**4. Role of stromal cells in regulation of mammary gland development**

### **4.1. Adipocytes**

**Figure 1.** Schematic representation of cells found within the structure of fully developed mammary gland. Scheme presents cross section of mammary alveolus surrounded by stromal components (cells and extracellular matrix).

and lobuloalveolar structures comprised of mammary epithelial cells of two primary lineages: myoepithelial (basal) cells and epithelial (luminal) cells, forming a bilayered structure, which is embedded in the stroma [12]. Mammary ducts consist of apically orientated luminal epithelial cells that line ducts with alveolar structures at the ends and of basally orientated myoepithelial cells surrounded by a laminin and collagen-rich basement membrane (BM). Luminal epithelial cells are separated from all kinds of stromal cells, laying on top of myoepithelial cells. The functionally distinct basal layer contains myoepithelial cells with contractile properties and cells with demonstrated stem cell activity, referred as mammary repopulating units (MRUs). These cells have an ability to regenerate the bilayered glandular structure of inner luminal and basal outer epithelial cells [12]. The myoepithelial and stromal cells produce the basement membrane, which is a thin sheet composed of collagen IV, laminins, entactin, and proteoglycans, and forms physical barrier separating the epithelial and stromal compartments [3]. The stromal compartment is composed of two mesenchymal lineages: adipocytes and fibroblasts, as well as infiltrating immune and vascular endothelial cells [5]. These cells synthesize extracellular matrix (ECM) components essential for three-dimensional microstructure of the stroma. Stromal ECM components include collagens, which are the major structural proteins, as well as proteoglycans, hyaluronic acid, fibronectins, and tenascins [13, 14] (**Figure 1**).

90 Stromal Cells - Structure, Function, and Therapeutic Implications

Stromal-epithelial interactions regulate mammary epithelial growth and differentiation during embryonic and postnatal development through soluble factors that are released into the environment, as well as through insoluble factors that are present in the stroma itself, referred as matrikines and matricryptins [14]. The stroma accounts for roughly 60% of the total tissue

> Adipocytes constitute the most abundant type of cells within the stroma of the mammary gland. Fat cells predominate in the stromal compartment of the mammary glands of rodents (mice and rats), whereas in the mammary glands of humans and ruminants adipocytes of white adipose tissue form the structure of a fibrous-adipose stroma along with fibroblasts. Adipocytes create a specific microenvironmental niche for MECs as the source of triglycerides and thus a source of energy, as well as a scaffold liable to invade, and a supply of various biologically active compounds.

> Adipose tissue modulates epithelial development, remodeling, and function in a statedependent manner. During embryonic morphogenesis, the fat pad together with the fibroblastic mesenchyme appears before ectoderm cell migration, creating environment and scaffold for mammary buds development. At this stage, each type of mesenchymal cells has different properties. It has been shown that fat pad mesenchyme induces elongation and branching of the mammary epithelium [5]. Lack of white adipose tissue in transgenic Z-ZIP/F1 female mice leads to compromised ductal growth during prenatal development, manifested by formation of only few underdeveloped ductal structures showing severe, abnormal distension [17]. Interestingly, these transgenic Z-ZIP/F1 mice produce a mass of lobuloalveolar structures in the mammary gland during pregnancy, which suggests that interactions between MECs and adipocytes are not essential for the functional differentiation of the mammary epithelium [17]. An alternative in vivo model of adipocytes depletion (FAT-ATTAC mice) allowed scientists to explore further the role of mammary-associated adipocytes. In FAT-ATTAC mice, elimination of adipocytes can be induced at any developmental

stage through induction of apoptotic cell death by administration of a FK1012 analog, which leads to the forced dimerization of a caspase-8 fusion protein uniquely expressed in adipose tissue [18]. This model allows for selective ablation of mammary adipocytes in female mice without affecting other fat pads. Under these conditions, Landskroner-Eiger and co-workers [18] demonstrated that the presence of adipocytes is necessary for proper formation of the extended ductal network in the mammary gland during puberty as well as for the maintenance of the normal alveolar structures that develop during adulthood. Ablation of adipocytes in mice starting from 2 weeks of age resulted in reduced ductal growth. Alterations in ductal features were caused by the loss of mechanical and physical support provided by adipocytes. However, when the loss of local adipocytes was initiated at 7 weeks of age in FAT-ATTAC mice model, an excessive lobulation was observed in the mammary gland. These observations indicate that adipocytes are critically involved in maintaining proper architecture and functionality of the mammary epithelium [18]. The important role of adipocytes in normal morphogenesis of the mammary epithelium was further confirmed in in vitro studies. MCF-10A human mammary epithelial cells co-cultured with human adipose-derived stem cells (hASCs) in Matrigel/collagen gels spread on silk scaffolds were able to create both alveolar- and duct-like structures. In contrast, monoculture of MCF-10A resulted in formation of only alveolar structures [19]. Consistently, EpH4 murine mammary epithelial cells cultured within adipose-rich collagen I formed branched mammary epithelial tubules within 24 h of culture [20]. It should be noted that the mammary-associated adipocytes also undergo massive morphological changes between the periods of lactation and involution. During lactation, adipose tissue serves as a major lipid store utilized as a source of energy for milk production. That is why in lactating mammary gland fat cells undergo lipid depletion and appear as long projections. At the time of involution, when milk synthesis ceases and mammary epithelium regresses, adipocytes regain their lipid stores, but some adipocytes undergo dedifferentiation into preadipocytes or are eliminated via apoptotic cell death [21].

