Loss of sfrp1 promotes ductal branching in the murine mammary gland
© Gauger et al.; licensee BioMed Central Ltd. 2012
Received: 28 February 2012
Accepted: 13 August 2012
Published: 28 August 2012
Skip to main content
© Gauger et al.; licensee BioMed Central Ltd. 2012
Received: 28 February 2012
Accepted: 13 August 2012
Published: 28 August 2012
Secreted frizzled-related proteins (SFRPs) are a family of proteins that block the Wnt signaling pathway and loss of SFRP1 expression is found in breast cancer along with a multitude of other human cancers. Activated Wnt signaling leads to inappropriate mammary gland development and mammary tumorigenesis in mice. When SFRP1 is knocked down in immortalized non-malignant mammary epithelial cells, the cells exhibit a malignant phenotype which resembles the characteristics observed in metastatic breast cancer stem-like cells. However, the effects of SFRP1 loss on mammary gland development in vivo are yet to be elucidated. The work described here was initiated to investigate the role of SFRP1 in mammary gland development and whether SFRP1−/− mice exhibit changes in mammary gland morphology and cell signaling pathways shown to be associated with SFRP1 loss in vitro.
10 week old nulliparous SFRP1−/− mammary glands exhibited branching with clear lobulo-alveolar development, which normally only occurs in hormonally stimulated mid-pregnant wt mammary glands. Explant cultures of SFRP1−/− mammary glands display increased levels of a well known Wnt signaling target gene, Axin2. Histomorphologic evaluation of virgin glands revealed that by 10 weeks of age, the duct profile is markedly altered in SFRP1−/− mice showing a significantly higher density of ducts with distinct alveoli present throughout the mammary gland, and with focal ductal epithelial hyperplasia. These findings persist as the mice age and are evident at 23 weeks of age. Changes in gene expression, including c-Myc, TGFβ-2, Wnt4, RANKL, and Rspo2 early in mammary gland development are consistent with the excessive hyper branching phenotype. Finally, we found that loss of SFRP1 significantly increases the number of mammary epithelial cells capable of mammosphere formation.
Our study indicates that SFRP1 gene is critical for maintaining proper mammary gland development, and that reduced levels of SFRP1 results in hyperplastic lesions and its loss may be a critical event in cancer initiation.
Members of the Wnt family of secreted proteins play crucial roles in the morphogenesis of most organ systems and regulate cellular proliferation, differentiation, migration and apoptosis [1–3]. The best characterized Wnt pathway is the canonical Wnt/β-catenin pathway whereby β-catenin stimulates the expression of specific target genes including the c-Myc oncogene. Atypical activation of the Wnt/β-catenin signaling pathway contributes to the genesis of a wide range of human cancers, including breast cancer . In the murine breast, aberrantly activated Wnt signaling leads to inappropriate mammary gland development and mammary tumorigenesis .
In addition to embryonic development, Wnt family members are common regulators of postnatal development and of homeostasis in adult tissues, including the growth and differentiation of the mammary gland [4–6]. The first stage of mammary gland development in mice is the specification of mammary rudiments, a prenatal event which occurs around embryonic day 10. The second stage of mammary development begins at around 3 weeks of age at the onset of puberty in response to ovarian hormones, resulting in a rapid expansion of the preexisting rudimentary ductal tree . Branching ductal morphogenesis proceeds across the entire mammary fat pad and is completed at approximately 10 weeks of age. During this phase of development, the growing tips of the ducts form the highly proliferative terminal end buds (TEBs), which house mammary stem cells [8, 9]. The major changes in morphology and function that are initiated at the start of pregnancy constitute a third stage in mammary gland development that is influenced and regulated by Wnt signaling. Development of the gland at this stage can be reduced to two principal elements: formation of extensive secondary branches of the ductal tree, and the appearance of alveolar units that constitute the secretory apparatus of the gland. Pregnancy dependent hormones normally trigger both of these aspects of mammary morphogenesis, but both can also be induced prematurely by ectopic expression of Wnt proteins in virgin mice [9–12].
