; only materials carried in the blood are exchanged. Therefore, maternal blood cells such as B lymphocytes are not normally transferred to the fetus, although the antibodies produced by B lymphocytes do cross the placenta (16-19). This separation of circulatory systems is very important for immunological reasons. It is known that half the fetus's genes come from the mother and half from the father. The father's genes are "foreign" to the mother, and this difference is potentially sufficient to trigger an immune response. The separation of the mother's and fetus's tissues and blood reduces the likelihood that the maternal immune cells will encounter fetal cells and launch an attack against the fetus.
As for the placenta itself, although white blood cells such as T lymphocytes and NK cells are plentiful in the endometrium, they do not react against the villi of the fetal chorion (20-22). The reason for this is not completely known, but it appears that proteins on the surface of the villi keep them safe.
It is known that fetal cells occasionally enter the maternal circulatory system, and the result is documented medical tragedy (23,24). The transfer of blood cells in the other direction, from mother to fetus, does not have the same effect, because the fetal immune system is not fully developed and cannot respond to any foreign antigens carried by maternal cells. In addition, the few maternal blood cells that may leak through the placenta to the fetus are not enough to launch an immune response against fetal antigens (25). Recent studies have shown that certain ante-partum conditions, such as placental insufficiency or chorioamnionitis, significantly increase the chances of maternal cells entering the fetus (26,27).
Studies on the role of the maternal immune system during pregnancy have focused mainly on the aspect of immune tolerance to the invading trophoblast and, therefore, embryo. While this is a critical aspect of reproductive immunology, it is also important to consider the function of the fetal immune system in the promotion of implantation and maintenance of pregnancy (28). The process of embryonic development is accompanied by the phenomenon of programmed cell death or apoptosis (29,30). Apoptosis is not the final stage in tissue development. The quick and effective removal of apoptotic cells by tissue macrophages is vital preventing "leakage" of self-antigens and promoting the production of proliferative/survival factors. One of the key requirements of apoptotic cell clearance is the resolution of inflammatory conditions, which, as in the case of pregnancy, may have lethal consequences (31).
6.1.1) Immune components in experimental fetal pathology
The role of the immune system in fetal pathology has been studied in many eperiments with laboratory animals. In nude mice, for example, it has been demonstrated that the bone-marrow-derived non-lymphoid thymus cells, most likely the Ia-antigen-positive thymic macrophages of dendritic cells, are responsible for the induction of tolerance to MHC antigens in developing T lymphocytes (32). The administration of drugs during pregnancy may result in potential long-term effects on the developing immune system. Offspring of such pregnancies may suffer an increased incidence or severity of autoimmune diseases as a result of placental transfer of deleterious agents (33).
Ovine abortions caused by Chlamydophila aborts resulted in the appearance, in the placental tissue, of a high number of cells expressing the macrophage-associated molecule CD14 and cells expressing MHC class II molecules (34). Many cells expressing mRNA encoding for tumor necrosis factor-alpha (TNFą) were observed. The fetal immune response included small numbers of CD4+ and CD8+ cells, γ/δ T cells and B cells. Production of TNFą by fetal macrophages expressing MHC II molecules may be of considerable significance in the pathogenesis of abortion.
Helicobacter pylori infection can induce activation of resident uterine immune cells and/or recruitment of cells at the endometrial level. H. pylori infection of pregnant mice was accompanied by macrophage activation in the endometrium, and by an increase in the number of CD4+ and CD8+ lymphocytes and of INFγ and MHC II expression (35). During pregnancy, preferential induction of Th2-type cytokines down regulates Th1-type responses, allowing fetal survival.
Microbial infections of pregnant Sprague-Dawley rats with Mycoplasma pulmonis chorioamnionitis were accompanied by accumulation of neutrophils in the capsular decidua, elevated mRNA levels of TNFą and IL-6 in the placental tissues, and the secretion of TNFą by the placenta during late gestation (36,37). Experimental inoculation of cattle with the parasite Neospora caninum in early gestation caused an immune response in the placenta (38). Pathological changes in the placenta consisted mainly of CD3+ lymphocytes, dominated by CD4+ and γ/δ TCR+ cells, with CD8+ cells present to a lesser extent. It is possible that a pro-inflammatory Th1 response early in gestation leads to destruction of the placental tissues themselves and is thus incompatible with fetal survival.
