| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cell and Tumor Biology |
1 Division of Endocrinology, Department of Medicine and 2 Department of Pathology, University of Colorado Health Sciences Center, Aurora, Colorado
Requests for reprints: Carol A. Sartorius, Division of Endocrinology, Department of Medicine, University of Colorado Health Sciences Center, 12801 East 17th Avenue MS8106, Aurora, CO 80045-7163. Phone: 303-724-3941; Fax: 303-724-3920; E-mail: Carol.Sartorius{at}uchsc.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ERß for estrogens and PRA and PRB for progesterone—that subserve different, at times, opposing functions depending on the target in question (1–4). Considerable evidence shows that the two hormones counter-regulate one another. For example, estrogen induces PR, whereas progesterone suppresses ER activity at least in some organs (5, 6). These complementary actions of estrogen and progesterone have often made it difficult to assign specific functions to each hormone. In the normal breast, growth of the mammary epithelium is maximal during the luteal phase of the menstrual cycle, which has led to the hypothesis that progesterone is involved in proliferation. Indeed, progesterone given in conjunction with estrogen as part of hormone replacement therapy in both women and monkeys increases the proliferative index of breast epithelium above that of estrogen alone (7, 8). However, other studies on estrogen plus progesterone, its effects in normal breast cells from women (9), and on normal human mammary epithelial cells both in vitro (10) and implanted into nude mice (11, 12) have concluded that progesterone alone cannot stimulate growth and either has no effect or inhibits estrogen-induced proliferation. Studies in PR knockout mice show that PR has both proliferative and differentiative functions in the murine mammary gland (13). In summary, progesterone can be stimulatory, inhibitory, or differentiative depending on the context of the experiment. The presence of other growth factors during these phases may help explain the various effects of progesterone in the normal breast.
Other studies address the role of progesterone in tumorigenesis. The seminal studies of Huggins et al. (14) underscore the complex actions of progesterone in carcinogen-induced mammary tumor development. Exogenous progesterone given before a carcinogen or the physiologically high progesterone levels of pregnancy are both protective (reviewed in ref. 15), whereas progesterone given after carcinogen exposure exacerbates tumor formation. Thus, progesterone has either inhibitory or stimulatory effects on breast cancer formation depending on dose and timing. The possible deleterious effects of progesterone recently resurfaced on demonstration in the Women's Health Initiative studies of hormone replacement therapy at menopause (16) that increased risk of breast cancer is associated with an estrogen plus progesterone regimen but not estrogen alone.
Also contentious is the effect of progesterone on proliferation of established breast cancers. The majority of primary human tumors are "hormone dependent" by virtue of the fact that they are ER+ and/or PR+ (17). It is assumed that such tumors proliferate in response to estrogen; indeed, antiestrogen or estrogen suppression therapies are highly successful (18). However, the proliferative effects of progesterone, if any, in established breast cancers remain unclear. Long-term in vivo studies of progesterone effects in tumor models are sparse. Progesterone treatment inhibits proliferation of established rat mammary tumors (19, 20) and PR-transfected MDA-231 (ER–PR–) human breast cancer cells grown into tumors in severe combined immunodeficient mice (21). On the other hand, Michna et al. (22) report inhibitory effects of antiprogestins, in which case progesterone would be growth stimulatory.
