Interaction of mouse mammary epithelial cells with collagen substrata: regulation of casein gene expression and secretion.

Mouse mammary epithelial cells (MMEC) secrete certain milk proteins only when cultured on floating collagen gels. We demonstrate here that modulation of milk proteins by substrata is manifested at several regulatory levels; (i) Cells cultured on floating collagen gels have 3- to 10-fold more casein mRNA than cells cultured on plastic or attached collagen gels. (ii) Cells on the latter two "flat" substrata, nevertheless, synthesize a significant amount of caseins, indicating that the remaining mRNA is functional. (iii) Cells on all substrata are inducible for casein mRNA and casein proteins by prolactin, but the extent of induction is greater on collagen than that on plastic--i.e., the substratum confers an altered degree of inducibility. (iv) Cells on all substrata synthesize casein proteins at rates proportional to the amount of casein mRNA, but the newly synthesized caseins in cells on plastic are degraded intracellularly, whereas those synthesized by cells on floating gels are secreted into the medium. (v) Cells on all substrata examined lose virtually all mRNA for whey acidic protein despite the fact that this mRNA is abundant in the mammary gland itself; we conclude that additional, as-yet-unknown, factors are necessary for synthesis and secretion of whey acidic protein in culture.     

Control of casein gene expression in isolated cultured rabbit epithelial mammary cells

Epithelial cells were isolated from mammary glands of pseudo-pregnant rabbits after a digestion of the tissue by collagenase and cell fractionation on a percoll gradient. These cells were cultured for various times in the presence of serum until growth was sufficient. The serum was then withdrawn for two days and hormones were added for one more day in the absence of serum. Cells were bound either directly to the plastic of the flask or to collagen spread in one or two layers. The concentration of β-casein mRNA was evaluated at the end of the culture by using a ³H-DNA complementary to β-casein mRNA.The concentration of β3-casein mRNA was increased when prolactin was present in the culture medium. Cortisol amplified this effect while being totally inefficient alone. This induction was obtained as well when cells were cultured on plastic or on monolayer, bilayer or floating collagen, and also when they were kept in suspension. However, they apparently responded to prolactin only when fibroblasts were eliminated by adding hydroxy-L-proline (allo) or cholera toxin to the culture medium or by using collagen. The lactose synthetase activity and casein synthesis always remained extremely low regardless of the presence of hormones in the culture medium. Cells cultured on collagen tended to become reorganized in a way somewhat similar to that in mammary tissue, whereas cells spread on plastic remained uniformly distributed.These results suggest that isolated rabbit mammary epithelial cells have kept the essence of their capacity to respond to prolactin regardless of the culture technique. Isolated cultured cells thus appear to be a viable tool for further studies on the mechanism of prolactin action.     

Insulin is essential for accumulation of casein mRNA in mouse mammary epithelial cells.

In the presence of cortisol and prolactin, insulin at concentrations as low as 1 ng/ml significantly stimulates casein synthesis in mammary explants from midpregnant mice; maximal synthesis occurs at 10 ng/ml. However, in the absence of insulin, no detectable immunoprecipitable casein is produced. Insulin also supports enhanced accumulation of casein mRNA in the presence of cortisol and prolactin; neither epidermal growth factor nor somatomedin C has this effect. These inductive actions of insulin are not secondary to a general maintenance effect on the mammary epithelial cell; insulin, epidermal growth factor, and somatomedin C can support the accumulation of RNA in rough endoplasmic reticulum equally well. In addition, these effects do not reflect a specific insulin requirement for prolactin sensitivity; epidermal growth factor can support prolactin-induced total RNA synthesis as well as insulin can. The results demonstrate that, although insulin, epidermal growth factor, and somatomedin C can all function as cell maintenance agents, only insulin, together with cortisol and prolactin, can induce casein mRNA accumulation.     

Hormonal effects on intracellular and secreted casein in cultures of mouse mammary epithelial cells on floating collagen membranes.

