Epigenetic underpinnings of developmental immunotoxicity and autoimmune disease

Abstract

The concordance rate for developing autoimmune disease in identical twins is around 50% demonstrating that gene and environmental interactions contribute to disease etiology. The environmental contribution to autoimmune disease is a wide-ranging concept including exposure to immunotoxic environmental chemicals. Because the immune system is immature during development suggests that adult-onset autoimmunity may originate when the immune system is particularly sensitive. Among the pollutants most closely associated with inflammation and/or autoimmunity include Bisphenol-A, mercury, TCDD, and trichloroethylene. These toxicants have been shown to impart epigenetic changes (e.g., DNA methylation) that may alter immune function and promote autoreactivity. Here we review these autoimmune-promoting toxicants and their relation to immune cell epigenetics both in terms of adult and developmental exposure.

1. Introduction

The immune system is designed to recognize and eliminate foreign antigens. If the immune system instead attacks self-antigens, autoimmune diseases may occur. Approximately 24 million Americans have one or more autoimmune disease. These chronic, incurable disorders disproportionately affect females, and are among the leading causes of death for young and middle-age women [1]. Twin studies have shown that although an individual’s genome may increase susceptibility, environmental triggers are required to initiate disease. Defining how the environment promotes autoimmunity will enhance understanding of so-called idiopathic autoimmune disease. The elevated prevalence and incidence rates of autoimmune disease parallel the documented increase in environmental pollutants leading to an appreciation of environmental toxicants common to industrialized nations as important riggers for autoimmunity [2].

Enhanced sensitivity of the immune system to environmental perturbations during key developmental events occurring prenatally and/or postnatally are critical for later life function [3–6]. The cells of the innate immune system (i.e. neutrophils, dendritic cells, NK cells, and macrophages) provide the first line of defense against pathogens. Their relative functional immaturity at birth means that innate immunity is weak in the newborn compared to an adult. The second line of defense is mediated by the cells of the adaptive immune system. T cells derived in the thymus are abundant at birth, and they need to undergo further maturation in the periphery to become fully functional. Peripheral B cells in the newborn are similarly immature, and require further maturation to respond to antigens. Thus, due to the vulnerability of the developing immune system, developmental exposures may influence adult autoimmunity [7].

When contemplating how developmental toxicant exposure “programs” the host for autoimmunity, one likely scenario involves epigenetic alterations such as aberrant DNA methylation. The epigenome consists of modifications of the genome that do not alter DNA base sequences, but can regulate gene expression and phenotype. While it is understood that the epigenome is regulated by several epigenetic mechanisms other than DNA methylation (e.g., histone acetylation and micro-RNA expression), of the various forms of epigenetic modifications, DNA methylation is the most thoroughly investigated. Maturation of immune cells are largely controlled by DNA methylation events that occur most often in early life that are functionally evident in later life and potentially to additional generations [8]. Auto-immune diseases [(lupus, rheumatoid arthritis, type 1 diabetes (T1D), and multiple sclerosis)] associated with environmental toxicants are also linked to abnormal methylation [9] and may represent a mechanism by which environmental triggers promote autoimmunity. Thus, the review will focus on toxicant-induced effects on DNA methylation in autoimmune disease.

2. DNA methylation changes at various stages of immune development

Autoimmune disease, whether antibody-or T cell-mediated is largely driven by CD4⁺ T cells. As shown in Fig. 1, epigenetic events play an important role in immune cell development and correspond to several key cellular maturational events. Prior to immune system development, genome-wide global epigenetic reprogramming in early embryonic development occurs immediately after fertilization to ensure loss of DNA methylation allowing for global repression and gene expression in all cells [10]. In later stages, CpG methylation coincides with general developmental life stages with a reported global trend of demethylation during Tcell development in the thymus closely related to the development of TCR function [11].

Fig. 1.

Key human immune system developmental checkpoints that correspond with important changes in DNA methylation. General steps of human immune system development spans from gestation to early postnatal life. These events in particular are epigenetically regulated via DNA methylation and represent a sensitive window for perturbation due to environmental insults that may later manifest in later life autoimmune disease in certain individuals with genetic susceptibility or lifestyle factors.

