Putting tumours in context.

The interactions between cancer cells and their micro- and macroenvironment create a context that promotes tumour growth and protects it from immune attack. The functional association of cancer cells with their surrounding tissues forms a new 'organ' that changes as malignancy progresses. Investigation of this process might provide new insights into the mechanisms of tumorigenesis and could also lead to new therapeutic targets.

Extracellular matrix signaling: integration of form and function in normal and malignant cells.

A growing number of studies have established reciprocal linkages between extracellular matrix (ECM)-integrins, growth factor signaling and cell-cell adhesion molecules. ECM-dependent tissue-specific gene expression has also been linked to chromatin remodeling. With respect to tissue morphogenesis and differentiation, crosstalk has been established between the ECM and the homeobox morphoregulatory genes. Each of these linkages is profoundly influenced by the cell's microenvironment and the resulting tissue form. Thus for a cell to achieve a differentiated phenotype, the ECM molecules and their receptors must integrate both form and function. In contrast, mutated genes and aberrant interactions with the microenvironment conspire to undermine this integration, often resulting in malignant transformation.

A hierarchy of ECM-mediated signalling regulates tissue-specific gene expression.

A dynamic and reciprocal flow of information between cells and the extracellular matrix contributes significantly to the regulation of form and function in developing systems. Signals generated by the extracellular matrix do not act in isolation. Instead, they are processed within the context of global signalling hierarchies whose constituent inputs and outputs are constantly modulated by all the factors present in the cell's surrounding microenvironment. This is particularly evident in the mammary gland, where the construction and subsequent destruction of such a hierarchy regulates changes in tissue-specific gene expression, morphogenesis and apoptosis during each developmental cycle of pregnancy, lactation and involution.

ErbB2, but not ErbB1, reinitiates proliferation and induces luminal repopulation in epithelial acini.

Both ErbB1 and ErbB2 are overexpressed or amplified in breast tumours. To examine the effects of activating ErbB receptors in a context that mimics polarized epithelial cells in vivo, we activated ErbB1 and ErbB2 homodimers in preformed, growth-arrested mammary acini cultured in three-dimensional basement membrane gels. Activation of ErbB2, but not that of ErbB1, led to a reinitiation of cell proliferation and altered the properties of mammary acinar structures. These altered structures share several properties with early-stage tumours, including a loss of proliferative suppression, an absence of lumen, retention of the basement membrane and a lack of invasive properties. ErbB2 activation also disrupted tight junctions and the cell polarity of polarized epithelia, whereas ErbB1 activation did not have any effect. Our results indicate that ErbB receptors differ in their ability to induce early stages of mammary carcinogenesis in vitro and this three-dimensional model system can reveal biological activities of oncogenes that cannot be examined in vitro in standard transformation assays.

From laminin to lamin: regulation of tissue-specific gene expression by the ECM.

Mammary epithelial cells need a laminin-rich extracellular matrix (ECM) to achieve a functionally differentiated phenotype that includes secretion of milk-specific proteins such as beta-casein. There is good evidence that ECM-induced expression of beta-casein involves an 'ECM-response element' in the promoter of the casein gene that is activated by integrin-mediated signalling. This article proposes that ECM-induced structural changes in the cytoskeleton, histone organization and the nuclear matrix contribute to this tissue-specific gene expression.

Localization of 9E3/CEF-4 in avian tissues: expression is absent in Rous sarcoma virus-induced tumors but is stimulated by injury.

