| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Department of Anatomy, University of California (H.P., P.Y., G.R.C.), San Francisco, California 94143-0130; Faculty of Medicine, Department of Anatomy, University of British Columbia (J.T.E.), Vancouver, British Columbia, Canada V6T 1Z3; Buck Institute for Age Research (R.M.N.), Novato, California 94945; and Geraldine Brush Cancer Research Institute, California Pacific Medical School (S.D.), San Francisco, California 94118
Address all correspondence and requests for reprints to: Dr. Gerald R. Cunha, Department of Anatomy, University of California, 513 Parnassus Avenue, HSW 1301, San Francisco, California 94143-0130. E-mail: grcunha{at}itsa.ucsf.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-lactalbumin (2). Thus, embryonic mammary mesenchyme induces the epidermis to undergo mammary differentiation, both morphologically and functionally. After formation of the mammary epithelial bud, it elongates to form the main duct(s), which subsequently undergoes branching morphogenesis. Although ductal branching morphogenesis is initiated in many species during the prenatal period, most ductal branching morphogenesis occurs during puberty in response to estrogen stimulation (3). In the mouse the process of ductal branching begins within the developing mammary fat pad (FP) on embryonic d 17. The embryonic mammary sprout invades the underlying mammary FP precursor and branches to form a rudimentary ductal tree at birth. The FP, a deeply situated population of sc mesenchymal cells (4), contains two cell types, adipocytes and fibroblasts, and develops independently of mammary mesenchyme.
The major functional unit of the human breast is the lobular structure arising from terminal ducts. This structure consists of several small blind-ending ductules lined by a continuous layer of luminal epithelial cells surrounded by a second layer of myoepithelial cells. The myoepithelial cells are in direct contact with the basement membrane, and the duct is surrounded by fibroblasts and intralobular stroma. Stromal-epithelial interactions play key roles in mammary gland development. Tissue recombination studies have elucidated the importance of estrogen receptor (ER), epidermal growth factor receptor, and PTH-related protein in the stroma of the developing mouse mammary gland (2, 5, 6). Unlike that in the mouse, investigation of the in vivo biology of human breast epithelium (huBrE) is limited. Endocrine regulation of human breast development in vivo is difficult to study directly, especially when assessing the specific roles of stromal vs. epithelial factors. Estrogen and progesterone in cooperation with pituitary hormones are the primary systemic hormones required for the induction of proliferation and differentiation of epithelial and stromal cells leading ultimately to the formation of ductal and alveolar structures during mammary gland development. Estrogen and progesterone, acting via their specific nuclear receptors, are essential for normal mammary gland development and differentiated function. In the mouse mammary gland estrogen is important for ductal growth, and estrogen and progesterone (E2+P) action is required for alveolar development (7, 8). The action of these hormones on human mammary epithelium has proven more difficult to study than that in the mouse.
To solve the technical and ethical problems associated with the study of huBrE, investigators have employed in vivo models in which normal human breast tissue or isolated huBrE was transplanted into nude mice. Such grafts of huBrE alone into nude mice have produced limited outgrowth and do not take into consideration the roles of stroma and epithelium in hormonally regulated mammary growth and differentiation.
McManus and Welsch (9) transplanted slices of normal human breast tissue sc into athymic nude mice. Treatment of the mice with estradiol stimulated the proliferation of ductal epithelium within the grafted human breast slices (10). These studies demonstrated that when the human epithelial component and the stroma are present, various hormonal manipulations of the host animal induced proliferation. Although transplants of human breast slices have been useful for investigating hormonal parameters, such studies do not lend themselves to an examination of stromal-epithelial interactions. To assess the growth of huBrE in vivo under experimental conditions, normal huBrE cells were transplanted into cleared fat pads of athymic nude mice, where they formed primitive glandular structures. Unfortunately only limited outgrowth of mammary ducts has been reported. Such ducts retain normal morphology and contain myoepithelial and luminal cells (11, 12). Estrogen responsiveness was not observed despite hormone administration to the host (13). A lactational response has been demonstrated in human tissue after the administration of lactogenic hormones to the hosts or after mating the xenograft-bearing female mice (11, 14).
