Whole mount preparation.

Inguinal mammary glands were isolated and processed as described previously (30). Ductal outgrowth was measured on the whole mounts as the distance from middle of the central lymph node to the leading edge of the ductal mass using Axiovision (Jena, Germany) 4.5 Software. The number of secondary and tertiary branches and TEB size and number were also evaluated from the whole mounts. Alveolar budding was quantified from whole mounted left inguinal mammary glands using Axiovision Software by gating three fixed 250 pixel² areas around the lymph node and manually counting the number of alveolar buds per tissue. A minimum of eight mice per genotype per time point was used for quantification.

Epithelial organoid preparation.

Mammary glands were incubated at 37°C using an orbital shaker for 1 h at 200 rpm in DMEM-F-12 medium (1:1) supplemented with collagenase type IA (0.5 mg/ml; Sigma-Aldrich, St. Louis, MO), hyaluronidase (0.055 mg/ml; Sigma-Aldrich), 1% penicillin-streptomycin (Thermo Scientific, Florence, KY), insulin (5 μg/ml), nystatin (60 U/ml), and gentamycin (50 μg/ml). Glands were centrifuged at 1,500 rpm to separate the fat cell layer from the ductal organoid fragments. The pellet containing the ducts was reconstituted in 1× PBS and was pulse centrifuged at 1,000 rpm for removal of red blood cells and fibroblasts. Epithelial ductal fragments were repeatedly pulse centrifuged for additional purification. Organoids were placed in lysis buffer or Tri-Reagent RT (Molecular Research Center, Cincinnati, OH) for protein or RNA isolation, respectively.

Histological staining.

Formalin-fixed, paraffin-embedded right inguinal mammary glands from 6- and 10-wk-old mice were cut into 5-μm tissue sections and stained with hematoxylin and eosin for gross morphological assessment. Additional 6-wk inguinal mammary gland sections were assessed for cellular proliferation and apoptosis within TEBs. Epithelial cell proliferation was evaluated by immunohistochemical Ki-67 staining. Briefly, antigen retrieval was carried out in citrate buffer, and slides were quenched for endogenous peroxide, blocked in goat serum, and incubated with the primary antibody (1:100; Fisher Scientific, Waltham, MA) at 4°C overnight. Slides were incubated with a biotinylated goat anti-rabbit secondary antibody (Vector Laboratories, Burlingame, CA) for 1 h prior to avidin conjugation (Vectastain ABC kit standard; Vector Laboratories). Epithelial cell apoptosis within TEBs was determined by terminal deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) staining using ApopTag Plus Peroxidase In Situ Apoptosis Kit (Millipore, Billerica, MA), following the manufacturer's instructions. Detection of Ki-67 and TUNEL staining was visible following treatment with 3,3′-diaminobenzidine (Sigma). Ki-67 and TUNEL staining were analyzed at ×40 magnification using ImageJ Software (National Institutes of Health, Bethesda, MD). All immunohistochemical staining for cell proliferation and apoptotic index was quantified within the body cell region of TEBs as a percentage, i.e., epithelial cells with positive incorporation over the total number of epithelial cells within each TEB. The range of epithelial cells per TEB was 213–692, with an average of ~356 ± 111 cells counted per TEB. A minimum of eight animals per genotype was evaluated with two sections per animal and three to six TEBs per section.

Western blotting.

Ex vivo mammary gland organoids, 21-day mammary gland fat pads, or whole mammary glands were homogenized in Laemmli lysis buffer, sonicated, and measured for protein concentration using a BCA protein assay. Immunoblotting was carried out using antibodies targeting VDR-D6 (Santa Cruz Biotechnology, Santa Cruz, CA), cytokeratin-18 (Epitomics, Burlingame, CA), perilipin-D418 (Cell Signaling Technology, Danvers, MA), ERα MC-20 (Santa Cruz Biotechnology), and actin I-19 (Santa Cruz Biotechnology).

Quantitative real-time PCR.