The avian gene 9E3/CEF-4, a member of the superfamily of genes that includes KC and gro, is expressed abundantly in exponentially growing cultures of chick embryo fibroblasts (CEFs) and at high levels in CEFs transformed with Rous sarcoma virus (RSV). The product of this gene is a secreted protein that has homologies and structural similarities to inflammatory mediators. The function of 9E3 is obscure and its expression in vivo has not yet been investigated. We studied by in situ hybridization and RNA blots the pattern of 9E3 mRNA distribution in the wings of normal, wounded, and RSV-infected newly hatched chicks. We found that the message for 9E3 is high in specific tissues in normal wings; whereas connective tissue, tendon, and bone express the gene, muscle fibers, endothelium, epidermis, and bone marrow do not. The distribution coincides with that of interstitial collagen. Wounding results in marked elevation of the mRNA within the granulation tissue formed during healing and in adjacent tissues, especially those showing neovascularization. Similar elevation of mRNA occurs immediately adjacent to RSV tumors but, surprisingly, the tumor tissue itself shows no detectable levels of this message. Cells explanted from the tumors and grown in culture also show no expression of 9E3, in marked contrast to the very high level found in similarly cultured RSV-transformed CEFs. These results show that there are intrinsic differences between transformed embryonic cells in tissue culture and RSV target cells in the hatched chick. However, the expression of the gene in the periphery of tumors leaves open the possibility that 9E3 may still be involved in RSV carcinogenesis. The abundant expression of 9E3 in normal tissues indicates that the product of this gene plays a normal physiological role in tissues growing by cell division, perhaps as a growth regulator. The elevated expression of 9E3 in areas of neovascularization, makes it possible that the product of this gene could act as an angiogenic factor. Finally, expression in conjunction with high collagen levels and in wounded tissues may point to a role in wound response and/or repair, possibly via alteration of extracellular matrix.

The embryonic environment strongly attenuates v-src oncogenesis in mesenchymal and epithelial tissues, but not in endothelia.

We demonstrate that the behavior of cells expressing v-src, a tyrosine kinase oncogene, differs profoundly between the embryonic and culture environments. V-src was introduced into avian embryo cells both in culture and in stage-24 embryo limbs, using replication-defective retroviral vectors. These vectors were used as single-hit, cellular markers to determine the environmental influences imposed by normal cells and tissues on clonal cell growth. The marker gene lacZ was coexpressed with v-src in order to locate the descendent cells. In culture, v-src induced rapid morphological transformation and anchorage-independent growth of embryo fibroblasts; the vectors were also tumorigenic in hatchling chickens. In contrast, most of the cell clones expressing v-src in the embryo grew normally without neoplasia. Expression of v-src vectors could be found in a wide range of cell types, demonstrating not only that neoplastic transformation is attenuated in ovo, but also that differentiation commitment in many lineages can be maintained concurrently with oncogene expression. Significantly, the embryonic control of cell growth could be perturbed by v-src under certain conditions. Rare, marked clones showed hyperplasia or dysplasia, and the primitive endothelium could succumb to rapid neoplasia; thus, these embryonic tissues are not inherently deficient in transformation factors. We propose that the environmental conditions imposed on cells in ovo are critical for the attenuation of neoplasia, while cultured cells lose this requisite environment.

Expression of extracellular matrix components is regulated by substratum.

Reconstituted basement membranes and extracellular matrices have been demonstrated to affect, positively and dramatically, the production of milk proteins in cultured mammary epithelial cells. Here we show that both the expression and the deposition of extracellular matrix components themselves are regulated by substratum. The steady-state levels of the laminin, type IV collagen, and fibronectin mRNAs in mammary epithelial cells cultured on plastic dishes and on type I collagen gels have been examined, as has the ability of these cells to synthesize, secrete, and deposit laminin and other, extracellular matrix proteins. We demonstrate de novo synthesis of a basement membrane by cells cultured on type I collagen gels which have been floated into the medium. Expression of the mRNA and proteins of basement membranes, however, are quite low in these cultures. In contrast, the levels of laminin, type IV collagen, and fibronectin mRNAs are highest in cells cultured on plastic surfaces, where no basement membrane is deposited. It is suggested that the interaction between epithelial cells and both basement membrane and stromally derived matrices exerts a negative influence on the expression of mRNA for extracellular matrix components. In addition, we show that the capacity for lactational differentiation correlates with conditions that favor the deposition of a continuous basement membrane, and argue that the interaction between specialized epithelial cells and stroma enables them to create their own microenvironment for accurate signal transduction and phenotypic function.

Coordinated expression of extracellular matrix-degrading proteinases and their inhibitors regulates mammary epithelial function during involution.