Another DNA methylation mechanism identified as an emerging concept in toxicology is epigenetic drift (i.e., drift) [12]. Drift is the divergence of the epigenome as a function of age due to stochastic changes in methylation. Under normal circumstances drift occurs because the fidelity of maintaining CpG methylation in mammalian cells (about 97–99% per cell division) is not absolute [13]. The small but significant error rate creates opportunity for changes in the methylome to occur and accumulate in constantly dividing cells, such as self-renewing effector/memory CD4⁺ T cells [14]. Drift involves both involves both hypo- and hyper-methylation events, and can encompass as much as 2.2% of total CpG sites, and 5–25% of specific genes over time [15,16]. Drift can impact promoter methylation status and gene expression, and has been used to explain the subset heterogeneity of memory CD4⁺ T cells that occurs during aging. In terms of autoimmune disease (e.g., T1D) results from twin studies suggest that drift causes heterogeneity in disease onset, severity, and predisposition to secondary complications [17]. The events that dysregulate drift are unclear, but appear to involve environmental exposures [18,19]. Importantly, although drift appears soon after birth, it occurs at a higher rate of change in children compared with adults [20]. Thus, although drift is still an understudied area of epigenetics, environmental influences may perturb this process in early life to promote autoimmunity.

One stage of vulnerability mediated by epigenetic changes is CD4⁺ T cell differentiation. Beginning in early life, the phenotype of differentiated CD4⁺ T cell subsets are normally controlled by carefully maintained levels of DNA methylation in the promoters of pertinent regulatory genes [21,22] (Fig. 2). The development of autoimmune disease can disrupt the methylation patterns of differentiated CD4⁺ T cells, resulting in the demethylation of genes that encode immunomodulatory factors as reported in juvenile arthritis [23]. Subsets of differentiated CD4⁺ Tcells (i.e., Th1/Th17) have been shown to promote autoimmune disease in part to their persistence as effector/memory CD4⁺ T cells. The dysregulated methylome in autoimmune disease is associated with increased heterogeneity or plasticity in these subsets [24,25].

Fig. 2.

The phenotype of differentiated CD4⁺ T cells are normally controlled by carefully maintained levels of DNA methylation in the promoters of pertinent genes. Development of autoimmune disease can disrupt methylation patterns of differentiated CD4⁺ T cells resulting in the demethylation of genes that encode pro-inflammatory cytokines, chemokines, adhesion molecules, or subset differentiation. Toxicants have been associated with DNA demethylation in CD4⁺ T cells tied to increased expression of critical pro-inflammatory genes, cell cycle molecules/regulatory proteins and adhesion molecules within biological pathways with known links to autoimmunity.

While several key maturational and differentiation events in Tcells are regulated by DNA methylation, it is not known whether these events promote autoreactivity. While studies have shown that the function of autoreactive CD4⁺ T cells can be mediated by epigenetic processes, most of this work has been done in lupus models. Normal activated CD4⁺ Tcells treated with the DNA methyltransferase drug/inhibitor 5-azacytidine in vitro hypomethylates certain genes important in autoimmunity that become autoreactive upon adoptive transfer [26]. Whether environmental toxicants drive autoimmune-promoting epigenetic events in CD4⁺ T cells similar to 5-azacytidine is not known.

Identification of environmental pollutants that promote autoimmunity has been studied extensively [27,28]. However, there are a few toxicants associated with autoimmunity in the context of developmental exposure that have been identified and that may also impart autoimmune-promoting effects, at least in part, by aberrant DNA methylation. These substances are briefly reviewed below.

3. Mercury

Human exposure to mercury is common due primarily to its anthropogenic release from industrial use. Evidence that mercury promotes autoimmunity appears to be more straightforward in mouse models where it stimulates ANAs and induces immune complex-mediated lupus nephritis. Developmental exposure to subclinical doses of HgCl₂ in maternal drinking water in mice has been shown to cause adverse immune effects in offspring. This includes increased number of activated CD4⁺ T cells and increased levels of brain-reactive antibodies [29].

In humans, mercury exposure was associated with sub-clinical autoimmunity as measured by increased ANAs among reproductive-age women [30]. Prenatal methyl-mercury exposure was associated with reduced levels of B cells and CD4⁺ T cells [31]. With regard to autoimmune disease, Sardinia has the second highest incidence of T1D in the world. The especially elevated levels of heavy metals, including mercury, in this country has led to the assumption that exposure to mercury, in the context of metals and other co-exposures during development promotes generation of this early-life autoimmune disease [32].

