The capacity of AhR agonists to cause endocrine disruption has been demonstrated in both the male and female reproductive systems, with effects observed in wildlife, experimental animals, and humans [11-15]. Hormone signaling pathways may be sensitive to extremely low dose exposure to various chemicals [16], and TCDD has been strongly implicated in disruption of estrogen-mediated signaling [17, 18]. There are various mechanisms by which AhR has been shown to affect estrogen receptor or other nuclear receptor signaling pathways, and these include modulation of hormone synthesis and/or metabolism, direct transcriptional regulation of nuclear receptors, stimulation of nuclear receptor degradation, as well as competition for coregulators necessary for transcriptional regulation by nuclear receptors [18]. Additionally, as a lipophilic, persistent organic pollutant, TCDD exposure may increase over time via bioaccumulation [19] [20].

The mammary gland serves as both a likely target for bioaccumulation of xenobiotic AhR ligands, as well as a model system to study how AhR regulates many aspects of development. The mammary gland develops rudimentary ductal structures during fetal and prepubertal stages, and these structures develop further at puberty in the female. However, it is not until pregnancy that the full developmental process proceeds with extensive cell proliferation and differentiation, resulting in further ductal branching and development of the milk-producing alveolar structures. Upon the weaning of offspring, the mammary gland undergoes involution, which is characterized by widespread apoptosis [21]. Our laboratory discovered that mice exposed to TCDD during pregnancy display impaired mammary gland differentiation and an inability to nutritionally support their offspring. Expression of the milk protein WAP gene (whey acidic protein) was also reduced in TCDD-exposed dams [22]. In mice, the coordinated induction of milk protein genes occurs around the 9th day of pregnancy [23], and we have found that the coordinated induction of these genes is impaired by TCDD (our unpublished data). These changes occurred in the absence of alterations in circulating estradiol, progesterone and prolactin levels [22], suggesting that AhR modulates other pathways important for mammary gland differentiation. Further evidence that AhR ligands disrupt mammary gland differentiation during pregnancy was provided in a study using mammary gland explants from estrogen and progesterone-primed mice that were then maintained under hormonal stimulation in culture. The explants were also exposed to the AhR agonist ligand 2,3,7,8-tetrachlorodibenzofuran, which acted as a negative regulator of mammary gland growth and differentiation by suppressing development of lobuloalveolar structures [24]. In order to determine the molecular mechanisms responsible for AhR-mediated disruption of mammary gland development and function, we have continued to study the effects of TCDD exposure in mice during pregnancy, and we have extended our studies using SCp2 mammary epithelial cells in culture.

Mammary gland development during pregnancy involves the differentiation of epithelial cells through formation of alveolar structures, cell polarization, and the production, secretion, and luminal sequestration of milk proteins [25, 26]. Primary mouse mammary epithelial cells and derived cell lines, cultured on extracellular matrix (ECM) in the presence of lactogenic hormones, have been shown to undergo morphological and functional changes that mirror those that occur in the mammary gland of the pregnant mouse [27-30]. The SCp2 mammary epithelial cell line is a clonal derivation of cells isolated from BALB/c mice at mid-pregnancy [29]. The original COMMA-1D cell line was heterogeneous, with only a subset of the cells expressing β-casein after the addition of ECM and lactogenic hormones [31]. COMMA-1D cells were then enriched for ECM-responsive, β-casein producing cells that were designated CID-9 [32]. Finally, the homogeneous SCp2 cell line was generated through expansion of a single cell-derived clone isolated from the CID-9 cell line [29]. SCp2 cells have a distinctly epithelial morphology, express cytokeratins, and lack the fibroblast-type filament protein vimentin. These characteristics, along with the capacity for β-casein expression, suggest that the SCp2 cell clone originated from mouse mammary epithelial cells.

The differentiation potential of SCp2 cells has been well-characterized, beginning with morphological and functional response to ECM and lactogenic hormones. When cultured with lactogenic hormones, but without ECM, less than 0.1% of cells express β-casein. With the addition of ECM, more than 90% of SCp2 cells form cell aggregates and express β-casein [29], demonstrating dependence on the addition of exogenous ECM for differentiation in culture. In contrast, the fibroblastic cell line SCg6, also derived from a CID-9 clone, did not form cell aggregates or express β-casein under any of the inductive culture conditions, indicating that a mixture of cell types was present in the parental cell line [29]. Further analysis showed that even in permissive cultures, SCp2 cells that remained at the periphery of cell clusters did not express β-casein, underscoring the importance of cell aggregation for functional differentiation [29].