Cultured on floating collagen membranes in the presence of lactogenic hormones, dissociated normal mammary epithelial cells from prelactating mice acquire the ultrastructural and biochemical characteristics of differentiated mammary secretory cells in vivo. The cells on floating collagen membranes in medium containing insulin alone have sparse secretory organelles, and a small amount of casein can be detected in these cells with a sensitive radioimmunoassay. These cells resemble counterpart cells in early-pregnant mice. When the cells are exposed to insulin, cortisol, and prolactin, the secretory apparatus is elaborated and significant increases in intracellular and extracellular casein are observed. In this environment, the intracellular casein content is generally four to eight times greater than in freshly dissociated cells or cells cultured in insulin alone. The amount of casein secreted into the medium by floating-collagen-membrane cultures in the three hormones is from 25 to 200 times greater than that secreted by cultures in insulin alone. Cells cultured on plastic substrates in either hormone combination fail to show any increase in intracellular or extracellular casein. On floating collagen membranes, the cells differentiate in response to hormones as they do in vivo and in organ culture. This cell-culture system provides an opportunity to study direct effects of environmental factors on mammary differentiation at the cellular level.     

Distinctive gene expression patterns in human mammary epithelial cells and breast cancers.

cDNA microarrays and a clustering algorithm were used to identify patterns of gene expression in human mammary epithelial cells growing in culture and in primary human breast tumors. Clusters of coexpressed genes identified through manipulations of mammary epithelial cells in vitro also showed consistent patterns of variation in expression among breast tumor samples. By using immunohistochemistry with antibodies against proteins encoded by a particular gene in a cluster, the identity of the cell type within the tumor specimen that contributed the observed gene expression pattern could be determined. Clusters of genes with coherent expression patterns in cultured cells and in the breast tumors samples could be related to specific features of biological variation among the samples. Two such clusters were found to have patterns that correlated with variation in cell proliferation rates and with activation of the IFN-regulated signal transduction pathway, respectively. Clusters of genes expressed by stromal cells and lymphocytes in the breast tumors also were identified in this analysis. These results support the feasibility and usefulness of this systematic approach to studying variation in gene expression patterns in human cancers as a means to dissect and classify solid tumors.     

Transfection of beta-casein chimeric gene and hormonal induction of its expression in primary murine mammary epithelial cells.

To study the regulatory sequence elements responsible for casein gene expression, we constructed a chimeric gene containing 5.3 kilobases (kb) of the 5'-flanking sequence and 1.6 kb of the 3'-flanking sequence of the mouse beta-casein gene fused to the bacterial chloramphenicol acetyl-transferase (CAT) gene. The chimeric gene was transfected by the calcium phosphate-precipitation procedure into primary mouse mammary epithelial cells prepared from pregnant mice. The transfection procedure had negligible effect on expression of the endogenous beta-casein gene. Expression of the beta-casein-CAT chimeric gene required the synergistic actions of insulin, hydrocortisone, and prolactin. Expression of the chimeric gene also depended on the appropriate substratum because the degree of hormonal induction of the chimeric gene was much higher in cells cultured on a reconstituted basement membrane (Matrigel) than in cells cultured on either type I collagen gel or plastic. On the other hand, the expression of a simian virus 40-CAT chimeric gene in which the CAT gene was driven by the early promoter of the virus was not influenced by the hormonal milieu and occurred at the highest level in cells cultured on plastic. Additional transfection experiments with a series of beta-casein-CAT constructs suggested the existence of regulatory elements responsible for hormonal induction and negative regulatory elements.     

Functional diversity of gro gene expression in human fibroblasts and mammary epithelial cells.