Disappointing initial data led to the conclusion that the mouse mammary FP was incapable of supporting expansive ductal growth of human mammary epithelium and that human and bovine mammary epithelia may require the presence of a more fibrous (collagenous) stroma than rodent mammary epithelia for proliferation and duct growth (13). In an attempt to improve ductal growth in vivo, human breast epithelial cells dissociated from reduction mammoplasties were embedded in extracellular matrices (type I collagen and Matrigel) before sc implantation into the nude mouse (15). Such collagen gel transplants responded to estrogen treatment (16). This system pioneered by the Nandi laboratory uses 10–30 million cells/gel. When such sc grafts were harvested, fibroblasts from the graft site had migrated into close proximity to the huBrE cells within the gel, creating an in vivo tissue recombination. Again, the epithelial cells were shown to have organized themselves into primitive tubules and branched structures.
Despite these encouraging results, robust ductal outgrowths have never been described in transplants of huBrE cells into nude mice. Given that the nature of the stroma is of considerable importance in determining the structure, function, and growth pattern of the resulting gland, we have modified the Nandi technique by incorporating mouse and/or human mammary fibroblasts in collagen gels containing huBrE. In so doing, a fibroblast-rich stroma is provided for the normal human breast epithelium. Instead of the relatively avascular sc graft site, our transplants are grafted under the renal capsule of nude mice. The renal site is highly vascular, thus optimizing the number of grafts successfully recovered (take rate) and growth of the graft, as previously described (17). With these modifications ductal growth is robust, with increased ductal density when huBrE is grown with either mouse or human fibroblasts. Immunohistochemical analysis of the grafts demonstrates the presence of myoepithelial cells and luminal epithelial cells that express appropriate steroid receptors. A response to hormonal stimuli was observed after treatment of the host with DES or E2+P pellets. A lactogenic response, induction of ß-casein and milk fat globule protein, has also been shown in the grafts grown in pregnant hosts. This improved method of growing huBrE in vivo allows study of mammary development and elucidation of the role of stromal-epithelial interactions during growth and differentiation of human mammary epithelium in vivo.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Hoechst 33258 dye staining
Formalin-fixed paraffin sections were stained with 4 µg/ml Hoechst 33258 (Calbiochem, La Jolla, CA) for 1 min at room temperature. This was followed by 1–5 min of washing under running tap water.
Production of human breast epithelial organoids and fibroblasts
Normal tissue from reduction mammoplasties was finely minced with scalpels and transferred to a dissociation flask (Bellco, Vineland, NJ) containing DMEM/Ham’s F-12 supplemented with 10 mM HEPES, 2% BSA (fraction V), 300 U/ml collagenase type 1, and 100 U/ml hyaluronidase. After 18 h at 37 C on a shaker, the resulting suspension was centrifuged at 40 x g for 30 sec to pellet epithelial cell clumps and organoids. This pellet was washed twice with fresh DMEM/Ham’s F-12 to remove residual enzyme, stromal cells, and tissue debris.
Alternatively, the suspension was filtered through a 51-µm pore size nylon filter to separate organoids that were retained on the filter from a fibroblast-enriched filtrate as described previously (18). Actively proliferating fibroblast cultures were maintained in DMEM/Ham’s F-12 and 10% fetal bovine serum.
Production and growth of mouse mammary fibroblasts
Number 4 mammary FPs were removed from 3-wk-old female wild-type C57/B6 mice. The medial portion containing the mammary duct was excised and discarded. The remaining lateral portion of the FP was chopped finely with a razor blade in a minimal amount of medium (DMEM/Ham’s F-12 + 10% fetal bovine serum) in a petri dish. Ten milliliters of medium were added to the dish, and the chopped tissue was transferred to a small 25-ml conical flask. After the addition of 10 U/ml collagenase (Sigma, St. Louis, MO), the mince was incubated for 20 min at 37 C. The tissue was then centrifuged for 10 min at 200 x g, the supernatant (containing fat) was removed, and the pellet was resuspended in 5 ml fresh medium. The mixture was centrifuged at 200 x g for 2 min, the supernatant was removed, and the pellet was resuspended in 5 ml medium. The final spin was for 30 sec, after which the pellet was resuspended in 1 ml medium and plated into a 75-ml flask. The resultant cultures were predominantly fibroblastic and negative for keratin expression.