Secreted frizzled-related proteins (SFRPs) are a family of Wnt antagonists which contain a cysteine-rich domain that is homologous to the Wnt-binding domain of frizzled receptor proteins . However, SFRPs do not contain a transmembrane domain and therefore are released into the extracellular compartment where they antagonize Wnt signaling by binding to Wnt ligands and preventing ligand-receptor interactions and signal transduction . Loss of SFRP1 expression is found in multiple human cancers including breast cancer [15–17]. When SFRP1 is knocked down in immortalized non-malignant mammary epithelial cells, the cells (TERT-siSFRP1) acquire a malignant phenotype characteristically observed in metastatic breast cancer stem-like cells [18, 19]. The work described here was initiated to investigate the role of SFRP1 in mammary gland development and whether SFRP1−/− mice exhibit changes in mammary gland morphology and cell signaling pathways shown to be associated with SFRP1 loss in vitro. Previously, analysis of 41 different non-skeletal tissues, including the mammary gland, in 20 or 40 week old mice has shown no overt morphologic differences between SFRP1+/+ and SFRP1−/− animals . However, considering that mammary gland development occurs around puberty, we chose to look at the morphological effects of SFRP1 loss beginning at this critical development stage, and at an adult stage. The significance of these studies is considerable because methylation of SFRP1 (resulting in suppression of SFRP1 expression) has been demonstrated to be an early change in premalignant breast lesions [21, 22].
Transforming growth factor (TGF)-β is a multifunctional cytokine that regulates a variety of physiological processes and also plays a dual role during mammary gland development and tumorigenesis. We have shown that loss of SFRP1 expression in vitro allows mammary cells to acquire sensitivity to TGF-β signaling . In an initial screen of genes affected by SFRP1 loss in HMECs we identified several upregulated genes within the TGF-β pathway, including the TGF-β2 ligand . The real-time PCR data presented here demonstrate that SFRP1 loss in vivo also results in significantly increased mRNA levels of TGF-β2 (Figure 4B). These findings may in part explain the altered mammary gland morphogenesis observed in SFRP1−/− mice. Specifically, it has been shown that TGF-β2, but not TGF-β1 or TGF-β3, is critical for lung branching in vivo  and synergizes with Wnt to promote mammary gland branching in vitro .
As discussed previously, overexpression of Wnt1, Wnt10b, and/or Wnt4 induces mammary gland hyper branching. Therefore, we wanted to establish whether these particular Wnt ligands are upregulated in SFRP1−/− animals. The expression of Wnt1 and Wnt10b was unaffected by SFRP1 loss (data not shown), however Wnt4 mRNA levels were significantly elevated in the mammary gland of SFRP1−/− mice (Figure 4C). Interestingly, Wnt4 is expressed during the period when side-branching occurs in early to mid-pregnancy [30, 31]. Brisken et. al. showed that Wnt4 null mammary glands were deficient in early lobulo-alveolar mammary outgrowth during pregnancy, and that Wnt4 is an effector for progesterone-induced mammary growth .
Critical to the Wnt4 downstream signaling for branching morphogenesis is the receptor of activated NF-κB ligand (RANKL) [33, 34]. RANKL was originally characterized for its role in the development, survival, and activation of osteoclasts during bone remodeling . Subsequent studies have shown that RANKL deficient mice exhibit a significant decrease in parity-induced mammary alveologenesis which results in a lactational defect . Furthermore, transgenic overexpression of RANKL or RANK (RANKL receptor) alone into the murine mammary gland elicits ductal side branching, alveologenesis, and mammary hyperplasia [35, 36]. Considering that SFRP1 has been shown to bind to and inhibit RANKL mediated action , we sought to determine whether the expression of RANKL is affected by SFRP1 loss. Indeed, we found that mRNA levels of RANKL were significantly elevated in the mammary gland of SFRP1−/− mice (Figure 4D). These data lend support to the notion that SFRP1 may play a role in tumor susceptibility since abrogation or accentuation of RANKL signaling renders the mammary epithelium markedly resistant or susceptible to mammary tumorigenesis respectively [36, 38]. Since RANKL has a role in bone remodeling and loss of SFRP1 increases the thickness of the trabecular bones , it has been suggested that SFRP1 inhibitors may be beneficial for the treatment and/or prevention of osteoporosis. However, SFRP1 inhibitors may not be beneficial to women who are genetically predisposed or who have pre-malignant lesions and may put them at risk for developing breast cancer.