induced branching morphogenesis of mammary tubules, with sites localized to the ends of the tubules, without appreciable lumen formation, which indicates that the biologically active

Stromal-Epithelial Interactions during Mammary Gland Development

http://dx.doi.org/10.5772/intechopen.80405

93

Adipokines detected in the mammary gland include hormones (leptin and adiponectin), growth factors (HGF, IGF), cytokines (interleukin 6—IL 6, tumor necrosis factor alpha— TNFα), as well as ECM components (collagen VI). It has been proven that HGF is especially important stimulator of branching morphogenesis [20]. HGF secreted by human predifferentiated hASCs affected the duct-like structure formation by mammary epithelial cells (MCF10A) in co-culture [19]. Moreover, systemic hormones (prolactin, GH) not only exert their action directly in epithelial cells but also can act indirectly via the stromal compartment of the mammary gland. Studies have shown that GH stimulates the mammary gland adipocytes to produce IGF-I [17]. It is also evident that pubertal branching morphogenesis in vivo is stimulated by steroid hormones, including estrogen, which act on receptors located in the

Leptin and adiponectin are the most extensively studied hormones synthesized by adipose tissue. They are found in higher concentrations in the mammary tissue than in blood and thus may be a part of an important paracrine or juxtacrine signaling system between adipose-rich stroma and epithelial cells [24]. Leptin, which was the first known adipokine discovered by Friedman and Coleman in 1994, is a 16 kDa nonglycosylated protein encoded by the *Ob* gene. This protein hormone is secreted mainly by adipose tissue to regulate body energy balance, suppressing food intake and thereby inducing weight loss. In the context of mammary gland physiology leptin actions are associated with regulation of the metabolic changes occurring during pregnancy and lactation, due to the fact that it is the key hormone regulating the metabolic adaptation of nutrient partitioning during the energy consuming processes [25]. MECs express leptin receptors (OB-Rb) and therefore may undergo direct regulation by leptin, whereas local production of leptin by mammary adipose tissue is under control of several hormones: insulin, glucocorticoids, and prolactin. Prolactin, the main lactogenic factor, was shown to regulate leptin and leptin receptor gene expression in the bovine mammary gland [26]. It is believed that prolactin may be the key signaling factor stimulating the mammary gland to interact with leptin in the regulation of milk synthesis during lactation [27]. In the presence of prolactin, leptin was shown to enhance the expression of α-casein gene (milk protein gene) in bovine mammary gland, indicating that leptin and prolactin interact to alter milk synthesis during lactation [27]. Estradiol, which is known to regulate ductal morphogenesis in the mammary gland, also plays an important role in the regulation of the extracellular

levels of leptin, as well as adiponectin in normal human mammary gland [28].

In contrast to leptin, circulating levels of adiponectin are inversely correlated with the body mass index (BMI). Adiponectin is a 240 amino acid protein of approximately 28–30 kDa existing as a monomer, although it forms dimmers and multimers, circulating as low, medium, and high molecular weight isoforms. Two types of receptors, adiponectin receptor 1 (AdipoR1) and adiponectin receptor 2 (AdipoR2), have distinct distribution patterns in different tissues. Both receptors were shown to be expressed in normal mammary epithelial cells [29, 30]. Binding of adiponectin to its receptor activates adenosine monophosphate-activated protein

molecules produced by adipocytes influence mostly the ductal growth [20].

stroma to induce production of mitogens including HGF [20].

#### *4.1.1. Adipokines*

Beyond the function of adipocytes as the energy storage depot, currently it is well accepted that these cells are actively producing and secreting a wide range of endocrine factors referred to as adipokines. Adipokines are signaling molecules that regulate various physiological processes in the body. In the context of the mammary gland, adipokines are thought to regulate normal development of this organ [22]. This group of compounds is also locally synthesized by adipocytes of the mammary stroma and act through juxtacrine or paracrine signals modulating epithelial cells proliferation. In vitro studies on normal human MECs (NMuMG cell line) elegantly demonstrated the effect of signaling molecules secreted by adipocytes. NMuMG cells were incubated for 24 or 48 h in the presence of conditioned medium derived from adipocytes (3T3-L1 cell line) at various degrees of differentiation: preadipocytes (preA), poorly differentiated adipocytes (pDA), and mature adipocytes (MA) [23]. After 24 h treatment human MECs showed significantly increased proliferative activity when cultured in conditioned media from pDA and MA, whereas after 48 h incubation the effect of increase proliferation was observed in the case of all conditioned media (preA, pDA, and MA) [23]. Another study revealed that 24 h treatment with conditioned medium from mature adipocytes induced branching morphogenesis of mammary tubules, with sites localized to the ends of the tubules, without appreciable lumen formation, which indicates that the biologically active molecules produced by adipocytes influence mostly the ductal growth [20].