We have developed in vivo human breast cancer models in which to test the effects of progesterone either alone or in the presence of estrogen. This is done in ER+ T47D cells that vary in their PR content (PR–, PRA+, PRB+, or PRA+PRB+; ref. 23). This is of interest because the equimolar PRA/PRB ratios of the normal mammary gland change during malignant progression so that >70% of invasive breast cancers express excessive levels of one or the other PR isoform (24, 25). This in turn influences hormone responsiveness (26). We show that regardless of whether the tumors are PR– or PR+ for either form their growth is unaffected by progesterone or medroxyprogesterone acetate (MPA) alone or together with estrogen. However, long-term treatment with progesterone or MPA leads to the appearance of cell subpopulations expressing the myoepithelial/basal markers cytokeratins (CK) 5 and 6. T47D cells originated from the pleural effusion of an ER+/PR+ luminal epithelial ductal cancer (27). Such tumors are marked by expression of glandular or secretory (luminal) intermediate filament CK8, CK18, and CK19 (28). Expression of these luminal CKs was decreased in progestin-treated tumors. Therefore, steroid hormones can influence the differentiation state of proliferating tumors as measured by their CK expression profile. In the case of myoepithelial CK5 and CK6, tumors expressing these markers are characterized by poor prognosis (29–31).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Experimental animals and xenograft tumor growth. All animal procedures were done under a protocol approved by the University of Colorado Institutional Animal Care and Use Committee. Ovariectomized female athymic nu/nu mice were obtained from Harlan Sprague-Dawley (Indianapolis, IN) at 5 to 6 weeks of age. Inoculation of T47D breast cancer cells into animals is essentially as described (32). Briefly, anesthetized animals were injected with 5 x 106 cells suspended in 50% Matrigel (Becton Dickinson, Bedford, MA) in MEM (Life Technologies). For estrogen plus progesterone or estrogen plus MPA experiments, progesterone or MPA were added to estrogen either at the start of the experiment or after 4 weeks of estrogen alone. Two tumors per animals were grown on the left and right flanks. All animals were implanted with silastic pellets containing hormone mixtures (described in ref. 32). These included placebo (10 mg cellulose), estrogen (2 mg 17ß-estradiol + 8 mg cellulose), progesterone (10 mg progesterone), or MPA (10 mg 6-methyl-17-hydroxyprogesterone acetate). Growth of tumors was monitored weekly for 8 weeks by measuring the length (l) and width (w) with a digital caliper, with l being the smaller measurement. Tumor volume was estimated by the formula l2w/2. At termination of the experiment, mice were euthanized by CO2 asphyxiation, and tumors were excised and weighed. Tumors that were overly necrotic were omitted from the study. In some cases, uteri (wet weight) and mammary glands (whole mount stains) were removed.
Serum levels of progesterone. Whole blood was collected from animals by aortic puncture at time of euthanasia and fractionated by centrifugation at 9,000 x g for 5 minutes at 4°C. Serum supernatants were removed and stored at –20°C. Serum from two animals were combined for each assay. Progesterone concentrations (ng/mL) in sera were determined by chemiluminescence microparticle immunoassay (Abbott Laboratories, North Chicago, IL) at the University of Colorado Hospital.
Mammary gland whole mounts. Whole mammary glands (no. 4) were removed from select animals at euthanasia and spread onto glass slides. Tissue was dried for 5 minutes before fixation in Carnoy's fixative (60% ethanol, 30% chloroform, 10% acetic acid) for 24 hours. After gradual rehydration for 1 hour each in 70% ethanol, 50% ethanol, and distilled water, slides were stained with carmine aluminum (0.2% carmine and 0.5% aluminum potassium sulfate) overnight. Slides were washed in 70%, 95%, and 100% ethanol for 1 hour each then cleared with xylene for 1 hour before mounting with Permount (Fisher, Fair Lawn, NJ). Photographs were taken at x 4 magnification.
5-Bromo-2'-deoxyuridine incorporation. Animals were injected i.p. with 5-bromo-2'-deoxyuridine (BrdUrd; Sigma, St. Louis, MO) at a concentration of 50 mg/kg in sterile PBS 2 hours before euthanasia. Tumors were removed and fixed in formalin, processed, and paraffin embedded. Sections (5 µm) taken from the middle portion of each tumor were stained with an anti-BrdUrd antibody (BD Biosciences, San Jose, CA) and counterstained with hematoxylin. Five random fields at high magnification (x40) from each tumor were counted for BrdUrd-positive cells in a blinded fashion. Small intestines were stained as a positive control with epithelial cells positive for BrdUrd.
Gene expression analysis. T47DPRA and T47DPRB tumors grown bilaterally on the same experimental animals were removed for gene expression analysis. Triplicate samples were obtained from three animals each treated with estrogen plus placebo, estrogen plus progesterone, or estrogen plus MPA. Total RNA was prepared from tumors using TRIzol reagent according to the manufacturer's instructions (Invitrogen, San Diego, CA). Polyadenylated RNA was prepared from total RNA and processed for hybridization to HG-U95Av2 arrays according to a detailed protocol by Affymetrix (Santa Clara, CA). Expression profiling was done by the University of Colorado Cancer Center Microarray Core Laboratory. Data were analyzed using Microarray Suite (Affymetrix) and GeneSpring software (Silicon Genetics, Redwood City, CA). Data were normalized for each chip, and statistical significance among expression levels in estrogen, estrogen plus progesterone, and estrogen plus MPA tumors was determined by one-way ANOVA followed by a Tukey post-test using a cutoff value of P < 0.05 (described in detail in ref. 33). Fold changes were determined by dividing the normalized intensities for the estrogen plus progesterone and estrogen plus MPA sets by the intensities in the estrogen set.