Epithelium and fibroblast recombinations
About 6–8 organoids of huBrE (
50,000–100,000 cells) were mixed with 250,000 fibroblasts in 20 µl neutralized collagen. After polymerization and overnight culture at 37 C, the collagen gels containing normal huBrE and fibroblasts were grafted under the renal capsule of female nude mouse hosts. Grafts were harvested after 1 month.
Animals
Forty- to 50-d-old female CD-1 nude mice were used as hosts for the grafts (Charles River Laboratory, Wilmington, MA). All food, water, cages, and bedding were sterilized before use. Aseptic techniques were used at all times for handling the mice. Renal grafting was carried out in laminar flow hoods. Before renal grafting, recipient hosts were anesthetized with Avertin at 0.02 ml/g body weight. A total of 40 mice were used in this study, and grafts were left in the animals for 1 month. The studies were carried out with approval from the University of California-San Francisco committee on animal care.
Renal grafting
The process of renal capsule grafting is described and illustrated in detail at the following website: http://mammary.nih.gov/tools/mousework/Cunha001/index/html.
Grafts were harvested after 1 month and could be identified as a white mass ranging from 0.25–0.5 cm in length on the kidney. After removal, the grafts were fixed overnight in 10% phosphate-buffered formalin (Fisher Scientific, Fairlawn, NJ), embedded in paraffin, sectioned, and stained by hematoxylin and eosin using standard procedures.
Hormone treatment
All pellets were made using a Parr pellet press. Mice were treated with a diethylstilbestrol (DES) pellet (2 mg DES and 18 mg cholesterol) for 30 d continuously from the time of grafting. Mice were also treated with 10 mg E2+P pellets (1:1000, E2:P). The pellets were placed sc at the time of grafting.
Immunohistochemistry
Histological examination was performed after fixation in 10% formalin, paraffin embedding, and sectioning at 6 µm. The immunohistochemical techniques used have been described previously (19). The antibodies used were mouse monoclonal keratins 8 and 14 (provided by Dr. Birgitte Lane, University of Dundee, Dundee, Scotland), Ki67 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), ER (DAKO Corp., Carpenteria, CA), PR (DAKO Corp.), human milk fat globule (Accurate Chemical and Scientific Corp., Westbury, NY), ß-casein (Harlan Sera-Lab, Loughborough, UK). The avidin-biotin-peroxidase procedure (20) was followed using a Vectastain ABC kit (Vector Laboratories, Inc., Burlingame, CA).
Mammary gland extraction and immunoblotting
Mammary glands from virgin and pregnant mice and huBrE and mouse mammary gland fibroblasts (mMgF) grafts from virgin and pregnant hosts were lysed in Nonidet P-40 lysis buffer [50 mM HEPES (pH 7.4), 1% Nonidet P-40, 150 mM NaCl, 25 mM ß-glycerol phosphate, 25 mM NaF, 5 mM EGTA, 1 mM EDTA, 10 µg/ml leupeptin and aprotinin, and 1 mM phenylmethylsulfonylfluoride], and tissue was pelleted at 10,000 x g for 30 min. Mammary gland and graft extract protein concentrations were determined by the method of Bradford, and protein was separated on 7.5–15% SDS-PAGE gels, blotted onto polyvinylidene difluoride, and detected by enhanced chemiluminescence as described previously (21).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Freshly prepared human breast organoids
Freshly prepared human breast ductal organoids before grafting retained little ductal morphology, and immunostaining with antibodies to myoepithelial and luminal cell markers demonstrated the presence of both of these cell types in the collagenase-isolated ductal organoids (data not shown). Histological analysis of the organoids showed a lack of contaminating human stroma. Thus, the organoids were essentially free of connective tissue elements.