Extracellular matrix (ECM) plays an important role in the maintenance of mammary epithelial differentiation in culture. We asked whether changes in mouse mammary specific function in vivo correlate with changes in the ECM. We showed, using expression of beta-casein as a marker, that the temporal expression of ECM-degrading proteinases and their inhibitors during lactation and involution are inversely related to functional differentiation. After a lactation period of 9 d, mammary epithelial cells maintained beta-casein expression up to 5 d of involution. Two metalloproteinases, 72-kD gelatinase (and its 62-kD active form), and stromelysin, and a serine proteinase tissue plasminogen activator were detected by day four of involution, and maintained expression until at least day 10. The expression of their inhibitors, the tissue inhibitor of metalloproteinases (TIMP) and plasminogen activator inhibitor-1, preceded the onset of ECM-degrading proteinase expression and was detected by day two of involution, and showed a sharp peak of expression centered on days 4-6 of involution. When involution was accelerated by decreasing lactation to 2 d, there was an accelerated loss of beta-casein expression evident by day four and a shift in expression of ECM-remodeling proteinases and inhibitors to a focus at 2-4 d of involution. To further extend the correlation between mammary-specific function and ECM remodeling we initiated involution by sealing just one gland in an otherwise hormonally sufficient lactating animal. Alveoli in the sealed gland contained casein for at least 7 d after sealing, and closely resembled those in a lactating gland. The relative expression of TIMP in the sealed gland increased, whereas the expression of stromelysin was much lower than that of a hormone-depleted involuting gland, indicating that the higher the ratio of TIMP to ECM-degrading proteinases the slower the process of involution. To test directly the functional role of ECM-degrading proteinases in the loss of tissue-specific function we artificially perturbed the ECM-degrading proteinase-inhibitor ratio in a normally involuting gland by maintaining high concentrations of TIMP protein with the use of surgically implanted slow-release pellets. In a concentration-dependent fashion, involuting mammary glands that received TIMP implants maintained high levels of casein and delayed alveolar regression. These data suggest that the balance of ECM-degrading proteinases and their inhibitors regulates the organization of the basement membrane and the tissue-specific function of the mammary gland.(ABSTRACT TRUNCATED AT 400 WORDS)

Control of mammary epithelial differentiation: basement membrane induces tissue-specific gene expression in the absence of cell-cell interaction and morphological polarity.

Functional differentiation in mammary epithelia requires specific hormones and local environmental signals. The latter are provided both by extracellular matrix and by communication with adjacent cells, their action being intricately connected in what appears to be a cascade of events leading to milk production. To distinguish between the influence of basement membrane and that of cell-cell contact in this process, we developed a novel suspension culture assay in which mammary epithelial cells were embedded inside physiological substrata. Single cells, separated from each other, were able to assimilate information from a laminin-rich basement membrane substratum and were induced to express beta-casein. In contrast, a stromal environment of collagen I was not sufficient to induce milk synthesis unless accompanied by cell-cell contact. The expression of milk proteins did not depend on morphological polarity since E-cadherin and alpha 6 integrin were distributed evenly around the surface of single cells. In medium containing 5 microM Ca2+, cell-cell interactions were impaired in small clusters and E-cadherin was not detected at the cell surface, yet many cells were still able to produce beta-casein. Within the basement membrane substratum, signal transfer appeared to be mediated through integrins since a function-blocking anti-integrin antibody severely diminished the ability of suspension-cultured cells to synthesize beta-casein. These results provide evidence for a central role of basement membrane in the induction of tissue-specific gene expression.

Modulation of secreted proteins of mouse mammary epithelial cells by the collagenous substrata.

It has been shown previously that cultures of mouse mammary epithelial cells retain their characteristic morphology and their ability to produce gamma-casein, a member of the casein gene family, only if they are maintained on floating collagen gels (Emerman, J.T., and D.R. Pitelka, 1977, In Vitro, 13:316-328). In this paper we show: (a) Cells on floating collagen gels secrete not only gamma-casein but also alpha 1-, alpha 2-, and beta-caseins. These are not secreted by cells on plastic and are secreted to only a very limited extent by cells on attached collagen gels. (b) The floating collagen gel regulates at the level of synthesis and/or stabilization of the caseins rather than at the level of secretion alone. Contraction of the floating gel is important in that cells cultured on floating glutaraldehyde cross-linked gels do not secrete any of the caseins. (c) The secretion of an 80,000-mol-wt protein, most probably transferrin, and a 67,000-mol-wt protein, probably butyrophilin, a major protein of the milk fat globule membrane are partially modulated by substrata. However, in contrast to the caseins, these are always detectable in media from cells cultured on plastic and attached gels. (d) Whey acidic protein, a major whey protein, is actively secreted by freshly isolated cells but is secreted in extremely limited quantities in cultured cells regardless of the nature of the substratum used. alpha-Lactalbumin secretion is also decreased significantly in cultured cells. (e) A previously unreported set of proteins, which may be minor milk proteins, are prominently secreted by the mammary cells on all substrata tested. We conclude that while the substratum profoundly influences the secretion of the caseins, it does not regulate the expression of every milk-specific protein in the same way. The mechanistic implications of these findings are discussed.