Immunohistochemistry and immunofluorescence. Whole tumors were removed from animals and fixed in 10% buffered formalin. Tissue was processed, paraffin embedded, and cut into 5-µm sections. After high-temperature antigen retrieval in citrate buffer, sections were stained with monoclonal antibodies specific for CK5 or CK8/CK18 (both from Novocastra, United Kingdom). Sections were counterstained with hematoxylin and mounted. Representative photographs were taken under a light microscope at x 40 magnification. For dual immunofluorescence, samples were processed as above and then stained simultaneously with a mouse monoclonal antibody specific for CK5 (Novocastra) and rabbit polyclonal antibodies to either CK18 (Calbiochem, La Jolla, CA) or human PR (DAKO, Carpinteria, CA). Sections were stained with secondary antibodies that fluoresce green (goat anti-mouse Alexa 488) and red (goat anti-rabbit Alexa 555; both from Molecular Probes, Eugene, OR). Nuclei were stained with 1 µg/mL 4',6-diamidino-2-phenylindole (DAPI) in methanol for 15 minutes at room temperature. Sections were mounted with Gelmount (Biomeda, Foster City, CA) and photographs were taken under x 100 magnification for the same field using the UV, FITC, and TRITC filters. Images were merged using ImagePro software (Media Cybernetics, Silver Springs, MD).
Statistical analyses. Statistics were done using GraphPad software (San Diego, CA) and statistical significance was determined by comparison of three or more groups using one-way ANOVA followed by a Tukey post-test. P < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Effects of hormones on the physiology of experimental animals. To confirm that progesterone and MPA were being delivered efficiently, the mice were monitored for several physiologic side effects of estrogen and progestin exposure. Figure 1A shows the circulating levels of progesterone in blood obtained by aortic puncture at the time of euthanasia. Serum fractions from two mice were combined per measurement. Control mice (cycling) had an average progesterone level of 4.7 ng/mL. Ovariectomy reduced this to less than half or 2.0 ng/mL. Addition of progesterone pellets for 8 weeks led to an 
4-fold increase in progesterone levels over controls to 20.6 ng/mL. Mice treated with estrogen for 8 weeks (plus addition of placebo for the last 4 weeks) had low levels of progesterone (3.1 ng/mL). In the estrogen plus progesterone group, in which animals received progesterone for the last 4 weeks, circulating progesterone levels were 22.1 ng/mL. These levels represent a high physiologic dose of the hormone. MPA levels in sera are not reported due to lack of a commercially available assay. However, the structural resemblance of MPA to progesterone, plus the ensuing physiologic data (see below), indicate that MPA is present at least at levels similar to progesterone. Circulating levels of estrogen (17ß-estradiol) were similar to those reported previously (32) with blood levels after 8 weeks of estrogen pellet implantation of 145 ± 47 pg/mL compared with 35 ± 9 pg/mL for ovariectomized controls (n = 4 each). Animals in the estrogen plus progesterone and estrogen plus MPA groups had average estrogen levels of 126 ± 37 and 133 ± 33 ng/mL, respectively (n = 4 each).