HuBrE grafted alone in the collagen gel
Organoids transplanted in collagen gel without the addition of fibroblasts formed small ducts and solid epithelial cords after 1 month of growth in intact female nude hosts (Fig. 1
). Typically, only three ductal profiles were observed per microscopic field using a x20 objective lens. These epithelial structures were surrounded by fibroblasts. Hoechst dye was used to determine the nature of the surrounding fibroblasts (22). Mouse fibroblasts were shown to contain intranuclear bodies and were distinguishable from human fibroblasts. Based on this Hoechst dye staining, the fibroblastic stroma was overwhelmingly of mouse origin (data not shown). Only a few human fibroblasts were detected, indicating that fibroblastic contamination of the organoids was minimal. The fibroblasts were presumably mouse renal fibroblasts from the graft site that had invaded the collagen gel. The epithelial structures and cords contained luminal cells expressing keratin 8 and myoepithelial cells expressing keratin 14 and smooth muscle actin. The ductal structures also expressed low levels of ER and very low almost undetectable levels of the progesterone receptor (PR). DES administration to the host resulted in an increase in the number of ductal structures observed to an average of five per microscopic field using a x20 objective lens. There was also loss of ER expression and induction of PR (Fig. 1). Anti-Ki67 antibody, which recognizes a human nuclear cell proliferation-associated antigen, was used as the marker for proliferation. The Ki67 labeling index increased by 100% after treatment with DES relative to that observed in the intact or ovariectomized host. This means that cell proliferation increased 2-fold in response to DES even in transplants of epithelium alone (Fig. 2B).
|
|
). The ducts observed in both huBrE+huBrF and huBrE+mMgF tissue TRs contained a continuous layer of luminal cells surrounded by a discontinuous layer of myoepithelial cells and resembled normal human breast tissue (Fig. 3). Ducts of huBrE+huBrF and huBrE+mMgF recombinations demonstrated the expression of keratin 8 in luminal epithelial cells. Myoepithelial cells expressed keratin 14 and -smooth muscle actin, as is the case for normal human breast epithelium (Fig. 3). In the normal human adult breast, ERs are expressed in an average of 7% of luminal epithelial cells (23, 24, 25) and are undetectable in the stroma. Eighty to 100% of the luminal epithelial cells in huBrE+mMgF and huBrE+huBrF recombinations expressed ER in untreated intact or ovariectomized female hosts (Fig. 4, A and B). PR was detected at low levels in huBrE+huBrF and huBrE+mMgF TRs grown in untreated intact hosts (Fig. 4, A and B). Thus, histodifferentiation of huBrE was comparable to that in normal human breast tissue when grown in association with normal human breast or mouse mammary gland fibroblasts. Fibroblasts from both sources allowed for the normal differentiation and robust ductal development of the huBrE in vivo after 1 month under the renal capsule.
|
|
) and huBrE+huBrF (Fig. 4B). In the untreated intact or ovariectomized host, an average of 8 ducts were observed/microscopic field at x20 objective, whereas an average of 11 ducts/microscopic field was observed in huBrE+mMgF treated with DES. This significantly increased the ductal number (Fig. 2A). Ki67 epithelial labeling was higher in the ducts in huBrE+mMgF after the addition of fibroblasts than after grafting the organoids alone in intact untreated hosts (Fig. 2B), although this increase was not statistically significant. This suggests that incorporation of mouse or human mammary fibroblasts into the collagen gels enhances the proliferation of huBrE cells. This effect on epithelial proliferation coincided with an increase in the number of ducts per microscopic field in grafts combined with mammary fibroblasts. The Ki67 labeling index was increased by 2-fold in huBrE+mMgF tissue recombinations from DES-treated hosts relative to that in grafts of organoids only grown in DES-treated hosts. The number of ducts per microscopic field in huBrE+mMgF increased in parallel with the Ki67 labeling index from 8 to an average of 11 after DES treatment. The huBrE+huBrF recombinations showed a similar increase in Ki67 labeling and ductal density after DES treatment (Fig. 4B). Treatment of hosts continuously with DES by sc pellet for 30 d induced a profound down-regulation of epithelial ER in huBrE grown in association with mouse or normal human mammary fibroblasts. PR was induced by DES treatment in luminal epithelial cells of huBrE+mMgF and huBrE+huBrF tissue recombinations (Fig. 4, A and B). Thus, both types of fibroblasts provide a favorable stromal environment for the growth of human mammary epithelium and allow for estrogenic response.