Total RNA was isolated from mammary organoids, fat pads, whole glands, or EpH4 cells using Tri-Reagent RT (Molecular Research Center) and subsequently reverse transcribed into cDNA (1.5 μg) using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). Between six and 12 independent control or experimental cDNA samples were analyzed for VDR, ERα, PGR, perilipin, cytokeratin-18, Cyp27b1, or Cyp24a1 mRNA expression in duplicate by real-time PCR analysis using FastStart SYBR Green Master (Roche Diagnostics, Indianapolis, IN). The following forward and reverse mouse primer sets were used: VDR 5′-GAAGCGCAAGGCCCTGTT-3′(forward) and 5′-CGCTGCACCTCCTCATCTGT-3′ (reverse); ERα 5′-GGCTGCAAGGCTTTCTTTAAGA-3′ (forward) and 5′-GCCTTTCATCATGCCCACTT-3′ (reverse); PGR 5′-TGTGGGAGCTGCAAGGTCTT-3′ (forward) and 5′-GACCATGCCAGCCTGACAA-3′ (reverse); perilipin 5′-GGAGAGTAAGGATGTCAATGAACAAG-3′ (forward) and 5′-CTTCTTTGGTGCTGTTGTAGGTCTT-3′ (reverse); cytokeratin-18 5′-GGGCTCAGATCTTTGCGAATT-3′ (forward) and 5′-TGCGGAGTCCATGGATGTC-3′ (reverse); Cyp27b1 5′GCGGGCTATGCTGGAACTC-3′ (forward) and 5′-GCACCGCGCCTATACTTTCT-3′ (reverse); Cyp24a1 5′CTGGCCTGGGACACCATTT-3′ (forward) and 5′-CTCCGTGACAGCAGCGTACA-3′ (reverse). Gene expression was normalized to 18S rRNA and reported as relative expression compared with the WT control. Error bars represent the standard error of the mean from three independent experiments.

Steroid hormone pellet implant.

Five-week control and experimental mice were surgically implanted with a 21-day-release hormone pellet containing 21 mg of progesterone and 21 μg of estradiol (Innovative Research of America, Sarasota, FL). The control or hormone pellet was implanted subcutaneously on the neck/upper back for 14 days. At 7 wk of age, the mice were euthanized and mammary glands removed for analysis. Left inguinal mammary glands were whole mounted for visibility of growing ducts and the onset of alveolar bud development and side branching in response to the hormone stimulation.

In vitro cell and ex vivo fat explant culture conditions.

The murine epithelial cell line EpH4 (ATCC, Manassas, VA), established from the mammary gland of a midpregnant BALB/c mouse (27, 36), was grown in DMEM supplemented with 10% FBS and puromycin. Cells were cultured to 80% confluency. Left and right inguinal mammary fad pads from 21-day-old WT C57BL/6J or global VDR-KO mice were extracted, lymph node and visible blood vessels were removed, and 50-mg (wet wt) pieces were added per well into a 24-well plate containing 400 μl of serum-free media. The explant medium contained DMEM-F-12 (1:1) and was supplemented with 1% penicillin-streptomycin and 200 nM adenosine. Vitamin D₃, in the form of 1,25D₃ (Sigma) or ethanol (vehicle control) was then added to the medium at the indicated concentration for 24 h prior to further use or analysis of the conditioned medium. For heat inactivation experiments, conditioned medium was heated to 100°C for 30 min and then cooled on ice to 37°C prior to usage.

MTT assay.

EpH4 cells were seeded in a 96-well clear plate at a density of 10,000 cells/well in culture medium and allowed to adhere overnight. Serum-free conditioned medium from fat explants was collected after 24 h in culture, diluted 1:1 in serum-free DMEM, and placed onto EpH4 cells. In designated experiments, recombinant mouse IL-6 (R & D Systems, Minneapolis, MN) was added to the media cocktail prior to incubation with EpH4 cells at the specified concentration. For IL-6 neutralization, mouse IL-6 monoclonal antibody (Clone MP5–20F3; R & D Systems) was added to the conditioned medium or placed onto EpH4 cells at a final concentration of 10 μg/ml. Confirmation of neutralization was carried out via ELISA analysis (data not shown). After 48 h the medium was aspirated, and 1 mg/ml thiazolyl blue tetrazolium bromide (Sigma) in 1× PBS was added at 100 μl/well for 60 min prior to incubation with DMSO. Absorbance was measured at 570 nm.

Secretome arrays.

Left and right inguinal mammary fat pads from four 21-day-old VDR-WT and VDR-KO mice were harvested, cleared of lymph nodes and visible blood vessels, cut into 50-mg (wet wt) pieces and treated with 10 nM 1,25D₃ for 24 h. Conditioned medium samples were pooled from eight wells and screened for secretory cytokines and adipokines using proteome profiler mouse antibody arrays (R & D Systems), following the manufacturer's instructions. Arrays were repeated three times, using conditioned medium from three separate experiments, and each experiment pooled conditioned medium from four mice (8 inguinal mammary glands). Results were quantified using ImageJ Software.