stage through induction of apoptotic cell death by administration of a FK1012 analog, which leads to the forced dimerization of a caspase-8 fusion protein uniquely expressed in adipose tissue [18]. This model allows for selective ablation of mammary adipocytes in female mice without affecting other fat pads. Under these conditions, Landskroner-Eiger and co-workers [18] demonstrated that the presence of adipocytes is necessary for proper formation of the extended ductal network in the mammary gland during puberty as well as for the maintenance of the normal alveolar structures that develop during adulthood. Ablation of adipocytes in mice starting from 2 weeks of age resulted in reduced ductal growth. Alterations in ductal features were caused by the loss of mechanical and physical support provided by adipocytes. However, when the loss of local adipocytes was initiated at 7 weeks of age in FAT-ATTAC mice model, an excessive lobulation was observed in the mammary gland. These observations indicate that adipocytes are critically involved in maintaining proper architecture and functionality of the mammary epithelium [18]. The important role of adipocytes in normal morphogenesis of the mammary epithelium was further confirmed in in vitro studies. MCF-10A human mammary epithelial cells co-cultured with human adipose-derived stem cells (hASCs) in Matrigel/collagen gels spread on silk scaffolds were able to create both alveolar- and duct-like structures. In contrast, monoculture of MCF-10A resulted in formation of only alveolar structures [19]. Consistently, EpH4 murine mammary epithelial cells cultured within adipose-rich collagen I formed branched mammary epithelial tubules within 24 h of culture [20]. It should be noted that the mammary-associated adipocytes also undergo massive morphological changes between the periods of lactation and involution. During lactation, adipose tissue serves as a major lipid store utilized as a source of energy for milk production. That is why in lactating mammary gland fat cells undergo lipid depletion and appear as long projections. At the time of involution, when milk synthesis ceases and mammary epithelium regresses, adipocytes regain their lipid stores, but some adipocytes undergo dedifferentiation into preadipocytes or are eliminated via apoptotic cell death [21].

92 Stromal Cells - Structure, Function, and Therapeutic Implications

Beyond the function of adipocytes as the energy storage depot, currently it is well accepted that these cells are actively producing and secreting a wide range of endocrine factors referred to as adipokines. Adipokines are signaling molecules that regulate various physiological processes in the body. In the context of the mammary gland, adipokines are thought to regulate normal development of this organ [22]. This group of compounds is also locally synthesized by adipocytes of the mammary stroma and act through juxtacrine or paracrine signals modulating epithelial cells proliferation. In vitro studies on normal human MECs (NMuMG cell line) elegantly demonstrated the effect of signaling molecules secreted by adipocytes. NMuMG cells were incubated for 24 or 48 h in the presence of conditioned medium derived from adipocytes (3T3-L1 cell line) at various degrees of differentiation: preadipocytes (preA), poorly differentiated adipocytes (pDA), and mature adipocytes (MA) [23]. After 24 h treatment human MECs showed significantly increased proliferative activity when cultured in conditioned media from pDA and MA, whereas after 48 h incubation the effect of increase proliferation was observed in the case of all conditioned media (preA, pDA, and MA) [23]. Another study revealed that 24 h treatment with conditioned medium from mature adipocytes

*4.1.1. Adipokines*

Adipokines detected in the mammary gland include hormones (leptin and adiponectin), growth factors (HGF, IGF), cytokines (interleukin 6—IL 6, tumor necrosis factor alpha— TNFα), as well as ECM components (collagen VI). It has been proven that HGF is especially important stimulator of branching morphogenesis [20]. HGF secreted by human predifferentiated hASCs affected the duct-like structure formation by mammary epithelial cells (MCF10A) in co-culture [19]. Moreover, systemic hormones (prolactin, GH) not only exert their action directly in epithelial cells but also can act indirectly via the stromal compartment of the mammary gland. Studies have shown that GH stimulates the mammary gland adipocytes to produce IGF-I [17]. It is also evident that pubertal branching morphogenesis in vivo is stimulated by steroid hormones, including estrogen, which act on receptors located in the stroma to induce production of mitogens including HGF [20].

Leptin and adiponectin are the most extensively studied hormones synthesized by adipose tissue. They are found in higher concentrations in the mammary tissue than in blood and thus may be a part of an important paracrine or juxtacrine signaling system between adipose-rich stroma and epithelial cells [24]. Leptin, which was the first known adipokine discovered by Friedman and Coleman in 1994, is a 16 kDa nonglycosylated protein encoded by the *Ob* gene. This protein hormone is secreted mainly by adipose tissue to regulate body energy balance, suppressing food intake and thereby inducing weight loss. In the context of mammary gland physiology leptin actions are associated with regulation of the metabolic changes occurring during pregnancy and lactation, due to the fact that it is the key hormone regulating the metabolic adaptation of nutrient partitioning during the energy consuming processes [25]. MECs express leptin receptors (OB-Rb) and therefore may undergo direct regulation by leptin, whereas local production of leptin by mammary adipose tissue is under control of several hormones: insulin, glucocorticoids, and prolactin. Prolactin, the main lactogenic factor, was shown to regulate leptin and leptin receptor gene expression in the bovine mammary gland [26]. It is believed that prolactin may be the key signaling factor stimulating the mammary gland to interact with leptin in the regulation of milk synthesis during lactation [27]. In the presence of prolactin, leptin was shown to enhance the expression of α-casein gene (milk protein gene) in bovine mammary gland, indicating that leptin and prolactin interact to alter milk synthesis during lactation [27]. Estradiol, which is known to regulate ductal morphogenesis in the mammary gland, also plays an important role in the regulation of the extracellular levels of leptin, as well as adiponectin in normal human mammary gland [28].