Primary avian tendon cells in culture. An improved system for understanding malignant transformation.

Primary avian tendon (PAT) cells which maintain their differentiated state in culture are rapidly transformed by Rous sarcoma virus. By criteria of morphology, increased rate of 2-deoxyglucose uptake, and loss of density dependent growth control, PAT cells transform as well as their less differentiated counterpart, chick embryo fibroblasts. In addition, the percentage of collagen produced by PAT cells drops on transformation by an order of magnitude, from 23 to 2.5%, but is unaffected by viral replication of a transformation-defective mutant. The responsiveness of normal and transformed PAT cells to various environmental factors changes dramatically upon transformation. Normal PAT cells respond to the presence of ascorbate and high cell density by raising the level of collagen synthesis from 5 to 23%. Transformed PAT cells are totally unresponsive. These and previously reported results lead us to postulate that the break-down in the normal regulatory mechanisms used by the cell to maintain the differentiated state is related to or is responsible for the onset of malignant transformation.

The tyrosine phosphorylation substrate p36 is developmentally regulated in embryonic avian limb and is induced in cell culture.

The 36-kD protein-tyrosine kinase substrate p36 has been variously postulated to be involved in membrane-cytoskeletal interactions, membrane traffic, and the regulation of phospholipase A2, and its phosphorylation may play some role in malignant transformation by avian sarcoma viruses. Because embryonic tissues are resistant to transformation by avian sarcoma viruses, we have examined the expression of p36 in the developing avian embryonic limb. The level of p36 increased progressively from day 5 to day 14 of development. It was largely absent from day-5 mesenchyme, and was induced during the differentiation of mesenchymal cells into connective tissue and cartilage, but was not induced in differentiating muscle. In contrast, p36 was detected in ectodermal cells at all developmental stages examined. When day-5 limbs were dissociated and cultured, p36 was induced in all adherent cells, beginning at 2-4 h after plating, and reaching levels comparable to those observed with intact day-14 limb tissue within 48 h. The accumulation of p36 in culture was dependent on substratum adherence, suggesting that its stability is regulated by cell attachment or spreading. These findings are consistent with a structural or mechanical role for p36.

Epimorphin functions as a key morphoregulator for mammary epithelial cells.

Hepatocyte growth factor (HGF) and EGF have been reported to promote branching morphogenesis of mammary epithelial cells. We now show that it is epimorphin that is primarily responsible for this phenomenon. In vivo, epimorphin was detected in the stromal compartment but not in lumenal epithelial cells of the mammary gland; in culture, however, a subpopulation of mammary epithelial cells produced significant amounts of epimorphin. When epimorphin-expressing epithelial cell clones were cultured in collagen gels they displayed branching morphogenesis in the presence of HGF, EGF, keratinocyte growth factor, or fibroblast growth factor, a process that was inhibited by anti-epimorphin but not anti-HGF antibodies. The branch length, however, was roughly proportional to the ability of the factors to induce growth. Accordingly, epimorphin-negative epithelial cells simply grew in a cluster in response to the growth factors and failed to branch. When recombinant epimorphin was added to these collagen gels, epimorphin-negative cells underwent branching morphogenesis. The mode of action of epimorphin on morphogenesis of the gland, however, was dependent on how it was presented to the mammary cells. If epimorphin was overexpressed in epimorphin-negative epithelial cells under regulation of an inducible promoter or was allowed to coat the surface of each epithelial cell in a nonpolar fashion, the cells formed globular, alveoli-like structures with a large central lumen instead of branching ducts. This process was enhanced also by addition of HGF, EGF, or other growth factors and was inhibited by epimorphin antibodies. These results suggest that epimorphin is the primary morphogen in the mammary gland but that growth factors are necessary to achieve the appropriate cell numbers for the resulting morphogenesis to be visualized.