Previous studies of gro and related genes that are overexpressed in transformed fibroblasts suggest that gro may encode a specific growth regulator. However, DNA and protein sequence comparisons reveal relatedness to platelet factor 4 and other proteins involved in the inflammatory response. In this paper, both growth-related and cytokine-induced responses in gro gene expression are described. Human foreskin fibroblasts are shown to express approximately 10-fold elevated gro, myc, and fos mRNAs in response to serum and to phorbol 12-myristate 13-acetate stimulation, with early response kinetics indicative of growth regulation. In response to interleukin 1, however, in growing cells gro mRNA is elevated at least 100-fold but myc remains constant and fos is not expressed, suggesting a second regulatory pathway. In normal cultured mammary epithelial cells, gro is constitutively expressed, and elevated mRNA levels are induced by phorbol 12-myristate 13-acetate, but not by interleukin 1. However, most carcinoma cell lines examined do not express gro mRNA, suggesting a third function of gro as a negative growth regulator in epithelial cells.     

Syndecan expression regulates cell morphology and growth of mouse mammary epithelial tumor cells.

S115 mouse mammary epithelial cells lose their epithelial morphology and become tumorigenic when exposed to steroids. We have recently reported that testosterone exposure results in the suppression of syndecan expression, suggesting that this cell surface proteoglycan may influence S115 cell phenotype. We now report that a similar suppression and morphological response of S115 cells can be achieved by glucocorticoid exposure. We introduced into S115 cells an exogenous gene construct containing the full-length human syndecan cDNA under the control of a glucocorticoid-inducible retroviral promoter, in order to study the effect of syndecan expression on S115 cell behavior. Glucocorticoid-induced re-expression of syndecan in S115 cells restored an epithelial phenotype, while control transfectants and parental S115 cells exhibited an altered, nonepithelial phenotype. Moreover, the S115 cells expressing exogenous syndecan revealed a reduced ability to form colonies in soft agar. Therefore, the maintenance of epithelial morphology and normal growth of S115 cells are dependent on syndecan expression.     

Basic fibroblast growth factor stimulates the growth and inhibits casein accumulation in mouse mammary epithelial cells in vitro

The effects of basic fibroblast growth factor (bFGF) on the growth and differentiation of mouse mammary epithelial cells in serum-free collagen gel culture were examined. Epithelial cells obtained from virgin or midpregnant mice grew when bFGF was added to medium containing either insulin at a concentration greater than or equal to 1 microgram/ml or somatomedin-C (Sm-C) at 150 ng/ml. This growth-promoting effect is of the same magnitude as, and additive with, the growth-promoting effect of epidermal growth factor (EGF) or mammogenic hormones. The sensitivity of the cells to EGF or mammogenic hormones was not altered by exposure to bFGF. The progeny cells resulting from growth stimulation by bFGF are capable of accumulating casein upon subsequent stimulation by prolactin (PRL), but accumulate less casein than cells grown in response to EGF. bFGF also appears to reduce casein accumulation if it is added to the cultures at the same time as PRL.     

Use of a reconstituted basement membrane to measure cell invasiveness and select for highly invasive tumor cells.

Malignant cells must traverse basement membranes during their migration to sites distant from the primary tumor. Since basement membranes are thought to be a critical barrier to the passage of tumor cells, we have constructed a model basement membrane-stromal matrix consisting of laminin and type IV collagen reconstituted onto a disk of type I collagen for use in an in vitro assay of invasiveness. Metastatic tumor cells and leukocytes are able to cross this barrier, whereas nonmetastatic tumor cells, fibroblasts, and epidermal cells cannot penetrate it. Those tumor cells that penetrate the barriers were found, when isolated and subcultured, to be more invasive and to produce more metastases than the parental population. This assay system should be useful for studying the invasiveness of tumor cells and for isolating highly invasive variants.     

Osmotic shock of cultured primary mammary cells amplifies the hormonal induction of casein gene expression

Primary cells from rabbit mammary gland cultured on floating collagen were transfected with various plasmids in different conditions. Conventional transfection methods using DEAE-dextran or calcium phosphate followed by an osmotic shock with dimethyl sulphoxide (DMSO), polyethylene glycol (PEG) or glycerol did not prevent lactogenic hormones to induce casein synthesis. On the contrary and unexpectedly, casein synthesis was markedly stimulated by transfection. This amplification was obtained as well with DMSO, PEG and glycerol alone or in the presence of DEAE-dextran, calcium phosphate or DNA. None of these compounds induced casein synthesis in the absence of prolactin. A shock by DMSO also amplified the accumulation of beta-casein mRNA in the presence of prolactin. These results show for the first time that primary cultured mammary cells can be efficiently transfected and still keep their capacity to respond to lactogenic hormones. They also indicate that the short osmotic shocks conventionally used in transfection have a potent long-term stimulatory effect on casein gene expression, which is mediated through an unknown mechanism.     