ELISA.

IL-6 levels in conditioned medium were measured by ELISA (R & D Systems). The assay was carried out according to the manufacturer's instructions using a 1:10 dilution of the conditioned medium. Samples were run in duplicate from three separate experiments.

Flow cytometry.

EpH4 cells were treated with conditioned medium from VDR-WT or VDR-KO mammary gland fat pads supplemented with 10 nM 1,25D3 or EtOH for 48 h, as described above. Cells were then stained for annexin V/propidium iodide (PI) as per the manufacturer's instructions (BD Biosciences, Franklin Lakes, NJ). Cells were analyzed using the FACS Aria machine and FACS Diva software (BD Biosciences) in two separate experiments with 10,000 events per treatment group.

Loss of VDR signaling in mammary epithelial cells or adipocytes leads to accelerated ductal outgrowth during pubertal development.

The VDR plays an important role in regulating the growth of the mammary epithelium during development. Global VDR-KO mice have heavier pubescent mammary glands that exhibit accelerated growth, an increased number of TEBs, and more secondary branch points compared with age-matched WT controls (55). To assess the significance of VDR signaling within different cell types of the mammary gland during pubertal development, mice with a conditional loss of VDR (Fabp4^{VDR fl/fl} and MMTV^{VDR fl/fl} mice) and control mice were euthanized between 5 and 10 wk of age. With the onset and conclusion of pubertal mammary gland development, no significant differences in ductal length were apparent at 5 or 10 wk of age between groups. However, Fabp4^{VDR fl/fl} and MMTV^{VDR fl/fl} mice displayed significantly increased ductal extension at 6, 7, and 8 wk of age compared with WT controls (Fig. 2, A and B). These time points encompass the peak of pubertal mammary gland development, where the most rapid growth is exhibited. Similarly to previous data using global VDR-KO mice (55), the greatest difference in ductal outgrowth for the cell type-specific VDR-ablated animals was observed at 6 wk of age, wherein a significant 1.5-fold increase in ductal extension was found (Fig. 2, A and B). Despite accelerated ductal outgrowth, Fabp4^{VDR fl/fl} and MMTV^{VDR fl/fl} mammary glands had no significant differences in TEB number or branching morphogenesis during pubertal development (Fig. 2C).

Open in a separate window

Fig. 2.

Increased mammary ductal extension with VDR ablation in adipose tissue or the mammary epithelium. A: representative whole mounts from 6-wk-old Fapb4^{VDR fl/fl} and MMTV^{VDR fl/fl} transgenic mice showing enhanced ductal outgrowth compared with age-matched controls. Scale bar, 2 mm. B: average ductal extension in mammary glands from 5-, 6-, 7-, 8-, and 10-wk-old transgenic mice compared with VDR-WT controls. VDR ablation from either cell type results in significantly augmented epithelial outgrowth at 6, 7, and 8 wk compared with controls. C: average no. of terminal end buds (TEBs), secondary branch points, and tertiary branch points in mammary glands from conditional mutant mice compared with age-matched VDR-WT controls. Data represent means ± SE from 8 to 40 inguinal mammary glands per group at each time point. *P < 0.05.

Loss of VDR in both the mammary epithelium and adipocytes led to reduced TEB clearing with increased epithelial cell proliferation and decreased apoptosis.

Further analysis conducted on 6-wk-old Fabp4^{VDR fl/fl} and MMTV^{VDR fl/fl} inguinal mammary glands revealed altered TEB morphology with poor epithelial cell clearing compared with VDR-WT glands (Fig. 3A, top). By 10 wk of age, ductal morphology was analogous to age-matched VDR-WT controls as assessed by hematoxylin and eosin staining (Fig. 3A, upper middle). These data are consistent with earlier studies demonstrating that TEB differentiation into terminal end ducts is VDR independent (43, 55). Importantly, postpubertal mammary gland functionality in Fabp4^{VDR fl/fl} and MMTV^{VDR fl/fl} parous females was indistinguishable from age-matched VDR-WT mice (data not shown), consistent with previous studies in global VDR-KO mice where dams were able to nurse to term and maintain healthy litters with similar pup size and weight (56).

Open in a separate window

Fig. 3.

