Progesterone Receptor Isoforms A and B: Temporal and Spatial Differences in Expression during Murine Mammary Gland Development
http://www.100md.com
内分泌学杂志 2005年第8期
Department of Physiology (S.Z.H., A.K.) and the Cell and Molecular Biology Program (M.D.A., K.T.S.), Michigan State University, East Lansing Michigan 48824
Address all correspondence and requests for reprints to: Sandra Z. Haslam, Ph.D., Department of Physiology, 2201 Biomedical and Physical Sciences Building, Michigan State University, East Lansing, Michigan 48824. E-mail: shaslam@msu.edu.
Abstract
Progesterone is a potent mitogen in the mammary gland. Based on studies using cells and animals engineered to express progesterone receptor (PR) isoforms A or B, PRA and PRB are believed to have different functions. Using an immunohistochemical approach with antibodies specific for PRA only or PRB only, we show that PRA and PRB expression in mammary epithelial cells is temporally and spatially separated during normal mammary gland development in the BALB/c mouse. In the virgin mammary gland when ductal development is active, the only PR protein isoform expressed was PRA. PRA levels were significantly lower during pregnancy, suggesting a minor role at this stage of development. PRB was abundantly expressed only during pregnancy, during alveologenesis. PRA and PRB colocalization occurred in only a small percentage of cells. During pregnancy there was extensive colocalization of PRB with 5-bromo-2'-deoxyuridine (BrdU) and cyclin D1; 95% of BrdU-positive cells and 83% of cyclin D1-positive cells expressed PRB. No colocalization of PRA with either BrdU or cyclin D1 was observed at pregnancy. In the virgin gland, PRA colocalization with BrdU or cyclin D1 was low; only 27% of BrdU-positive cells and 4% of cyclin D1-positive cells expressed PRA. The implication of these findings is that different actions of progesterone are mediated in PRB positive vs. PRA-positive cells in vivo. The spatial and temporal separation of PR isoform expression in mouse mammary gland provides a unique opportunity to determine the specific functions of PRA vs. PRB in vivo.
Introduction
THE RELATIVE ROLES of estrogen (E) and progesterone (P) in regulating epithelial cell proliferation of the normal human breast and their contributions to breast cancer risk have been controversial. Originally it was presumed that because P antagonizes E-induced proliferation in the uterus, it would also antagonize E-induced proliferation in the breast (1). However, P in combination with E has more potent proliferative activity than E alone in the adult mammary gland in animal models (monkey and rodent) (2, 3) and the adult human breast (4). In humans this is the case for premenopausal cycling women and postmenopausal women receiving hormone replacement therapy (HRT). In postmenopausal women, combined continuous E+P HRT is associated with the highest proliferative index and the highest increase in breast epithelial density when compared with no HRT or E alone HRT (4). Furthermore, a significantly greater breast cancer risk is associated with E+P HRT (5, 6, 7, 8). Thus, P can contribute significantly to breast cancer risk.
P action is mediated through binding to the P receptor (PR). The PR consists of two isoforms, PRA and PRB, which are expressed from a single gene in both humans and rodents (9). Two promoters, one specific for PRA and the other specific for PRB, have been identified for human (10) and rat (11) PR. Initiation of translation at two distinct AUG signals produces the B and A forms of PR. PRB differs from PRA by an amino terminal extension of 164 amino acids. Studies to identify the functional roles of PRA and PRB in the mammary gland have been carried out in vivo using transgenic mice (PRA or PRB transgenes) (12, 13) and PR gene-deleted mice [total PR, PRA only (PRAKO) or PRB only] (14, 15, 16). From these studies it has been inferred that PRB is required for alveologenesis during pregnancy. The specific function of PRA has not yet been identified. In vitro studies using cell lines have shown that the unique amino terminal region of PRB encodes a transactivation function that plays an important role in specifying target genes that can be activated by PRB but not by PRA (17). Therefore, PRA and PRB can have different functions in the same cell, and the activity of the individual isoforms of the receptor may also vary among different types of cells.
The mouse is currently the most extensively studied and best understood model of P action in the normal mammary gland. Genetically altered mice have provided some insights into the functions of the two PR isoforms in mouse mammary gland. These genetically altered mice have an altered mammary gland phenotype (12, 13, 14, 15, 16); this suggests that mammary gland development is abnormal. Our approach in the present study was to investigate specific PR isoforms in mammary gland of genetically unaltered, wild-type mice as a function of development.
Biochemical methods to analyze PR isoform expression and function in the mouse mammary gland have provided limited information about the functional roles of PRA and PRB because they do not provide insight into the cellular distribution or colocalization of the isoforms. The most direct approach to address this question is immunohistochemical analysis of PR isoform-specific expression. It was generally assumed that if an anti-PR antibody detected both isoforms in immunoblot analysis, then it also detected both isoforms in immunohistochemical analysis (16, 18, 19). The report of Mote et al. (20) showed that this assumption is not correct. Mote et al. (20) analyzed a panel of 11 antihuman PR antibodies for their ability to detect PRA and/or PRB in human cells engineered to express specific isoforms of PR. To determine antibody specificity, MCF-7 breast cancer cell sublines that express only PRA, only PRB, or both PRA and PRB were analyzed (20). By immunoblot analysis, 10 of the antibodies detected both PRA and PRB; only one antibody detected only PRB. By contrast, by immunohistochemistry, eight of the antibodies detected only PRA. These eight antibodies were unable to detect PRB in MCF-7 cells expressing only PRB. Two of the antibodies detected both PRA and PRB. Only one antibody detected PRB only.
The findings of Mote et al. (20) demonstrate the importance of using anti-PR antibodies with well-defined immunohistochemical PRA or PRB isoform specificity. Previous studies of PR in mouse mammary gland used anti-PR antibodies that had not been characterized for immunohistochemical PR isoform specificity (16, 18, 19). The purpose of the present study was to determine the in vivo expression pattern of PRA and PRB proteins in mouse mammary gland by immunohistochemistry using well-characterized, PR isoform-specific antibodies. We used antibodies that detect only PRA or only PRB by immunohistochemistry in human tissues and have also been shown to have the same isoform specificity in mouse ovary (21). Using these PR isoform-specific antibodies, we analyzed PR isoform expression and colocalization in various structures of the normal mouse mammary gland (ducts, end buds, side branches, alveoli) at different developmental stages that are known to exhibit different proliferative and morphological responses to P (22, 23, 24). We also investigated colocalization of PRA, PRB, 5-bromo-2'-deoxyuridine (BrdU), and cyclin D1.
Materials and Methods
Animals
BALB/c female mice from our own colony were the source of mammary glands at the following ages and developmental stages: virgin immature (3 or 6 wk), virgin adult (10–12 or 17–20 wk), pregnant (7 or 14 d), lactating (10 d), or postpartum involuting (9 wk). To simulate mammary gland development during pregnancy, ovary intact virgin mice received sc beeswax pellets containing 17?-estradiol (20 μg) plus progesterone (20 mg) (E+P) for 13 d. C57BL PRA null mice were obtained from Dr. Orla Conneely (Baylor College of Medicine, Houston TX). All animal experimentation was conducted in accord with accepted standards of humane animal care and approved by the All University Committee on Animal Use and Care at Michigan State University.
Immunohistochemistry with anti-PR isoform-specific antibodies
Mouse monoclonal antibodies specific in immunohistochemistry for PRA only (hPRa7; referred to as anti-PRA antibody) or PRB only (hPRa6; referred to as anti-PRB antibody) (20, 21) were a generous gift from Dr. Christine Clark (University of Sydney, Sydney, Australia) or were purchased from Neomarkers (Fremont, CA). Mammary tissues were fixed in 10% phosphate-buffered formalin [0.4% sodium phosphate monobasic and 0.65% sodium phosphate dibasic (anhydrous) in 10% formalin] overnight at 4 C, dehydrated, cleared, and embedded in paraffin. Five-micrometer sections were mounted onto coverslips to which 3-aminopropyl triethoxysilane had been applied and allowed to dry for 24 h at room temperature. Tissue sections were immersed in 10 mM sodium citrate solution (pH 6.0) and exposed to a combination of heat and pressure for antigen retrieval as previously described (25). The protocol used to detect PRA or PRB in mouse mammary gland was similar to that used in human breast tissue (26) and mouse ovary (21) as described. To block nonspecific background staining, sections were incubated with goat antimouse IgG Fab fragments (Jackson ImmunoResearch Laboratories, West Grove, PA) [1:100 in PBS containing 1% BSA (1% PBSA), 60 min], rinsed with PBS, and then blocked with normal goat serum (Vector Laboratories, Burlingame, CA) (1:1 dilution in PBS, 30 min). Incubation with primary mouse anti-PRA or anti-PRB monoclonal antibody (1:100 dilution in PBS/0.5% Triton X-100) was for 1 h followed by 30 min with a biotinylated goat antimouse antibody (Dako, Carpinteria, CA) (1:400) and ABC reagent (Vector Laboratories). Two PBS rinses were performed between incubation with each antibody. Immunoperoxidase localization of antibody staining was obtained using 3'-3'-diaminobenzidene. The sections were counterstained with hematoxylin. Sections were visualized using a Nikon Eclipse 400 microscope and a SPOT RT color camera with SPOT software (Diagnostic Instruments, Sterling Heights, MI).
Double labeling with PRA and PRB isoform-specific antibodies
When we labeled with either anti-PRA or anti-PRB antibody alone, virgin and pregnant mammary glands yielded the same isoform-specific staining patterns whether detection was by immunoperoxidase or immunofluorescence. However, when we double labeled virgin or pregnant mammary gland with the anti-PRA antibody plus anti-PRB antibody, the PRA- and PRB-specific patterns were not maintained, and all PR-positive cells were positive for both PRA and PRB. We overcame this antibody staining artifact in double-labeling experiments by using a rabbit polyclonal anti-PR antibody, sc-538 (Santa Cruz Biotechnology, Santa Cruz, CA) that we demonstrated in this study recognizes only PRA (see Fig. 7). With this method the PR isoform-specific patterns were maintained in double-labeling experiments. After antigen retrieval, sections were incubated overnight at 4 C with sc-538 (1:400 in 2% PBSA), rinsed twice with PBS, and incubated with goat antirabbit antibody conjugated to Alexa 488 (green) (Molecular Probes, Eugene, OR) (1:100 in PBS, 30 min). Sections were then blocked with goat antimouse IgG Fab fragments (Jackson ImmunoResearch Laboratories) (1:100 in 1% PBSA, 60 min), blocked with normal goat serum (Vector Laboratories) (1:1 in PBS, 30 min), and incubated overnight at 4 C with mouse monoclonal primary antibody (anti-PRB,1:50 in PBS-0.5% Triton X-100). PRB localization was detected with goat antimouse secondary antibody conjugated to Alexa 546 (red) (Molecular Probes) (1:100 in PBS, 30 min). In some experiments the fluorochromes used to detect PRA and PRB were reversed. Nuclei were counterstained with TOPRO-3 Iodide (blue) (Molecular Probes), and sections were visualized and images captured using a Pascal laser scanning confocal microscope (Zeiss, Thornwood, NY).
FIG. 7. Immunodetection of PRA by sc-538 anti-PR antibody. Tissue sections from 6-wk-old virgin (A) or 14-d pregnant (B) mammary glands were double labeled with anti-PRA antibody (red nuclei) and sc-538 antibody (green nuclei); nuclei were counterstained with TOPRO-3 (dark blue). In the virgin and pregnant gland, the anti-PRA and sc-538 antibody staining show complete colocalization and are visualized as white nuclei in the merged images. Scale bar, 20 μm.
Immunoblot analysis
In the 6-wk-old virgin mammary gland, there is a high ratio of stroma to epithelium. To overcome the problem of dilution of epithelial cell proteins, mammary epithelial cells were obtained from pooled mammary glands of seven 6-wk-old mice and enriched by an enzymatic dissociation method used to obtain epithelial cells for primary culture, as previously described (27). Whole mammary glands were obtained from 14-d pregnant mice. Uteri were obtained from 6-wk-old virgin mice. Whole mammary glands or uteri were minced and homogenized in a buffer of 50 mM potassium phosphate (pH 7.0), 10 nM EGTA, 10 mM sodium molybdate, 12 mM thioglycerol, 10% glycerol (1 ml/g mammary tissue, 0.5 ml/uterus) containing protease inhibitor cocktail (Sigma, St. Louis, MO) using a Polytron homogenizer. Epithelial cells were sonicated in 400 μl of buffer of 50 mM potassium phosphate (pH 7.0), 10 nM EGTA, 10 mM sodium molybdate, 12 mM thioglycerol, 10% glycerol. Homogenates were centrifuged at 14,000 x g for 30 min and supernatants were used for immunoblots. Mammary gland extract (35 μl) or uterine extract (15 μl) was mixed with NuPAGE LDS sample buffer and NuPAGE sample reducing reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruction and boiled for 10 min at 70 C. Protein samples were resolved on 4–20%NuPAGE Bis-Tris gel (Invitrogen) under denaturing conditions and transferred onto Protran nitrocellulose membranes (Schleicher & Schuell, Keene, NH). Membranes were treated with Qentix Western blot signal enhancer (Pierce, Rockford, IL), blocked in 5% milk in Tris-buffered saline with 0.5% Tween 20 overnight at 4 C and incubated with primary antibodies for at least 2 h at room temperature.