In contrast to leptin, circulating levels of adiponectin are inversely correlated with the body mass index (BMI). Adiponectin is a 240 amino acid protein of approximately 28–30 kDa existing as a monomer, although it forms dimmers and multimers, circulating as low, medium, and high molecular weight isoforms. Two types of receptors, adiponectin receptor 1 (AdipoR1) and adiponectin receptor 2 (AdipoR2), have distinct distribution patterns in different tissues. Both receptors were shown to be expressed in normal mammary epithelial cells [29, 30]. Binding of adiponectin to its receptor activates adenosine monophosphate-activated protein kinase (AMPK), a nutrient-sensing enzyme, which regulates several key pathways involved in protein synthesis and cellular energy metabolism. One of the few researches on bovine mammary gland disclosed that adiponectin expression in the mammary gland decreases in the peak and late-lactation period, although adiponectin receptor 1 (AdipoR1) expression increases in the same period [30]. Moreover, leptin/adiponectin ratio is directly proportional to the size of stem cell population in vivo. It was evidenced that leptin alone is sufficient to stimulate mammary stem cell self-renewal, leading to significant increase in the stem cell population. In contrast, unopposed adiponectin decreases the size of the mammary stem cell pool in vitro. It is believed that leptin and adiponectin may function as both endocrine and paracrine/juxtacrine factors to modulate the size of the normal stem cell pool [24].

differentiation, and apoptosis in human MECs, regulating growth in normal and transformed cells [34–36]. The ability of human MECs to synthesize 1,25D3 locally within the mammary epithelium to regulate cellular growth and differentiation may constitute a potential mechanism by which elevated serum 25D3 is associated with a decreased risk of developing breast cancer or metastatic progression [37]. Ching and co-workers [38] investigated the hypothesis that adipocytes from the mammary stroma express the signaling components necessary to participate in vitamin D3 synthesis and act via VDR, potentially modulating ductal epithelial cell growth and differentiation. Mammary adipocytes expressing VDR were shown to participate in bioactivating 25-hydroxyvitamin D3 (25D3) to the active ligand, 1,25D3, and secrete it to the surrounding microenvironment. Active vitamin D3 in turn was able to inhibit the ductal epithelial cell growth [38]. Similar results were obtained by Matthews and co-workers, who used a different animal model in their studies [39]. This group generated CVF transgenic mice with adipose-specific *Vdr* gene deletion and noted that adipose deletion of *Vdr* significantly enhanced mammary epithelial density and branching, supporting the hypothesis that vitamin D receptor

Stromal-Epithelial Interactions during Mammary Gland Development

http://dx.doi.org/10.5772/intechopen.80405

95

in mature adipocytes exerts anti-proliferative effects on the mammary epithelium [39].

In terms of investigating interactions between epithelial and stromal compartments of the mammary gland, it is important to expand our knowledge about reciprocal cell-cell interactions within the gland. There are still relatively few studies focused on the influence of MECs on the adipocytes population. A vivid example is an in vitro model of three-dimensional (3D) collagen gels containing differentiated adipocytes, which were used to investigate the mutual interactions between adipocytes and MECs during branching morphogenesis [20]. In this research, 3T3-L1 mouse preadipocytes were embedded in collagen, differentiated, and then treated with MECs-derived conditioned medium. Samples treated with conditioned media formed fewer and smaller fatty clusters and showed lower expression of lipoprotein lipase (LPL) and adipogenic transcription factor PPARγ2. These data suggest that MECs either inhibited or delayed differentiation of the preadipocytes [20]. In vivo, during embryonic mammary gland development, the fat pad is present before the epithelium invades, and epithelial compartment invades the stroma causing its reduction [20]. Similar conclusion was made by investigators who demonstrated that MECs produce the enzyme galactose 3-*O*-sulfotransferase 2 (GAL3STS2), which was able to inhibit the expression of adipogenic transcription factor C/EBPb and fatty acid-binding protein 4 (FABP4)—a marker of adipocytes differentiation [40]. In addition, accumulation of triglycerides was also inhibited under the influence of GAL3STS2. The authors postulate that GAL3ST2 may generate multiple signals related to integrin activation, including its effect on preadipocyte differentiation [40]. Taken together, it seems that epithelial compartment reduces the adipose tissue during mammary gland morphogenesis and works as negative feedback creating an appropriate/ favorable microenvironment for itself.

Stromal adipocytes play a profound role in regulation of mammogenesis during both embryonic and postnatal development of the mammary gland. These cells are necessary for proper ductal elongation and branching and are critically involved in maintaining proper architecture and