The toxic effects of exposure to environmental contaminants are often linked to the timing of the exposure. Timing is particularly important in tissues during periods of development, which typically involve widespread cell proliferation, apoptosis, and tissue reorganization, as well as structural and functional differentiation. Virgin female rodents prenatally exposed to TCDD have persistent mammary gland abnormalities as well as predisposition to mammary cancer as adults [38-41]. In the adult, few tissues undergo the same level of development and differentiation that occurs during fetal development. The exception to this is the mammary gland, which undergoes extensive development at the onset of pregnancy, producing a functional milk-producing structure for the nourishment of offspring. It is the later stage of epithelial cell clustering and differentiation to form the milk protein-producing alveolar structures [26] that is modeled by prolactin and hydrocortisone treatment of SCp2 cells in culture [29]. Few studies have initiated exposure to AhR ligands at the onset of pregnancy, or in conjunction with hormonal treatment to model the effects of pregnancy, for the purpose of analyzing effects on lobuloalveolar development and lactation in the exposed dam [22, 24]. Our model of TCDD exposure using SCp2 cells allowed variation in the timing of exposure while focusing the analysis on effects that occur in epithelial cells. We found that delaying addition of TCDD reduced the detrimental effects of exposure on the functional differentiation of mammary epithelial cells in culture. Our results suggest that a window of time very early in the process of alveolar development and functional differentiation contains critical molecular signaling events that are disrupted upon activation of AhR by an exogenous ligand. An alternative explanation is that decreased β-casein is due to the duration of TCDD treatment, not absence of AhR-mediated signals during early stages of cellular differentiation. However, the critical cellular and morphological changes occur within 3 days of plating on Matrigel and induction with lactogenic hormones [30]. With TCDD-exposure initiated prior to accumulation of significant levels of β-casein protein and continued exposure for at least 5 days, cultures with delayed addition of TCDD are able to produce and accumulate β-casein. This supports the idea of an early window of exposure during which AhR activation is able to interfere with the production and/or accumulation of milk protein in SCp2 cells.

Lactogenic differentiation occurs during mid-pregnancy in the mouse, beginning when epithelial cell clusters, or alveoli, form at the ends of the mammary ducts. In late pregnancy, the alveoli become dilated due to pressure from milk protein secretions produced by the functional alveolar epithelial cells [26]. E-cadherin is a cell adhesion molecule critical for the formation of epithelial cell layers as well as the polarization and function of epithelial cells that make up various tissues, including the mammary gland [42]. Deletion of E-cadherin in a knockout mouse model resulted in embryonic lethality [43]; therefore, alternative methods have been used to study the specific function of this protein in the mammary gland. Transgenic expression of a truncated form of E-cadherin lacking its extracellular domain resulted in a dominant-negative effect on cell-cell adhesion, cell polarity, and cell-matrix interactions that disrupted the structure and function of the fully differentiated mammary epithelium [44]. Additionally, a conditional knockout of the E-cadherin gene in differentiating alveolar epithelial cells of the mammary gland resulted in dramatic reduction in milk protein production with concomitant loss of prolactin-induced STAT5a activation [45]. Activation of STAT5a, a protein essential for pregnancy-induced mammary gland development and lactogenesis [33, 46], is necessary for maximal transcription of multiple milk protein genes [47-49]. Thus, our observation that E-cadherin levels and STAT5 activation are altered by TCDD activation of AhR in the mammary gland and mammary epithelial cells in culture is of interest. E-cadherin protein levels are reduced following TCDD exposure in the MCF-7 breast cancer cell line [50], however this is the first study showing the same effect in untransformed mammary epithelial cells and mammary glands from exposed mice. The lack of cell clustering and fewer, as well as smaller, alveolar-like structures observed in SCp2 mammary epithelial cells in our experiments may then result from the loss of cell-cell adhesion due to the reduction in E-cadherin expression. In addition to fewer clusters forming, the loss of E-cadherin expression may affect cell-cell signaling, limiting cellular polarization and differentiation that ultimately reduces the ability of these cells to produce β-casein and other milk proteins.

In parallel with the reduced levels of E-cadherin, which is linked with reduced activation of STAT5 in vivo [45], our studies show altered post-translational modification of STAT5 in the mammary glands of mice exposed to TCDD throughout pregnancy. STAT5 is functionally activated by phosphorylation [51], and our results show reduced phosphorylation of STAT5 following TCDD exposure, with no alteration in total levels of STAT5 protein. Once activated, STAT5 dimerization and nuclear translocation prepares this transcription factor for binding to promoter regions of target genes [52, 53]. Less phosphorylation, and therefore less activation, of STAT5 may result in reduced induction of transcription of the β-casein gene as well other milk-protein target genes, ultimately resulting in overall suppression of milk-protein expression. The paucity of putative AHREs in the upstream promoter region of the β-casein gene itself makes it an unlikely candidate for direct transcriptional regulation by TCDD-activated AhR. However, the alterations we observed in STAT5 phosphorylation in the mammary gland after exposure to TCDD raise the possibility that AhR regulates expression of kinases responsible for the activation of STAT5. The kinase JAK2 is a component of the well-characterized JAK-STAT signaling pathway leading to β-casein protein expression [51, 54], and is responsible for phosphorylation-mediated activation of STAT5 during mammary gland development and differentiation in pregnancy. JAK2 is also a phospho-protein [55], and it is possible that activation of AhR signaling by TCDD inhibits the phosphorylation of JAK2, alters the total amount of JAK2 protein present in the mammary gland, or impacts molecular targets upstream of the JAK-STAT pathway.