Workshop on perinatal exposure to dioxin-like compounds. V. Immunologic effects.

The immune system comprises a highly integrated network of multiple tissues and cell types with complicated interactions and effects. It is modulated by the endocrine and nervous systems and there is growing realization of its multifunctionality. The session focusing on immunologic effects of dioxin and related compounds following prenatal exposure involved a review of the immunotoxic effects that have been reported for polyhalogenated aromatic hydrocarbons (PHAHs), a discussion of species differences in responses, and development of the immune system, and data from two ongoing epidemiological studies comparing the immune status of children exposed to higher-than-average concentrations of PHAHs both prenatally and lactationally.


Introduction
Dr. Kerkvliet reviewed the role of the immune system in health and disease (1,2). Immunosuppression may result in an increased incidence of disease from bacteria, viruses, and parasites as well as an increase in tumors. Inappropriate activation of the immune system or loss of normal suppressor cell control is associated with allergies, hypersensitivity, and autoimmunity. The immune system involves a complex network of cells (macrophages, lymphocytes, other white blood cells) that communicate via soluble mediators such as the interleukins and other cytokines. The interactions are highly regulated. The macrophage initiates certain immune responses by taking up foreign materials nonspecifically and presenting antigens on its surface to T cells, which become activated. Various T-cell subtypes secrete different cytokines, which are involved in the func-tion of other T cells as well as controlling the activity of B cells.
Studies in the 1970s demonstrated that TCDD treatment resulted in dramatic effects on the thymus (3)(4)(5). In fact, at relatively high doses of TCDD in adults, dioxin treatment resulted in atrophy of the lymphoid tissue. However, the ability of dioxin to modulate a number of immune responses has been shown to occur at doses much lower than those resulting in thymic atrophy. Nevertheless, prenatal effects on the thymus and/or T-cell-mediated immunity may be critical. The immunotoxic effects of TCDD have been shown to be mediated by binding to the Ah receptor, based on structure/activity studies and use of mice genetically different in their responsiveness to TCDD (6)(7)(8)(9). The cellular targets of TCDD appear to be multiple, as suggested by the different assays and end points that have been assessed. While B cells are clearly targets of TCDD action, they appear less sensitive than macrophages and T cells (10). Approaches to studying the immunotoxicity of TCDD have involved in vivo, ex vivo, and in vitro studies (11)(12)(13). The in vitro experiments are very sensitive to culture conditions, and dioxin's effects appear dependent upon unknown serum factors (11).
A decrease in the ability to respond to a primary antibody challenge, often measured by responsiveness to sheep red blood cells (SRBC), is a sensitive and reproducible response to dioxin exposure in both mice and nonhuman primates (8,13,14). However, the immunotoxicity of dioxin in people is unclear. There have only been a limited number of investigations involving different study designs and parameters. Furthermore, there is an inherent difficulty in measuring the immune status of humans noninvasively, and for any assay that has been used there is a broad range of "normal" within the human population. Some of the assays that have been successfully used in animals have not been applied to humans. Often the cohort exposure is not validated and the immune status often has been examined long after exposure. The outstanding issues involve understanding the mechanism in animal models and basis for species specificity and sensitivity (15). The recent reports of alterations in lymphocyte subsets induced by TCDD require validation, both in animal models and in humans.
Dr. Smialowicz compared the immunotoxic effects of TCDD in rats vs mice. While thymic involution occurs in both species, the response to the SRBC is clearly different (16). The ED50 for immunosuppression in the mouse is approximately 0.7 jig/kg. In contrast, no immunosuppression is observed in the rat; instead, an increase in response to the antigen is observed at 3 pg/kg. Examination of T-cell phenotypes revealed no apparent effect due to TCDD in the mouse, even when the ability to respond to the SRBC was completely suppressed. In contrast, in two strains of rats in which response to the antigen was enhanced by TCDD, there was a decrease in the T-suppressor population (CD4-CD8+), an increase in double negative T cells (CD4-CD8-), and an increase in IgM+ B cells. These alterations are consistent with enhanced response to a primary antigen challenge. Whether there are Environmental Health Perspectives changes in other subsets remains to be determined. Preliminary work on effects of in utero exposure have also demonstrated a reduction in thymic cellularity in the rat, accompanied by a decrease in CD4-CD8+ and CD4+CD8+ cells.
There are also species differences in the ability of TCDD to compromise host defenses. Exposure of the mouse to TCDD prior to infection by the parasite trichinella leads to a decrease in T-cell proliferation in response to trichinella antigen (17). This response is comparable to the immunosuppression induced by SRBC. This depression of T-cell proliferation may serve as a marker for the delayed expulsion of the parasite that occurs at higher doses. In contrast, in the rat there was a suggestion of an enhanced T-cell proliferative response and no evidence of enhanced parasitemia (18). Response to viral infection appears also to demonstrate differences between rats and mice. Doses as low as 10 ng/kg of TCDD 1 week prior to challenge with influenza virus resulted in significantly enhanced mortality in mice (19). However, few if any effects were seen on viral-induced pulmonary NK activity in mice. In contrast, in rats, using a nonlethal strain of influenza, effects on viral-augmented NK were observed at doses of 3 pg/kg (20). Which rodent is the best model for humans? The answer is unclear. However, the similarity in the mouse and nonhuman primate response to SRBC suggests that the mouse may be an appropriate surrogate for the immunotoxic effects of TCDD in people.