|
6 weeks of age ranged from 20 to 22 g. Ovariectomized mice 7 to 10 days postsurgery were slightly larger than intact age-matched controls, but these differences were not significant. After 8 weeks, intact female nude mice weighed 23.7 g on average compared with 30.0 g for ovariectomized mice. Progesterone and MPA alone maintained the higher body mass of ovariectomized mice. However, estrogen treatment suppressed growth of ovariectomized mice. The lower mass of estrogen-only mice (23.6 g) was partially reversed by estrogen plus progesterone or estrogen plus MPA after only 4 weeks of the progestins to 28.5 and 28.3 g, respectively. Hormone treatment also affected uterine wet weight (Fig. 1C). At the time of dissection, uteri from intact animals (including ovaries) weighed 74.0 mg, which was reduced to 23.2 mg by ovariectomy. Progestins alone increased uterine mass to 29.5 mg (progesterone) and 40.1 mg (MPA), which was only statistically significant for MPA (P < 0.01). Estrogen treatment led to extensive hypertrophy (126.0 mg), which was partially reduced by estrogen plus progesterone (109.3 mg) or estrogen plus MPA (95.2 mg). Whole mammary gland ductal branching and lobuloalveolar development at the time of euthanasia is shown in Fig. 1D. Ovariectomized animals have underdeveloped mammary glands. Estrogen alone minimally increases ductal branching, whereas progesterone or MPA alone had no noticeable effect. However, estrogen plus progesterone or estrogen plus MPA led to extensive branching and lobuloalveolar development, indicating that the levels of circulating progesterone and MPA achieved in mice by the implants were physiologic.
Progestins alone do not promote tumor growth. Ovariectomized mice bearing bilateral T47DPR– and T47Dwt or T47DPRA and T47DPRB tumors were implanted with pellets containing placebo (10 mg cellulose), progesterone (10 mg), or estrogen (2 mg) and tumor volumes were measured weekly for 8 weeks (Fig. 2A). Cells inoculated into ovariectomized mice implanted with placebo pellets remain viable but do not grow, with final average volumes ranging from 80 to 130 mm3. Importantly, neither progesterone nor MPA alone promoted tumor growth regardless of the PR content of the cells. In contrast, estrogen alone led to rapid growth with maximum average tumor volumes at 8 weeks ranging from 250 to 600 mm3 depending on tumor type. Note the smaller average size of PRA+ tumors (T47DPRA and T47Dwt; discussed in ref. 32).
|
Progestins do not alter estrogen-dependent tumor growth. We next tested whether progesterone or MPA modify estrogen-dependent tumor growth in vivo (Fig. 3). T47DPR– and T47Dwt or T47DPRA and T47DPRB cells were implanted on opposite flanks of the same mice and tumors were established for 4 weeks under estrogen stimulation. A second pellet containing placebo, progesterone, or MPA (10 mg each) was then implanted (arrows), while estrogen was continued, and tumors were grown for an additional 4 weeks. Changes in tumor volume per week are shown in Fig. 3A. T47DPR– tumors grew to an average final volume of 400 mm3 and were unaffected by 4 weeks of progesterone or MPA treatment. T47DPRB tumors grew to an average final volume of 667 mm3, which was increased 20% by progesterone or MPA. Estrogen-treated T47DPRA tumors grew to 217 mm3, which decreased by 10% to 20% when progesterone or MPA were added. Although these subtle differences in growth were reproducible in separate experiments, none reached statistical significance. Estrogen-dependent growth of T47Dwt (193 mm3) was not significantly altered by estrogen plus progesterone or estrogen plus MPA.
|
To assess proliferation rates, several mice bearing T47D tumors treated with estrogen, estrogen plus progesterone, or estrogen plus MPA described above were injected with BrdUrd 2 hours before sacrifice. Sections were stained with an anti-BrdUrd antibody, and five fields from separate areas of each tumor were photographed and scored for the number of BrdUrd-positive cells. Figure 3C shows representative sections from estrogen and estrogen plus MPA–treated tumors (T47DPR–, T47DPRA, T47DPRB, and T47Dwt) used for BrdUrd analysis. Mean positive cells per field ranged from 61 to 90 as indicated, with no significant differences noted between tumors from estrogen and estrogen plus MPA–treated animals. T47DPRA and T47DPRB tumors treated with estrogen plus progesterone had 89 ± 13 and 81 ± 11 BrdUrd-positive cells per field, respectively, which also do not differ from estrogen alone (data not shown).
Progestins increase expression of myoepithelial cytokeratins in a subset of progesterone receptor–positive tumor cells. To investigate other potential physiologic changes in progestin-treated tumors, we determined gene expression profiles in three pairs of T47DPRA and T47DPRB tumors treated with estrogen, estrogen plus progesterone, or estrogen plus MPA using Affymetrix HuFL-U95Av2 arrays. Data were normalized, comparisons among hormone treatments were made, and genes were identified that were significantly changed at least 2-fold (up or down) by the progestins compared with the estrogen control. Genes with the highest fold regulation by progestins in both T47DPRA and T47DPRB tumors included CK5 and CK6-markers of myoepithelial cells. We then assessed gene expression levels of other CK family members on the array.