Altered TEB morphology in mice with conditional VDR ablation in the mammary epithelium or adipose tissue. A: representative hematoxylin and eosin (H & E) staining on 6- and 10-wk-old mammary glands. H & E-stained tissue sections from 6-wk-old transgenic mice (top) showing an inhibition of TEB clearing with loss of VDR expression from either the mammary epithelium or adipose tissue. H & E-stained mammary tissues sections from 10-wk-old glands display a normal ductal phenotype (upper middle). Ki-67 and TUNEL staining of TEBs showing increased epithelial cell proliferation and decreased apoptosis, respectively. H & E pictures were taken at ×10 magnification (6 and 10 wk), and Ki-67 and TUNEL images were taken at ×40 magnification. B: quantification of Ki-67 immunohistochemical staining from 6-wk-old inguinal mammary glands. A minimum of 10 glands/genotype were analyzed, each consisting of 2 to 5 TEBs to generate the mean percent for each mouse. C: quantification of epithelial TUNEL staining in the body cell region of TEBs. A minimum of 8 glands/genotype each consisting of 2 to 6 TEBs were analyzed to generate the mean percent for each mouse. Graphs represent %means ± SE. *P < 0.05.

To further evaluate the epithelial cell-dense TEB phenotype in Fapb4^{VDR fl/fl} and MMTV^{VDR fl/fl} mice, mammary sections from 6-wk-old mice were stained for Ki-67 to assess epithelial cell proliferation. Similarly, the apoptotic potential was determined via TUNEL analysis (Fig. 3A,lower middle and bottom). The percentage of Ki-67-positive epithelial cells within TEBs of Fapb4^{VDR fl/fl} and MMTV^{VDR fl/fl} mammary glands was significantly higher than the age-matched VDR-WT controls (Fig. 3B). TEBs from Fabp4^{VDR fl/fl} and MMTV^{VDR fl/fl} mammary glands also exhibited significantly less epithelial cell death (Fig. 3C).

Mammary epithelial-specific VDR signaling modulates alveolar precursors during pubertal development.

Despite similarities in the numbers of TEBs as well as in secondary and tertiary branching (Fig. 2C), precocious alveolar budding was suspected in nulliparous MMTV^{VDR fl/fl} mammary glands compared with age-matched VDR-WT controls throughout pubertal development (Fig. 4A). Alveolar buds are precursors to mature lobuloalveoli and are composed of epithelial and myoepithelial cells that eventually direct milk production and transport for proper lactation (17). Alveolar budding in the mammary gland occurs predominantly during pregnancy and lactation, and it is not commonly found in the developing mammary gland. However, previous work has demonstrated that global VDR-KO mice are more sensitive to hormonally induced mammary alveologenesis in vivo and ex vivo (55). Figure 4B depicts the average number of alveolar precursors per area and shows that the MMTV^{VDR fl/fl} mice exhibit a significantly higher number at 5, 6, 7, 8, and 10 wk compared with age-matched Fapb4^{VDR fl/fl} and VDR-WT controls. To further elucidate the cell type-specific contribution of VDR to hormonal hypersensitivity, estrogen and progesterone pellets were implanted subcutaneously into pubescent mice for 2 wk. Hormone exposure stimulated secondary and tertiary branching morphogenesis, stunted ductal outgrowth, and promoted precocious alveolar budding in developing mammary glands of all genotypes (Fig. 4C). However, the number of alveolar precursors per gland was significantly higher in hormone-stimulated MMTV^{VDR fl/fl} mice compared with VDR WT control and Fapb4^{VDR fl/fl} mammary glands (Fig. 4D).

Open in a separate window

Fig. 4.

Increased sensitivity to hormonally induced alveolar growth in the mammary gland of MMTV^{VDR fl/fl} mice. A: representative carmine-alum stained mammary whole mounts from MMTV^{VDR fl/fl} mice show an increased incidence of alveolar buds compared with VDR-WT mice at 6 wk of age. Scale bar, 2 mm. B: quantification of alveolar precursor incidence from 5- to 10-wk-old transgenic mice compared with age-matched VDR-WT controls. Alveolar precursors were counted manually and averaged from 3 fixed locations per gland gated to a 250 pixel² area. C: representative carmine-stained whole mounts from 7-wk-old mice given an estrogen and progesterone (E + P) hormone pellet at 5 wk of age to induce glandular growth in vivo. D: alveolar precursors and alveolar buds from mice with hormonal implants were counted and averaged from 3 fixed locations per gland gated to a 250 pixel² area. Data represent the average number of alveolar precursors and buds for all mice (n = 8–40). *P < 0.05.