*4.1.3. Influence of mammary epithelium on stromal adipocytes*

*4.1.4. Summary*

Recent studies have shown chemerin as a novel adipokine, which may actively take part in regulation of the mammary gland lactogenesis. Chemerin, also called retinoic acid receptor responder protein 2 (RARRES2), is a 16 kDa chemoattractant cytokine (chemokine) mainly expressed in and secreted from white adipose tissue. Chemerin is secreted as a 143-amino acid inactive precursor, pro-chemerin, and is activated by proteolytic removal of six to seven amino acids from its C-terminus by proteases such as elastase or cathepsin G. Three G proteincoupled receptors are able to bind chemerin with high affinity, namely chemokine receptorlike 1 (CMKLR1), G protein-coupled receptor 1 (GPR1), and C-C chemokine receptor-like 2 (CCRL2). Chemerin inhibits cAMP production and promotes phospholipase C activation, IP3 release, calcium mobilization as well as activation of PI3K and MAPK pathways [31]. In bovine mammary gland, the expression of chemerin was greater in adipose tissue of postpartum dairy cows versus pregnant cows, and two out of three chemerin receptors (CMKLR1 and CCRL2) were expressed in bovine MECs [31]. Studies with immortalized bovine MECs treated with chemerin revealed upregulated expression of genes associated with fatty acid synthesis, glucose uptake, and casein synthesis; thus, it is postulated that chemerin may play a role of lactogenesis regulator in bovine mammary epithelium. Surprisingly, adiponectin reduced the expression of CMKLR1 receptor, without altering CCRL2 expression [30]. These results imply that adiponectin is not only able to counteract the effects of leptin but also able to regulate the influence of chemerin on mammary epithelial cells.

#### *4.1.2. Other adipocyte-related molecular regulators of mammogenesis*

Adipocytes of the mammary stroma also express retinoids (RARs), which are potent transcription regulators [32]. Co-cultures of primary adipocytes, or in vitro differentiated adipocyte cell line, with mammary epithelium showed that when activated, adipocyte-RARs contribute to generation of secreted proliferative and pro-migratory factors affecting branching morphogenesis [33]. RARs expressed by adipocytes were shown to be important regulators of secreted growth factor—pleiotrophin (PTN), involved in paracrine regulation of epithelial ductal tree development [33]. Adipocyte-RARs induced parathyroid hormone receptor (PTHR) expression leading to increased expression of PTN, which in turn regulated mammary epithelial migration.

Adipocytes also express vitamin D receptor (VDR), which is expressed in both epithelial and stromal compartment of the mammary gland and is known to participate in regulation of hormone-induced growth and differentiation throughout development [34]. VDR complexes with the active ligand, 1a,25-dihydroxyvitamin D3 (1,25D3), to induce cell cycle arrest, differentiation, and apoptosis in human MECs, regulating growth in normal and transformed cells [34–36]. The ability of human MECs to synthesize 1,25D3 locally within the mammary epithelium to regulate cellular growth and differentiation may constitute a potential mechanism by which elevated serum 25D3 is associated with a decreased risk of developing breast cancer or metastatic progression [37]. Ching and co-workers [38] investigated the hypothesis that adipocytes from the mammary stroma express the signaling components necessary to participate in vitamin D3 synthesis and act via VDR, potentially modulating ductal epithelial cell growth and differentiation. Mammary adipocytes expressing VDR were shown to participate in bioactivating 25-hydroxyvitamin D3 (25D3) to the active ligand, 1,25D3, and secrete it to the surrounding microenvironment. Active vitamin D3 in turn was able to inhibit the ductal epithelial cell growth [38]. Similar results were obtained by Matthews and co-workers, who used a different animal model in their studies [39]. This group generated CVF transgenic mice with adipose-specific *Vdr* gene deletion and noted that adipose deletion of *Vdr* significantly enhanced mammary epithelial density and branching, supporting the hypothesis that vitamin D receptor in mature adipocytes exerts anti-proliferative effects on the mammary epithelium [39].

#### *4.1.3. Influence of mammary epithelium on stromal adipocytes*

In terms of investigating interactions between epithelial and stromal compartments of the mammary gland, it is important to expand our knowledge about reciprocal cell-cell interactions within the gland. There are still relatively few studies focused on the influence of MECs on the adipocytes population. A vivid example is an in vitro model of three-dimensional (3D) collagen gels containing differentiated adipocytes, which were used to investigate the mutual interactions between adipocytes and MECs during branching morphogenesis [20]. In this research, 3T3-L1 mouse preadipocytes were embedded in collagen, differentiated, and then treated with MECs-derived conditioned medium. Samples treated with conditioned media formed fewer and smaller fatty clusters and showed lower expression of lipoprotein lipase (LPL) and adipogenic transcription factor PPARγ2. These data suggest that MECs either inhibited or delayed differentiation of the preadipocytes [20]. In vivo, during embryonic mammary gland development, the fat pad is present before the epithelium invades, and epithelial compartment invades the stroma causing its reduction [20]. Similar conclusion was made by investigators who demonstrated that MECs produce the enzyme galactose 3-*O*-sulfotransferase 2 (GAL3STS2), which was able to inhibit the expression of adipogenic transcription factor C/EBPb and fatty acid-binding protein 4 (FABP4)—a marker of adipocytes differentiation [40]. In addition, accumulation of triglycerides was also inhibited under the influence of GAL3STS2. The authors postulate that GAL3ST2 may generate multiple signals related to integrin activation, including its effect on preadipocyte differentiation [40]. Taken together, it seems that epithelial compartment reduces the adipose tissue during mammary gland morphogenesis and works as negative feedback creating an appropriate/ favorable microenvironment for itself.