E2+P induced proliferation of huBrE grown in association with mouse mammary fibroblasts
Tissue recombinations were prepared with huBrE+mMgF and treated continuously for 30 d with 10 mg E2+P pellets in intact or ovariectomized hosts. The E2+P was administered by sc pellet as described by Popnikolov and colleagues (15), which produces systemic levels of about 250 pg/ml E2 and 40 ng/ml P (both 10-fold above normal). This hormonal treatment has been shown to stimulate the proliferation of grafted huBrE (16). For the most part the effects of E2+P were identical in most aspects to those of treatment with DES alone. Treatment of huBrE+mMgF grafts with E2+P down-regulated epithelial ER (Fig. 4A) and stimulated epithelial proliferation. The epithelial Ki67 labeling index in E2+P-treated grafts was increased more than 2-fold (from 8 to 18 cells labeled) compared with that for untreated tissue recombinations (Fig. 2B). In parallel, E2+P increased ductal density to an average of 11 ducts/microscopic field, similar to that observed with DES (Fig. 2A). There was no significant increase in Ki67 or ductal density by combining progesterone with estrogen. E2+P did not induce epithelial PR expression as was observed with DES (Fig. 4A).
Effect of host pregnancy on huBrE+mMgF recombinations
For the human breast it has been difficult to study the roles of various hormones and growth factors on the proliferation of epithelial cells during pregnancy. Many endocrinological changes occur during pregnancy, including variations in estrogen, progesterone, glucocorticoid, PRL, and placental lactogen. Due to the complexity of the endocrine profile in pregnancy it has been difficult to mimic pregnancy conditions in organ culture or primary cell culture. In vivo, a lactogenic response has been shown previously in organoids injected into the mouse mammary gland of nude mice after host matings (14). To elicit a lactogenic response in the grafts of huBrE grown in association with mouse or human mammary fibroblasts, the host mice were mated 3 wk after grafting. The grafts were removed before delivery of the pups at about 18 d gestation. Histological analysis of the ducts showed a clear lactogenic response. Ducts were distended with secretions, and the apical cytoplasm of luminal cells was vacuolated (Fig. 5A).
|
). Western analysis of virgin and pregnant mouse mammary glands and huBrE+mMgF grafts from virgin and pregnant hosts was also carried out (Fig. 5C). Using an antibody against human ß-casein gave a specific and a nonspecific band in the pregnant mouse mammary gland sample (Fig. 5C, lane 2). A significant increase in ß-casein was observed in huBrE+mMgF recombinations harvested from a pregnant host, where a single specific band of 28 kDa was observed (Fig. 5C, lane 4). This demonstrates the ability of the graft to undergo changes in morphology and function associated with pregnancy. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our studies of the growth of huBrE in nude mice differ from all previous reports in several respects. Our experiments are the first to graft mixtures of huBrE plus fibroblasts (not adipose tissue) as the stromal element, and thus differ from previous reports of injecting huBrE cells into mouse cleared FPs. Based upon previous transplantation experiments, the stromal requirements of mouse and human mammary epithelium are species specific. Mouse mammary gland epithelium undergoes extensive growth and ductal branching in an adipose stroma provided by the mammary FP. In contrast, human mammary epithelium transplanted into the mouse mammary FP does not grow, but, rather, forms rounded hollow structures encapsulated by connective tissue (13, 15, 26). Although exogenous hormones stimulated DNA synthesis in these structures, expansive ductal outgrowth of the huBrE did not occur (13). Almost identical morphological and proliferative responses were recorded when primary bovine mammary epithelial cells were transplanted into the mouse mammary FP, strongly suggesting that the stromal component provided by the mouse mammary FP is not optimal for ductal growth and branching of both human and bovine mammary epithelial tissues (27). Sheffield suggested that "human mammary epithelia may require the presence of a more fibrous (collagenous) tissue than rodent mammary epithelia for normal proliferation" (13). Thus, combining the mammary epithelium with mammary stromal fibroblasts may more appropriately reconstitute the normal stromal environment of the human mammary gland.