A low number of T cells (at most, 10% of those seen in the adult airways) was found in different parts of the respiratory tract in horse fetuses (39). The low level of MHC class II expression in the fetus, together with the reduced number of T cells, was consistent with the suggestion that the fetal immune system requires exposure to airborne stimuli for the full development. The low level of MHC II expression in the mare may have been reflecting the immunosuppression that accompanies pregnancy.
6.1.2) The immune components in human fetal pathology
Characteristics of the fetal immune system under pathogenic effects are very similar between experimentally caused pathology and that which develops as a result of moternal or fetal illness. Immunosuppressive and other drugs administered to mothers during pregnancy and lactation might affect the development of the fetal and neonatal immune system (40,41). Chorioamnionitis or intrauterine fetal pneumonia caused by Chlamydia trachomatis (42) or Candida colonies (43) or other infections can be considered examples of fetal disorders in humans. Fetuses younger than 13 weeks showed no inflammatory response or cells positive for Igs and proliferating Candida colonies which were evident in the lungs. The 16- and 22-week-old cases revealed a unique giant cell response in the terminal airways and increasing numbers of Ig-positive cells, with an increased proportion of IgA-positive cells in the older cases.
Severe chorioamnionitis is associated with a nonspecific inflammatory response comparable to that of neonatal sepsis (44) characterized by shrinkage of the thymus and spleen depletion, involving both B and T lymphocytes (45). Chorioamnionitis accompanied by an intrauterine inflammatory response of the fetal lungs is characterized by a severe infiltration of macrophages, neutrophils, and lymphocytes, and by sharply increased expression of IL-8 mRNA (46). Chorioamnionitis increases fetal intrahepatic myelopoiesis as one defense mechanism and induces a fetal extramedullary hematopoietic response in the second trimester of gestation (47). Fetal myelopoiesis significantly increases with leukocyte clustering.
One of potentially life-threatening disease for the both mother and fetus is pre-eclampsia, during which increased production of chemokines and leukocyte activation have been described in the fetal circulation (48). The activation of neutrophils and monocytes in the fetus involves enhanced chemokine activation, possibly contributing to the fetal morbidity in this disorder. The activation of neutrophils and monocytes is accompanied by raised plasma levels of the chemokines IL-8 and growth-related oncogene-ą. The NK cell counts of umbilical blood in preeclampsic fetuses, as well as the proliferative and killing abilities of these cells, are significantly increased (49).
In fetuses with Down's syndrome, a statistically significant depletion in the total number of CD3+ T cells and a significant increase in the CD8/CD4 ratio during the second trimester of gestation has been found (50). The increased number of B cells along with primary follicles in fetuses with Down's syndrome, implies that at least part of the thymic medulla works and behaves like a peripheral lymphoid organ, receiving mature lymphocytes and turning them from inactive to immunoefficient cells (51). The inflamed thyroid gland was shown to be capable of accumulating fetal cells, including T cells and dendritic cells (52 ). Such active immune cells may have a profound regulatory influence on autoimmune thyroiditis in pregnancy and the postpartum period.
An important cause of respiratory distress in newborn infants is meconium aspiration syndrome (MAS) (53). Approximately 12% to 15% of human infants are born through meconium-contaminated amniotic fluid (54), and these infants are much more likely to develop respiratory distress and require respiratory support (55). When meconium is present in the amniotic fluid, about 5% of neonates will develop MAS, and about 5% or more of these infants will die (54). The pathophysiology of MAS involves airway obstruction, surfactant dysfunction, and pulmonary inflammation (56). Meconium may interfere with the function of alveolar macrophages by decreasing their phagocytic activity (57) and inducing oxidative stress and apoptosis (58).