#### *4.1.4. Summary*

kinase (AMPK), a nutrient-sensing enzyme, which regulates several key pathways involved in protein synthesis and cellular energy metabolism. One of the few researches on bovine mammary gland disclosed that adiponectin expression in the mammary gland decreases in the peak and late-lactation period, although adiponectin receptor 1 (AdipoR1) expression increases in the same period [30]. Moreover, leptin/adiponectin ratio is directly proportional to the size of stem cell population in vivo. It was evidenced that leptin alone is sufficient to stimulate mammary stem cell self-renewal, leading to significant increase in the stem cell population. In contrast, unopposed adiponectin decreases the size of the mammary stem cell pool in vitro. It is believed that leptin and adiponectin may function as both endocrine and

Recent studies have shown chemerin as a novel adipokine, which may actively take part in regulation of the mammary gland lactogenesis. Chemerin, also called retinoic acid receptor responder protein 2 (RARRES2), is a 16 kDa chemoattractant cytokine (chemokine) mainly expressed in and secreted from white adipose tissue. Chemerin is secreted as a 143-amino acid inactive precursor, pro-chemerin, and is activated by proteolytic removal of six to seven amino acids from its C-terminus by proteases such as elastase or cathepsin G. Three G proteincoupled receptors are able to bind chemerin with high affinity, namely chemokine receptorlike 1 (CMKLR1), G protein-coupled receptor 1 (GPR1), and C-C chemokine receptor-like 2 (CCRL2). Chemerin inhibits cAMP production and promotes phospholipase C activation, IP3 release, calcium mobilization as well as activation of PI3K and MAPK pathways [31]. In bovine mammary gland, the expression of chemerin was greater in adipose tissue of postpartum dairy cows versus pregnant cows, and two out of three chemerin receptors (CMKLR1 and CCRL2) were expressed in bovine MECs [31]. Studies with immortalized bovine MECs treated with chemerin revealed upregulated expression of genes associated with fatty acid synthesis, glucose uptake, and casein synthesis; thus, it is postulated that chemerin may play a role of lactogenesis regulator in bovine mammary epithelium. Surprisingly, adiponectin reduced the expression of CMKLR1 receptor, without altering CCRL2 expression [30]. These results imply that adiponectin is not only able to counteract the effects of leptin but also able

Adipocytes of the mammary stroma also express retinoids (RARs), which are potent transcription regulators [32]. Co-cultures of primary adipocytes, or in vitro differentiated adipocyte cell line, with mammary epithelium showed that when activated, adipocyte-RARs contribute to generation of secreted proliferative and pro-migratory factors affecting branching morphogenesis [33]. RARs expressed by adipocytes were shown to be important regulators of secreted growth factor—pleiotrophin (PTN), involved in paracrine regulation of epithelial ductal tree development [33]. Adipocyte-RARs induced parathyroid hormone receptor (PTHR) expression leading

to increased expression of PTN, which in turn regulated mammary epithelial migration.

Adipocytes also express vitamin D receptor (VDR), which is expressed in both epithelial and stromal compartment of the mammary gland and is known to participate in regulation of hormone-induced growth and differentiation throughout development [34]. VDR complexes with the active ligand, 1a,25-dihydroxyvitamin D3 (1,25D3), to induce cell cycle arrest,

paracrine/juxtacrine factors to modulate the size of the normal stem cell pool [24].

94 Stromal Cells - Structure, Function, and Therapeutic Implications

to regulate the influence of chemerin on mammary epithelial cells.

*4.1.2. Other adipocyte-related molecular regulators of mammogenesis*

Stromal adipocytes play a profound role in regulation of mammogenesis during both embryonic and postnatal development of the mammary gland. These cells are necessary for proper ductal elongation and branching and are critically involved in maintaining proper architecture and function of the mammary epithelium. This effect is exerted through direct cell-cell contact with the mammary epithelial cells as well as through paracrine signals induced by secreted adipokines. This group of biologically active molecules includes HGF supporting ductal morphogenesis, leptin and adiponectin that may modulate the size of the mammary stem cell pool within the glandular tissue, as well as chemerin, which may be a novel, local regulator of lactogenesis, as it is involved in regulation of fatty acids and milk protein synthesis and glucose uptake (**Figure 2**).

**4.2. Fibroblasts**

cells [41].

stimuli from the blood supply must pass [44].

Fibroblasts, together with adipocytes, are the major cellular components of the mammary stroma and play an integral role in regulating mammary gland development. As mentioned previously, during prenatal period of mammogenesis, the fat pad and fibroblastic mesenchyme appear before ectoderm cell migration, creating the environment and scaffold for emerging mammary buds [4]. Fibroblastic cells of the mesenchyme are in direct contact with the developing epithelial rudiment, and their signals first determine the identity of MECs [41]. In parallel, the epithelium also influences mesenchymal maturation. Research done on murine model of mammary gland morphogenesis revealed that by day 14 of mouse embryonic development the mammary mesenchyme condenses to form a few layers of fibroblastrich cells closely surrounding the epithelial rudiment, and it is distinct from the fat pad precorsor tissue, which develops from more deeply located subcutaneous mesenchymal