Rescue of mammary epithelial cell apoptosis and entactin degradation by a tissue inhibitor of metalloproteinases-1 transgene.

We have used transgenic mice overexpressing the human tissue inhibitor of metalloproteinases (TIMP)-1 gene under the control of the ubiquitous beta-actin promoter/enhancer to evaluate matrix metalloproteinase (MMP) function in vivo in mammary gland growth and development. By crossing the TIMP-1 transgenic animals with mice expressing an autoactivating stromelysin-1 transgene targeted to mammary epithelial cells, we obtained a range of mice with genetically engineered proteolytic levels. The alveolar epithelial cells of mice expressing autoactivating stromelysin-1 underwent unscheduled apoptosis during late pregnancy. When stromelysin-1 transgenic mice were crossed with mice overexpressing TIMP-1, apoptosis was extinguished. Entactin (nidogen) was a specific target for stromelysin-1 in the extracellular matrix. The enhanced cleavage of basement membrane entactin to above-normal levels was directly related to the apoptosis of overlying mammary epithelial cells and paralleled the extracellular MMP activity. These results provide direct evidence for cleavage of an extracellular matrix molecule by an MMP in vivo.

Matrix metalloproteinase stromelysin-1 triggers a cascade of molecular alterations that leads to stable epithelial-to-mesenchymal conversion and a premalignant phenotype in mammary epithelial cells.

Matrix metalloproteinases (MMPs) regulate ductal morphogenesis, apoptosis, and neoplastic progression in mammary epithelial cells. To elucidate the direct effects of MMPs on mammary epithelium, we generated functionally normal cells expressing an inducible autoactivating stromelysin-1 (SL-1) transgene. Induction of SL-1 expression resulted in cleavage of E-cadherin, and triggered progressive phenotypic conversion characterized by disappearance of E-cadherin and catenins from cell-cell contacts, downregulation of cytokeratins, upregulation of vimentin, induction of keratinocyte growth factor expression and activation, and upregulation of endogenous MMPs. Cells expressing SL-1 were unable to undergo lactogenic differentiation and became invasive. Once initiated, this phenotypic conversion was essentially stable, and progressed even in the absence of continued SL-1 expression. These observations demonstrate that inappropriate expression of SL-1 initiates a cascade of events that may represent a coordinated program leading to loss of the differentiated epithelial phenotype and gain of some characteristics of tumor cells. Our data provide novel insights into how MMPs function in development and neoplastic conversion.

Extracellular matrix regulates whey acidic protein gene expression by suppression of TGF-alpha in mouse mammary epithelial cells: studies in culture and in transgenic mice.

Whey acidic protein (WAP) is an abundant rodent milk protein. Its expression in mouse mammary epithelial cell cultures was previously found to require the formation of an extracellular matrix (ECM)-induced three-dimensional alveolar structure. In the absence of such structures, cells were shown to secrete diffusible factors leading to suppression of WAP expression. We demonstrate here that (a) TGF-alpha production and secretion by mammary cells is downregulated by the basement membrane-dependent alveolar structure, and (b) compared with beta-casein, WAP expression is preferentially inhibited both in culture and in transgenic mice when TGF-alpha is added or overexpressed. Thus, (c) the enhanced TGF-alpha production when cells are not in three-dimensional structures largely accounts for the WAP-inhibitory activity found in the conditioned medium. Since this activity can be abolished by incubating the conditioned medium with a function blocking antibody to TGF-alpha. The data suggest that ECM upregulates WAP by downregulating TGF-alpha production. We also propose that changes in TGF-alpha activity during mouse gestation and lactation could contribute to the pattern of temporal expression of WAP in the gland. These results provide a clear example of cooperation among lactogenic hormones, ECM, and locally acting growth factors in regulation of tissue-specific gene expression.

Extracellular matrix regulates expression of the TGF-beta 1 gene.