Use of the renal capsule as a graft site is another technical modification relative to previous work. The vascularity of the renal site far exceeds that of the sc site and dramatically increases cellular survival in the first 24 h postgrafting. This maximizes the effective inoculum size and take rate. The effective inoculum size is the number of cells surviving the anoxic conditions in the first 24–48 h before vascularity of the graft is established. During this critical period, a certain percentage of cells die, thus decreasing the effective inoculum size. The percentage of surviving cells is then available to generate mammary ductal tissue. The number of cells lost immediately postgrafting can be exceedingly high, especially in the relatively avascular sc graft site. We see this advantage of the renal graft site in our current mammary studies insofar as the number of ductal profiles observed when 50,000–100,000 huBrE cells (6–8 organoids) are transplanted with fibroblasts appears to be more than or equal to that observed when 10–30 million huBrE cells are transplanted sc in collagen gels (12).
Most important we have verified that human or mouse mammary fibroblasts can support normal histodifferentiation and functional differentiation of human breast epithelium. The normal human mammary gland shows a double-layered epithelium consisting of a single layer of columnar luminal cells and a basal cell layer of myoepithelial cells. Xenografted organoids and mammary fibroblasts also showed this organization, as luminal cells were positive for keratin 8, and the basal cells were positive for keratin 14 and -actin. From the huBrE+mMgF and huBrE+huBrF tissue recombinations analyzed, TRs containing mouse mammary fibroblasts consistently gave better results than TRs containing human fibroblasts, as the grafts were larger, the take rate was higher, and more ductal structures were observed per microscopic field. The mouse mammary fibroblasts were derived from female donors 3 wk of age; therefore, the fibroblasts were young and from growing mammary gland. The human fibroblasts were isolated from reduction mammoplasty patients and thus were derived from mature glands. This factor may have influenced the growth and developmental potential of the human mammary ducts. Ducts in both huBrE+huBrF and huBrE+mMgF TRs were DES responsive, and after the host was mated, milk proteins were detected in the ductal lumen. The hormone response is a key feature of this model, as ducts behave similarly to the normal human mammary gland.
The organoids when transplanted alone under the renal capsule, i.e. without addition of fibroblasts, were DES responsive, as shown by a 2-fold increase in the Ki67 labeling index and an increase in the ductal density per microscopic field. This suggests that human mammary epithelium alone is able to respond directly to DES. Combining the organoids with fibroblasts increased the number of ducts per microscopic field and increased the percentage of proliferating cells (in untreated hosts), providing a more optimal condition for growth of the epithelium. After treatment with the DES pellet for 1 month, epithelial ER was down-regulated, and epithelial PR was induced, confirming previous reports of huBrE grafted alone in collagen gel (28). These effects on ER and PR were observed when the organoids were implanted alone or in combination with mouse or human mammary fibroblasts. In the huBrE+mMgF tissue recombinations grown in intact untreated hosts, ER was detected in both the epithelial and stromal compartments. In huBrE+huBrF TRs grown in an intact untreated host, ER was detected only in the luminal epithelium and not in the stroma. Eighty percent of luminal huBrE cells were ER positive, which is vastly higher than the 7% found in normal adult human breast tissue (29, 30). This may be due in part to the lower systemic levels of estrogen found in the nude mouse vs. the adult human. As estrogen down-regulates ER, the lower systemic estrogen levels provided by the nude mouse host may allow more of the epithelial cells to express