Figure 4A shows normalized expression levels of transcripts for the myoepithelial CK5 and CK6 in T47DPRA or T47DPRB tumors treated with estrogen, estrogen plus progesterone, and estrogen plus MPA. Whereas expression of CK5 and CK6 was nearly absent in estrogen-treated tumors, their levels were increased substantially by both progesterone and MPA (8.5- and 16.5-fold for CK5 and 13.7 and 30.7-fold for CK6 in T47DPRA; 4.5- and 5.5-fold for CK5 and 9.9- and 18.1-fold for CK6 in T47DPRB). Figure 4B shows expression levels of the luminal epithelial CK8, CK18, and CK19. Transcripts for these CKs were highly expressed in estrogen-treated T47DPRA and T47DPRB tumors. Expression was decreased by progestins 1.3- to 2-fold (significant only for estrogen plus MPA at 1.5-fold cutoff; P < 0.05). Of the 20 known epithelial CKs, transcripts for CK10 and CK17 were the only others called "present" in the estrogen-treated tumors, and their levels (10% that of CK18) were not significantly altered by progesterone or MPA. CK7 (a luminal cell marker) and CK14 (a myoepithelial cell marker) were not represented on the array.
|
5% to 10% of tumor cells. CK6 showed a similar pattern of expression (data not shown). Virtually all tumor cells stain positive for the luminal epithelial CK8/CK18 regardless of tumor type or hormone treatment (Fig. 4D). Tumors cells expressing cytokeratin 5 contain less cytokeratin 18. To specifically determine if individual CK5+ cells have altered expression of luminal CKs (not detectable by peroxidase staining), dual immunofluorescence studies were done. Sections from estrogen plus MPA–treated T47DPRA tumors were dually stained with antibodies to CK5 (monoclonal) and CK18 (polyclonal). Sections were then stained with specific green and red fluorescing secondary antibodies, and after antibody incubations, nuclei were stained with DAPI. Figure 5A shows a representative field of a T47DPRA tumor at high magnification containing a cluster of CK5+ cells (green). CK18+ cells in the same field are shown in red. Most cells positive for CK5 have lost expression of CK18 (arrows) compared with neighboring CK5–/CK18+ cells. Images were merged with or without the nuclear stain DAPI. As indicated in the merged image (arrow), there were always a few CK5+ cells in each cluster that retained some expression of CK18. This same pattern of CK5/CK18 expression occurs in T47DPRB tumors. Dual immunofluorescence was also done with antibodies to CK5 and PR. A representative image of an estrogen plus MPA–treated T47DPRB tumor is shown in Fig. 5B. Individual CK5+ cells are shown in green and PR in red. Merged images (±DAPI) show nuclear PR staining in the same cells with cytoplasmic CK5 staining. Results were identical in T47DPRA tumors.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Despite the physiologic rationale, progesterone therapy is not widely used for breast cancer and we therefore sought to address in detail the ability of progesterone to suppress the proliferative effects of estrogen. We now report that neither 4 nor 8 weeks of progesterone or MPA treatment was able to substantially alter estrogen-dependent proliferation of our ER+PR+ human tumor xenograft models. The circulating hormone concentrations achieved by implants in our experimental animals represent a high physiologic dose (20 ng/mL or 64 nmol/L) and have a marked effect on the physiology of target organs in these mice. We therefore sought to address other aspects of tumor biology that progestins might affect.
Progestins initiate a luminal to myoepithelial switch in a subset of tumor cells. Tumor cells were originally thought to maintain the differentiation properties of their normal cell of origin. However, studies now show that tumor cells are not rigidly programmed to a single differentiation state but rather that they are capable of undergoing morphologic transitions over time. These include luminal-myoepithelial, epithelial-mesenchymal, and mesenchymal-epithelial transitions (40–42). The theory that epithelial cancers arise from progenitor stem cells also argues in favor of tumor cell plasticity (43). It is postulated that, in the breast, luminal cells can give rise to myoepithelial cells but that the reverse cannot occur, suggesting a linear order of differentiation (40).