Transforming growth factor-beta (TGF-beta) is a potent regulator of cell proliferation and modulates the interactions of cells with their extracellular matrix (ECM), in part by inducing the synthesis of various ECM proteins. Three different isoforms of TGF-beta are synthesized in a defined pattern in specific cell populations in vivo. In the specific case of TGF-beta 1, this well-defined and limited expression stands in sharp contrast to its synthesis by virtually all cells in culture. Using mammary epithelial cells as a model system, we evaluated the substratum dependence of the expression of TGF-beta 1. The level of TGF-beta 1 expression is high in cells on plastic, but is strongly downregulated when cells are cultured on a reconstituted basement membrane matrix. In contrast, TGF-beta 2 mRNA levels in cells on either substratum remain unchanged. Using the chloramphenicol acetyl transferase gene as reporter gene under the control of the TGF-beta 1 promoter, we show that transcription from this promoter is suppressed when the cells are in contact with either endogenously synthesized or exogenously administered basement membrane. TGF-beta 1 promoter activity is strongly induced by the absence of basement membrane, i.e., by direct contact of the cells with plastic. This modulation of transcription from the TGF-beta 1 promoter occurs in the absence of lactogenic hormones which allow full differentiation. Our results thus indicate that basement membrane is an important regulator of TGF-beta 1 synthesis, and explain why most cells in culture on plastic express TGF-beta 1 in contrast with the more restricted TGF-beta 1 synthesis in vivo. We propose that there is a feedback loop whereby TGF-beta 1-induced synthesis of basement membrane components is repressed once a functional basement membrane is present. Finally, these results together with our current knowledge of regulation of TGF-beta 1 and TGF-beta 2 synthesis, suggest that, in vivo, TGF-beta 1 may play a major role in regulating the ECM synthesis and the cell-ECM interactions, whereas TGF-beta 2 may be more important in morphogenetic processes.

Epimorphin mediates mammary luminal morphogenesis through control of C/EBPbeta.

We have shown previously that epimorphin (EPM), a protein expressed on the surface of myoepithelial and fibroblast cells of the mammary gland, acts as a multifunctional morphogen of mammary epithelial cells. Here, we present the molecular mechanism by which EPM mediates luminal morphogenesis. Treatment of cells with EPM to induce lumen formation greatly increases the overall expression of transcription factor CCAAT/enhancer binding protein (C/EBP)beta and alters the relative expression of its two principal isoforms, LIP and LAP. These alterations were shown to be essential for the morphogenetic activities, since constitutive expression of LIP was sufficient to produce lumen formation, whereas constitutive expression of LAP blocked EPM-mediated luminal morphogenesis. Furthermore, in a transgenic mouse model in which EPM expression was expressed in an apolar fashion on the surface of mammary epithelial cells, we found increased expression of C/EBPbeta, increased relative expression of LIP to LAP, and enlarged ductal lumina. Together, our studies demonstrate a role for EPM in luminal morphogenesis through control of C/EBPbeta expression.

Reversion of the malignant phenotype of human breast cells in three-dimensional culture and in vivo by integrin blocking antibodies.

In a recently developed human breast cancer model, treatment of tumor cells in a 3-dimensional culture with inhibitory beta1-integrin antibody or its Fab fragments led to a striking morphological and functional reversion to a normal phenotype. A stimulatory beta1-integrin antibody proved to be ineffective. The newly formed reverted acini re-assembled a basement membrane and re-established E-cadherin-catenin complexes, and re-organized their cytoskeletons. At the same time they downregulated cyclin D1, upregulated p21(cip,wat-1), and stopped growing. Tumor cells treated with the same antibody and injected into nude mice had significantly reduced number and size of tumors in nude mice. The tissue distribution of other integrins was also normalized, suggesting the existence of intimate interactions between the different integrin pathways as well as adherens junctions. On the other hand, nonmalignant cells when treated with either alpha6 or beta4 function altering antibodies continued to grow, and had disorganized colony morphologies resembling the untreated tumor colonies. This shows a significant role of the alpha6/beta4 heterodimer in directing polarity and tissue structure. The observed phenotypes were reversible when the cells were disassociated and the antibodies removed. Our results illustrate that the extracellular matrix and its receptors dictate the phenotype of mammary epithelial cells, and thus in this model system the tissue phenotype is dominant over the cellular genotype.