Vitamin D₃-dependent VDR signaling in mammary adipose tissue modulates mammary epithelial cell turnover through promotion of cell death.

Given the novel data documenting the role of VDR signaling in adipose tissue on epithelial ductal extension in the mammary gland, we sought to examine VDR-modulated cross-talk between these two cell types in vitro. EpH4 cells, a nontransformed murine mammary epithelial cell line, were utilized as the source of mammary epithelial cells. EpH4 cells express mRNA levels of Vdr and Cyp27b1 similar to mammary epithelial organoids from VDR-WT mice (Fig. 5A) and are responsive to vitamin D₃ treatment, as judged by the upregulation of Cyp24a1 mRNA (a VDR target gene) in a dose-dependent manner (Fig. 5B). To examine the effect of VDR signaling in adipocytes on mammary epithelial cell growth, mammary fat pad explants from 21-day-old VDR-WT mice were cultured ex vivo, and conditioned medium was applied to EpH4 cells. After 48 h, the change in EpH4 cell number was calculated by 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assays, and values were normalized to the control medium, which was set at 1. As shown in Fig. 5C, conditioned medium from fat was able to significantly augment EpH4 cell number in the absence of exogenous vitamin D₃. However, conditioned medium from VDR-WT fat explants treated with 1,25D₃ dose-dependently reduced EpH4 cell number to a greater extent than direct treatment of cells with 1,25D₃ alone (Fig. 5D), suggesting that epithelial cell growth inhibition occurs through vitamin D₃-mediated paracrine signaling from adipose tissue. At 10 nM 1,25D₃, conditioned medium from VDR-WT mammary fat pads significantly reduced EpH4 cell number by roughly 60% compared with the ethanol-treated (0 nM 1,25D₃) VDR-WT fat explant conditioned medium control (Fig. 5E). In sharp contrast, conditioned medium from VDR-KO fat explants had a minimal effect on EpH4 cell number regardless of 1,25D₃ supplementation. Heat inactivation of the VDR-WT conditioned medium rescued EpH4 viability, suggesting that a heat-labile protein produced by the fat in response to 1,25D₃ was responsible for the growth inhibition of the mammary epithelial cells (Fig. 5E). Analysis of EpH4 cells by flow cytometry showed that conditioned medium from VDR-WT mammary fat pads treated with vitamin D₃ induced death in ~40% of the cells (Fig. 5F), whereas cell proliferation measured by EDU incorporation was not altered with any treatment (data not shown).

Open in a separate window

Fig. 5.

VDR-regulated cross-talk between mammary adipocytes and epithelial cells. A: qRT-PCR mRNA expression of Vdr and Cyp27b1 in EpH4 cells compared with epithelial organoids from VDR-WT mature mammary glands. B: Cyp24a1 mRNA expression in EpH4 cells is induced by increasing concentrations of 1,25-dihydroxyvitamin D₃ (1,25D₃) or 25-hydroxyvitamin D₃ (25OHD₃) at 48 h. Data represent the means ± SE of 3 individual experiments. C: MTT assay for EpH4 cell no. after 48-h exposure to mammary gland fat pad explant conditioned media from VDR WT (WT CM) mice. Data represent the mean of 3 experiments, each performed with 8 replicates. Values were normalized to the media control (media). D: MTT assay for EpH4 cell no. after 48-h exposure to media plus the specified concentration of 1,25D₃ (media) or mammary gland fat pad explant CM from VDR-WT (WT CM) mice treated with increasing concentrations of 1,25D₃. Data represent the mean value from 2 experiments, each performed with 6 replicates. Values are normalized to the respective ethanol-treated control (0 nM 1,25D₃). E: MTT assay for EpH4 cell no. after 48-h exposure to VDR-WT (WT CM) or VDR-knockout (KO CM) mammary gland fat pad explant CM treated with or without 10 nM 1,25D₃. Medium on the right was heat inactivated prior to incubation with EpH4 cells. Data represent the mean of 4 experiments, each performed with 6 replicates. Values were normalized to the respective ethanol-treated control (0 nM 1,25D₃), which is set at 1. F: annexin V/PI flow cytometry analysis of EpH4 cells cultured for 48 h in VDR-WT (WT CM) or global VDR-KO (KO CM) mammary fat pad CM supplemented with or without 10 nM 1,25D₃. Cells gated for PI are noted as “dead,” annexin V and annexin V/PI are noted as “apoptotic,” and unstained cells are “living.” Data represent the mean of 2 experiments, each performed with 10,000 events. *P < 0.05.