Rheumatic autoimmune diseases are high prevalent in women, particularly during their childbearing years. If the maternal disease is characterized by the presence of IgG-isotype autoantibodies, they can cross the placenta and potentially cause antibody-mediated damage to the fetus (59). This is typically the case in the so-called neonatal lupus erythematosus. A similar mechanism has been shown in infants of patients with immune thrombocytopenic purpura and, less frequently, in those from mothers with anti-phospholipid syndrome.
6.2. The secretory immune system in human fetuses
6.2.1) Lymphoid-epithelial components of the secretory immune system in self-protection of human fetuses
Some researchers are of the opinion that the SIS evolves after birth as a reaction to massive microbial invasion and the introduction of large amounts of different foreign antigens through the mucous membranes of the upper respiratory, digestive and urogenital tracts (60,61). However, as has been shown by the other authors and stated in previous chapters of this book, components of the SIS, such as SC, J chain, IgM, IgA and lymphocytes, are already present not only in human fetuses in the third trimester of gestation (62,63), but also in embryos of the first trimester (64,65).
Maturation of the immune system is well known to start early in human fetal life (65). As noted at the beginning of this chapter, B lymphocytes develop in the liver by 9 weeks of gestation and are present in the blood and spleen by 12 weeks. From 14 weeks, T lymphocytes leave the thymus, and subsequently cells with helper and suppressor phenotypes appear in the spleen. The lack of secondary lymphoid tissues in healthy fetuses most probably reflects a lack of antigen stimulus. On the other hand, newborn plasma contains adult levels of IgG which are acquired across the placenta from the mother.
In human fetuses, components of the SIS have been found in the epithelium of the salivary glands and mouth (66,67), trachea and lungs (68,69), and digestive tract (70,71). Small amounts of SC appear in the intestinal mucosa before week 29 of gestation, and its quantity increases rapidly thereafter (72). Secretory IgA-containing epithelial cells have been found in the respiratory tract and intrahepatic bile ducts of fetuses at 20 to 21 weeks of gestation (73). Immunocompetent mucosa-associated cells (dendritic cells, T lymphocytes, B lymphocytes, and macrophages) have been found in the human fetal larynx after week 14 of gestation (74). SC of fetal urogenital origin has been found in the amniotic fluid (75). Lymphoid cells expressing IgA and IgM, as well as other immunocompetent cells, have been described in fetuses of the second trimester of gestation (76), particularly, in the gut and amniotic fluid (77). From 11 to 14 weeks of gestation, CD68+ and CD40+ cells are present throughout the lamina propria. With the emergence of lymphoid aggregates (14-16 weeks), dendritic cells and B lymphocytes are detected in the fetal gut; however, their expression is restricted to the lymphoid aggregates. Lymphoid follicles forming after 16 weeks of gestation contain MHC II-positive cells of different subtypes of T lymphocytes.
Components of the SIS during the second trimester of gestation (weeks 13-25) were studied in 36 human fetuses obtained as a result of medically or socially recommended abortions (group I) or which had died of different causes -- abruptio placentae, placenta previa, and chorioamnionitis (group II) (78). The first group included 21 cases with no signs of foreign antigenic influences. The second group included 15 cases of acute chorioamnionitis with sepsis, aspiration syndrome or meningitis. In both groups, fetuses were of similar gestational ages.
Fetuses without antigenic effects (group I)
Elements of the SIS were widespread in the different organs of these fetuses. SC, J chain, IgA, IgM and IgG were found in the epithelium and glands of the digestive organs such as the mouth cavity, pharynx, esophagus, stomach and intestine, throughout the respiratory tract (larynx, trachea, lungs) and urinary tracts (kidneys, ureters, bladder), in hepatocytes and the epithelium of the bile duct, and in acini and ducts of the pancreas, in the follicular epithelium of the thyroid and the ovaries, in the epithelium of the Fallopian tubes and uterus, in the epididymis and rete testes, in the epithelium of the choroid plexuses of the cerebral ventricles, in the mesothelium of the pleura, epicardium and peritoneum, and in the epithelium of the skin, sweat and sebaceous glands. A few macrophages, B cells and different subsets of T lymphocytes were observed in these structures. All of these SIS elements were already observed in 13-week-old fetuses and were maintained during the whole second trimester of gestation.