To detect PR, mouse monoclonal antihuman PR hPRa7 (dilution 1:100) or hPRa6 (dilution 1:100) (Neomarkers) or rabbit polyclonal antihuman PR sc-538+sc-539 (dilution 1:100 for each) (Santa Cruz Biotechnology) primary antibodies were used. The combination of sc- 538+sc-539 was used for immunoblot analysis of pregnant mammary gland in an attempt to enhance detection of PRA because PRA expression was reduced during pregnancy. The secondary antibodies were horseradish peroxidase-labeled sheep antimouse antibody (dilution 1:2000) (Amersham, Buckinghamshire, UK) or donkey antirabbit antibody (dilution 1:2000) (Santa Cruz Biotechnology), respectively. After 1 h incubation with secondary antibodies, membranes were washed, incubated with Super Signal West Pico chemiluminescent substrate (Pierce), and exposed to x-ray film for 2–10 min.
Colocalization of PRA, PRB, cyclin D1, and BrdU
For these studies mouse monoclonal anti-BrdU antibody (provided as a kit from Amersham Biosciences, Piscataway, NJ) and mouse monoclonal anti-cyclin D1 antibody (Cell Signaling Technology, Beverly, MA) were used. After antigen retrieval tissue sections were incubated overnight at 4 C with mouse monoclonal anti-PRA or anti-PRB antibody. PRA or PRB localization was detected with goat antimouse secondary antibody conjugated to Alexa 546 (red) (Molecular Probes) (1:100 in PBS, 30 min). Sections were then blocked with goat antimouse IgG Fab fragments (Jackson ImmunoResearch Laboratories) (1:200 in 1% PBSA, 60 min), blocked with normal goat serum (Vector Laboratories) (1:1 in PBS, 30 min), and incubated for 1 h at room temperature with anti-BrdU antibody or overnight at 4 C with the anti-cyclin D1 antibody (1:200 in 2% PBSA). BrdU and cyclin D1 localization were detected with a biotinylated goat antimouse secondary antibody (Dako) (1:400 in PBS, 30 min), which was recognized by streptavidin-conjugated Alexa 488 (green) (Molecular Probes) (1:100 in PBS, 45 min).
For all dual-immunofluorescent labeling, nuclei were counterstained with TOPRO-3 iodide (blue) (Molecular Probes), and sections were visualized and images captured using a Zeiss Pascal laser scanning confocal microscope.
PR quantitation and statistical analyses
Sections treated for PRA and/or PRB detection by immunoperoxidase or immunofluorescence methods were quantitated for the number of PRA- and/or PRB-positive cells with the aid of a light microscope (immunoperoxidase) or from captured images (immunofluorescence). Three to 10 mice per developmental stage were analyzed; a minimum of 1000 total cells and three independent sections per mouse were analyzed. PR-positive cells are expressed as a percentage of total epithelial cells counted. Results are expressed as mean ± SEM, and differences are considered significant at P < 0.05 by using Student’s t test or ANOVA where appropriate.
Results
Immunoperoxidase localization and quantitation of PRA at different stages of mammary gland development
The earliest age examined for PRA expression was 3 wk of age. At this age ovarian cycles have not yet started, and the prepubertal mammary gland exists as a small epithelial rudiment similar to the one present at birth; the percentage of PRA-positive cells was 55 ± 2% (Fig. 1). By 6 wk of age, ovarian cycles have started and in the pubertal 6-wk-old virgin mammary gland, 58 ± 3% of mammary epithelial cells were PRA positive (Fig. 1). At 6 wk of age, the PRA-positive cells were observed in end buds (Fig. 2, A and E) and ducts (Fig 2, B and F). PRA-positive cells in end buds were localized in the internal layer of cells; the cap cell layer of end buds was negative for PRA (Fig. 2E). At 10–12 wk of age, the mammary glands of most mice had grown to the limits of the fat pad; however, the glands of some mice (23%) still contained end buds. In the mammary glands of 10- to 12-wk-old virgin mice, the percentage of PRA-positive epithelial cells in ducts was 50 ± 2% (Figs. 1 and 2, C and G), which was not significantly different from 3- or 6-wk-old virgin. We also examined the effect of estrus cycle stage on PRA expression; no difference in the percentage of PRA-positive cells was observed at estrus vs. diestrus (52 ± 3% estrus vs. 51 ± 4% diestrus). At 17–20 wk of age, in all cases, end buds were no longer detected, and the ductal tree had grown to the limits of the mammary fat pad. The percentage of PRA-positive cells decreased significantly to 28 ± 3% (P < 0.05) (Fig. 1). At 7 d of pregnancy, PRA was detected in 25 ± 1% of cells (Fig. 1). However, at 14 days of pregnancy, PRA was detected in only 11 ± 2% of ductal epithelial cells and 6.0 ± 0.3% of alveolar cells (P < 0.001) (Figs. 1 and 2, D and H). No PRA-positive cells were detected during lactation (Fig. 1). After lactational involution, at 9 wk post weaning, PRA was detected in 12 ± 1% of the ductal cells and 10 ± 1% regressed alveolar cells (Fig. 1). Notably, the percentage of PRA-positive cells was significantly lower after pregnancy, compared with age-matched virgin mice, at 17–20 wk of age (P < 0.01) (Fig. 1). Antibody staining of PRA was always localized to the nucleus of epithelial cells and was not detected in myoepithelial cells or stromal cells at any of the developmental stages studied (Fig. 2).
FIG. 1. Quantitation of PRA at different stages of mammary gland development. Immunoperoxidase localization of PRA was carried out using anti-PRA antibody on tissue sections from 3-, 6-, 10- to 12, or 17- to 20-wk-old virgin, 7- or 14-d pregnant, 10-d lactating (Lact) mice, and at 9 wk post weaning [lactational involution (Invol)] as described in Materials and Methods. The values represent the mean ± SEM from three to five mice per group with a minimum of 1000 cells/mouse analyzed. PRA-positive cells decreased significantly with age (3 and 6 wk > 10–12 wk > 17–20 wk) in virgin mice and further during pregnancy and lactation (7 d > 14 d > lactating). The 9-wk involuted mammary gland had fewer PRA-positive cells than age-matched virgin mammary gland (17–20 wk old) (P < 0.01). No PRA staining was detected (ND) during lactation.
FIG. 2. Immunoperoxidase localization of PRA at different stages of mammary gland development. Representative sections from 6-wk-old immature (A and E, end bud; B and F, duct), 12-wk-old adult (C and G, duct), and 14-d pregnant (D and H, alveoli) mouse mammary gland were treated with anti-PRA antibody (A-H) as described in Materials and Methods and control sections without antibody (I, immature end bud; J, immature duct; K, adult duct; L, pregnant alveoli). Higher-magnification images of boxed areas in A–D are shown in E–H. Brown-stained PRA-positive nuclei are indicated by black arrowheads and PRA-negative cells by red arrowheads. End bud cap cells (E) or myoepithelial cells (F and G) are indicated by arrows (scale bar, 50 μm).
Immunoperoxidase localization and quantitation of PRB at different stages of mammary gland development
No PRB-positive cells were detected in 3-, 6-, 10- to 12-, or 17- to 20-wk-old virgin mammary glands (Figs. 3 and 4, A, B, D, and E). During pregnancy, no PRB was detected at 7 d, but PRB was abundantly expressed by 14 d in 48 ± 4% of epithelial cells (Fig. 3). PRB was localized mainly in alveolar cells (Fig. 4, C and F). PRB staining was seen in both the cytoplasm and nucleus of epithelial cells (Fig. 4F). PRB was not detected in myoepithelial cells or stromal cells (Fig. 4F). No PRB was detected in the lactating mammary gland. After lactational involution PRB staining was observed in 6 ± 1% of cells in remaining alveolar structures (Fig. 3); less than 1% of ductal cells were PRB positive.
FIG. 3. Quantitation of PRB at different stages of mammary gland development. Immunofluorescence localization of PRB was carried out using anti-PRB antibody on tissue sections from 3-, 6-, 10- to 12, or 17- to 20-wk-old virgin, 7- or 14-d pregnant, 10-d lactating mice, and at 9 wk post weaning [lactational involution (Invol)] as described in Materials and Methods. The values represent the mean ± SEM from three to five mice per group with a minimum of 1000 cells/mouse analyzed. No PRB staining was detected (ND) in the virgin mammary gland (3, 6, 10–12, or 17–20 wk old), in the 7-d pregnant mammary gland or during lactation. PRB was detected at 14 d of pregnancy and in a smaller percentage of cells in the 9-wk involuted mammary gland (P < 0.001).
FIG. 4. Immunoperoxidase localization of PRB at different stages of mammary gland development. Representative sections from 6-wk-old immature (A and D, end bud), 12-wk-old adult (B and E, duct), and 14-d pregnant (C and F, alveoli) mice were treated with anti-PRB antibody (A–F) as described in Materials and Methods and control sections without antibody (G, immature end bud; H, adult duct; I, pregnant alveoli). Higher-magnification images of boxed areas in A and B are shown in D and E, and a higher magnification of pregnant mammary gland is shown in F. Brown-stained PRB-positive nuclei (F) are indicated by black arrowheads, PRB-positive cytoplasmic staining (F) by double arrows, and PRB-negative nuclei by red arrowheads. End bud cap cells (D) are indicated by arrows (scale bar, 50 μm).
PR isoform specificity of antibodies for immunohistochemistry
Although Mote et al. (20) had already demonstrated PRA and PRB isoform specificity of monoclonal anti-PRA (hPRa7) and anti-PRB (hPRa6) antibodies, respectively, we sought further confirmation using PRAKO mice (15). Figure 5A shows that no staining was detected with the anti-PRA antibody in virgin 8-wk-old PRAKO mice, whereas PRA-positive cells were detected in wild-type 8-wk-old virgin mice. PRAKO mice cannot become pregnant; however, E+P treatment induces pregnancy-like lobuloalveolar development (15). Figure 5A also shows that no PRA staining was detected in E+P-treated PRAKO mice, whereas PRA staining was detected in E+P-treated wild-type mice.
FIG. 5. Immunodetection of PRA and PRB in wild-type vs. PRA null mice. Immunofluorescence localization of PRA (A) or PRB (B) was carried out on sections from 8-wk-old virgin or 13-d E+P-treated 12-wk-old virgin wild-type (WT) and PRAKO mice. Antibody staining was carried out with anti-PRA antibody (red nuclei) or anti-PRB antibody (light blue nuclei); nuclei were counterstained with TOPRO-3 (dark blue nuclei). Positive staining is indicated by white arrowheads and negative nuclei are indicated by yellow arrowheads. Scale bar, 20 μm.
Figure 5B shows that PRB staining was observed in 12-wk-old E+P-treated PRAKO mice with the anti-PRB antibody, and the pattern of staining was the same as seen in wild-type E+P-treated mice. No PRB staining was detected in 8-wk-old virgin PRAKO or wild-type mice (Fig. 5B). These results demonstrate that the anti-PRB antibody detects PRB only in PRAKO mice (under conditions of simulated pregnancy) similar to wild-type mice. Thus, the staining patterns obtained in PRAKO mice confirmed the specificity of the anti-PRA antibody to detect only PRA and the specificity of the anti-PRB antibody to detect only PRB.
Immunoblot analysis of PRA and PRB expression
The apparent absence of PRB in virgin mammary gland, based on immunohistochemistry, was explored further by immunoblot analysis, using an antibody that detects both PRA and PRB isoforms (Fig. 6A). As expected, both PRA and PRB were detected in mouse uterus immunoblot (lane 1), which is known to express both isoforms (28). By contrast, this antibody detected only PRA in virgin mammary gland (Fig. 6A, lane 2). PRB was not detected in virgin mammary gland using antibody specific for PRB (hPRa6) (Fig. 6B, lane 2); the same antibody detected PRB in mouse uterus (Fig. 6B, lane 1). These findings are consistent with our immunohistochemical finding of only PRA in virgin mammary gland and indicate that absence of PRB is not due to epitope masking.
FIG. 6. Immunoblot analysis of PR in mammary gland. A, Extracts from uterus (lane 1) and isolated epithelial cells from 6-wk-old mammary glands (lane 2) were subjected to SDS-PAGE, and blots were probed with hPRa7 anti-PR antibody as described in Materials and Methods. PRA was detected as a single band at 91 kDa in uterus and isolated epithelial cells (lanes 1 and 2); PRB was detected as a single band at 119 kDa in uterus only (lane 1). B, Extracts from uterus (lane 1) and isolated epithelial cells from 6-wk-old mammary glands (lane 2) were subjected to SDS-PAGE, and blots were probed with hPRa6 anti-PR antibody, which detects only PRB, as described in Materials and Methods. PRB was detected as a single band at 119 kDa in uterus (lane 1); no PRB was detected in isolated epithelial cells (lane 2). C, Extracts from uterus (lane 1) and whole 14-d pregnant mammary glands (lane 2) were subjected to SDS-PAGE, and blots were probed with a mixture of sc-538 and sc-539 anti-PR antibodies described in Methods. PRA was detected as a single band at 91 kDa in uterus only (lane 1); PRB was detected as a single band at 119 kDa in uterus and mammary gland (lanes 1 and 2).
The low level of PRA in pregnant mammary gland, based on immunohistochemistry, was explored further by immunoblot analysis using antibody that detects both isoforms (Fig. 6B). Immunoblot analysis showed only a PRB band (Fig. 6B, lane 2), consistent with immunohistochemistry showing a predominance of PRB over PRA. Failure to detect a PRA band is likely due to the low percentage of PRA-positive cells in pregnant mammary gland.
Immunofluorescence colocalization of PRA and PRB
The different patterns of PRA and PRB expression observed during pregnancy suggested that PRA and PRB are present in different cells. To test this hypothesis, we undertook colocalization studies with the anti-PRA and anti-PRB antibodies. Immunofluorescent labeling of virgin and pregnant mammary glands with either anti-PRA or anti-PRB antibody alone yielded the same isoform-specific staining patterns that were obtained by immunoperoxidase detection. However, when we double labeled virgin or pregnant mammary gland with the anti-PRA antibody plus anti-PRB antibody, the PRA- and PRB-specific patterns were not maintained, and all PR-positive cells were positive for both PRA and PRB. To overcome this artifact, we sought to identify another antibody that was specific for only PRA in immunohistochemistry.