R-spondin (RSPO) family proteins activate the canonical WNT signaling pathway by binding to the Wnt co-receptor LRP6 leading to β-catenin-dependent gene activation [39–41]. Interestingly, the expression of Rspo2 and other Rspo family members is induced when mouse mammary tumor virus (MMTV) is inserted into mammary epithelial cells, and these proteins have been associated with malignant transformation and tumorigenesis [42–44]. Loss of Rspo1 hinders the development of the murine mammary gland leading to lack of side branching along secondary ducts . At 5 weeks of age in SFRP1-/- mice, we started observing changes in the expression of genes involved in branching morphogenesis. Although we did not see Rspo1 changes at this stage (Figure 4E), the expression of Rspo2 is significantly elevated (Figure 4F). Additionally enhanced side branching at 10 wks of age in response to SFRP1 loss is associated with significantly elevated expression of both Rspo1 and Rspo2 (data not shown). These findings may partially explain how SFRP1 loss increases the expression of Wnt4, through Rspo2 enhancement of the β-catenin-dependent activity of Wnt4 in murine epithelial cells . Furthermore, the expression pattern of Rspo1 parallels that of Wnt4 during mammary gland development which indicates that there is functional relationship between these Wnt signaling proteins .
Ductal outgrowth is driven by TEBs, which are believed to house a small population of cells that are essential for mammary gland development, mammary stem cells (MaSCs) [8, 9]. Importantly, MaSCs are capable of regenerating all cell-lineages that comprise the murine mammary gland and support for their existence comes from cell enrichment studies using specific cell surface markers as well as mammary gland transplantation experiments [47, 48]. Similar to stem cells in either the hematopoietic system or neuronal system, the self-renewal and differentiation of MaSCs is regulated by multiple genes and pathways. Several studies have shown that Wnt/β-catenin signaling is activated in several stem cell types, including in mammary stem cells. Liu et. al. demonstrated that transgenic mice with activated Wnt/β-catenin signaling in the mammary gland have a significantly higher stem cell-enriched population compared to wild type mice . Additionally, the stem cell-enriched population of cultured primary mammary epithelial cells was significantly increased when cells were treated with a Wnt agonist and this effect could be abrogated by the addition of a Wnt pathway inhibitor .
The work described herein clearly demonstrate that SFRP1−/− mice exhibit precocious mammary gland side branching with clear lobulo-alveolar development, which normally only occurs in hormonally stimulated mid-pregnant wt mammary glands. These changes may be partially explained by alterations in the expression of genes critical for mammary gland development and by an increase in the number of mammosphere forming cells. Previously, it has been demonstrated that suppression of SFRP1 expression is an early change in human premalignant breast lesions . Taken together, our study indicates that the SFRP1 gene is critical for maintaining proper mammary gland development, that reduced levels of SFRP1 results in hyperplastic lesions, and its loss may be a critical event in cancer initiation. Therefore, caution should be exercised in the potential use of SFRP1 inhibitors for the treatment of osteroporosis until the effects of such inhibition on breast tumorigenesis are fully elucidated.
All procedures were performed in accordance with the NIH guidelines for the ethical treatment of animals and were approved by the Baystate Medical Center Institutional AnimalCare and Use Committee before initiating these studies. Female 129/C57Blk6 mice (n = 6) were individually housed in plastic cages with food and water provided continuously, and maintained on a 12:12 light cycle.
Mice were euthanized with carbon dioxide and fifth inguinal mammary glands were excised from 5 wk virgin, 10 wk virgin, pregnant day 8, and pregnant day 15 animals, spread on microscope slides, and fixed overnight in methacarn (60% methanol, 30% chloroform, 10% acetic acid). The fixed glands were washed in 70% ethanol for 15 min, rinsed in water for 5 min, and stained overnight at 4°C in carmine alum stain (1 g carmine and 2.5 g aluminum potassium sulfate in 500 ml water). The glands were then dehydrated progressively in 70%-95%-100% ethanol, cleared in xylene for 1 hr, and mounted on slides with Cytoseal™XYL mounting medium (Richard-Allan Scientific). Mammary whole mounts where photographed using an Olympus BX41 light microscope using SPOTSOFTWARE (Diagnostic Instruments, Inc, Sterling Heights, MI).