Stromal-Epithelial Interactions during Mammary Gland Development

http://dx.doi.org/10.5772/intechopen.80405

97

Moreover, it has been shown that more than one phenotype of normal fibroblasts can be distinguished within the stromal compartment of the mammary gland, and each has the potential for various epigenetic effects on normal epithelial cells depending on their proximity to the parenchyma [42]. Intralobular fibroblasts can be distinguished from interlobular fibroblast as they differ in the expression patterns of several proteins such as collagen type XIV and CD13 [43]. Morsing and co-workers conducted a study using fluorescence-activated cell sorting analysis, by which they were able to isolate and characterize two lineages of stromal fibroblasts from human mammary gland, and showed their different impact on the mammary epithelium [44]. Lobular fibroblasts were characterized by high expression of a surface glycoprotein CD105 (which is a part of the TGF beta receptor complex) and low expression of CD26 surface marker, also known as dipeptidyl peptidase-4. In terms of biological properties, CD105high/CD26low lobular fibroblasts resembled mesenchymal stem cells and supported luminal epithelial growth and branching morphogenesis. On the other hand, the second identified fibroblastic cells subpopulation, termed interlobular fibroblasts, showed low expression of CD105 and high expression of CD26 and did not exert such impact on the branching morphogenesis of epithelial progenitors [44]. It has been suggested that the interstitial stroma serves mainly to form a barrier between capillaries and epithelium, across which epitheliotropic

It is worth noting that contrary to the overall structure of the mammary parenchyma, which is similar among mammalian species being composed of bilayered luminal and basal epithelial cells, the relative abundance of connective tissue is more species-specific. Stroma surrounding the lobules and ducts (intra and interlobular stroma) in mice is sparse, and there is little non-cellular fibrous connective tissue between ducts, whereas the white adipose tissue is abundant. In humans, the ratio of fibrous connective tissue to adipose is opposite, with an abundance of stroma surrounding the alveoli and ducts, predominance of fibrous connective tissue between ducts, and reduced adipose content [14]. Interestingly, in the case of outbred Sprague Dawley female rats, the organization of the mammary stroma is intermediate between mice and humans, and it is thought that its histological pattern is more similar to the one observed in humans than mice. The mammary stroma in cattle is also more fibrous

**Figure 2.** Schematic representation of different types of interactions between mammary epithelial cells [forming terminal end bud (TEB)] and stromal cells during mammogenesis.

#### **4.2. Fibroblasts**

function of the mammary epithelium. This effect is exerted through direct cell-cell contact with the mammary epithelial cells as well as through paracrine signals induced by secreted adipokines. This group of biologically active molecules includes HGF supporting ductal morphogenesis, leptin and adiponectin that may modulate the size of the mammary stem cell pool within the glandular tissue, as well as chemerin, which may be a novel, local regulator of lactogenesis, as it is involved in regulation of fatty acids and milk protein synthesis and glucose uptake (**Figure 2**).

96 Stromal Cells - Structure, Function, and Therapeutic Implications

**Figure 2.** Schematic representation of different types of interactions between mammary epithelial cells [forming terminal

end bud (TEB)] and stromal cells during mammogenesis.

Fibroblasts, together with adipocytes, are the major cellular components of the mammary stroma and play an integral role in regulating mammary gland development. As mentioned previously, during prenatal period of mammogenesis, the fat pad and fibroblastic mesenchyme appear before ectoderm cell migration, creating the environment and scaffold for emerging mammary buds [4]. Fibroblastic cells of the mesenchyme are in direct contact with the developing epithelial rudiment, and their signals first determine the identity of MECs [41]. In parallel, the epithelium also influences mesenchymal maturation. Research done on murine model of mammary gland morphogenesis revealed that by day 14 of mouse embryonic development the mammary mesenchyme condenses to form a few layers of fibroblastrich cells closely surrounding the epithelial rudiment, and it is distinct from the fat pad precorsor tissue, which develops from more deeply located subcutaneous mesenchymal cells [41].

Moreover, it has been shown that more than one phenotype of normal fibroblasts can be distinguished within the stromal compartment of the mammary gland, and each has the potential for various epigenetic effects on normal epithelial cells depending on their proximity to the parenchyma [42]. Intralobular fibroblasts can be distinguished from interlobular fibroblast as they differ in the expression patterns of several proteins such as collagen type XIV and CD13 [43]. Morsing and co-workers conducted a study using fluorescence-activated cell sorting analysis, by which they were able to isolate and characterize two lineages of stromal fibroblasts from human mammary gland, and showed their different impact on the mammary epithelium [44]. Lobular fibroblasts were characterized by high expression of a surface glycoprotein CD105 (which is a part of the TGF beta receptor complex) and low expression of CD26 surface marker, also known as dipeptidyl peptidase-4. In terms of biological properties, CD105high/CD26low lobular fibroblasts resembled mesenchymal stem cells and supported luminal epithelial growth and branching morphogenesis. On the other hand, the second identified fibroblastic cells subpopulation, termed interlobular fibroblasts, showed low expression of CD105 and high expression of CD26 and did not exert such impact on the branching morphogenesis of epithelial progenitors [44]. It has been suggested that the interstitial stroma serves mainly to form a barrier between capillaries and epithelium, across which epitheliotropic stimuli from the blood supply must pass [44].