ER. Alternatively, the increased number of ER-positive cells may be due to the temporary reversion of the epithelium to a more primitive developmental state when ER expression is known to be much higher (30). Administration of DES via sc pellet at the time of grafting resulted in a down-regulation of the ER and an induction of the PR in huBrE+mMgF and huBrE+huBrF TRs. The reduction in ER levels and the simultaneous induction of PR levels after treatment with estrogen have also been observed in human breast tissue implanted into female intact nude mice treated with estrogen (31). This is probably a direct effect of DES on the mammary gland epithelium not requiring a specific stromal environment. ER also has important interrelationships with the PR system in the modulation of biological responses. PR expression is regulated by estrogen via estrogen response elements on the progesterone receptor (32). By binding of estrogen to progesterone and estrogen response elements, the PR is induced. In most tissues coexpressing PR, estrogen controls the regulation of PR, thereby also controlling sensitivity to progestins. Thus, PR expression is up-regulated by estrogen and down-regulated by progesterone in most target tissues (33), including huBrE grown in vivo in combination with mMgF or huBrF. Administering both estrogen and progesterone to huBrE+mMgF at the time of grafting also showed a proliferative response. However, although DES alone induced PR, epithelial PR was undetectable in huBrE+mMgF tissue recombinants treated with E2+P. The regulation of PR gene expression is complex. In various cell types PR is induced via different mechanisms. For many cell types PR is directly induced by estrogens in the responding cells and decreased by progestins. However, in mouse uterine epithelium PR is constitutively expressed in the absence of signaling through ER. Treatment with E2 down-regulates uterine epithelial PR, an effect mediated through ER in the stroma (34). PR is also up-regulated in the mouse uterus as a result of decidualization through mechanisms not understood (34), but not involving ER. The mechanism of down-regulation of epithelial PR in huBrE+mMgF TR treated with E2+P clearly involves the action of P. This may involve the direct action of P on the mammary epithelial cells. Alternatively, paracrine activity may be involved. Ongoing and planned experiments using mouse mammary fibroblasts from ERKO and PRKO mice may resolve this issue. In the context of the assays performed in the present study, the induction of PR was the only end point that distinguished the effect of DES alone from that of E2+P. The Ki67 index and the ductal profiles of huBrE+mMgF and huBrE+HuBrF TRs treated with DES alone or E2+P were virtually identical, with no additive effect of the E2+P.
The induction of human casein in the huBrE+mMgF grafts after 18 d of mouse pregnancy demonstrates the ability of the human breast epithelium to undergo the complex morphological and differentiation changes that occur in the normal mammary gland. A variety of endocrinological changes occur during pregnancy, including variations in estrogen, progesterone, glucocorticoid, PRL, and placental lactogen. Histological analysis of grafts grown in pregnant hosts revealed distended vacuolated epithelial cells and secretions in ductal lumen. Mammary epithelial cells grafted in collagen gels do not exhibit this response under similar conditions (35), suggesting that the specific reconstituted stroma employed in this study facilitate functional differentiation. Functional differentiation culminated in increased expression of ß-casein protein, as shown by Western analysis. The lactation observed in the huBrE+mMgF recombinations was a response exclusively to murine hormones exerting their influence over the 3-wk span of a murine pregnancy. Although the hormonal conditions provided by the mouse host were probably not optimal, it is evident that human mammary epithelium is responsive to mouse PRL. These results are consistent with studies using organoids injected into the nude mouse (14).