Intermediate filament CKs are often used as markers to determine the epithelial origin and differentiation state of tumors. These consist of 20 proteins, both the acidic type I and their neutral-basic type II partners (44). In the human breast, luminal epithelial cells are distinguished by expression of CK8, CK18, CK19, and sometimes CK7, whereas myoepithelial cells express CK5, CK6, CK14, and sometimes CK17 (28). Several studies have correlated the pattern of CK5/CK6 expression in clinical cases of ductal breast carcinoma with tumor grade and prognosis (29–31). In general, >95% of all tumors expressed luminal epithelial CK markers. Within these luminal CK+ tumors, 16% to 27% also contained cells positive for myoepithelial CKs (called "mixed" or "bimodal"). Cancers expressing purely myoepithelial CKs or no CKs were rare (<1%). Malzahn et al. found that expression of myoepithelial CKs was linked to poorly differentiated grade 3 cancers and shorter overall and disease-free survival (29). van de Rijn et al. evaluated >600 tumors and found that expression of CK5/CK6 and/or CK17 was linked to poor clinical outcome (30). In a study of >1,900 cases, Abd El-Rehim et al. found that CK5/CK6 expression in mixed luminal/myoepithelial breast cancers was associated with shorter disease-free interval and ER negativity (31). Conversely, expression of luminal CKs is linked to a more favorable prognosis (31, 45).
We now report that progestin treatment leads to a marked increase in a subpopulation of cells expressing myoepithelial CK5 and CK6. This is compared with estrogen alone, in which virtually all tumor cells are CK5–/CK6–. Five percent to 15% of cells exhibited this response to progesterone depending on the individual tumor. Microarray and dual-immunofluorescence studies confirmed that many of these CK5+ cells are negative for or express reduced levels of the luminal CK8, CK18, and CK19. Interestingly, we could not detect smooth muscle actin (SMA) expression, a marker of terminally differentiated myoepithelial cells, in CK5+ tumor cells (data not shown). Bocker et al. have described such CK5+/CK6+ cells that are negative for luminal CKs and SMA (46). They propose that these are progenitor cells that may later become fully differentiated luminal (CK18+) or myoepithelial (SMA+) cells. Cells containing dual expression of CK5 and CK18 have also been described in both mammary and prostate epithelia and are postulated to be in an "intermediate" state of differentiation (47, 48). Thus, CK5+ cells in our tumors may be myoepithelial precursor cells, which have lost expression of CK18 but which lack expression of the terminal marker SMA. Because phenotypic transitions in tumors are postulated to be long-term processes (40), longer progesterone treatment times or additional exogenous factors may be required to induce complete transition to the myoepithelial state.
Implications. We conclude that progesterone induces a subpopulation of cells within the tumor to gradually switch to a myoepithelial differentiation state. This occurs without any significant changes in overall tumor growth and in a PR-dependent manner. It remains to be determined whether progesterone targets unique cells within the tumor directly, which then expand clonally by cell division, or whether random cells within the tumor exhibit heterogeneous sensitivity progesterone. In this regard, it is interesting that a very minor cell fraction (<0.1%) constitutively expresses CK5. The expression of significant amounts of CK5/CK6 has certainly been correlated in primary human breast cancers with higher grade and worse prognosis (29–31). On the other hand, a few investigators hypothesize that some luminal cells in the normal breast express CK5/CK6, which is then lost during carcinogenesis (49, 50). In this regard, progesterone could be conferring a more "normal" morphology on tumor cells. It remains to be seen whether the differentiation state of tumor cells can be altered by exogenous factors in women; our data show, however, that such a phenotypic switch can be hormonally driven.
|
Acknowledgments |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Fatima Nawaz for technical assistance.