The potential autoimmune-promoting capacity of mercury is undoubtedly complex and likely involves other factors that enhance disease risk. Mercury can promote oxidative stress by depletion of anti-oxidant glutathione in immune cells [33]. This aspect of toxicity is functionally relevant to DNA methylation effects because glutathione pathway intermediates direct methionine metabolism to increase or decrease methyl donors to execute cellular methylation reactions. Direct evidence that mercury impacts DNA methylation in immune cells was shown in humans, where prenatal exposure to mercury detected in infant toenails correlated with hypermethylation of CpG islands of cord blood leukocytes [34]. These data highlight the potential for DNA methylation events in mercury-induced immunotoxicity or autoimmunity.

4. Bisphenol A

Bisphenol A (BPA) is a xenoestrogenic compound used to manufacture polycarbonate plastics and epoxy resins that can be detected in human blood and tissues [35]. There is evidence that developmental BPA exposure alters the immune system later in life and may be a potential autoimmune trigger. Studies report associations between urinary levels of BPA and allergic asthma [36]. In mice gestation, only exposure to BPA increased the development of adult experimental autoimmune encephalomyelitis (EAE) [37]. In a different study, gestational and lactational exposure to BPA failed to exacerbate adult EAE [38]. Diabetic (NOD) mice given BPA during gestation generated female offspring with an increased incidence of diabetes [39]. Perinatal BPA exposure accelerated inflammation in a mouse model of virally-induced multiple sclerosis [40].

Although the effects of BPA in mouse models were potentially relevant to human disease, most used concentrations of BPA that were higher than the typical concentration found in drinking water (<1 ppb). Developmental exposure to BPA at the US EPA oral reference dose (50 μg/kg/day) did not increase the severity of experimental inflammatory colitis in mice [41]. In humans, serum BPA in adults has shown consistent association with autoantibodies associated with thyroid- and neuron-specific antigens [42,43]. An association between urinary BPA concentrations and asthma in a cohort of inner-city children was reported implying developmental/early life exposures promoted inflammatory disease [44]. These results suggest that BPA may enhance autoimmunity depending on dose, model, and window of exposure.

DNA methylation changes with BPA exposure have been reported in the actual tissues that are targeted by the autoimmune response as in perinatal BPA exposure and altered DNA methylation in liver [45]. In NOD mice, BPA exposure increased the number of T_REGS, while also increasing production of IL-17 [39]. Thus, BPA-mediated DNA methylation events may play a role in effector cell generation since differentiation of naïve CD4 Tcells into these subsets is regulated in part by DNA methylation. Even if toxicants do not induce uni-directional changes in the methylation state of specific genes, they may support epigenetic drift as shown in a longitudinal study of early life exposure to BPA [46]. These data suggest the potential role for DNA methylation in BPA-mediated immunotoxicity.

5. Trichloroethylene

One toxicant linked to the development of autoimmunity is the industrial solvent and water pollutant, trichloroethylene (TCE). Chronic TCE exposure (mostly occupational, but sometimes environmental) has been linked to a variety of autoimmune diseases including lupus [47], autoimmune hepatitis [48], and scleroderma [49]. Signs of immune activation and alterations in lymphocyte subsets have been associated with chronic environmental exposure to drinking water contaminated by TCE or via dermal or inhalational occupational exposure [50–54]. We and others have reported that chronic TCE exposure in drinking water during adulthood induces autoimmune hepatitis in autoimmune-prone MRL +/+ mice, and that this disease development is associated with several changes in CD4⁺ T cells [55,56].

Developmental exposure to TCE is a concern based on the ability of TCE and its metabolites to cross the placenta and its detection in breastmilk [57]. We and others have shown that continuous developmental exposure to TCE in mice (gestation, lactation, and early life) generated CD4⁺ Tcell alterations and/or early signs of tissue inflammation in both normal and autoimmune-susceptible mouse strains [58,59]. Similar immune-altering effects were observed with gestation- or postnatal-only exposure in young adult mice [60,61]. We recently reported continuous chronic exposure to low-level TCE beginning at gestation generated autoimmune hepatitis at postnatal day 259 even when TCE exposure was stopped 15 weeks earlier [62]. Because the effects persisted after TCE was removed from the drinking water suggested that programming events played a role in disease pathology. Indeed TCE altered the DNA methylation profile of the IFN-γ promoter in mouse CD4⁺ T cells [63]. In a genome-wide DNA methylation and gene expression study, TCE exposure in vivo differentially methylated CpGs in regions that bind polycomb group proteins [64] whose function is to regulate T effector cell expansion and differentiation [65]. Chronic exposure to TCE increased epigenetic drift in CD4⁺ T cells that corresponded with immune pathology [66]. These results support evidence that epigenetic events may play a role in TCE-induced immunotoxicity and autoimmunity.