The pattern of PRA expression that we observed in the virgin mammary gland with the anti-PRA antibody was similar to antibody staining patterns reported by others who used the sc-538 anti-PR antibody (18). This led us to surmise that the sc-538 antibody might in fact be PRA isoform specific. To directly test this hypothesis, we carried out double-labeling experiments with the anti-PRA antibody plus sc-538 antibody and used immunofluorescence confocal microscopy to investigate colocalization of the antibodies. The results presented in Fig. 7A show complete colocalization of the anti-PRA antibody with the sc-538 anti-PR antibody in the virgin mammary gland. In the 14-d pregnant gland (Fig. 7B) the sc-538 antibody also showed complete colocalization with the anti-PRA antibody and the same low level of expression (relative to the virgin) that was observed with the monoclonal anti-PRA antibody (Fig. 2, D and H). PRA was exclusively localized in the nucleus with the sc-538 antibody in both virgin and pregnant mammary glands. Thus, it appears that the sc-538 antibody is specific for the PRA isoform in immunohistochemistry. The sc-538 has also been shown to be specific for PRA in immunohistochemistry in human cells (20).
Having established the specificity of sc-538 to detect only PRA, we carried out colocalization studies of PRA and PRB in double-labeling experiments with sc-538 and the anti-PRB antibody. At 14 d of pregnancy, three subsets of cells were found: cells that expressed PRA only, PRB only, or both PRA and PRB (Fig. 8, A and B). Forty-three percent of cells were positively labeled for PRB (Fig. 8A). Of the PRA-positive cells (8%), about half were also positive for PRB (Fig. 8A). Thus, colocalization of PRA and PRB occurred in only 4% of cells during pregnancy.
FIG. 8. Colocalization of PRA and PRB in pregnancy. Dual-immunofluorescence detection of PRA and PRB was carried out and visualized by laser scanning confocal microscopy as described in Materials and Methods. A, Quantitation of PRA and PRB colocalization; the values represent the mean ± SEM of the percentage of epithelial cells expressing one isoform only (PRA or PRB) or both isoforms (PRAB); values were obtained using five mice with a minimum of 1000 cells/mouse analyzed. B, Photomicrograph of PRA and PRB colocalization (PRA, green nuclei; PRB, red nuclei); nuclei were counterstained with TOPRO-3 (blue nuclei). Three subsets of PR-positive cells are seen in the merged image: those expressing both isoforms (white nuclei in square), PRA only (green nucleus in circle) or PRB only (red nuclei in oval). Scale bar, 20 μm.
PR isoform expression and colocalization with cyclin D1 or BrdU
A role for P has been implicated in ductal development in the virgin mammary gland (19). PRB and cyclin D1 are required for alveologenesis during pregnancy (16, 29). Epithelial cell proliferation is common to both ductal development and alveologenesis. Having found that PRA and PRB are present in different cells and at different stages of mammary gland development, it was of interest to determine how PR isoform expression was related to proliferation and cyclin D1 expression. To accomplish this, mammary glands were obtained from 14-d pregnant and 6-wk-old mice injected with a pulse of BrdU 2 h before being killed to label cells in S phase. Tissue sections were double labeled with anti-BrdU plus anti-PRA antibody or with anti-BrdU antibody plus anti-PRB antibody. Additional tissue sections were also double labeled with anti-cyclin D1 antibody plus anti-PRA or anti-PRB antibody. Immunofluorescence confocal microscopy was used to determine the colocalization of PRA and/or PRB with BrdU or cyclin D1.
In the pregnant mammary gland, 16% of cells were BrdU positive at 2 h after BrdU injection, and 46% of cells were PRB positive (Fig. 9A). Fifteen percent of cells were BrdU and PRB positive; thus, 95% of BrdU-positive cells were PRB positive (Figs. 9A and 10A). In pregnant mammary gland, PRA and BrdU were not colocalized in the same cells (Figs. 9A and 10B).
FIG. 9. Quantitation of colocalization of PRB or PRA with BrdU in pregnant and virgin mammary glands. Dual-immunofluorescence detection of PRB or PRA and BrdU was carried out on tissue sections from 14-d pregnant (A) and 6-wk-old virgin (B) mammary glands and visualized by laser scanning confocal microscopy as described in Materials and Methods. A minimum of 1000 cells were counted for each antibody combination tested, i.e. PRB and BrdU or PRA and BrdU in pregnant mammary gland and PRA and BrdU in virgin mammary gland. The values represent the mean ± SEM from three to five mice with a minimum of 1000 cells/mouse analyzed.
FIG. 10. Detection of colocalization of PRB or PRA with BrdU in pregnant and virgin mammary glands. Dual-immunofluorescence detection was carried out in 14-d pregnant (A and B) or 6-wk-old virgin (C and D) mammary gland using anti-PRB (A) or anti-PRA (B–D) antibodies and TOPRO-3 nuclear stain and were visualized by laser-scanning confocal microscopy as described in Materials and Methods. A, PRB (red nuclei, white arrows) and BrdU (green nuclei, white arrowheads) staining were extensively colocalized (white nuclei, yellow arrows) in merged images. B, PRA (red nuclei, white arrows) and BrdU (green nuclei, white arrowheads) staining did not colocalize in merged images and were seen as red (white arrows) and light blue (white arrowheads) nuclei. C, In 6-wk-old virgin mammary gland duct, PRA (green nuclei, white arrow), BrdU (red nuclei, white arrowhead) staining did not colocalize in merged images and were seen as red (white arrowheads) and light blue (white arrows) nuclei. D, In 6-wk-old mammary gland end bud, most PRA (green nuclei, white arrow), BrdU (red nuclei, white arrowhead) staining did not colocalize and in merged images and were seen as red (white arrowheads) and light blue (white arrows) nuclei. End bud cap cells were prominently labeled by BrdU (red nuclei, white arrowheads). Instances of colocalization of PRA and BrdU are seen in merged image as white nuclei (yellow arrows). Scale bar, 20 μm.
We also analyzed PRA and BrdU colocalization in the 6-wk-old, virgin mammary gland. We chose this age and stage of development because there is extensive proliferation and a high percentage of PRA-positive cells in the virgin mammary gland. We found that 15% of cells were BrdU positive, 56% of cells were PRA positive, and 4% were PRA and BrdU positive (Fig. 9B). Thus, only 27% of BrdU-positive cells were PRA positive and only 7% of PRA-positive cells were BrdU positive. Most BrdU-positive cells were located in the cap cell layer of end buds (Fig. 10D), which is a region of the end bud that is devoid of PRA-positive cells (Fig. 2, A and E). Fewer BrdU-positive cells were present in ducts (Fig. 10C).
In the pregnant mammary gland, 56% of cells were cyclin D1 positive and 49% were PRB positive (Fig. 11A). Forty-six percent of cells were positive for both PRB and cyclin D1; thus, 83% of cyclin D1-positive cells were PRB positive and 94% of PRB-positive cells were cyclin D1 positive (Fig. 12A). There was no colocalization of PRA with cyclin D1 (Figs. 11A and 12B).
FIG. 11. Quantitation of colocalization of PRB or PRA with cyclin D1 in pregnant and virgin mammary glands. Dual-immunofluorescence detection of PRB or PRA and cyclin D1 was carried out on tissue sections from 14-d pregnant (A) and 6-wk-old virgin (B) mammary glands and visualized by laser-scanning confocal microscopy as described in Materials and Methods. The values represent the mean ± SEM from three mice per group (virgin and pregnant) with a minimum of 1000 cells/mouse analyzed for each antibody combination tested, i.e. PRB and cyclin D1 or PRA and cyclin D1 in pregnant mammary gland and PRA and cyclin D1 in virgin mammary gland.
FIG. 12. Detection of colocalization of PRB or PRA with cyclin D1 in pregnant and virgin mammary glands. Dual-immunofluorescence detection of PRB or PRA and cyclin D1 was carried out on tissue sections from 14-d pregnant (A and B) and 6-wk-old virgin (C and D) mammary glands and visualized by laser-scanning confocal microscopy as described in Materials and Methods. Nuclei were counterstained with TOPRO-3 (A–D, blue). Examples of PRB- (A; red nuclei) or PRA (B–D; red nuclei)-positive cells are indicated with white arrows, and examples of cyclin D1-positive cells (A–C; green nuclei) are indicated with white arrowheads. PRB and cyclin D1 colocalization is seen as white nuclei in the merged image (A), and examples are indicated with yellow arrows. In pregnant mammary gland (B), there was no colocalization of PRA and cyclin D1 in the merged image, and PRA-positive nuclei stain red (white arrows) and cyclin D1-positive nuclei stain light blue (white arrowheads). In virgin mammary gland (C) when colocalization of PRA and cyclin D1 was observed, it was seen as white nuclei in the merged image; examples are indicated with yellow arrows. D, An example of a virgin duct without cyclin D1-positive cells. Scale bar, 20 μm.
In the 6-wk-old virgin mammary gland, the percentage of cyclin D1-positive cells was significantly less than in pregnant mammary gland (18 vs. 56%; P < 0.05) (Fig. 11, A and B). Fifty-four percent of cells were PRA positive, and cyclin D1 and PRA were colocalized in 1% of cells; thus, 4% of cyclin D1-positive cells were also PRA positive and 2% of PRA-positive cells were cyclin D1 positive (Fig. 11B). Figure 12C illustrates colocalization in a duct that is cyclin D1 positive. Many ducts had no cyclin D1-positive cells, yet PRA was highly expressed (Fig. 12D).
Discussion
The results presented in this paper demonstrate that PRA and PRB expression are temporally and spatially separated during murine mammary gland development. Only PRA was highly expressed in the immature and adult virgin mammary gland. By contrast, PRB was seen only during pregnancy, mainly in alveolar epithelial cells. During pregnancy, the majority of PR-positive cells contained only PRB, and colocalization of PRA and PRB occurred in a small proportion of epithelial cells. During pregnancy PRB colocalized extensively with the proliferation marker BrdU and with cyclin D1. In contrast, PRA did not colocalize with BrdU or cyclin D1 during pregnancy and was infrequently colocalized with BrdU or cyclin D1 in the virgin gland. The implication of these findings is that different actions of P are mediated in PRB positive vs. PRA-positive cells in vivo.
P action in the virgin mammary gland: predominant role of PRA
In the 6-wk-old immature virgin gland, whereas 54% of epithelial cells were PRA positive, only 2% of PRA-positive cells were cyclin D1 positive and only 4% of PRA-positive cells were BrdU positive. These results indicate that the majority of PRA-positive cells were not in S phase during our 2-h labeling period. Most BrdU-positive cells were in the cap cell layer of end buds, which is recognized to be a major growth point. The cap cell layer was devoid of PRA-positive cells, supporting the concept that PRA-positive cells do not constitute the major pool of proliferating cells. We cannot rule out the possibility that P may play a role in proliferation via a paracrine mechanism in which PRA-positive cells produce a factor that affects the proliferation of neighboring PR negative cap cells.
Proliferation leading to ductal elongation occurs via cap cell proliferation and is mediated by E and growth factors such as epidermal growth factor, hepatocyte growth factor, and IGF-I (3, 30). The requirement for E is supported by the complete absence of ductal elongation in ER gene-deleted mice (31). In contrast, ductal elongation does occur in total PR gene-deleted mice (14). These results indicate that the presence of PR is not an absolute requirement for ductal elongation in the virgin gland.
Organogenesis during embryonic development results from the net effect of the precise spatial patterning of proliferation and apoptosis. Similarly, postnatal ductal development in the mammary gland can be considered to be the result of spatially organized proliferation and apoptosis. Proliferation occurs in the cap cell layer of the end bud, giving rise to a multilayered internal mass of cells below the cap cell layer (32). Formation of the ductal lumen requires the removal of this internal cell mass. Apoptotic cells have been observed in this internal layer of cells of the end bud (32), suggesting that apoptosis may play a key role in lumen formation in ducts. We have previously shown in vitro that mammary organoids derived from virgin mammary gland respond to the synthetic progestin, R5020, by forming a lumen (30). Treatment of organoids with R5020 induces apoptosis that is spatially localized within mammary organoids and centrally within luminal structures; R5020 does not induce proliferation in these organoids (30). Based on these observations, we have hypothesized that one of the actions of P in mammary gland development is to facilitate lumen formation through P-induced apoptosis (30). In the present study, we showed that only PRA was expressed in the virgin gland, and within end buds PRA-positive cells were localized in the internal layer of cells. This raises the possibility that one way that P promotes ductal development in the virgin gland, at least in part, is by facilitating lumen formation through a proapoptotic mechanism mediated by PRA. The observation that ductal development can occur in total PR deleted as well as PRA gene-deleted mice indicates that there are additional mechanisms that promote lumen formation, and that these mechanisms are operative in PR gene-deleted mice and may compensate for the lack of PR.
P action in pregnancy: predominant role of PRB
PRB-positive cells were seen only in mammary glands of pregnant mice (Figs. 4C and 8B) or alveolar structures of adult E+P-treated mice (Fig. 5B). In pregnant mice PRB was abundantly expressed and the PRB-positive cells were localized mainly in alveolar structures. We found extensive colocalization of PRB with BrdU and cyclin D1 in pregnant mammary gland. This indicates that PRB-positive cells are in the proliferative pool of cells and express cyclin D1. Our results indicate that PRB has the primary role in inducing alveologenesis. Other studies have inferred the same conclusion based on different approaches, namely that there is no defect in alveologenesis in the PRAKO mouse (15), that there is a lack of alveologenesis in the PRB-only gene-deleted mouse (16), and that precocious alveologenesis occurs in PRB overexpressing transgenic mice (13).
