Objective
Regulatory T lymphocytes mediate homeostasis of the immune system and differentiate under the control of the transcription factor FoxP3 in the fetal thymus. We asked whether fetal inflammation caused by chorioamnionitis would modulate thymus development.
Study Design
Fetal sheep were exposed to an intraamniotic injection of 10 mg lipopolysaccharide at 5 hours, 1 day, 2 days, or 5 days before delivery at 123 gestation days. Cord blood lymphocytes, plasma cortisol, and thymus weight were measured. Glucocorticoid receptor–, activated caspase-3–, Ki-67–, proliferating cell nuclear antigen–, nuclear factor-κB–, and FoxP3-positive cells were immunohistochemically evaluated in thymus.
Results
Intraamniotic lipopolysaccharide exposure decreased the number of circulating lymphocytes by 40% after 1 day. Thymus-to-body weight ratios were reduced in all lipopolysaccharide groups by a maximum of 40% at 5 days. Lipopolysaccharide exposure modestly increased plasma cortisol concentration, increased nuclear factor-κB immunostaining in fetal thymus and reduced the number of FoxP3-positive cells by 40% at 1 day.
Conclusion
Intraamniotic exposure to lipopolysaccharide induced thymic changes and influenced thymic FoxP3 expression.
T-lymphocyte development and selection in the thymus is a complex process that integrates endogenous and exogenous stimuli during fetal and postnatal life. There may be a “window of opportunity/vulnerability” in fetal life at which a variety of factors, which include infection, may modulate the development of the immune system. With fetal stress, the thymus can involute rapidly and decrease in size, probably because of glucocorticoid-induced apoptosis of immature thymocytes. Recovery from these processes may be important for the normal development and function of the immune system later in life.
Histologic chorioamnionitis is present in 60% of very preterm deliveries. Many fetal organs may be affected by exposure to antenatal inflammation, which suggests that chorioamnionitis causes a “multiorgan disease of the fetus.” Fetal thymic involution is associated with the fetal inflammatory response syndrome in women with preterm labor and rupture of membranes. Clinical and experimental data indicate that exposure to antenatal inflammation can have both risk and benefit for the fetus. For example, chorioamnionitis can induce lung maturation but also may cause brain and gut injury. The unifying link between the affected organs may be the fetal immune response. Little is known about the effect of preterm birth and chorioamnionitis on T-lymphocyte development in the thymus. However, prematurity and chorioamnionitis are associated with increased early childhood wheezing and physician-diagnosed asthma. We previously showed, in a surgical model of chorioamnionitis, that lipopolysaccharide contact with the fetal respiratory tract or gut changed the percentage of CD4, CD8, and increased CD4/CD25 positive cells in the thymus.
Natural regulatory T (T reg ) lymphocytes, a subgroup of T lymphocytes, regulate immune responses and participate in tolerance responses. T reg lymphocytes are also involved in human diseases such as cancer, infections, autoimmunity, allergy, and asthma. The transcription factor FoxP3 is a marker for T reg lymphocytes in mice and humans. In the mouse, natural FoxP3 + T reg lymphocytes are formed in the thymus and contribute to peripheral T reg lymphocytes. Proinflammatory cytokines and microbial pathogens initiate innate immune responses by nuclear factor (NF)-κB signaling. The physiologic function of NF-κB in adaptive immune pathways is less clear. However, NF-κB is essential for thymocyte development and activation.
For this study, we tested the hypothesis that chorioamnionitis that is caused by lipopolysaccharide exposure in fetal sheep would induce changes in the fetal thymus. We measured cortisol concentrations and lymphocyte counts in the systemic circulation, thymus weight, apoptosis and proliferation, and expression of NF-κB, Foxp3 and glucocorticoids receptor in the fetal thymus .
Material and Methods
Animal-model and sample collection
The animal studies were performed in Western Australia with date-mated Merino ewes with singleton pregnancies. All studies were approved by the animal care and use committees at the Department of Agriculture and Food, Western Australia, and at the Cincinnati Children’s Hospital Medical Center, OH. In groups of 4-7 animals, chorioamnionitis was induced 5 hours, 1 day, 2 days, or 5 days before caesarean delivery at 123 days gestation (term, 150 days) by intraamniotic injection of 10 mg lipopolysaccharide ( Escherichia coli 055:B5; Sigma Chemical Company, St. Louis, MO; Figure 1 ). Seven animals were used for the 5-hour time point; 6 animals were used for the 1-day time point; 4 animals were used for the 2-day time point, and 6 animals were used for the 5-day time point. The control group received intraamniotic saline solution injections at either 5 hours (n = 2), 1 day (n = 2), 2 days (n = 2), and 5 days (n = 2) before delivery. All animals were studied at the same gestational age. Thus, we used the composite control. There were no differences between the control animals that received saline solution injections at different time points before delivery (data not shown). We therefore have combined the animals as 1 control group. Results for lung inflammation and maturation that were associated with chorioamnionitis were reported previously. Cord blood samples were collected at delivery for complete white blood counts. Fetal plasma cortisol concentrations were measured with a commercial radioimmunoassay kit (ICN, Irvine, CA). The inter- and intraassay coefficients of variation were 0.159 and 0.112, respectively. The chest was opened, and the intrathoracic thymus was removed. The thymus was weighed, snap frozen, and fixed in formalin.