6. TCDD/AHR ligands

While several environmental pollutants bind the aryl hydrocarbon receptor (AHR), 2,3,7,8-tetrachlorodibenzo-p-dioxin or TCDD is the prototypical ligand. Developmental exposure to TCDD and other AHR-binding xenobiotics alter the immune system. In contrast to the primarily immunosuppressive nature of adult TCDD exposure, developmental exposure to TCDD generated immune dysfunction with a different result. Adult C57BL/6 and autoimmune-prone SNF (1) mice exposed to TCDD on gestational day 12 demonstrated autoimmune glomerulonephritis including increased levels of anti-dsDNA in adulthood [67,68]. Neonatal exposure to TCDD increased pro-inflammatory cytokine and autoantibody production in a mouse model of Sjogren’s syndrome through signaling events in the neonatal thymus [69]. More recently, the novel autoimmune-prone Gnaq± mice were used to examine the autoimmune-promoting effects of TCDD on CD4⁺ Tcells. Exposure to TCDD generated offspring which developed lupus-like disease sooner and at a higher frequency than the offspring of vehicle-exposed dams [70]. Thus, unlike adult exposure, developmental exposure to AHR ligands appear to promote autoimmunity. It is plausible to hypothesize that epigenetic events which are more relevant during developmental periods may play a role in this disparity.

Studies have shown that DNA methylation plays a role in immunoregulation of AHR ligands. When the methylation status of CpG islands present in Foxp3 and IL-17 promoters following AHR activation were examined for their methylation/demethylation status, demethylation of CpG islands present in the Foxp3 promoter and increased methylation of CpG islands of the IL-17 promoter, following activation of AHR during colitis was observed [71]. In a different model, TCDD changed DNA methylation patterns in CD8⁺ T cells in the context of host resistance to viral infection [72].

7. Conclusion and future research needs

This review summarized several environmental chemicals studied for their potential to alter immune function, DNA methylation, and promote autoimmunity after developmental exposure. Although many autoimmune-promoting immunotoxicants appear to target CD4⁺ T cells, end organ pathology may also occur from toxicant-induced damage that enhances immunogenicity of proteins in certain target tissues or by disrupting repair or anti-oxidant systems designed to promote regeneration/recovery. Autoimmunity is complex and may be affected by lifestyle factors and genetic susceptibility (Fig. 3). In particular, the role of co-exposures, whether other toxicants or lifestyle factors that also impair immune function/DNA methylation should be an important research consideration for future study. For example, 50% of American women of childbearing age are overweight or obese [73]. Data from animal studies demonstrated adverse outcomes in models of autoimmunity and allergic sensitization models of maternal and postnatal obesity [74,75]. In humans, maternal obesity influenced programing of the neonatal immune system [76] potentially enhancing risk to inflammatory disease such as asthma in offspring [77]. Thus certain risk factors may synergistically act on an already sensitive developing immune system by modifying the methylome and downstream gene expression leading to enhanced susceptibility to autoimmunity. There are only a few studies with toxicants outlined in this review that addressed diet/obesity co-exposures [(e.g., BPA [12] and TCE (manuscript in preparation)]. Future studies should aim to interrogate DNA methylation and gene expression events to confirm the role of epigenetic modifications when considering both exogenous and/or endogenous risk factors in the development of autoimmune disease.

Fig 3.

Conceptual diagram of hypothesized factors associated with autoimmunity. Autoimmune diseases result from complex interplay of gene–environmental interactions. The developmental period represents an enhanced time of epigenetic plasticity that may facilitate functional changes in an already sensitive maturing immune system. Multiple exposures to several factors together with impaired toxicant-induced regeneration/repair mechanisms converge with these factors to promote autoreactivity and disease.