In contrast to PRB, there was no PRA colocalization with either BrdU or cyclin D1 in the pregnant gland, suggesting that PRA-positive cells do not constitute the major proliferative pool in the pregnant mammary gland. These observations do not discount the possibility that PRA nevertheless plays a role in pregnancy because expansion of the epithelium and side branching are detected as early as d 7 of pregnancy (Aupperlee, M. D., and S. Z. Haslam, unpublished observations), when 27% of the epithelial cells were PRA positive and none were PRB positive (Figs. 1 and 3).
Previous studies have reported a lack of colocalization of PR with markers of proliferation (16, 33). However, in those studies the PR isoform specificity of the antibody used (Dako A0098) was not identified. We have determined that the Dako A0098 anti-PR antibody colocalizes with PRA and not with PRB. This was determined in studies carried out as shown for the sc-538 anti-PR antibody (Fig. 7) (Aupperlee, M. D., and S. Z. Haslam, unpublished observations). Our own studies using PR isoform-specific antibodies demonstrate a lack of colocalization of PRA with BrdU in pregnant mammary gland, but we find extensive colocalization of PRB with BrdU and cyclin D1 in pregnant mammary gland.
The lack of PRA and PRB staining during lactation is in agreement with previous reports of the absence of specific P ligand binding and lack of detectable PR mRNA in lactating mouse mammary gland (28, 34). Although PRA-positive cells were detected again after involution, the percentage of PRA-positive cells never returned to the prepregnancy virgin level. This was not due to aging because the percent of PRA-positive cells was significantly higher in 20-wk-old virgin mammary glands than age-matched parous mice. A low level of PRB (6% PRB-positive cells) was detected in alveolar structures after lactational involution but not in age-matched virgin mammary gland. These results demonstrate that expression of both PRA and PRB is permanently altered by pregnancy. Pregnancy is protective against carcinogen-induced mammary tumors in mice and rats (35). Our results show two important changes caused by pregnancy: a reduction in PRA-positive cells relative to age-matched virgins and presence of PRB after lactation relative to the virgin state. Further studies to elucidate the specific functional roles of PR isoforms in the mammary gland before, during, and after pregnancy may provide new insights about the mechanism(s) underlying differences in susceptibility to tumorigenesis of virgin vs. parous mice.
PR isoform subcellular localization and progesterone action
Previous studies using cell lines have shown that if expressed in the same cells, PRA and PRB proteins can dimerize and bind to DNA as three different species: AA or BB homodimers or AB heterodimers (9). The specific contribution of each of the dimers to the effects of P may be dependent on the transactivation properties contributed to the complexes by the PRB-specific domain. It has also been reported that PRB transcriptional activity is inhibited by PRA. During pregnancy we found that the vast majority of PRB-positive cells contained only PRB, and only a small percentage of cells (4%) contained both PRA and PRB. Our results indicate that the prevailing situation in the mouse mammary gland is that cells contain AA or BB homodimers and that the potential for AB heterodimer formation is limited to a small number of cells during pregnancy. This suggests that in the mouse, heterodimer formation does not play a major role in progesterone action in the mammary gland.
In our study, PRB was detected primarily in the nucleus and some cells faintly in the cytoplasm (see Fig. 4F). In contrast, using the anti-PRA and sc-538 anti-PR antibodies, PRA was detected only in the nucleus. We cannot rule out the possibility that there may also be a cytoplasmic form of PRA not detected by the PR antibodies we used. PR localization in the mammary gland in both cytoplasm and nucleus has been detected using a PR antibody of unknown PR isoform specificity (18) and in human T47D breast cancer cells overexpressing either PRA or PRB (36). It is conceivable that different anti-PR antibodies may detect epitopes that are exposed on the cytoplasmic, nuclear, or both forms of the receptor.
PR isoform expression in the human vs. mouse
In the human breast, immunohistochemical analysis of PR isoform expression has been carried out on normal tissue from premenopausal cycling women (26). In that study PRA or PRB expression and colocalization were determined by dual immunofluorescence with the same antibodies used herein (26). PRA vs. PRB was expressed at a ratio of 1:1, and patterns of expression were similar. The proportion of PR-positive cells was 10–20% with marked variability throughout a section, with PR positivity in individual ducts or lobules ranging from 0 to 90%. Dual-immunofluorescence studies revealed uniform colocalization of PRA with PRB (26). Our study indicates an interesting difference in PR isoform expression between the mouse and the human mammary gland. In the mouse there is PRB expression only during pregnancy, and colocalization of PRA with PRB occurs in only a small percentage of cells. One possible explanation for this difference may be the predominance of a ductal organization of mammary epithelium in the adult nonpregnant mouse. This is particularly true in BALB/c strain mice, used in our study. In contrast, in the adult nonpregnant human there is a higher ratio of lobules to ducts. Studies of PRB null mice have shown that PRB expression is required for alveologenesis and lobule formation (16). Therefore, PRB expression may be a defining characteristic of mammary lobule formation and/or maintenance and may explain why PRB-positive cells are more abundant in the human breast. In this regard, the maintenance of some alveolar structures in mouse mammary gland after pregnancy may also be due to the continued, albeit reduced, expression of PRB after pregnancy. Analysis of other mouse strains, such as the C3H strain, which develop a more lobular morphology in the virgin state (compared with BALB/c strain) may provide additional insights into the relationship between alveolar morphogenesis and PRB expression. It is also important to note that PR isoform expression in the human has only been studied in the adult premenopausal breast. There is no information on PR isoform expression at other stages of human breast development such as puberty or pregnancy. It remains to be seen what analysis of these other stages may reveal about the pattern of PR isoform expression and/or colocalization in the human breast. Clearly more information is needed about PR isoform expression in the human breast.
Understanding the specific functions of PRA and PRB isoforms in vivo is critical to understanding their respective roles in the normal breast and in the etiology of breast cancer. The spatial and temporal separation of PRA and PRB isoform expression in mouse mammary gland offers a unique opportunity to explore further the specific functions and mechanisms of action PRA vs. PRB in vivo.
Acknowledgments
The authors thank Alexis Drolet, Jeff Leipprandt, Nityanand Sunil, and Yvette Gross for technical support of these studies.
References
Haslam SZ, Osuch JR, Raafat AM, Hofseth LJ 2002 Postmenopausal hormone replacement therapy: effects on normal mammary gland in humans and in a mouse postmenopausal model. J Mammary Gland Biol Neoplasia 7:93–105
Cline JM, Soderqvist G, von Schoultz E, Skoog L, von Schoultz B 1996 Effects of hormone replacement therapy on the mammary gland of surgically postmenopausal cynomolgus macaques. Am J Obstet Gynecol 174:93–100
Fendrick JL, Raafat AM, Haslam SZ 1998 Mammary gland growth and development from the postnatal period to postmenopause: ovarian steroid receptor ontogeny and regulation in the mouse. J Mammary Gland Biol Neoplasia 3:7–22
Hofseth LJ, Raafat AM, Osuch JR, Pathak DR, Slomski CA, Haslam SZ 1999 Hormone replacement therapy with estrogen or estrogen plus medroxyprogesterone acetate is associated with increased epithelial proliferation in the normal postmenopausal breast. J Clin Endocrinol Metab 84:4559–4565
Magnusson C, Persson I, Adami HO 2000 More about: effect of hormone replacement therapy on breast cancer risk: estrogen versus estrogen plus progestin. J Natl Cancer Inst 92:1183–1184
Ross RK, Paganini-Hill A, Wan PC, Pike MC 2000 Effect of hormone replacement therapy on breast cancer risk: estrogen versus estrogen plus progestin. J Natl Cancer Inst 92:328–332
Schairer C, Lubin J, Troisi R, Sturgeon S, Brinton L, Hoover R 2000 Menopausal estrogen and estrogen-progestin replacement therapy and breast cancer risk. JAMA 283:485–491
Writing Group for the Women’s Health Initiative Investigators 2002 Risks and benefits of estrogen plus progestin in healthy postmenopausal women. JAMA 288:321–333
Lydon JP, Sivaraman L, Conneely OM 2000 A reappraisal of progesterone action in the mammary gland. J Mammary Gland Biol Neoplasia 5:325–338
Kastner P, Krust A, Turcotte B, Stropp U, Tora L, Gronemeyer H, Chambon P 1990 Two distinct estrogen-regulated promoters generate transcripts encoding the two functionally different human progesterone receptor forms A and B. EMBO J 9:1603–1614
Kraus WL, Montano MM, Katzenellenbogen BS 1993 Cloning of the rat progesterone receptor gene 5'-region and identification of two functionally distinct promoters. Mol Endocrinol 7:1603–1616
Shyamala G, Yang X, Silberstein G, Barcellos-Hoff MH, Dale E 1998 Transgenic mice carrying an imbalance in the native ratio of A to B forms of progesterone receptor exhibit developmental abnormalities in mammary glands. Proc Natl Acad Sci USA 95:696–701
Shyamala G, Yang X, Cardiff RD, Dale E 2000 Impact of progesterone receptor on cell-fate decisions during mammary gland development. Proc Natl Acad Sci USA 97:3044–3049
Lydon JP, DeMayo FJ, Funk CR, Mani SK, Hughes AR, Montgomery Jr CA, Shyamala G, Conneely OM, O’Malley BW 1995 Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev 9:2266–2278
Mulac-Jericevic B, Mullinax RA, DeMayo FJ, Lydon JP, Conneely OM 2000 Subgroup of reproductive functions of progesterone mediated by progesterone receptor-B isoform. Science 289:1751–1754
Mulac-Jericevic B, Lydon JP, DeMayo FJ, Conneely OM 2003 Defective mammary gland morphogenesis in mice lacking the progesterone receptor B isoform. Proc Natl Acad Sci USA 100:9744–9749
Takimoto GS, Tung L, Abdel-Hafiz H, Abel MG, Sartorius CA, Richer JK, Jacobsen BM, Bain DL, Horwitz KB 2003 Functional properties of the N-terminal region of progesterone receptors and their mechanistic relationship to structure. J Steroid Biochem Mol Biol 85:209–219
Silberstein GB, Van Horn K, Shyamala G, Daniel CW 1996 Progesterone receptors in the mouse mammary duct: distribution and developmental regulation. Cell Growth Differ 7:945–952
Atwood CS, Hovey RC, Glover JP, Chepko G, Ginsburg E, Robison WG, Vonderhaar BK 2000 Progesterone induces side-branching of the ductal epithelium in the mammary glands of peripubertal mice. J Endocrinol 167:39–52
Mote PA, Johnston JF, Manninen T, Tuohimaa P, Clarke CL 2001 Detection of progesterone receptor forms A and B by immunohistochemical analysis. J Clin Pathol 54:624–630
Gava N, Clarke CL, Byth K, Arnett-Mansfield RL, deFazio A 2004 Expression of progesterone receptors A and B in the mouse ovary during the estrous cycle. Endocrinology 145:3487–3494
Haslam SZ 1988 Progesterone effects on deoxyribonucleic acid synthesis in normal mouse mammary glands. Endocrinology 122:464–470
Haslam SZ 1989 The ontogeny of mouse mammary gland responsiveness to ovarian steroid hormones. Endocrinology 125:2766–2772
Haslam SZ, Counterman LJ, St. John AR 1993 Hormonal basis for acquisition of estrogen-dependent progesterone receptors in the normal mouse mammary gland. Steroid Biochem (Life Sci Adv) 12:27–34
Mote PA, Balleine RL, McGowan EM, Clarke CL 1999 Colocalization of progesterone receptors A and B by dual immunofluorescent histochemistry in human endometrium during the menstrual cycle. J Clin Endocrinol Metab 84:2963–2971
Mote PA, Bartow S, Tran N, Clarke CL 2002 Loss of co-ordinate expression of progesterone receptors A and B is an early event in breast carcinogenesis. Breast Cancer Res Treat 72:163–172
Haslam SZ, Levely ML 1985 Estrogen responsiveness of normal mouse mammary cells in primary cell culture: association of mammary fibroblasts with estrogenic regulation of progesterone receptors. Endocrinology 116:1835–1844
Shyamala G, Schneider W, Schott D 1990 Developmental regulation of murine mammary progesterone receptor gene expression. Endocrinology 126:2882–2889
Sicinski P, Weinberg RA 1997 A specific role for cyclin D1 in mammary gland development. J Mammary Gland Biol Neoplasia 2:335–342
Sunil N, Bennett JM, Haslam SZ 2002 Hepatocyte growth factor is required for progestin-induced epithelial cell proliferation and alveolar-like morphogenesis in serum-free culture of normal mammary epithelial cells. Endocrinology 143:2953–2960
Bocchinfuso WP, Korach KS 1997 Mammary gland development and tumorigenesis in estrogen receptor knockout mice. J Mammary Gland Biol Neoplasia 2:323–334
Hovey RC, Trott JF, Vonderhaar BK 2002 Establishing a framework for the functional mammary gland: from endocrinology to morphology. J Mammary Gland Biol Neoplasia 7:17–38
Seagroves TN, Lydon JP, Hovey RC, Vonderhaar BK, Rosen JM 2000 C/EBP? (CCAAT/enhancer binding protein) controls cell fate determination during mammary gland development. Mol Endocrinol 14:359–368
Haslam SZ, Shyamala G 1979 Effect of oestradiol on progesterone receptors in normal mammary glands and its relationship with lactation. Biochem J 182:127–131
Medina D 2004 Breast cancer: the protective effect of pregnancy. Clin Cancer Res 10:380S–384S
Jacobsen BM, Schittone SA, Richer JK, Horwitz KB 2005 Progesterone-independent effects of human progesterone receptors (PRs) in estrogen receptor-positive breast cancer: PR isoform-specific gene regulation and tumor biology. Mol Endocrinol 19:574–587(Mark D. Aupperlee, Kyle T)
Address all correspondence and requests for reprints to: Sandra Z. Haslam, Ph.D., Department of Physiology, 2201 Biomedical and Physical Sciences Building, Michigan State University, East Lansing, Michigan 48824. E-mail: shaslam@msu.edu.