|
Footnotes |
|---|
Received 2/14/05. Revised 6/24/05. Accepted 8/17/05.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ERß. Mol Interv 2003;3:281–92. and -ß in reproductive tissues. Reproduction 2003;125:143–9.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  K. B. Horwitz Commentary: The Year in Basic Science: Update of Estrogen Plus Progestin Therapy for Menopausal Hormone Replacement Implicating Stem Cells in the Increased Breast Cancer Risk Mol. Endocrinol., December 1, 2008; 22(12): 2743 - 2750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B. Horwitz and C. A. Sartorius Progestins in Hormone Replacement Therapies Reactivate Cancer Stem Cells in Women with Preexisting Breast Cancers: A Hypothesis J. Clin. Endocrinol. Metab., September 1, 2008; 93(9): 3295 - 3298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B. Horwitz, W. W. Dye, J. C. Harrell, P. Kabos, and C. A. Sartorius Rare steroid receptor-negative basal-like tumorigenic cells in luminal subtype human breast cancer xenografts PNAS, April 15, 2008; 105(15): 5774 - 5779. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Liang, C. Besch-Williford, R. A. Brekken, and S. M. Hyder Progestin-Dependent Progression of Human Breast Tumor Xenografts: A Novel Model for Evaluating Antitumor Therapeutics Cancer Res., October 15, 2007; 67(20): 9929 - 9936. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kariagina, M. D. Aupperlee, and S. Z. Haslam Progesterone Receptor Isoforms and Proliferation in the Rat Mammary Gland during Development Endocrinology, June 1, 2007; 148(6): 2723 - 2736. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. M. Jacobsen, J. C. Harrell, P. Jedlicka, V. F. Borges, M. Varella-Garcia, and K. B. Horwitz Spontaneous Fusion with, and Transformation of Mouse Stroma by, Malignant Human Breast Cancer Epithelium Cancer Res., August 15, 2006; 66(16): 8274 - 8279. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. E. Harvell, J. K. Richer, D. C. Allred, C. A. Sartorius, and K. B. Horwitz Estradiol Regulates Different Genes in Human Breast Tumor Xenografts Compared with the Identical Cells in Culture Endocrinology, February 1, 2006; 147(2): 700 - 713. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, P < 0.01, final body mass significantly was smaller than ovariectomized, progesterone, MPA, estrogen + progesterone, and estrogen + MPA. C, uteri were removed and weighed from experimental animals at necropsy. Individual masses are depicted for normal (n = 9), ovariectomized (n = 8), progesterone (n = 8), MPA (n = 10), estrogen + placebo (n = 29), estrogen + progesterone (n = 31), and estrogen + MPA (n = 28) animals. Bar, mean uterine mass for each group. *, P < 0.01, MPA uteri are significantly larger than ovariectomized uteri; 
), progesterone only (
), or estrogen only (
) and tumor growth was monitored for 8 weeks. Points, tumor volume (mm3) for each cell line; bars, SE. Sample numbers (n) for the placebo, progesterone, and estrogen groups were 12, 12, and 10 for T47DPR– (T47DPRneg); 14, 11, and 10 for T47DPRB; 14, 11, and 10 for T47DPRA; and 13, 14, and 10 for T47Dwt, respectively. *, P < 0.001, after 8 weeks of growth, estrogen-treated tumors are significantly larger than placebo or progesterone tumors for each tumor type. B, mean individual tumor masses for T47DPR–, T47DPRB, T47DPRA, or T47Dwt treated with placebo, progesterone, or estrogen. Bar, mean. Note that scales are different for larger tumors (T47DPR– and T47DPRB) for both growth and mass. *, P < 0.001, estrogen-treated tumors are significantly larger than placebo or progesterone-treated tumors for each cell type.
). Points, mean tumor volume (mm3) for each cell line; bars, SE. Sample numbers (n) for the estrogen + placebo, estrogen + progesterone, and estrogen + MPA groups are 10, 10, and 10 for T47DPR–; 20, 17, and 19 for T47DPRB; 21, 17, and 21 for T47DPRA; and 10, 10, and 10 for T47Dwt, respectively. B, individual tumor masses for T47DPR–, T47DPRB, T47DPRA, and T47Dwt tumors treated with estrogen + placebo, estrogen + progesterone, or estrogen + MPA. Bar, mean. Note that scales are different depending on tumor size. No significant difference in tumor volumes (A) or masses (B) were noted between estrogen-, estrogen + progesterone–, or estrogen + MPA–treated groups for any tumor type (P > 0.05). C, mice were given i.p. BrdUrd injections (50 mg/kg) 2 hours before harvesting tumors. Sections from each tumor were stained with an anti-BrdUrd antibody. The number of BrdUrd-positive cells per field was counted in five fields per tumor; total tumors (n = 3). Representative fields used for quantitation for estrogen- and estrogen + MPA–treated T47DPR–, T47Dwt, T47DPRA, and T47DPRB tumors. Magnification, x40. Bar, 100 µm. Mean ± SD BrdUrd-positive cells per field are indicated.