It is worth noting that contrary to the overall structure of the mammary parenchyma, which is similar among mammalian species being composed of bilayered luminal and basal epithelial cells, the relative abundance of connective tissue is more species-specific. Stroma surrounding the lobules and ducts (intra and interlobular stroma) in mice is sparse, and there is little non-cellular fibrous connective tissue between ducts, whereas the white adipose tissue is abundant. In humans, the ratio of fibrous connective tissue to adipose is opposite, with an abundance of stroma surrounding the alveoli and ducts, predominance of fibrous connective tissue between ducts, and reduced adipose content [14]. Interestingly, in the case of outbred Sprague Dawley female rats, the organization of the mammary stroma is intermediate between mice and humans, and it is thought that its histological pattern is more similar to the one observed in humans than mice. The mammary stroma in cattle is also more fibrous and contains less adipose tissue than the fatty mouse mammary stroma. The morphology of the bovine mammary gland resembles that of the human breast, because the mammary epithelium is generally closely associated with fibrous connective tissue, which in this case is extensively developed [45].

lacking either FGF10 or FGFR2b fail to form mammary placodes 1, 2, 3, and 5 [51]. In mouse embryos lacking *Fgf10* gene, an epithelial sprout derived from placode 4 failed to branch, which completely inhibited the formation of a primitive epithelial network in the neonatal mice after birth [51]. In humans, a birth defect known as Poland syndrome, which is characterized by the underdevelopment of the somite-derived pectoral muscle on one side of the body and a corresponding hypoplasia of the overlying breast on the same side, arises from disruption in FGF10 signaling, because *Fgf10*+/− glands show reduced thickening of the ectoderm along the mammary line and subsequent loss of buds 3 and 5 [6]. Furthermore, secreted FGFs are known to stimulate TEBs promoting luminal epithelial cell expansion, ductal branching, and their differentiation into myoepithelial cells. The majority of FGFs is involved in branching process and involution, both of which require ECM rearrangement. In the case of pregnancy, signals through FGFR2-IIIb are essential to stimulate normal lobuloalveolar development [48]. Recent studies revealed that *Spry2* gene, which encodes an inhibitor of signaling via receptor tyrosine kinases, is essential for regulation of both FGF2-based ductal elongation and FGF10-mediated epithelial invasion during normal mammary gland development. For example, loss of *Spry*2 expression results in increased FGF signaling activities, causing more rapid ductal elongation and epithelial invasion, which leads to accelerated epithelial invasion during pubertal branching. Conversely, a decrease of FGF signaling leads to slower than normal ductal elongation and invasion, resulting in stunted epithelial invasion during postnatal branching of the mammary gland [52]. It was also revealed that basal epithelial cells lacking *Fgfr2* gene did not generate an epithelial network due to failure in luminal differentiation, and *Fgfr2−/−* epithelium was unable to undergo ductal branching initiation, which depends on directional epithelial migration [53]. The results of the abovementioned studies demonstrated that distinct types of FGFs stimulate epithelial cells on different levels. FGF2 controls the ductal elongation process, which depends on cell proliferation and expansion, while FGF10

Stromal-Epithelial Interactions during Mammary Gland Development

http://dx.doi.org/10.5772/intechopen.80405

99

regulates the branch initiation process depended on directional epithelial migration.

pletely abolished by a neutralizing antibody against HGF [41].

Other fibroblast-derived bioactive compounds like TGF-β1, HGF, or stroma cell-derived factor-1 (SDF-1) also known as CXCL12, were shown to influence mammary parenchyma development in a paracrine manner [54, 55]. HGF is a multi-functional cytokine stimulating invasion, motility, and morphogenesis. Its presence was found in conditioned media from human mammary fibroblasts [56, 57]. Fibroblast-derived conditioned media containing HGF were shown to induce tubulogenesis and branching morphogenesis of TAC-2 mouse mammary epithelial cell line [20]. In addition, it is well documented that fibroblastic HGF mediates the proliferation of estrogen receptor positive (ER+) mammary epithelial cells [43]. HGF was identified as one of the major mediators of this effect, because in in vitro experiments the proliferative activity of MECs cultured in fibroblast-derived conditioned medium was com-

Another important growth factor—TGF-β1, secreted by the mammary stroma, acts in an auto/ paracrine manner to regulate glandular morphogenesis and remodeling by preventing inappropriate side branching. The presence of TGF-β1 was detected in mature periductal ECM in mice, and it was specifically downregulated at sites where side branches were being initiated [58]. Furthermore, TGF-β1 plays an important role in regulation of growth and activity of fibroblasts. This growth factor functions by signaling to cell surface type II receptors, which recruit

#### *4.2.1. Fibroblast-mammary epithelial cell interactions during mammogenesis*

The composition of the mammary stroma largely determines the progression of glandular epithelium development. Attempts to recapitulate human breast epithelial morphogenesis by introducing human MECs into the cleared mammary fat pads of mice were unsuccessful for a long time, due to improper composition of murine stroma comprising mainly adipocytes. Kuperwasser and co-workers used a different approach, creating a model of humanized mouse mammary gland by injecting immortalized human mammary stromal fibroblasts labeled with green fluorescence protein (GFP) into the cleared mice mammary fat pad prior to injection of human breast organoids. Addition of human fibroblasts to the murine fat pad effectively stimulated human MECs proliferation and promoted organization of differentiated acini structures [46]. This experiment pointed to tight stromal-epithelial species affinity [46]. A follow-up study was made, in which human macrophages were also injected. This procedure intensified humanization of the murine fat pad by enhancing fibroblast proliferation and engraftment of the mammary fat pad, thereby providing a larger stromal scaffold for breast epithelial growth and acini formation [47].