Linoleic acid, but not cortisol, stimulates accumulation of casein by mouse mammary epithelial cells in serum-free collagen gel culture.

A two-step culture system has been developed to analyze the role of hormones in casein accumulation by mammary epithelial cells obtained from adrenalectomized and ovariectomized adult virgin mice. In the first step cells are grown inside collagen gel in medium containing insulin, epidermal growth factor (EGF), and linoleic acid for 9 days; these conditions stimulate very little casein accumulation. Following this growth phase the gels are released to float in medium containing insulin, prolactin, and linoleic acid. During this second phase the mammary cells will accumulate large amounts of casein, but only in the simultaneous presence of insulin, prolactin, and linoleic acid; in the absence of linoleic acid casein accumulation is greatly reduced. The casein accumulation is not dependent on the presence of the glucocorticoid cortisol and will occur in spite of the presence of the antiglucocorticoid agent RU 38 486. To determine if the response to cortisol observed in organ culture by other investigators might be mediated by stromal cells, epithelial cells were grown in collagen gel under fatty acid-free conditions and then cocultured with explants of mammary fat pads from adult virgin mice with or without mammary parenchyma. The cocultures were performed in fatty acid-free medium containing insulin and prolactin with or without cortisol. In the majority of experiments the mammary epithelial cells in the collagen gel accumulate more casein in the presence of cortisol than in its absence, irrespective of the presence of mammary parenchyma in the explant. Thus, mammary epithelial cells are directly dependent on insulin and prolactin for casein accumulation and indirectly dependent on cortisol by means of its effect on the stromal cells. This cortisol effect may be to cause release into the medium of linoleic acid or a metabolic product of linoleic acid from the stromal cells.     

Biosynthesis of basement membrane collagen in cultures of renal glomerular and tubular epithelial cells.

Confluent cultures of renal glomerular or tubular epithelial cells were incubated with [14C] proline and [3H] lysine. The incorporation rate of both radioactive precursors was found to be linear for up to 12 h. The synthesis and secretion of basement membrane collagenous polypeptides was demonstrated by the presence in the culture media of non-dialyzable 4-hydroxy [14C] proline and hydroxy [3H] lysine. After gel filtration of the culture media on Sephadex G-100 columns, glomerular and tubular basement membrane polypeptides were identified in the chromatographic fractions by radioimmunoassay. They were further purified by affinity chromatography, using Sepharose cyanogen-bromide coupled with specific rabbit anti-human glomerular or tubular basement membrane antibodies. Absorbed labelled membrane polypeptides were eluted from the Sepharose by acidic medium at 4 degrees C. This membrane material represented 3-4% of the total proteins synthesized by glomerular and tubular cells. The glomerular and tubular basement membrane polypeptides purified by affinity chromatography exhibited a molecular weight of approximately 140,000; 80% of the total hydroxy [3H] lysine was recovered as glucosyl-galactosyl-hydroxy [3H] lysine. Analysis of the carbohydrate content of labelled basement membrane polypeptide chains originating from glomerular or tubular cells incubated with [14C] glucose of [14C glucosamine indicated the presence of glucose, galactose, mannose, glucosamine, and galactosamine. No fucose, mannosamine or sialic acid were detectable. The data demonstrate that glomerular and tubular epithelial cells are able to synthesize basement membrane collagenous polypeptides in culture. This property might provide a useful tool for the study of the biosynthesis of similar material by diseased kidneys.     

Expression of a lactogen-dependent insulin-like growth factor-binding protein in cultured mouse mammary epithelial cells.