The immunoreactivity of the different SIS components varied in different organs. The SC and J chain usually reacted intensively, except for cells in the distal part of the hypophysis and in pancreatic islands, where the SC was weakly reactive. In Leydig's cells of the testes and some other organs, where only the J chain was found, the SC was absent. Expression of IgA, IgG and especially of IgM, was very low in the large intestine and hepatocytes in the center of liver lobules. IgM was sometimes absent in the choroid plexuses of the brain. SIS components were not seen in the gray substance of the brain, myocardium, intestinal goblet cells, skeletal muscles, fibroblasts, chondrocytes or osteoblasts. A small number of B lymphocytes expressing IgA and IgM was seen in different organs (Table X).
Fetuses with antigenic effects (group II)
In fetuses which had been subjected to massive antigenic effects at chorioamnionitis, expression of the SC and J chain was no different from that observed in group I (Fig. 7). Immunoreactivity of IgA, IgG and especially IgM was weak or even absent in the epithelium of the skin, respiratory, digestive and urinary tracts, hepatocytes, tubules of the kidneys, and choroid plexuses of the brain. The number of IgA- and IgM-positive lymphocytes increased in the spleen and lymph nodes, lungs, and in the mucous membranes of the stomach and intestine (Table X).
It has been shown that the whole SIS-protein-complex is already present in many tissues of the human fetus in the second trimester of gestation (78). Such tissues include mainly the widely spread border tissues covered with the epithelium or its analog, such as the mesothelium of serous cavities. Our findings of the SIS protein components in the fetal respiratory, digestive and urogenital tracts are very similar to those observed in children and adults (63,79).
In fetuses, there are cellular components of the SIS, such as lymphocytes of different subsets, including IgA- and IgM-containing B cells, and macrophages. However, their amounts are small (Table XI), and they are diffusely located in subepithelial tissues and intraepithelial spaces. Special lymphoid-epithelial structures, such as tonsils, solitary follicles (nodules) and aggregated follicles (Peyer's patches), which are typical of the adult SIS (62), were not found in second-trimester fetuses. Such peculiarities in the cellular structure of the fetal SIS as part of the common lymphoid system of fetuses can be explained by its "immaturity".
The total mass of lymphoid tissue in second-trimester fetuses is minute. The total weight of the main lymphoid organs such as the spleen and thymus is 0.092±0.02% of mean body weight in 13- to 15-week-old fetuses, 0.36±0.07% in 23- to 25-week-old fetuses, and 0.73±0.09% in 38- to 40-week-old fetuses. Lymph nodules are few and are of microscopic size. Some lymphoid tissue structures are absent. Reactive centers of lymph nodules are not seen, even in the infected fetuses. The process of transformation of Ig-synthesized B lymphocyte stops at the immunoblast stage, and these never become mature plasma cells in cases with acute infections (80).
The massive antigenic effects under chorioamnionitis (group II fetuses) caused a two- to fourfold increase in the number of IgA- and IgM-synthesizing lymphocytes in the spleen, lymph nodes, and respiratory and digestive tract organs where SIS is localized and where the antigenic effect is manifested at its highest level (Table X). As a result, Ig synthesis was activated in the common immune system and in the SIS. The immunoreactivity of IgA, IgM and IgG in the epithelial cells sharply decreased or disappeared altogether. Sometimes, IgA and IgG immunoreactivity was seen in the fibrin in the bronchial cavity or in the stomach mucus as a manifestation of the exocrine secretion of Igs by the epithelium. The described changes could be considered a morphological manifestation of SIS functional activity in human fetuses in the second trimester of gestation.