Abstract
Progesterone is a potent mitogen in the mammary gland. Based on studies using cells and animals engineered to express progesterone receptor (PR) isoforms A or B, PRA and PRB are believed to have different functions. Using an immunohistochemical approach with antibodies specific for PRA only or PRB only, we show that PRA and PRB expression in mammary epithelial cells is temporally and spatially separated during normal mammary gland development in the BALB/c mouse. In the virgin mammary gland when ductal development is active, the only PR protein isoform expressed was PRA. PRA levels were significantly lower during pregnancy, suggesting a minor role at this stage of development. PRB was abundantly expressed only during pregnancy, during alveologenesis. PRA and PRB colocalization occurred in only a small percentage of cells. During pregnancy there was extensive colocalization of PRB with 5-bromo-2'-deoxyuridine (BrdU) and cyclin D1; 95% of BrdU-positive cells and 83% of cyclin D1-positive cells expressed PRB. No colocalization of PRA with either BrdU or cyclin D1 was observed at pregnancy. In the virgin gland, PRA colocalization with BrdU or cyclin D1 was low; only 27% of BrdU-positive cells and 4% of cyclin D1-positive cells expressed PRA. The implication of these findings is that different actions of progesterone are mediated in PRB positive vs. PRA-positive cells in vivo. The spatial and temporal separation of PR isoform expression in mouse mammary gland provides a unique opportunity to determine the specific functions of PRA vs. PRB in vivo.
Introduction
THE RELATIVE ROLES of estrogen (E) and progesterone (P) in regulating epithelial cell proliferation of the normal human breast and their contributions to breast cancer risk have been controversial. Originally it was presumed that because P antagonizes E-induced proliferation in the uterus, it would also antagonize E-induced proliferation in the breast (1). However, P in combination with E has more potent proliferative activity than E alone in the adult mammary gland in animal models (monkey and rodent) (2, 3) and the adult human breast (4). In humans this is the case for premenopausal cycling women and postmenopausal women receiving hormone replacement therapy (HRT). In postmenopausal women, combined continuous E+P HRT is associated with the highest proliferative index and the highest increase in breast epithelial density when compared with no HRT or E alone HRT (4). Furthermore, a significantly greater breast cancer risk is associated with E+P HRT (5, 6, 7, 8). Thus, P can contribute significantly to breast cancer risk.
P action is mediated through binding to the P receptor (PR). The PR consists of two isoforms, PRA and PRB, which are expressed from a single gene in both humans and rodents (9). Two promoters, one specific for PRA and the other specific for PRB, have been identified for human (10) and rat (11) PR. Initiation of translation at two distinct AUG signals produces the B and A forms of PR. PRB differs from PRA by an amino terminal extension of 164 amino acids. Studies to identify the functional roles of PRA and PRB in the mammary gland have been carried out in vivo using transgenic mice (PRA or PRB transgenes) (12, 13) and PR gene-deleted mice [total PR, PRA only (PRAKO) or PRB only] (14, 15, 16). From these studies it has been inferred that PRB is required for alveologenesis during pregnancy. The specific function of PRA has not yet been identified. In vitro studies using cell lines have shown that the unique amino terminal region of PRB encodes a transactivation function that plays an important role in specifying target genes that can be activated by PRB but not by PRA (17). Therefore, PRA and PRB can have different functions in the same cell, and the activity of the individual isoforms of the receptor may also vary among different types of cells.
The mouse is currently the most extensively studied and best understood model of P action in the normal mammary gland. Genetically altered mice have provided some insights into the functions of the two PR isoforms in mouse mammary gland. These genetically altered mice have an altered mammary gland phenotype (12, 13, 14, 15, 16); this suggests that mammary gland development is abnormal. Our approach in the present study was to investigate specific PR isoforms in mammary gland of genetically unaltered, wild-type mice as a function of development.
Biochemical methods to analyze PR isoform expression and function in the mouse mammary gland have provided limited information about the functional roles of PRA and PRB because they do not provide insight into the cellular distribution or colocalization of the isoforms. The most direct approach to address this question is immunohistochemical analysis of PR isoform-specific expression. It was generally assumed that if an anti-PR antibody detected both isoforms in immunoblot analysis, then it also detected both isoforms in immunohistochemical analysis (16, 18, 19). The report of Mote et al. (20) showed that this assumption is not correct. Mote et al. (20) analyzed a panel of 11 antihuman PR antibodies for their ability to detect PRA and/or PRB in human cells engineered to express specific isoforms of PR. To determine antibody specificity, MCF-7 breast cancer cell sublines that express only PRA, only PRB, or both PRA and PRB were analyzed (20). By immunoblot analysis, 10 of the antibodies detected both PRA and PRB; only one antibody detected only PRB. By contrast, by immunohistochemistry, eight of the antibodies detected only PRA. These eight antibodies were unable to detect PRB in MCF-7 cells expressing only PRB. Two of the antibodies detected both PRA and PRB. Only one antibody detected PRB only.
The findings of Mote et al. (20) demonstrate the importance of using anti-PR antibodies with well-defined immunohistochemical PRA or PRB isoform specificity. Previous studies of PR in mouse mammary gland used anti-PR antibodies that had not been characterized for immunohistochemical PR isoform specificity (16, 18, 19). The purpose of the present study was to determine the in vivo expression pattern of PRA and PRB proteins in mouse mammary gland by immunohistochemistry using well-characterized, PR isoform-specific antibodies. We used antibodies that detect only PRA or only PRB by immunohistochemistry in human tissues and have also been shown to have the same isoform specificity in mouse ovary (21). Using these PR isoform-specific antibodies, we analyzed PR isoform expression and colocalization in various structures of the normal mouse mammary gland (ducts, end buds, side branches, alveoli) at different developmental stages that are known to exhibit different proliferative and morphological responses to P (22, 23, 24). We also investigated colocalization of PRA, PRB, 5-bromo-2'-deoxyuridine (BrdU), and cyclin D1.
Materials and Methods
Animals
BALB/c female mice from our own colony were the source of mammary glands at the following ages and developmental stages: virgin immature (3 or 6 wk), virgin adult (10–12 or 17–20 wk), pregnant (7 or 14 d), lactating (10 d), or postpartum involuting (9 wk). To simulate mammary gland development during pregnancy, ovary intact virgin mice received sc beeswax pellets containing 17?-estradiol (20 μg) plus progesterone (20 mg) (E+P) for 13 d. C57BL PRA null mice were obtained from Dr. Orla Conneely (Baylor College of Medicine, Houston TX). All animal experimentation was conducted in accord with accepted standards of humane animal care and approved by the All University Committee on Animal Use and Care at Michigan State University.
Immunohistochemistry with anti-PR isoform-specific antibodies
Mouse monoclonal antibodies specific in immunohistochemistry for PRA only (hPRa7; referred to as anti-PRA antibody) or PRB only (hPRa6; referred to as anti-PRB antibody) (20, 21) were a generous gift from Dr. Christine Clark (University of Sydney, Sydney, Australia) or were purchased from Neomarkers (Fremont, CA). Mammary tissues were fixed in 10% phosphate-buffered formalin [0.4% sodium phosphate monobasic and 0.65% sodium phosphate dibasic (anhydrous) in 10% formalin] overnight at 4 C, dehydrated, cleared, and embedded in paraffin. Five-micrometer sections were mounted onto coverslips to which 3-aminopropyl triethoxysilane had been applied and allowed to dry for 24 h at room temperature. Tissue sections were immersed in 10 mM sodium citrate solution (pH 6.0) and exposed to a combination of heat and pressure for antigen retrieval as previously described (25). The protocol used to detect PRA or PRB in mouse mammary gland was similar to that used in human breast tissue (26) and mouse ovary (21) as described. To block nonspecific background staining, sections were incubated with goat antimouse IgG Fab fragments (Jackson ImmunoResearch Laboratories, West Grove, PA) [1:100 in PBS containing 1% BSA (1% PBSA), 60 min], rinsed with PBS, and then blocked with normal goat serum (Vector Laboratories, Burlingame, CA) (1:1 dilution in PBS, 30 min). Incubation with primary mouse anti-PRA or anti-PRB monoclonal antibody (1:100 dilution in PBS/0.5% Triton X-100) was for 1 h followed by 30 min with a biotinylated goat antimouse antibody (Dako, Carpinteria, CA) (1:400) and ABC reagent (Vector Laboratories). Two PBS rinses were performed between incubation with each antibody. Immunoperoxidase localization of antibody staining was obtained using 3'-3'-diaminobenzidene. The sections were counterstained with hematoxylin. Sections were visualized using a Nikon Eclipse 400 microscope and a SPOT RT color camera with SPOT software (Diagnostic Instruments, Sterling Heights, MI).
Double labeling with PRA and PRB isoform-specific antibodies
When we labeled with either anti-PRA or anti-PRB antibody alone, virgin and pregnant mammary glands yielded the same isoform-specific staining patterns whether detection was by immunoperoxidase or immunofluorescence. However, when we double labeled virgin or pregnant mammary gland with the anti-PRA antibody plus anti-PRB antibody, the PRA- and PRB-specific patterns were not maintained, and all PR-positive cells were positive for both PRA and PRB. We overcame this antibody staining artifact in double-labeling experiments by using a rabbit polyclonal anti-PR antibody, sc-538 (Santa Cruz Biotechnology, Santa Cruz, CA) that we demonstrated in this study recognizes only PRA (see Fig. 7). With this method the PR isoform-specific patterns were maintained in double-labeling experiments. After antigen retrieval, sections were incubated overnight at 4 C with sc-538 (1:400 in 2% PBSA), rinsed twice with PBS, and incubated with goat antirabbit antibody conjugated to Alexa 488 (green) (Molecular Probes, Eugene, OR) (1:100 in PBS, 30 min). Sections were then blocked with goat antimouse IgG Fab fragments (Jackson ImmunoResearch Laboratories) (1:100 in 1% PBSA, 60 min), blocked with normal goat serum (Vector Laboratories) (1:1 in PBS, 30 min), and incubated overnight at 4 C with mouse monoclonal primary antibody (anti-PRB,1:50 in PBS-0.5% Triton X-100). PRB localization was detected with goat antimouse secondary antibody conjugated to Alexa 546 (red) (Molecular Probes) (1:100 in PBS, 30 min). In some experiments the fluorochromes used to detect PRA and PRB were reversed. Nuclei were counterstained with TOPRO-3 Iodide (blue) (Molecular Probes), and sections were visualized and images captured using a Pascal laser scanning confocal microscope (Zeiss, Thornwood, NY).
FIG. 7. Immunodetection of PRA by sc-538 anti-PR antibody. Tissue sections from 6-wk-old virgin (A) or 14-d pregnant (B) mammary glands were double labeled with anti-PRA antibody (red nuclei) and sc-538 antibody (green nuclei); nuclei were counterstained with TOPRO-3 (dark blue). In the virgin and pregnant gland, the anti-PRA and sc-538 antibody staining show complete colocalization and are visualized as white nuclei in the merged images. Scale bar, 20 μm.
Immunoblot analysis
In the 6-wk-old virgin mammary gland, there is a high ratio of stroma to epithelium. To overcome the problem of dilution of epithelial cell proteins, mammary epithelial cells were obtained from pooled mammary glands of seven 6-wk-old mice and enriched by an enzymatic dissociation method used to obtain epithelial cells for primary culture, as previously described (27). Whole mammary glands were obtained from 14-d pregnant mice. Uteri were obtained from 6-wk-old virgin mice. Whole mammary glands or uteri were minced and homogenized in a buffer of 50 mM potassium phosphate (pH 7.0), 10 nM EGTA, 10 mM sodium molybdate, 12 mM thioglycerol, 10% glycerol (1 ml/g mammary tissue, 0.5 ml/uterus) containing protease inhibitor cocktail (Sigma, St. Louis, MO) using a Polytron homogenizer. Epithelial cells were sonicated in 400 μl of buffer of 50 mM potassium phosphate (pH 7.0), 10 nM EGTA, 10 mM sodium molybdate, 12 mM thioglycerol, 10% glycerol. Homogenates were centrifuged at 14,000 x g for 30 min and supernatants were used for immunoblots. Mammary gland extract (35 μl) or uterine extract (15 μl) was mixed with NuPAGE LDS sample buffer and NuPAGE sample reducing reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruction and boiled for 10 min at 70 C. Protein samples were resolved on 4–20%NuPAGE Bis-Tris gel (Invitrogen) under denaturing conditions and transferred onto Protran nitrocellulose membranes (Schleicher & Schuell, Keene, NH). Membranes were treated with Qentix Western blot signal enhancer (Pierce, Rockford, IL), blocked in 5% milk in Tris-buffered saline with 0.5% Tween 20 overnight at 4 C and incubated with primary antibodies for at least 2 h at room temperature.
To detect PR, mouse monoclonal antihuman PR hPRa7 (dilution 1:100) or hPRa6 (dilution 1:100) (Neomarkers) or rabbit polyclonal antihuman PR sc-538+sc-539 (dilution 1:100 for each) (Santa Cruz Biotechnology) primary antibodies were used. The combination of sc- 538+sc-539 was used for immunoblot analysis of pregnant mammary gland in an attempt to enhance detection of PRA because PRA expression was reduced during pregnancy. The secondary antibodies were horseradish peroxidase-labeled sheep antimouse antibody (dilution 1:2000) (Amersham, Buckinghamshire, UK) or donkey antirabbit antibody (dilution 1:2000) (Santa Cruz Biotechnology), respectively. After 1 h incubation with secondary antibodies, membranes were washed, incubated with Super Signal West Pico chemiluminescent substrate (Pierce), and exposed to x-ray film for 2–10 min.