The results show that huBrE develops, grows, responds to hormones, and differentiates in association with normal mouse and human mammary gland fibroblasts when grafted under the renal capsule in a collagen gel. The similarity in results of using both human and mouse stroma demonstrate that processes in the human breast are sufficiently identical to those in the mouse so that normal mMgF can substitute for human mammary stroma. Because wild-type mouse mammary fibroblasts facilitate or promote the growth and development of human breast epithelium, future studies can use mouse mammary fibroblasts derived from a variety of knockout mice to assess the roles of various paracrine pathways in growth, differentiation, and hormone response of huBrE. Also, experimental knockouts of genes in human fibroblasts can be used. Using this methodology, it will be possible to address the effects of specific stromal factors on human mammary gland development in vivo. This novel approach provides an excellent model in which to study the development and epithelial-stromal interactions that take place in the development, growth, and differentiation of human mammary epithelium. The use of tumor fibroblasts or tumorigenic epithelium in this model will also allow for the study of breast cancer.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: DES, Diethylstilbestrol; E2+P, estrogen and progesterone; ER, estrogen receptor ; FP, fat pad; huBrE, human breast epithelium; huBrF, human breast fibroblasts; mMgF, mouse mammary gland fibroblasts; PR, progesterone receptor; TR, tissue recombination.
Received May 31, 2002.
Accepted for publication August 23, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. Loo, N. Pryer, P. Young, T. Liang, S. Coberly, K. L. King, K. Kang, P. Roberts, M. Tsao, X. Xu, et al. The glycotope-specific RAV12 monoclonal antibody induces oncosis in vitro and has antitumor activity against gastrointestinal adenocarcinoma tumor xenografts in vivo Mol. Cancer Ther., March 1, 2007; 6(3): 856 - 865. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cvoro, S. Paruthiyil, J. O. Jones, C. Tzagarakis-Foster, N. J. Clegg, D. Tatomer, R. T. Medina, M. Tagliaferri, F. Schaufele, T. S. Scanlan, et al. Selective Activation of Estrogen Receptor-{beta} Transcriptional Pathways by an Herbal Extract Endocrinology, February 1, 2007; 148(2): 538 - 547. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. K.C. Tsai, E. Y.-Y. Chuang, J. B. Little, and Z.-M. Yuan Cellular Mechanisms for Low-Dose Ionizing Radiation-Induced Perturbation of the Breast Tissue Microenvironment Cancer Res., August 1, 2005; 65(15): 6734 - 6744. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Behbod and J. M. Rosen Will cancer stem cells provide new therapeutic targets? Carcinogenesis, April 1, 2005; 26(4): 703 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Blauer, P.K. Heinonen, P.M. Martikainen, E. Tomas, and T. Ylikomi A novel organotypic culture model for normal human endometrium: regulation of epithelial cell proliferation by estradiol and medroxyprogesterone acetate Hum. Reprod., April 1, 2005; 20(4): 864 - 871. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Medina and F. Kittrell Stroma is not a major target in DMBA-mediated tumorigenesis of mouse mammary preneoplasia J. Cell Sci., January 1, 2005; 118(1): 123 - 127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Parmar and G. R Cunha Epithelial-stromal interactions in the mouse and human mammary gland in vivo Endocr. Relat. Cancer, September 1, 2004; 11(3): 437 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Medina Stromal fibroblasts influence human mammary epithelial cell morphogenesis PNAS, April 6, 2004; 101(14): 4723 - 4724. [Full Text] [PDF]
|
|
|
|
|
|
C. Kuperwasser, T. Chavarria, M. Wu, G. Magrane, J. W. Gray, L. Carey, A. Richardson, and R. A. Weinberg From The Cover: Reconstruction of functionally normal and malignant human breast tissues in mice PNAS, April 6, 2004; 101(14): 4966 - 4971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Paruthiyil, H. Parmar, V. Kerekatte, G. R. Cunha, G. L. Firestone, and D. C. Leitman Estrogen Receptor {beta} Inhibits Human Breast Cancer Cell Proliferation and Tumor Formation by Causing a G2 Cell Cycle Arrest Cancer Res., January 1, 2004; 64(1): 423 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Novaro, D. C. Radisky, N. E. R. Castro, A. Weisz, and M. J. Bissell Malignant Mammary Cells Acquire Independence from Extracellular Context for Regulation of Estrogen Receptor {alpha} Clin. Cancer Res., January 1, 2004; 10(1): 402S - 409S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Schmeichel and M. J. Bissell Modeling tissue-specific signaling and organ function in three dimensions J. Cell Sci., June 15, 2003; 116(12): 2377 - 2388. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||