Dr. Gasiewicz reviewed the development of the immune system (21), beginning with the yolk sac as the source of the stem cells that invade the developing mouse thymus on gestation day 11. The yolk sac is also the source of the cells that migrate into the fetal liver and thence to the bone marrow and spleen. There are self-renewing progenitor cells only in the fetus which, if altered or eliminated results in changes in the immune repertoire throughout life. A series of studies has demonstrated that prenatal and perinatal exposure to TCDD affects mainly T-cell responses; few B cell effects are observed, and only at high doses (4). A number of investigators have suggested that the developing embryo/fetus may be more sensitive than the adult to immunosuppression induced by TCDD (22)(23)(24). It is important to note that the immune system of the newborn rodent is much more immature than that of the human. Early postnatal rodent immune system development would occur prenatally in humans.
The fetal thymus also appears to be more sensitive than the adult thymus. Dioxin alters thymic differentiation, causing changes in the total number of cells in each lymphocyte subset (25). There may be an effect on immature populations of intrathymic cells due to a defect in differentiation. Alternatively, there may be changes in migration of cells from the bone marrow to the thymus (25)(26)(27). Recent studies have examined markers such as terminal deoxynucleotidyl transferase (TdT) and recombinase activating gene (RAG) that are present only in lymphoid stem cell populations and involved in gene rearrangements. TdT protein and mRNA decrease following TCDD exposure both in the fetal liver and bone marrow, while no effects are noted in the thymus. This decrease correlates with thymic atrophy as does a decrease in RAG mRNA. Studies have demonstrated that the TCDD-treated prothymocytes are unable to repopulate the thymus of irradiated mice. This suggests that a direct effect of TCDD on bone marrow stem cell populations may contribute to the elicited thymic atrophy. Future studies are needed to examine the issue of dose-response relationships, chronic lowdose effects, effects on B cells, role of cytokines, and hormonal involvement. Furthermore, it is necessary to determine if or how an effect of TCDD on bone marrow stem cell populations may affect other, more subtle, aspects of the immune system other than thymic size. It is important to note that while estrogen may also produce thymic atrophy via effects on prothymocytes, the effects of TCDD cannot be blocked by estrogen receptor antagonists.
Arctic Quebec is the site of an ambitious series of studies discussed by Dr. Dewailly (28). Infectious disease (e.g., meningitis, measles) incidence, as well as that of otitis, is 20-fold higher in the first year of life among the Inuit than in individuals living in southern Quebec. This appears to be associated with immune dysfunction as measured by a low immunization take rate, and raises issues of altered host resistance. The Inuits have elevated levels of PCBs, PCDDS, and PCDFs (29,30). Current investigations are focused on immunologic examinations of the babies at 2, 6, and 12 months, and comparison of breast-fed to nonbreast-fed infants. Is breast feeding protective against disease? Is breast feeding, due to the high level of lactational transfer of PHAHs, associated with elevated disease incidence? There is a suggestion that babies with acute otitis have been nursed by mothers with higher levels of PHAHs or have nursed longer than nonaffected infants. The T-helper/T-suppressor cell ratio may also decrease with increased exposure, although it is still within the normal range. A negative correlation was detected between the decrease in CD4/CD8 ratio and the total toxic equivalency. Prenatal exposure of rodents to TCDD also appears to result in a decrease in the relative population of T-helper versus T-suppressor cells in the thymus. Further studies are clearly warranted in this population, with emphasis on lymphocyte subset information and determination of immunization-take rates. In addition, it would be helpful to have a control population whose mothers have "normal" PHAH levels to examine whether the immunosuppressive effects noted are due to prenatal or lactational exposure.
Dr. Helge stressed the importance of conducting parallel studies in animals and humans. In studies of TCDD-exposed marmosets, changes in the ratio of CD4+CDW29+ helper cells to CD8+ CD56+ cytotoxic cells were noted following extremely low dose (10 ng/kg) treatment of adults (31). However, at even lower doses, the helper cells actually increased (32). The meaning of the biphasic nature of this response is not clear. In vitro treatment of marmoset cells results in suppression of the normal stimulatory effect of pokeweed mitogen on B cells (33).
Bottle-and breast-fed human infants are being examined and compared for lymphocyte subsets as well as their response to stimulation by pokeweed mitogen, PMA, Con A, and anti-CD3. Lower stimulation by all of these mitogens was observed in breast-fed infants. This could be associated with colostrum, which is known to inhibit stimulation because the lack of mitogen responsiveness disappeared by the time the babies were 5 months old. The T-cell subsets are being examined in cord blood and in the children to determine whether breast feeding is associated with changes similar to those observed in the dioxin-treated marmosets. This population study would also benefit from a comparison to babies whose mothers have lower levels of PHAHs than in the group currently under examination for the effects of breast feeding.
It is clear that there is a need to develop sensitive and feasible methods to examine immune responses in children exposed to TCDD prenatally, lactationally, or both. More effort must also be directed at examining the immune effects that have been noted to occur in animals, such as the suppression of the primary antibody response.

Environmental Health Perspectives
Alterations in lymphocyte subsets is clearly a promising area for future investigation, with potential for use as biomarkers. One conclusion that appears clear is that TCDD and related compounds have the ability to alter differentiation of cells in the immune system, as well as in other systems of the body. Viewing dioxin as a chemical that has the ability to disrupt cellular differentiation puts into perspective not only its effects on the immune system but also its dysregulatory role in development of the nervous and reproductive systems.