Fourth or sixth inguinal mammary glands from 5 wk virgin, 10 wk virgin, 23 wk virgin, pregnant day 8, and pregnant day 15 animals (n = 6 animals/genotype) were formalin-fixed and paraffin-embedded. Tissue sections were cut on a Leica microtome at a thickness of 4 μm on Superfrost-plus slides, and stained with hematoxylin and eosin (H&E) per standard protocol. Briefly, slides were dried in a microwave for 1 min and then at 62°C for 15 minutes. Slides were subsequently de-paraffinized with xylene 3X, cleared with graded ETOH (100% X2, 95%, 70%), and rinsed in dH2O. Sections were stained for 3 minutes in Mayer’s hematoxylin, washed with glacial acetic acid water for 15 sec, and then ammonia water to blue, and stained with eosin. Finally, slides were dehydrated in ETOH and xylene before manual coverslipping. Stained images were captured with an Olympus BX41 light microscope using SPOTSOFTWARE.
Mice were euthanized with carbon dioxide and fourth inguinal mammary glands were excised (n = 6 animals/genotype). The tissue was aseptically minced and for each genotype, a representation from all 6 animals was placed on 2 different Surgifoam gelatin sponges (Ferrosan, Sueborg, Denmark) in 60 mm tissue culture dishes containing 5 mL of control or Wnt3a medium prepared as described . Explant cultures were maintained for 24 hours in 5% CO2 air and subsequently formalin-fixed and paraffin-embedded.
Immunohistochemistry (IHC) was performed on a DakoCytomation autostainer using the Envision HRP Detection system (Dako, Carpinteria, CA). Each mammary tissue block was sectioned at 4 μm on a graded slide, deparaffinized in xylene, rehydrated in graded ethanols, and rinsed in Tris-phosphate-buffered saline (TBS). Heat induced antigen retrieval was performed in a microwave at 98°C in 0.01 M citrate buffer. After cooling for 20 minutes, sections were rinsed in TBS and subjected to the primary rabbit polyclonal anti-Axin2 antibody (1:100, Abcam, ab32197) for 45 minutes. Immunoreactivity was visualized by incubation with chromogen diaminobenzidine (DAB) for 5 minutes. Tissue sections were counterstained with hematoxylin, dehydrated through graded ethanols and xylene, and cover-slipped. Images were captured with an Olympus BX41 light microscope using SPOTSOFTWARE.
PCR primer sequences
5’- CGACATGAACAAATGCATCA -3’
5’- CTCCTGACACTTGGTGCAGA -3’
5’- GCCCATCAGGGTATTATGGA -3’
5’- TCACAGTTTTCTATTCTGCATCG -3’
5’- CTAAGGCCAACCGTGAAAAG -3’
5’- ACCAGAGGCATACAGGGACA -3’
Eight 8–10 week old virgin mice were euthanized with carbon dioxide (4 SFRP1+/+ and 4 SFRP1−/−) and fourth mammary glands were harvested, minced, and finally dissociated in DMEM:F12 (Sigma, St. Louis, MO) supplemented with 5 % fetal bovine serum (Gibco, Paisleley, UK), 2 mg/ml collagenase (Worthington Biochemicals, Freehold, NJ), 100u/ml hyaluronidase (Sigma), 100u/ml pen/strep (Gibco) and 100 μg/ml gentamicin (Gibco) for 6 hours. The cell pellet was collected and further dissociated with 1mlpre-warmed 0.05% Trypsin-EDTA (Gibco) and 200μl1mg/ml Dnase I (Roche, Mannheim, Germany). Cell suspensions were sieved through a 40 μm cell strainer to obtain single cell suspensions. Primary single cells were seeded onto ultra-low attachment dishes (Corning, Corning, NY) at a density of 20,000 viable cells/ml. Cells were grown in a serum-free mammary growth medium (EpiCult®B for Mouse Mammary Epithelial Cell Culture, Vancouver, BC) supplemented with 10ng/ml EGF (Sigma), 10 ng/ml FGF (Sigma), 4 μg/ml heparin, 100u/ml pen/strep (Gibco) and 100 μg/ml gentamicin (Gibco) . On day 21, the total number of mammospheres was quantified by counting spheres that were at least 100 μm in size. Images of mammosphere formation were captured with a Nikon Eclipse TE2000-Uusing Metaview™ software.
secreted frizzled related protein
terminal end buds
transforming growth factor-β
receptor of activated NF-κB ligand
mammary stem cell.
This work was funded by a grant from the Rays of Hope Foundation.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.