Colocalization of PRA, PRB, cyclin D1, and BrdU
For these studies mouse monoclonal anti-BrdU antibody (provided as a kit from Amersham Biosciences, Piscataway, NJ) and mouse monoclonal anti-cyclin D1 antibody (Cell Signaling Technology, Beverly, MA) were used. After antigen retrieval tissue sections were incubated overnight at 4 C with mouse monoclonal anti-PRA or anti-PRB antibody. PRA or PRB localization was detected with goat antimouse secondary antibody conjugated to Alexa 546 (red) (Molecular Probes) (1:100 in PBS, 30 min). Sections were then blocked with goat antimouse IgG Fab fragments (Jackson ImmunoResearch Laboratories) (1:200 in 1% PBSA, 60 min), blocked with normal goat serum (Vector Laboratories) (1:1 in PBS, 30 min), and incubated for 1 h at room temperature with anti-BrdU antibody or overnight at 4 C with the anti-cyclin D1 antibody (1:200 in 2% PBSA). BrdU and cyclin D1 localization were detected with a biotinylated goat antimouse secondary antibody (Dako) (1:400 in PBS, 30 min), which was recognized by streptavidin-conjugated Alexa 488 (green) (Molecular Probes) (1:100 in PBS, 45 min).
For all dual-immunofluorescent labeling, nuclei were counterstained with TOPRO-3 iodide (blue) (Molecular Probes), and sections were visualized and images captured using a Zeiss Pascal laser scanning confocal microscope.
PR quantitation and statistical analyses
Sections treated for PRA and/or PRB detection by immunoperoxidase or immunofluorescence methods were quantitated for the number of PRA- and/or PRB-positive cells with the aid of a light microscope (immunoperoxidase) or from captured images (immunofluorescence). Three to 10 mice per developmental stage were analyzed; a minimum of 1000 total cells and three independent sections per mouse were analyzed. PR-positive cells are expressed as a percentage of total epithelial cells counted. Results are expressed as mean ± SEM, and differences are considered significant at P < 0.05 by using Student’s t test or ANOVA where appropriate.
Results
Immunoperoxidase localization and quantitation of PRA at different stages of mammary gland development
The earliest age examined for PRA expression was 3 wk of age. At this age ovarian cycles have not yet started, and the prepubertal mammary gland exists as a small epithelial rudiment similar to the one present at birth; the percentage of PRA-positive cells was 55 ± 2% (Fig. 1). By 6 wk of age, ovarian cycles have started and in the pubertal 6-wk-old virgin mammary gland, 58 ± 3% of mammary epithelial cells were PRA positive (Fig. 1). At 6 wk of age, the PRA-positive cells were observed in end buds (Fig. 2, A and E) and ducts (Fig 2, B and F). PRA-positive cells in end buds were localized in the internal layer of cells; the cap cell layer of end buds was negative for PRA (Fig. 2E). At 10–12 wk of age, the mammary glands of most mice had grown to the limits of the fat pad; however, the glands of some mice (23%) still contained end buds. In the mammary glands of 10- to 12-wk-old virgin mice, the percentage of PRA-positive epithelial cells in ducts was 50 ± 2% (Figs. 1 and 2, C and G), which was not significantly different from 3- or 6-wk-old virgin. We also examined the effect of estrus cycle stage on PRA expression; no difference in the percentage of PRA-positive cells was observed at estrus vs. diestrus (52 ± 3% estrus vs. 51 ± 4% diestrus). At 17–20 wk of age, in all cases, end buds were no longer detected, and the ductal tree had grown to the limits of the mammary fat pad. The percentage of PRA-positive cells decreased significantly to 28 ± 3% (P < 0.05) (Fig. 1). At 7 d of pregnancy, PRA was detected in 25 ± 1% of cells (Fig. 1). However, at 14 days of pregnancy, PRA was detected in only 11 ± 2% of ductal epithelial cells and 6.0 ± 0.3% of alveolar cells (P < 0.001) (Figs. 1 and 2, D and H). No PRA-positive cells were detected during lactation (Fig. 1). After lactational involution, at 9 wk post weaning, PRA was detected in 12 ± 1% of the ductal cells and 10 ± 1% regressed alveolar cells (Fig. 1). Notably, the percentage of PRA-positive cells was significantly lower after pregnancy, compared with age-matched virgin mice, at 17–20 wk of age (P < 0.01) (Fig. 1). Antibody staining of PRA was always localized to the nucleus of epithelial cells and was not detected in myoepithelial cells or stromal cells at any of the developmental stages studied (Fig. 2).
FIG. 1. Quantitation of PRA at different stages of mammary gland development. Immunoperoxidase localization of PRA was carried out using anti-PRA antibody on tissue sections from 3-, 6-, 10- to 12, or 17- to 20-wk-old virgin, 7- or 14-d pregnant, 10-d lactating (Lact) mice, and at 9 wk post weaning [lactational involution (Invol)] as described in Materials and Methods. The values represent the mean ± SEM from three to five mice per group with a minimum of 1000 cells/mouse analyzed. PRA-positive cells decreased significantly with age (3 and 6 wk > 10–12 wk > 17–20 wk) in virgin mice and further during pregnancy and lactation (7 d > 14 d > lactating). The 9-wk involuted mammary gland had fewer PRA-positive cells than age-matched virgin mammary gland (17–20 wk old) (P < 0.01). No PRA staining was detected (ND) during lactation.
FIG. 2. Immunoperoxidase localization of PRA at different stages of mammary gland development. Representative sections from 6-wk-old immature (A and E, end bud; B and F, duct), 12-wk-old adult (C and G, duct), and 14-d pregnant (D and H, alveoli) mouse mammary gland were treated with anti-PRA antibody (A-H) as described in Materials and Methods and control sections without antibody (I, immature end bud; J, immature duct; K, adult duct; L, pregnant alveoli). Higher-magnification images of boxed areas in A–D are shown in E–H. Brown-stained PRA-positive nuclei are indicated by black arrowheads and PRA-negative cells by red arrowheads. End bud cap cells (E) or myoepithelial cells (F and G) are indicated by arrows (scale bar, 50 μm).
Immunoperoxidase localization and quantitation of PRB at different stages of mammary gland development
No PRB-positive cells were detected in 3-, 6-, 10- to 12-, or 17- to 20-wk-old virgin mammary glands (Figs. 3 and 4, A, B, D, and E). During pregnancy, no PRB was detected at 7 d, but PRB was abundantly expressed by 14 d in 48 ± 4% of epithelial cells (Fig. 3). PRB was localized mainly in alveolar cells (Fig. 4, C and F). PRB staining was seen in both the cytoplasm and nucleus of epithelial cells (Fig. 4F). PRB was not detected in myoepithelial cells or stromal cells (Fig. 4F). No PRB was detected in the lactating mammary gland. After lactational involution PRB staining was observed in 6 ± 1% of cells in remaining alveolar structures (Fig. 3); less than 1% of ductal cells were PRB positive.
FIG. 3. Quantitation of PRB at different stages of mammary gland development. Immunofluorescence localization of PRB was carried out using anti-PRB antibody on tissue sections from 3-, 6-, 10- to 12, or 17- to 20-wk-old virgin, 7- or 14-d pregnant, 10-d lactating mice, and at 9 wk post weaning [lactational involution (Invol)] as described in Materials and Methods. The values represent the mean ± SEM from three to five mice per group with a minimum of 1000 cells/mouse analyzed. No PRB staining was detected (ND) in the virgin mammary gland (3, 6, 10–12, or 17–20 wk old), in the 7-d pregnant mammary gland or during lactation. PRB was detected at 14 d of pregnancy and in a smaller percentage of cells in the 9-wk involuted mammary gland (P < 0.001).
FIG. 4. Immunoperoxidase localization of PRB at different stages of mammary gland development. Representative sections from 6-wk-old immature (A and D, end bud), 12-wk-old adult (B and E, duct), and 14-d pregnant (C and F, alveoli) mice were treated with anti-PRB antibody (A–F) as described in Materials and Methods and control sections without antibody (G, immature end bud; H, adult duct; I, pregnant alveoli). Higher-magnification images of boxed areas in A and B are shown in D and E, and a higher magnification of pregnant mammary gland is shown in F. Brown-stained PRB-positive nuclei (F) are indicated by black arrowheads, PRB-positive cytoplasmic staining (F) by double arrows, and PRB-negative nuclei by red arrowheads. End bud cap cells (D) are indicated by arrows (scale bar, 50 μm).
PR isoform specificity of antibodies for immunohistochemistry
Although Mote et al. (20) had already demonstrated PRA and PRB isoform specificity of monoclonal anti-PRA (hPRa7) and anti-PRB (hPRa6) antibodies, respectively, we sought further confirmation using PRAKO mice (15). Figure 5A shows that no staining was detected with the anti-PRA antibody in virgin 8-wk-old PRAKO mice, whereas PRA-positive cells were detected in wild-type 8-wk-old virgin mice. PRAKO mice cannot become pregnant; however, E+P treatment induces pregnancy-like lobuloalveolar development (15). Figure 5A also shows that no PRA staining was detected in E+P-treated PRAKO mice, whereas PRA staining was detected in E+P-treated wild-type mice.
FIG. 5. Immunodetection of PRA and PRB in wild-type vs. PRA null mice. Immunofluorescence localization of PRA (A) or PRB (B) was carried out on sections from 8-wk-old virgin or 13-d E+P-treated 12-wk-old virgin wild-type (WT) and PRAKO mice. Antibody staining was carried out with anti-PRA antibody (red nuclei) or anti-PRB antibody (light blue nuclei); nuclei were counterstained with TOPRO-3 (dark blue nuclei). Positive staining is indicated by white arrowheads and negative nuclei are indicated by yellow arrowheads. Scale bar, 20 μm.
Figure 5B shows that PRB staining was observed in 12-wk-old E+P-treated PRAKO mice with the anti-PRB antibody, and the pattern of staining was the same as seen in wild-type E+P-treated mice. No PRB staining was detected in 8-wk-old virgin PRAKO or wild-type mice (Fig. 5B). These results demonstrate that the anti-PRB antibody detects PRB only in PRAKO mice (under conditions of simulated pregnancy) similar to wild-type mice. Thus, the staining patterns obtained in PRAKO mice confirmed the specificity of the anti-PRA antibody to detect only PRA and the specificity of the anti-PRB antibody to detect only PRB.
Immunoblot analysis of PRA and PRB expression
The apparent absence of PRB in virgin mammary gland, based on immunohistochemistry, was explored further by immunoblot analysis, using an antibody that detects both PRA and PRB isoforms (Fig. 6A). As expected, both PRA and PRB were detected in mouse uterus immunoblot (lane 1), which is known to express both isoforms (28). By contrast, this antibody detected only PRA in virgin mammary gland (Fig. 6A, lane 2). PRB was not detected in virgin mammary gland using antibody specific for PRB (hPRa6) (Fig. 6B, lane 2); the same antibody detected PRB in mouse uterus (Fig. 6B, lane 1). These findings are consistent with our immunohistochemical finding of only PRA in virgin mammary gland and indicate that absence of PRB is not due to epitope masking.
FIG. 6. Immunoblot analysis of PR in mammary gland. A, Extracts from uterus (lane 1) and isolated epithelial cells from 6-wk-old mammary glands (lane 2) were subjected to SDS-PAGE, and blots were probed with hPRa7 anti-PR antibody as described in Materials and Methods. PRA was detected as a single band at 91 kDa in uterus and isolated epithelial cells (lanes 1 and 2); PRB was detected as a single band at 119 kDa in uterus only (lane 1). B, Extracts from uterus (lane 1) and isolated epithelial cells from 6-wk-old mammary glands (lane 2) were subjected to SDS-PAGE, and blots were probed with hPRa6 anti-PR antibody, which detects only PRB, as described in Materials and Methods. PRB was detected as a single band at 119 kDa in uterus (lane 1); no PRB was detected in isolated epithelial cells (lane 2). C, Extracts from uterus (lane 1) and whole 14-d pregnant mammary glands (lane 2) were subjected to SDS-PAGE, and blots were probed with a mixture of sc-538 and sc-539 anti-PR antibodies described in Methods. PRA was detected as a single band at 91 kDa in uterus only (lane 1); PRB was detected as a single band at 119 kDa in uterus and mammary gland (lanes 1 and 2).
The low level of PRA in pregnant mammary gland, based on immunohistochemistry, was explored further by immunoblot analysis using antibody that detects both isoforms (Fig. 6B). Immunoblot analysis showed only a PRB band (Fig. 6B, lane 2), consistent with immunohistochemistry showing a predominance of PRB over PRA. Failure to detect a PRA band is likely due to the low percentage of PRA-positive cells in pregnant mammary gland.
Immunofluorescence colocalization of PRA and PRB
The different patterns of PRA and PRB expression observed during pregnancy suggested that PRA and PRB are present in different cells. To test this hypothesis, we undertook colocalization studies with the anti-PRA and anti-PRB antibodies. Immunofluorescent labeling of virgin and pregnant mammary glands with either anti-PRA or anti-PRB antibody alone yielded the same isoform-specific staining patterns that were obtained by immunoperoxidase detection. However, when we double labeled virgin or pregnant mammary gland with the anti-PRA antibody plus anti-PRB antibody, the PRA- and PRB-specific patterns were not maintained, and all PR-positive cells were positive for both PRA and PRB. To overcome this artifact, we sought to identify another antibody that was specific for only PRA in immunohistochemistry.
