Finally! A Model for Progesterone Receptor Action in Normal Human Breast

Substantial clinical and experimental data over the years have supported the conclusion that estradiol-17β is the ovarian hormone primarily responsible for growth and progression of breast cancer (1). Progesterone was assumed to have negligible effects on breast tumorigenesis, in part because of its well established role to inhibit the proliferative effect of estrogen in the epithelial compartments of the endometrium and promote differentiation of the uterine glands (2). However, much publicized observational and clinical studies have more recently implicated progesterone as a proliferative hormone in the normal human breast and as a life-long risk factor for breast cancer, suggesting progesterone also contributes to early stages of breast tumorigenesis (3,4,5,6,7,8,9). Paradoxically, in established breast cancers, the presence of progesterone receptor (PR) in the primary tumor is an independent marker of favorable prognosis and is associated with a more differentiated tumor phenotype, suggesting progesterone has a protective role against progression and invasion once a tumor has developed (10,11,12). Indeed, in most PR-positive breast cancer cell lines, progestins induce a transient proliferation followed by cell cycle arrest such that the long-term effect is growth inhibition (13). This apparent dichotomy suggests that progesterone signaling is altered during disease progression. Uncertainties concerning the role of progesterone in the normal human breast and tumorigenesis have been due in part to the absence of a suitable PR-positive progesterone-responsive experimental system for the normal human breast.

Because there are many common developmental and hormonal features in rodent and human mammary morphogenesis and tumorigenesis, the mouse has served as an important in vivo surrogate to model the normal human breast. The PR knockout mouse has clearly established a proliferative role for progesterone in the mammary epithelium (2,14). In the postpubertal nulliparous animal, progesterone stimulates ductal side branching and is required for the massive epithelial cell expansion and alveologenesis that occurs during early pregnancy (2,14,15). The PR knockout mouse has also underscored the critical role of PR-mediated signaling in mammary tumorigenesis (16,17), a finding supported by independent rodent mammary tumor models (18,19,20). Despite significant advancements in our understanding of the cellular and molecular principles that underlie progesterone action in the rodent mammary epithelium, translating these findings to the human has also been difficult without a suitable experimental system for the normal human breast.

In this issue of Endocrinology, Graham et al. (21) report on an improved basement membrane-embedded system for three-dimensional (3D) culture of primary normal human breast cells that retains estrogen receptor (ER) and PR expression and responsiveness to progesterone. This for the first time establishes an in vitro system for sex-steroid hormone action in the normal human breast. This builds on the pioneering work of Bissell and colleagues (22) and Debnath and Brugge (23), who showed that primary and functionally normal mammary epithelial cell lines, when grown in growth factor-reduced reconstituted basement membrane (Matrigel), form acini with a single layer of polarized epithelial cells and a hollowed out lumen that recapitulates lobuloalveolar structures of the normal breast. Furthermore, signals transmitted from components of the reconstituted basement membrane were found to be sufficient and required to maintain ER expression and responsiveness to estradiol-17β in 3D cultures of primary mouse mammary epithelial cells (22). By immunohistochemical analysis of human 3D breast acini, ER and PR were found to be heterogeneously coexpressed in luminal cells but not in myoepithelial cells. After progesterone administration, a subset of luminal epithelial cells were found to undergo proliferation, and the majority of the proliferating cells were PR negative, localized nearby PR-positive cells that were largely nonproliferating. This patterning of PR-positive and PR-negative epithelial cells recapitulates the rodent normal mammary gland and earlier immunohistochemical findings with human breast biopsy samples (24,25,26) (Fig. 1). Together these studies suggest an evolutionary conserved paracrine mechanism of progesterone regulation of mammary epithelial proliferation and that the 3D culture system is suitable to study paracrine mechanisms in the normal human breast.

Figure 1.