The components of the fetal SIS, such as IgA and IgM, are of fetal origin (60). Maternal IgG has been found in the fetal epithelium, and this is considered an additional proof of insufficient functional activity of both the common and secretory fetal immune systems (78). Under massive antigenic attack, immunoreactivity of Igs in the fetal epithelial cells decreases to complete disappearance, whereas reactivity of the SC and J chain does not change (Fig. 7). It can be supposed that SC and J chain do not leave the cells during SIS function, and their discovery in the cavities of some organs (such as the amnion, 75) may result from cell shedding and destruction. Cells without exocrine secretion in some organs, such as the distal lobe of the hypophysis, pancreatic islets, etc., contain only the J chain without SC. Our study demonstrated that the SIS is widely present and functionally active in fetuses in the second trimester of gestation. It is not restricted to mucosal membranes, it is present in the fetal organs (Fig. 8) and it plays an important role in the immune defense of the entire fetuses and its strategically important organs against foreign antigenic influences.
(Color Figs.)
Fig. 7.
A. A 22-week-old fetus. SC in the epithelium of sebaceous glands of the skin (brown staining). ´200.
B. A 19-week-old fetus. J chain in the epidermis of the skin and the epithelium of sebaceous glands. ´200.
C. A 23-old fetus. Chorioamnionitis. SC in the epithelium of the bronchus (b), esophagus (e), thymus and thyroid (t). ´100.
D. A 17-week-old fetus. Chorioamnionitis and aspiration syndrome. SC in the epithelial cells of the bronchiole. ´200.
Fig. 8.
A. A 17-week-old fetus. The kidneys. SC in the epithelium covering of glomerulus (heads of arrows), loop of Henle (small arrows), collecting tubules (ct). x100.
B. The same case. IgA in the epithelium of convoluted (small arrows) and collecting (a large arrow) tubules. x200.
C. A 20-week-old fetus. The bladder. SC in the superficial epithelium. x400.
D. A 21-week-old fetus. The uterus. J chain in the epithelium. x400.
Table X.
The number of IgA- and IgM-positive lymphocytes in different fetal organs
(per 50,000 µm², mean ± SE) (After ref. 65)
Organs studied |
IgA |
IgM |
||||
|
Group I fetuses |
Group II fetuses |
Group I fetuses |
Group II fetuses |
|
|
Spleen |
0.8±0.3 |
2.3±0.5 a |
3.0±0.6 |
5.4±0.9 a |
|
|
Lymph nodes |
1.1±0.4 |
2.6±0.6 a |
2.3±0.5 |
5.2±1.2a |
|
|
Lungs |
0.2±0.1 |
0.9±0.3a |
0.3±0.1 |
1.6±0.5a |
|
|
Stomach, Small intestine |
0.4±0.2 |
1.8±0.5a |
1.9±0.7 |
6.9±1.8a |
|
|
Liver |
1.2±0.4 |
2.4±0.7 |
1.9±0.4 |
4.8±1.2a |
|
|
Pancreas |
0.2±0.1 |
0.8±0.4 |
0.6±0.4 |
2.5±0.8a |
|
|
Kidneys |
0.2±0.2 |
0.6±0.3 |
1.3±0.7 |
1.9±0.8 |
|
|
Choroid plexus |
0.3±0.2 |
0.6±0.3 |
1.3±0.7 |
1.9±0.8 |
|
|
a Significantly different from Group I, p < 0.05.
Groups of fetuses: I, without antigenic attacks; II, with antigenic attacks.
Table XI. The number of immunocompetent cells in the liver of embryos and fetuses (mean±SE in 50,000 μm2) (After ref. 65)
Groups of patientsa |
Age of pregnancy (weeks) |
Macrophages |
CD3+ T cells |
CD20+ B cells |
IgA+ B cells |
I |
3.5-4 to 6 |
11.7±1.5 |
- |
- |
- |
|
7 to 8 |
12.7±1.7 |
single |
single |
- |
|
9 to 12 |
19.3±2.4b |
4.6±0.8b |
3.9±0.7b |
0.8±0.4 |
II |
3.5-4 to 6 |
16.3±2.1 |
- |
- |
- |
|
7 to 8 |
8.5±1.3 b |
single |
single |
- |
|
9 to 12 |
27.4±2.7 b |
5.6±1.1b |
2.9±0.7b |
0.6±0.3 |
a Groups of patients: see footnotes to X.
הרשמה לניוזלטר