The pattern of PRA expression that we observed in the virgin mammary gland with the anti-PRA antibody was similar to antibody staining patterns reported by others who used the sc-538 anti-PR antibody (18). This led us to surmise that the sc-538 antibody might in fact be PRA isoform specific. To directly test this hypothesis, we carried out double-labeling experiments with the anti-PRA antibody plus sc-538 antibody and used immunofluorescence confocal microscopy to investigate colocalization of the antibodies. The results presented in Fig. 7A show complete colocalization of the anti-PRA antibody with the sc-538 anti-PR antibody in the virgin mammary gland. In the 14-d pregnant gland (Fig. 7B) the sc-538 antibody also showed complete colocalization with the anti-PRA antibody and the same low level of expression (relative to the virgin) that was observed with the monoclonal anti-PRA antibody (Fig. 2, D and H). PRA was exclusively localized in the nucleus with the sc-538 antibody in both virgin and pregnant mammary glands. Thus, it appears that the sc-538 antibody is specific for the PRA isoform in immunohistochemistry. The sc-538 has also been shown to be specific for PRA in immunohistochemistry in human cells (20).
Having established the specificity of sc-538 to detect only PRA, we carried out colocalization studies of PRA and PRB in double-labeling experiments with sc-538 and the anti-PRB antibody. At 14 d of pregnancy, three subsets of cells were found: cells that expressed PRA only, PRB only, or both PRA and PRB (Fig. 8, A and B). Forty-three percent of cells were positively labeled for PRB (Fig. 8A). Of the PRA-positive cells (8%), about half were also positive for PRB (Fig. 8A). Thus, colocalization of PRA and PRB occurred in only 4% of cells during pregnancy.
FIG. 8. Colocalization of PRA and PRB in pregnancy. Dual-immunofluorescence detection of PRA and PRB was carried out and visualized by laser scanning confocal microscopy as described in Materials and Methods. A, Quantitation of PRA and PRB colocalization; the values represent the mean ± SEM of the percentage of epithelial cells expressing one isoform only (PRA or PRB) or both isoforms (PRAB); values were obtained using five mice with a minimum of 1000 cells/mouse analyzed. B, Photomicrograph of PRA and PRB colocalization (PRA, green nuclei; PRB, red nuclei); nuclei were counterstained with TOPRO-3 (blue nuclei). Three subsets of PR-positive cells are seen in the merged image: those expressing both isoforms (white nuclei in square), PRA only (green nucleus in circle) or PRB only (red nuclei in oval). Scale bar, 20 μm.
PR isoform expression and colocalization with cyclin D1 or BrdU
A role for P has been implicated in ductal development in the virgin mammary gland (19). PRB and cyclin D1 are required for alveologenesis during pregnancy (16, 29). Epithelial cell proliferation is common to both ductal development and alveologenesis. Having found that PRA and PRB are present in different cells and at different stages of mammary gland development, it was of interest to determine how PR isoform expression was related to proliferation and cyclin D1 expression. To accomplish this, mammary glands were obtained from 14-d pregnant and 6-wk-old mice injected with a pulse of BrdU 2 h before being killed to label cells in S phase. Tissue sections were double labeled with anti-BrdU plus anti-PRA antibody or with anti-BrdU antibody plus anti-PRB antibody. Additional tissue sections were also double labeled with anti-cyclin D1 antibody plus anti-PRA or anti-PRB antibody. Immunofluorescence confocal microscopy was used to determine the colocalization of PRA and/or PRB with BrdU or cyclin D1.
In the pregnant mammary gland, 16% of cells were BrdU positive at 2 h after BrdU injection, and 46% of cells were PRB positive (Fig. 9A). Fifteen percent of cells were BrdU and PRB positive; thus, 95% of BrdU-positive cells were PRB positive (Figs. 9A and 10A). In pregnant mammary gland, PRA and BrdU were not colocalized in the same cells (Figs. 9A and 10B).
FIG. 9. Quantitation of colocalization of PRB or PRA with BrdU in pregnant and virgin mammary glands. Dual-immunofluorescence detection of PRB or PRA and BrdU was carried out on tissue sections from 14-d pregnant (A) and 6-wk-old virgin (B) mammary glands and visualized by laser scanning confocal microscopy as described in Materials and Methods. A minimum of 1000 cells were counted for each antibody combination tested, i.e. PRB and BrdU or PRA and BrdU in pregnant mammary gland and PRA and BrdU in virgin mammary gland. The values represent the mean ± SEM from three to five mice with a minimum of 1000 cells/mouse analyzed.
FIG. 10. Detection of colocalization of PRB or PRA with BrdU in pregnant and virgin mammary glands. Dual-immunofluorescence detection was carried out in 14-d pregnant (A and B) or 6-wk-old virgin (C and D) mammary gland using anti-PRB (A) or anti-PRA (B–D) antibodies and TOPRO-3 nuclear stain and were visualized by laser-scanning confocal microscopy as described in Materials and Methods. A, PRB (red nuclei, white arrows) and BrdU (green nuclei, white arrowheads) staining were extensively colocalized (white nuclei, yellow arrows) in merged images. B, PRA (red nuclei, white arrows) and BrdU (green nuclei, white arrowheads) staining did not colocalize in merged images and were seen as red (white arrows) and light blue (white arrowheads) nuclei. C, In 6-wk-old virgin mammary gland duct, PRA (green nuclei, white arrow), BrdU (red nuclei, white arrowhead) staining did not colocalize in merged images and were seen as red (white arrowheads) and light blue (white arrows) nuclei. D, In 6-wk-old mammary gland end bud, most PRA (green nuclei, white arrow), BrdU (red nuclei, white arrowhead) staining did not colocalize and in merged images and were seen as red (white arrowheads) and light blue (white arrows) nuclei. End bud cap cells were prominently labeled by BrdU (red nuclei, white arrowheads). Instances of colocalization of PRA and BrdU are seen in merged image as white nuclei (yellow arrows). Scale bar, 20 μm.
We also analyzed PRA and BrdU colocalization in the 6-wk-old, virgin mammary gland. We chose this age and stage of development because there is extensive proliferation and a high percentage of PRA-positive cells in the virgin mammary gland. We found that 15% of cells were BrdU positive, 56% of cells were PRA positive, and 4% were PRA and BrdU positive (Fig. 9B). Thus, only 27% of BrdU-positive cells were PRA positive and only 7% of PRA-positive cells were BrdU positive. Most BrdU-positive cells were located in the cap cell layer of end buds (Fig. 10D), which is a region of the end bud that is devoid of PRA-positive cells (Fig. 2, A and E). Fewer BrdU-positive cells were present in ducts (Fig. 10C).
In the pregnant mammary gland, 56% of cells were cyclin D1 positive and 49% were PRB positive (Fig. 11A). Forty-six percent of cells were positive for both PRB and cyclin D1; thus, 83% of cyclin D1-positive cells were PRB positive and 94% of PRB-positive cells were cyclin D1 positive (Fig. 12A). There was no colocalization of PRA with cyclin D1 (Figs. 11A and 12B).
FIG. 11. Quantitation of colocalization of PRB or PRA with cyclin D1 in pregnant and virgin mammary glands. Dual-immunofluorescence detection of PRB or PRA and cyclin D1 was carried out on tissue sections from 14-d pregnant (A) and 6-wk-old virgin (B) mammary glands and visualized by laser-scanning confocal microscopy as described in Materials and Methods. The values represent the mean ± SEM from three mice per group (virgin and pregnant) with a minimum of 1000 cells/mouse analyzed for each antibody combination tested, i.e. PRB and cyclin D1 or PRA and cyclin D1 in pregnant mammary gland and PRA and cyclin D1 in virgin mammary gland.
FIG. 12. Detection of colocalization of PRB or PRA with cyclin D1 in pregnant and virgin mammary glands. Dual-immunofluorescence detection of PRB or PRA and cyclin D1 was carried out on tissue sections from 14-d pregnant (A and B) and 6-wk-old virgin (C and D) mammary glands and visualized by laser-scanning confocal microscopy as described in Materials and Methods. Nuclei were counterstained with TOPRO-3 (A–D, blue). Examples of PRB- (A; red nuclei) or PRA (B–D; red nuclei)-positive cells are indicated with white arrows, and examples of cyclin D1-positive cells (A–C; green nuclei) are indicated with white arrowheads. PRB and cyclin D1 colocalization is seen as white nuclei in the merged image (A), and examples are indicated with yellow arrows. In pregnant mammary gland (B), there was no colocalization of PRA and cyclin D1 in the merged image, and PRA-positive nuclei stain red (white arrows) and cyclin D1-positive nuclei stain light blue (white arrowheads). In virgin mammary gland (C) when colocalization of PRA and cyclin D1 was observed, it was seen as white nuclei in the merged image; examples are indicated with yellow arrows. D, An example of a virgin duct without cyclin D1-positive cells. Scale bar, 20 μm.
In the 6-wk-old virgin mammary gland, the percentage of cyclin D1-positive cells was significantly less than in pregnant mammary gland (18 vs. 56%; P < 0.05) (Fig. 11, A and B). Fifty-four percent of cells were PRA positive, and cyclin D1 and PRA were colocalized in 1% of cells; thus, 4% of cyclin D1-positive cells were also PRA positive and 2% of PRA-positive cells were cyclin D1 positive (Fig. 11B). Figure 12C illustrates colocalization in a duct that is cyclin D1 positive. Many ducts had no cyclin D1-positive cells, yet PRA was highly expressed (Fig. 12D).
Discussion
The results presented in this paper demonstrate that PRA and PRB expression are temporally and spatially separated during murine mammary gland development. Only PRA was highly expressed in the immature and adult virgin mammary gland. By contrast, PRB was seen only during pregnancy, mainly in alveolar epithelial cells. During pregnancy, the majority of PR-positive cells contained only PRB, and colocalization of PRA and PRB occurred in a small proportion of epithelial cells. During pregnancy PRB colocalized extensively with the proliferation marker BrdU and with cyclin D1. In contrast, PRA did not colocalize with BrdU or cyclin D1 during pregnancy and was infrequently colocalized with BrdU or cyclin D1 in the virgin gland. The implication of these findings is that different actions of P are mediated in PRB positive vs. PRA-positive cells in vivo.
P action in the virgin mammary gland: predominant role of PRA
In the 6-wk-old immature virgin gland, whereas 54% of epithelial cells were PRA positive, only 2% of PRA-positive cells were cyclin D1 positive and only 4% of PRA-positive cells were BrdU positive. These results indicate that the majority of PRA-positive cells were not in S phase during our 2-h labeling period. Most BrdU-positive cells were in the cap cell layer of end buds, which is recognized to be a major growth point. The cap cell layer was devoid of PRA-positive cells, supporting the concept that PRA-positive cells do not constitute the major pool of proliferating cells. We cannot rule out the possibility that P may play a role in proliferation via a paracrine mechanism in which PRA-positive cells produce a factor that affects the proliferation of neighboring PR negative cap cells.
Proliferation leading to ductal elongation occurs via cap cell proliferation and is mediated by E and growth factors such as epidermal growth factor, hepatocyte growth factor, and IGF-I (3, 30). The requirement for E is supported by the complete absence of ductal elongation in ER gene-deleted mice (31). In contrast, ductal elongation does occur in total PR gene-deleted mice (14). These results indicate that the presence of PR is not an absolute requirement for ductal elongation in the virgin gland.
Organogenesis during embryonic development results from the net effect of the precise spatial patterning of proliferation and apoptosis. Similarly, postnatal ductal development in the mammary gland can be considered to be the result of spatially organized proliferation and apoptosis. Proliferation occurs in the cap cell layer of the end bud, giving rise to a multilayered internal mass of cells below the cap cell layer (32). Formation of the ductal lumen requires the removal of this internal cell mass. Apoptotic cells have been observed in this internal layer of cells of the end bud (32), suggesting that apoptosis may play a key role in lumen formation in ducts. We have previously shown in vitro that mammary organoids derived from virgin mammary gland respond to the synthetic progestin, R5020, by forming a lumen (30). Treatment of organoids with R5020 induces apoptosis that is spatially localized within mammary organoids and centrally within luminal structures; R5020 does not induce proliferation in these organoids (30). Based on these observations, we have hypothesized that one of the actions of P in mammary gland development is to facilitate lumen formation through P-induced apoptosis (30). In the present study, we showed that only PRA was expressed in the virgin gland, and within end buds PRA-positive cells were localized in the internal layer of cells. This raises the possibility that one way that P promotes ductal development in the virgin gland, at least in part, is by facilitating lumen formation through a proapoptotic mechanism mediated by PRA. The observation that ductal development can occur in total PR deleted as well as PRA gene-deleted mice indicates that there are additional mechanisms that promote lumen formation, and that these mechanisms are operative in PR gene-deleted mice and may compensate for the lack of PR.
P action in pregnancy: predominant role of PRB
PRB-positive cells were seen only in mammary glands of pregnant mice (Figs. 4C and 8B) or alveolar structures of adult E+P-treated mice (Fig. 5B). In pregnant mice PRB was abundantly expressed and the PRB-positive cells were localized mainly in alveolar structures. We found extensive colocalization of PRB with BrdU and cyclin D1 in pregnant mammary gland. This indicates that PRB-positive cells are in the proliferative pool of cells and express cyclin D1. Our results indicate that PRB has the primary role in inducing alveologenesis. Other studies have inferred the same conclusion based on different approaches, namely that there is no defect in alveologenesis in the PRAKO mouse (15), that there is a lack of alveologenesis in the PRB-only gene-deleted mouse (16), and that precocious alveologenesis occurs in PRB overexpressing transgenic mice (13).