An emerging model for progesterone action in the human breast shares common features with a proposed paracrine mechanism of action model for progesterone in the rodent mammary gland (24,25,26). Based on findings of Graham et al., (21), the schematic outlines a hypothetical model to explain the role of progesterone (P4) signaling in the expansion and morphogenesis of the epithelial compartment in the human breast. A subset of mammary epithelial cells (red), which are PR positive, receive, transduce, and subsequently relay, in a paracrine fashion, the proliferative signal imparted by P4 hormone to a juxtacrine cell subgroup (green) that is PR negative. Gene profiling studies implicate the notch pathway as possibly one of a number of paracrine signals that project the proliferative effects of progesterone from PR-positive cells to a subset of PR-negative cells (green) that exhibit bipotent progenitor properties. Upon receiving paracrine signal inputs, the progenitor cell subtype reprograms and proliferates to generate both luminal and myoepithelial cell lineages. Luminal epithelial descendants can be positive or negative for PR expression, whereas myoepithelial cell are PR negative. Together, the expansion and differentiation of these cell lineages contribute to increased lobuloalveologenesis. Induction of DNA replication licensing factors, minichromosome maintenance proteins, and kinetochore formation proteins are possible early molecular changes that precede progesterone-induced progenitor cell reprogramming and subsequent proliferation.

Gene profiling of progesterone-induced genes in these 3D acini revealed an enrichment of targets associated with proliferation including the cell cycle and components of the DNA replication licensing machinery, minichromosome maintenance, and kinetochore formation. Interestingly, minimal overlap was observed with progesterone-induced gene targets in a breast cancer cell line grown as a 3D culture, suggesting the presence of distinct progesterone/PR signaling pathways in normal breast and breast cancer. Gene targets (Wnt-4, receptor of activated nuclear factor-κB ligand, and amphiregulin) known to be functionally important paracrine mediators of progesterone proliferation in the rodent mammary gland (27,28,29), however, were not affected by progesterone in human breast 3D cultures, suggesting either a species difference or that the microenvironment in vivo is required for progesterone regulation of these paracrine factors.

Using various immunohistochemical markers of progenitor cells and a mammosphere assay that selects for survival of progenitor cells as spheres in suspension, Graham et al. (21) further showed that a bipotent progenitor cell, which is capable of generating both luminal and myoepithelial descendants in these 3D cultures, is a likely cell type that receives the progesterone paracrine signal(s) to direct expansion of the mammary epithelial compartment (Fig. 1). Although previously suspected, these studies are the first to provide a functional link between progesterone exposure and mammary progenitor cell-fate determination (30). From molecular profiling, genes in the Notch signaling pathway were also identified as progesterone up-regulated targets, suggesting that Notch is a potential paracrine mediator of progesterone in the human breast epithelium. This proposal is significant because Notch signaling has been associated with control of tissue stem/progenitor cell fate decisions (31) and deregulation of Notch signaling has been suggested to underlie a number of epithelial cancers (32).

The progesterone-responsive human breast epithelial 3D culture system developed by Graham et al. (21) provides a promising new tool to address a number of important issues and questions. First is the influence of genetic alterations on individual and combined roles of ER- and PR-mediated signaling during the early stages of mammary epithelial neoplastic transformation and progression, including the role of the A and B isoforms of PR. Changes in normal 1:1 ratios of PR-A to PR-B have been observed to occur frequently in certain benign breast disease lesions and breast cancer, suggesting that a switch in PR isoform signaling could contribute to breast disease progression (33,34). Second is the identification of paracrine factors and mechanisms that mediate progesterone control of proliferation and how deregulation of this system contributes to transformation. A switch from a paracrine to an autocrine mode for ER/PR signaling in the human breast has been previously postulated to herald a breakdown in normal hormone responses that can lead to tumorigenesis (24,25,26). Also, these steroid signaling pathways have been predicted to influence cancer stem cell fate (30). Third is a comparison of the progesterone-induced target gene network, or transcriptome, in the normal human breast with that of the recently identified progesterone transcriptome in the normal mouse mammary gland (15) to disclose common and divergent pathways between these two species. The fourth issue is to determine whether cross talk occurs between progesterone and prolactin signaling pathways in the human breast as has been observed in the murine mammary gland (15,35). Prolactin has been shown to exert an important role in not only murine mammary morphogenesis but also breast cancer susceptibility (36). The final issue is to investigate the role of extranuclear actions of PR in mediating progestin-induced activation of cytoplasmic/cell membrane signal transduction pathways that has been studied so far almost exclusively in breast cancer cells and established immortalized cell lines (37,38). Are the signaling pathways and role of extranuclear PR the same or different in normal breast and cancer cells? Because breast cancer risk is associated with the cumulative exposure of the breast epithelium to ovarian steroids (1), the above studies promise to not only broaden our mechanistic understanding of the role of progesterone (and estrogens) in normal breast proliferation and cancer progression but also may well accelerate the formulation of more effective diagnostic, prognostic, and/or therapeutic approaches for hormone-dependent breast cancers.