The ability of normal mouse mammary epithelial cells (MECs) to express insulin-like growth factor-binding proteins (IGFBPs) was examined. MECs were isolated from day 11 pregnant mice and cultured on floating collagen gels in serum-free basal medium. After 24 h, the medium was replaced with fresh medium with/or without mouse PRL (mPRL), mouse placental lactogen-I (mPL-I), mPL-II, mouse GH (mGH), IGF-I, and IGF-II, either alone or in combinations. The MECs were cultured for an additional 5 days before collection of conditioned medium (CM). The relative amount of IGFBPs present in the CM was determined by Western ligand blotting, and alpha-lactalbumin content was determined with a specific RIA. The CM of the MECs contained two IGFBPs, with approximate mol wt of 29K and 40-45K. The 40-45K IGFBP appears to be the mouse equivalent of IGFBP-3, but the identity of the 29K IGFBP is not presently known. The 29K IGFBP was not N-glycosylated and did not cross-react with antiserum to rodent IGFBP-2 or human IGFBP-1. Basal IGFBP expression was very low, but the addition of mPL-I, or mPL-II stimulated a marked increase in the amount of 29K IGFBP that was released into the CM and a lesser increase in the release of IGFBP-3. This increase in the release of 29K IGFBP was dose dependent, with increases found at concentrations as low as 1 ng/ml lactogen. mGH also stimulated the release of 29K IGFBP, but was less potent than any of the three lactogens. Treatment of MECs with either IGF-I or IGF-II increased the amount of both the 29K IGFBP and IGFBP-3 in the CM, with relative potencies similar to those of the lactogenic hormones. However, when either IGF-I or IGF-II was added together with one of the lactogenic hormones, the release of 29K IGFBP was increased in an additive manner. While the IGFs acted additively with the lactogenic hormones on the expression of 29K IGFBP, they did not stimulate alpha-lactalbumin production by the MECs or act to enhance the effects of the lactogenic hormones in stimulating alpha-lactalbumin production. This study demonstrates that IGFBPs are expressed in normal mouse MECs, and the release of these IGFBPs into the CM is hormonally regulated by both lactogenic hormones and IGFs.     

Extracellular matrix-dependent tissue-specific gene expression in mammary epithelial cells requires both physical and biochemical signal transduction.

Extracellular matrix (ECM) profoundly influences the growth and differentiation of the mammary gland epithelium, both in culture and in vivo. Utilizing a clonal population of mouse mammary epithelial cells that absolutely requires an exogenous ECM for function, we developed a rapid assay to study signal transduction by ECM. Two components of the cellular response to a basement membrane overlay that result in the expression of the milk protein beta-casein were defined. The first component of this response involves a rounding and clustering of the cells that can be physically mimicked by plating the cells on a nonadhesive substratum. The second component is biochemical in nature, and it is associated with beta 1 integrin clustering and increased tyrosine phosphorylation. The second component is initiated in a morphology-independent manner, but the proper translation of this biochemical signal into a functional response requires cell rounding and cell clustering. Thus, physical and biochemical signal transduction events contribute to the ECM-dependent regulation of tissue-specific gene expression in mouse mammary epithelial cells.     

Glucocorticoid-induced formation of tight junctions in mouse mammary epithelial cells in vitro.

Phenotypically stable cultures of untransformed mouse mammary epithelial cells (denoted 31EG4) were established and utilized to investigate the lactogenic hormone (glucocorticoids, insulin, and prolactin) regulation of tight junction formation. When 31EG4 cells were grown on permeable supports for 4 days in medium containing the synthetic glucocorticoid dexamethasone and insulin, confluent cell monolayers obtained a transepithelial electrical resistance (TER) of 1000-3000 omega.cm2. In contrast, over the same time period, confluent monolayers treated with insulin or insulin and prolactin maintained a low TER (35-150 omega.cm2). Consistent with the formation of tight junctions, apical to basolateral paracellular permeability was decreased from 12% to 1% for [14C]mannitol and 3.3% to 0.3% for [3H]inulin when cells were cultured in dexamethasone. This effect of dexamethasone on TER required extracellular calcium, de novo protein synthesis, dose-dependently correlated with glucocorticoid receptor occupancy, and was not due to an increase in cell density. As shown by direct and indirect immunofluorescence microscopy, dexamethasone treatment did not modulate the production or location of filamentous actin, the tight junction protein ZO-1, or the cell adhesion protein E-cadherin. Our results suggest that glucocorticoids play a fundamental role in the function and maintenance of cell-cell contact in the mammary epithelia by inducing the formation of tight junctions.     