In contrast to PRB, there was no PRA colocalization with either BrdU or cyclin D1 in the pregnant gland, suggesting that PRA-positive cells do not constitute the major proliferative pool in the pregnant mammary gland. These observations do not discount the possibility that PRA nevertheless plays a role in pregnancy because expansion of the epithelium and side branching are detected as early as d 7 of pregnancy (Aupperlee, M. D., and S. Z. Haslam, unpublished observations), when 27% of the epithelial cells were PRA positive and none were PRB positive (Figs. 1 and 3).
Previous studies have reported a lack of colocalization of PR with markers of proliferation (16, 33). However, in those studies the PR isoform specificity of the antibody used (Dako A0098) was not identified. We have determined that the Dako A0098 anti-PR antibody colocalizes with PRA and not with PRB. This was determined in studies carried out as shown for the sc-538 anti-PR antibody (Fig. 7) (Aupperlee, M. D., and S. Z. Haslam, unpublished observations). Our own studies using PR isoform-specific antibodies demonstrate a lack of colocalization of PRA with BrdU in pregnant mammary gland, but we find extensive colocalization of PRB with BrdU and cyclin D1 in pregnant mammary gland.
The lack of PRA and PRB staining during lactation is in agreement with previous reports of the absence of specific P ligand binding and lack of detectable PR mRNA in lactating mouse mammary gland (28, 34). Although PRA-positive cells were detected again after involution, the percentage of PRA-positive cells never returned to the prepregnancy virgin level. This was not due to aging because the percent of PRA-positive cells was significantly higher in 20-wk-old virgin mammary glands than age-matched parous mice. A low level of PRB (6% PRB-positive cells) was detected in alveolar structures after lactational involution but not in age-matched virgin mammary gland. These results demonstrate that expression of both PRA and PRB is permanently altered by pregnancy. Pregnancy is protective against carcinogen-induced mammary tumors in mice and rats (35). Our results show two important changes caused by pregnancy: a reduction in PRA-positive cells relative to age-matched virgins and presence of PRB after lactation relative to the virgin state. Further studies to elucidate the specific functional roles of PR isoforms in the mammary gland before, during, and after pregnancy may provide new insights about the mechanism(s) underlying differences in susceptibility to tumorigenesis of virgin vs. parous mice.
PR isoform subcellular localization and progesterone action
Previous studies using cell lines have shown that if expressed in the same cells, PRA and PRB proteins can dimerize and bind to DNA as three different species: AA or BB homodimers or AB heterodimers (9). The specific contribution of each of the dimers to the effects of P may be dependent on the transactivation properties contributed to the complexes by the PRB-specific domain. It has also been reported that PRB transcriptional activity is inhibited by PRA. During pregnancy we found that the vast majority of PRB-positive cells contained only PRB, and only a small percentage of cells (4%) contained both PRA and PRB. Our results indicate that the prevailing situation in the mouse mammary gland is that cells contain AA or BB homodimers and that the potential for AB heterodimer formation is limited to a small number of cells during pregnancy. This suggests that in the mouse, heterodimer formation does not play a major role in progesterone action in the mammary gland.
In our study, PRB was detected primarily in the nucleus and some cells faintly in the cytoplasm (see Fig. 4F). In contrast, using the anti-PRA and sc-538 anti-PR antibodies, PRA was detected only in the nucleus. We cannot rule out the possibility that there may also be a cytoplasmic form of PRA not detected by the PR antibodies we used. PR localization in the mammary gland in both cytoplasm and nucleus has been detected using a PR antibody of unknown PR isoform specificity (18) and in human T47D breast cancer cells overexpressing either PRA or PRB (36). It is conceivable that different anti-PR antibodies may detect epitopes that are exposed on the cytoplasmic, nuclear, or both forms of the receptor.
PR isoform expression in the human vs. mouse
In the human breast, immunohistochemical analysis of PR isoform expression has been carried out on normal tissue from premenopausal cycling women (26). In that study PRA or PRB expression and colocalization were determined by dual immunofluorescence with the same antibodies used herein (26). PRA vs. PRB was expressed at a ratio of 1:1, and patterns of expression were similar. The proportion of PR-positive cells was 10–20% with marked variability throughout a section, with PR positivity in individual ducts or lobules ranging from 0 to 90%. Dual-immunofluorescence studies revealed uniform colocalization of PRA with PRB (26). Our study indicates an interesting difference in PR isoform expression between the mouse and the human mammary gland. In the mouse there is PRB expression only during pregnancy, and colocalization of PRA with PRB occurs in only a small percentage of cells. One possible explanation for this difference may be the predominance of a ductal organization of mammary epithelium in the adult nonpregnant mouse. This is particularly true in BALB/c strain mice, used in our study. In contrast, in the adult nonpregnant human there is a higher ratio of lobules to ducts. Studies of PRB null mice have shown that PRB expression is required for alveologenesis and lobule formation (16). Therefore, PRB expression may be a defining characteristic of mammary lobule formation and/or maintenance and may explain why PRB-positive cells are more abundant in the human breast. In this regard, the maintenance of some alveolar structures in mouse mammary gland after pregnancy may also be due to the continued, albeit reduced, expression of PRB after pregnancy. Analysis of other mouse strains, such as the C3H strain, which develop a more lobular morphology in the virgin state (compared with BALB/c strain) may provide additional insights into the relationship between alveolar morphogenesis and PRB expression. It is also important to note that PR isoform expression in the human has only been studied in the adult premenopausal breast. There is no information on PR isoform expression at other stages of human breast development such as puberty or pregnancy. It remains to be seen what analysis of these other stages may reveal about the pattern of PR isoform expression and/or colocalization in the human breast. Clearly more information is needed about PR isoform expression in the human breast.
Understanding the specific functions of PRA and PRB isoforms in vivo is critical to understanding their respective roles in the normal breast and in the etiology of breast cancer. The spatial and temporal separation of PRA and PRB isoform expression in mouse mammary gland offers a unique opportunity to explore further the specific functions and mechanisms of action PRA vs. PRB in vivo.
Acknowledgments
The authors thank Alexis Drolet, Jeff Leipprandt, Nityanand Sunil, and Yvette Gross for technical support of these studies.
References
Haslam SZ, Osuch JR, Raafat AM, Hofseth LJ 2002 Postmenopausal hormone replacement therapy: effects on normal mammary gland in humans and in a mouse postmenopausal model. J Mammary Gland Biol Neoplasia 7:93–105
Cline JM, Soderqvist G, von Schoultz E, Skoog L, von Schoultz B 1996 Effects of hormone replacement therapy on the mammary gland of surgically postmenopausal cynomolgus macaques. Am J Obstet Gynecol 174:93–100
Fendrick JL, Raafat AM, Haslam SZ 1998 Mammary gland growth and development from the postnatal period to postmenopause: ovarian steroid receptor ontogeny and regulation in the mouse. J Mammary Gland Biol Neoplasia 3:7–22
Hofseth LJ, Raafat AM, Osuch JR, Pathak DR, Slomski CA, Haslam SZ 1999 Hormone replacement therapy with estrogen or estrogen plus medroxyprogesterone acetate is associated with increased epithelial proliferation in the normal postmenopausal breast. J Clin Endocrinol Metab 84:4559–4565
Magnusson C, Persson I, Adami HO 2000 More about: effect of hormone replacement therapy on breast cancer risk: estrogen versus estrogen plus progestin. J Natl Cancer Inst 92:1183–1184
Ross RK, Paganini-Hill A, Wan PC, Pike MC 2000 Effect of hormone replacement therapy on breast cancer risk: estrogen versus estrogen plus progestin. J Natl Cancer Inst 92:328–332
Schairer C, Lubin J, Troisi R, Sturgeon S, Brinton L, Hoover R 2000 Menopausal estrogen and estrogen-progestin replacement therapy and breast cancer risk. JAMA 283:485–491
Writing Group for the Women’s Health Initiative Investigators 2002 Risks and benefits of estrogen plus progestin in healthy postmenopausal women. JAMA 288:321–333
Lydon JP, Sivaraman L, Conneely OM 2000 A reappraisal of progesterone action in the mammary gland. J Mammary Gland Biol Neoplasia 5:325–338
Kastner P, Krust A, Turcotte B, Stropp U, Tora L, Gronemeyer H, Chambon P 1990 Two distinct estrogen-regulated promoters generate transcripts encoding the two functionally different human progesterone receptor forms A and B. EMBO J 9:1603–1614
Kraus WL, Montano MM, Katzenellenbogen BS 1993 Cloning of the rat progesterone receptor gene 5'-region and identification of two functionally distinct promoters. Mol Endocrinol 7:1603–1616
Shyamala G, Yang X, Silberstein G, Barcellos-Hoff MH, Dale E 1998 Transgenic mice carrying an imbalance in the native ratio of A to B forms of progesterone receptor exhibit developmental abnormalities in mammary glands. Proc Natl Acad Sci USA 95:696–701
Shyamala G, Yang X, Cardiff RD, Dale E 2000 Impact of progesterone receptor on cell-fate decisions during mammary gland development. Proc Natl Acad Sci USA 97:3044–3049
Lydon JP, DeMayo FJ, Funk CR, Mani SK, Hughes AR, Montgomery Jr CA, Shyamala G, Conneely OM, O’Malley BW 1995 Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev 9:2266–2278
Mulac-Jericevic B, Mullinax RA, DeMayo FJ, Lydon JP, Conneely OM 2000 Subgroup of reproductive functions of progesterone mediated by progesterone receptor-B isoform. Science 289:1751–1754
Mulac-Jericevic B, Lydon JP, DeMayo FJ, Conneely OM 2003 Defective mammary gland morphogenesis in mice lacking the progesterone receptor B isoform. Proc Natl Acad Sci USA 100:9744–9749
Takimoto GS, Tung L, Abdel-Hafiz H, Abel MG, Sartorius CA, Richer JK, Jacobsen BM, Bain DL, Horwitz KB 2003 Functional properties of the N-terminal region of progesterone receptors and their mechanistic relationship to structure. J Steroid Biochem Mol Biol 85:209–219
Silberstein GB, Van Horn K, Shyamala G, Daniel CW 1996 Progesterone receptors in the mouse mammary duct: distribution and developmental regulation. Cell Growth Differ 7:945–952
Atwood CS, Hovey RC, Glover JP, Chepko G, Ginsburg E, Robison WG, Vonderhaar BK 2000 Progesterone induces side-branching of the ductal epithelium in the mammary glands of peripubertal mice. J Endocrinol 167:39–52
Mote PA, Johnston JF, Manninen T, Tuohimaa P, Clarke CL 2001 Detection of progesterone receptor forms A and B by immunohistochemical analysis. J Clin Pathol 54:624–630
Gava N, Clarke CL, Byth K, Arnett-Mansfield RL, deFazio A 2004 Expression of progesterone receptors A and B in the mouse ovary during the estrous cycle. Endocrinology 145:3487–3494
Haslam SZ 1988 Progesterone effects on deoxyribonucleic acid synthesis in normal mouse mammary glands. Endocrinology 122:464–470
Haslam SZ 1989 The ontogeny of mouse mammary gland responsiveness to ovarian steroid hormones. Endocrinology 125:2766–2772
Haslam SZ, Counterman LJ, St. John AR 1993 Hormonal basis for acquisition of estrogen-dependent progesterone receptors in the normal mouse mammary gland. Steroid Biochem (Life Sci Adv) 12:27–34
Mote PA, Balleine RL, McGowan EM, Clarke CL 1999 Colocalization of progesterone receptors A and B by dual immunofluorescent histochemistry in human endometrium during the menstrual cycle. J Clin Endocrinol Metab 84:2963–2971
Mote PA, Bartow S, Tran N, Clarke CL 2002 Loss of co-ordinate expression of progesterone receptors A and B is an early event in breast carcinogenesis. Breast Cancer Res Treat 72:163–172
Haslam SZ, Levely ML 1985 Estrogen responsiveness of normal mouse mammary cells in primary cell culture: association of mammary fibroblasts with estrogenic regulation of progesterone receptors. Endocrinology 116:1835–1844
Shyamala G, Schneider W, Schott D 1990 Developmental regulation of murine mammary progesterone receptor gene expression. Endocrinology 126:2882–2889
Sicinski P, Weinberg RA 1997 A specific role for cyclin D1 in mammary gland development. J Mammary Gland Biol Neoplasia 2:335–342
Sunil N, Bennett JM, Haslam SZ 2002 Hepatocyte growth factor is required for progestin-induced epithelial cell proliferation and alveolar-like morphogenesis in serum-free culture of normal mammary epithelial cells. Endocrinology 143:2953–2960
Bocchinfuso WP, Korach KS 1997 Mammary gland development and tumorigenesis in estrogen receptor knockout mice. J Mammary Gland Biol Neoplasia 2:323–334
Hovey RC, Trott JF, Vonderhaar BK 2002 Establishing a framework for the functional mammary gland: from endocrinology to morphology. J Mammary Gland Biol Neoplasia 7:17–38
Seagroves TN, Lydon JP, Hovey RC, Vonderhaar BK, Rosen JM 2000 C/EBP? (CCAAT/enhancer binding protein) controls cell fate determination during mammary gland development. Mol Endocrinol 14:359–368
Haslam SZ, Shyamala G 1979 Effect of oestradiol on progesterone receptors in normal mammary glands and its relationship with lactation. Biochem J 182:127–131
Medina D 2004 Breast cancer: the protective effect of pregnancy. Clin Cancer Res 10:380S–384S
Jacobsen BM, Schittone SA, Richer JK, Horwitz KB 2005 Progesterone-independent effects of human progesterone receptors (PRs) in estrogen receptor-positive breast cancer: PR isoform-specific gene regulation and tumor biology. Mol Endocrinol 19:574–587(Mark D. Aupperlee, Kyle T)