Immunohistochemistry
Sections (2 μm, transverse) were cut from formalin-fixed, paraffin-embedded thymus, and were mounted on APES-(3-amino-propyltriethoxy-silane; Roth, Karlsruhe, Germany) coated slides. Slides were dewaxed and dehydrated in ethanol. Antigen retrieval was performed when the slides were boiled for 10 minutes in citric acid buffer, pH 6.0, in a microwave (750 W). Endogenous peroxidase activity was inhibited by 3% hydrogen peroxide in methanol. Phosphate-buffered saline solution with 5% goat serum was used to block nonspecific binding. Slides were incubated overnight at 4°C in a humidified chamber with primary antibodies at appropriate dilution in phosphate-buffered saline solution. Caspase-3 activation, through cleavage of its regulatory NH 2 -terminal domain, is a well-characterized biochemical marker of apoptosis. Antibodies against proliferating cell nuclear antigen (PCNA), a nonhistone protein, which is elevated in the S, G2, and M phases of mitosis, and Ki-67, which is expressed in the S, G1, G2, and M phases of mitosis, were used as proliferation-associated antibodies. The following primary antibodies were used: (1) polyclonal rabbit anti–glucocorticoid receptor α and β (diluted 1:250, sc-8992; Santa Cruz Biotechnology Inc, Santa Cruz, CA), (2) monoclonal rabbit anti–activated caspase-3 (diluted 1:100; #9664; Cell Signaling Technology Inc, Danvers, MA), (3) monoclonal mouse anti-Ki-67 (diluted: 1:50; Dako M7240; Dako Corp, Glostrup, Denmark), (4) monoclonal mouse anti-PCNA (diluted: 1:400; Dako M0879; Dako Corp), (5) polyclonal anti-human RelBp68 (component of NF-κB; diluted: 1:1000; sc-226; Santa Cruz Biotechnology Inc), and (6) monoclonal mouse anti-FoxP3 (diluted 1:400, 14-7979-82; NatuTec, Frankfurt, Germany). Detection was with anti-rabbit (glucocorticoid receptor and activated caspase-3) or anti-mouse (Ki-67, PCNA, FoxP3) secondary antibodies that was conjugated to horseradish peroxidase (Dako Corp) and developed with 3,3′ diaminobenzidine (brown color). After being counterstained with hematoxylin, samples were dehydrated and coverslipped. Omission of the primary antibody acted as a negative control. An average for each sample was calculated from 5 fields that were chosen at random. Numeric analysis was done for apoptotic cells and proliferative cells. Six high-power fields were photographed for the medulla and cortex. The intensity of staining was measured by a semiquantitative scale (magnification, ×200); no staining (0), weak (+1), moderate (+2), or strong (+3) by a blinded examiner and results are expressed as percentages of all cells counted.
Statistical analysis
Results are given as means ± SEM. Comparisons between the groups were performed by analysis of variance with the Student-Newman test as the post hoc analysis. Significance was accepted at probability value of < .05.
Results
Lymphocytes in cord blood and thymus weight
Blood lymphocyte count and thymus weight are markers for stress-induced thymus involution. In this model of intraamniotic exposure to lipopolysaccharide, blood lymphocytes were lower than control by 40% after 1 day and elevated at 5 days ( Figure 2 , A). The thymus/body weight ratios were lower in all lipopolysaccharide-exposed groups, compared with control ( Figure 2 , B).
Plasma cortisol and expression of corticosteroid receptors in thymus
At 2 days after lipopolysaccharide exposure, plasma cortisol levels (1.3 ± 0.3 μg/dL) were higher than control (0.5 ± 0.1 μg/dL; Figure 3 ). Immunohistochemistry for the glucocorticoid receptor was performed to evaluate whether the small increase in cortisol altered expression of the corticosteroid receptor in the thymus. No differences in immunostaining of the glucocorticoid receptor were detected between the controls and the lipopolysaccharide-exposed groups in the thymic cortex ( Figure 3 , B) and in the thymic medulla (data not shown). The distribution of immunostaining was also not different (data not shown).