Growth factors and cytokines upregulate gelatinase expression in bone marrow CD34(+) cells and their transmigration through reconstituted basement membrane

The mechanism(s) underlying the release of stem/progenitor cells from bone marrow into the circulation is poorly understood. We hypothesized that matrix metalloproteinases (MMPs), especially gelatinases, which are believed to participate in the proteolysis of basement membranes and in the migration of leukocytes, may facilitate this process. First, we investigated whether CD34(+) stem/progenitor cells express gelatinases A (MMP-2) and/or B (MMP-9) and whether growth factors and cytokines (granulocyte colony-stimulating factor [G-CSF], granulocyte-macrophage colony-stimulating factor [GM-CSF], stem cell factor [SCF], macrophage colony-stimulating factor [M-CSF], interleukin-3 [IL-3], IL-6, IL-8, and tumor necrosis factor-alpha [TNF-alpha]) are able to modulate their expression. Next, we examined the transmigration of these stem/progenitor cells through reconstituted basement membrane (Matrigel) and its modulation by growth factors and cytokines. CD34(+) cells were obtained from steady-state bone marrow and peripheral blood (from leukapheresis products collected either in steady-state hematopoiesis or after mobilization with G-CSF plus chemotherapy or G-CSF alone). We found that peripheral blood CD34(+) cells, regardless of whether they were mobilized or not, strongly expressed both gelatinases (MMP-2 and MMP-9) in contrast to steady-state bone marrow CD34(+) cells, which did not. However, all the growth factors and cytokines tested could induce MMP-2 and MMP-9 secretion by the latter cells. Moreover, the stimulatory effects of G-CSF and SCF on both MMP-2 and MMP-9 secretion were found to be significantly higher in CD34(+) cells isolated from bone marrow than in those from peripheral blood. In addition TNF-alpha, GM-CSF, and IL-6 increased the secretion of a partially active form of MMP-2. Basal transmigration of bone marrow CD34(+) cells through Matrigel was lower than that of peripheral blood CD34(+) cells (P <.0001), but growth factors and cytokines increased it by 50% to 150%. Positive correlations were established between expression of gelatinases and CD34(+) cell migration (r >.9). The stimulatory effect of G-CSF was significantly greater on the migration of CD34(+) cells from bone marrow than on those from peripheral blood (P =.004). Moreover, CD34(+) cell migration was reduced to approximately 50% by antibodies to MMP-2 and MMP-9, tissue inhibitors of metalloproteinases (rhTIMP-1 and -2), and o-phenanthroline. TNF-alpha-induced gelatinase secretion and migration of CD34(+) cells and of clonogenic progenitors (colony-forming unit-granulocyte-macrophage [CFU-GM], burst-forming unit-erythroid [BFU-E], colony-forming unit granulocyte, erythroid, monocyte, megakaryocyte [CFU-GEMM], and colony-forming unit-megakaryocyte [CFU-MK]) were dose-dependent. Therefore, this study demonstrated that CD34(+) cells that are circulating in peripheral blood express both MMP-2 and MMP-9 and transmigrate through Matrigel. In contrast, CD34(+) cells from steady-state bone marrow acquire similar properties after exposure to growth factors and cytokines, which upregulate expression of gelatinases and transmigration of these cells when they enter the bloodstream. Hence, we suggest that growth factors and cytokines induce release of stem/progenitor cells from bone marrow into peripheral blood during mobilization, as well as during steady-state hematopoiesis, by signaling through gelatinase pathways.