Epigenetics, Cancer and the Microenvironment - Springer Link

P1: VENDOR/GCQ/LOV/FHX

P2: GCO/LOV

Tally: FTK

Journal of Mammary Gland Biology and Neoplasia (JMGBN)

PP133-300576

April 10, 2001

9:32

Style file version Nov. 07, 2000

Journal of Mammary Gland Biology and Neoplasia, Vol. 6, No. 2, 2001

It Takes a Tissue to Make a Tumor: Epigenetics, Cancer and the Microenvironment Mary Helen Barcellos-Hoff

How do normal tissues limit the development of cancer? This review discusses the evidence that normal cells effectively restrict malignant behavior, and that such tissue forces must be subjugated to establish a tumor. The action of ionizing radiation will be specifically discussed regarding the disruption of the microenvironment that promotes the transition from preneoplastic to neoplastic growth. Unlike the highly unpredictable nature of genetic mutations, the response of normal cells to radiation damage follows an epigenetic program similar to wound healing and other damage responses. Our hypothesis is that the persistent disruption of the microenvironment in irradiated tissue compromises its ability to suppress carcinogenesis. KEY WORDS: Carcinogenesis; epigenetics; mammary; ionizing radiation; stroma; tissue.

INTRODUCTION

unraveling the subterfuges by which tumors secure the support of normal cells can be the basis for strategies that strangle tumor growth. Three decades ago, Pierce proposed that cancer was a problem of developmental biology, and was among the first to propose that how the genome is controlled in cancer is as important as genomic aberrations (2). Further studies directed toward understanding the molecular basis of the concepts raised by Pierce and others continue to challenge cancer researchers today. Additional evidence, though often not acknowledged, indicates that normal tissue interactions actively inhibit cancer formation. How do normal tissue interactions control aberrant behavior? Probably not by altering the genome of the initiated cell. As in society, the most effective tactics are “social norms” and “peer pressure.” Various extracellular signals, which are poorly defined at this point, press upon every cell in the body, dictating its behavior and function. Clearly such epigenetic signals are dominant over genetic scripts given that humans contain but one genome, and yet are composed of at least 300 distinct cell types. For example, epithelia are “specified” via inductive interactions with stroma (3, 4). Further, the recent success in cloning entire animals from mature epithelial cell nuclei and the remarkable plasticity of cell fate evidenced by multipotent stem cells support

In that all tissues are complex communities of cooperating cells that perform distinct functions, the emphasis of recent cancer research on events occurring in single cells, such as mutations and growth regulation, may be misplaced for understanding the development of the disease. Instead, we can understand the factors that control single cells so that we understand how cancer develops from individual cells into multicellular, heterogeneous tumors. The evolution of multicellular organisms into purpose-specific tissues is obtained through differential expression of the genome. Cells receive information about how they should behave from their microenvironment, which consists of other cells, insoluble extracellular matrix proteins, soluble hormones and cytokines (1). Thus, the recent cell-centric view in cell biology has begun to shift toward understanding the integration of multiple extrinsic signals. This view is paralleled in cancer research with a new appreciation that tumor formation requires the complicity of normal cells. As a result

1

Life Sciences Division, Bldg. 74-174, 1 Cyclotron Rd., Lawrence Berkeley National Laboratory, Berkeley, California 94720. e-mail: [email protected]

213 C 2001 Plenum Publishing Corporation 1083-3021/01/0400-0213$19.50/0 °


P2: GCO/LOV

Tally: FTK


PP133-300576

April 10, 2001

214 the contention that phenotype is as much a function of environment as an expression of genotype (5). Finally, the recent notion that cells live or die by virtue of the presence of extrinsic survival signals from extracellular proteins and growth factors, suggests that there is little intrinsic ‘will to live’ attributable to the cell per se (6). Given the overwhelming evidence, which is beyond the scope of this review, that cell behavior is a consequence of the microenvironment, what, if anything, precludes these influences from affecting the development of cancer?

HOW DO CELLS BECOME CANCERS? Three stages of tumor development, i.e., initiation, promotion and progression, have been functionally defined in experimental animals. This multistep model of carcinogenesis defines cancer initiation as genomic change, and promotion as the series of events leading to proliferation of initiated cells. Agents that promote carcinogenesis are generally thought to increase the probability that a cell will acquire the additional mutations necessary for neoplastic progression (7–9). The questions that have dominated the literature for many years are: how many, and which, genetic changes are necessary for a cell to become a tumor? Nevertheless, in recent years, interdependence between genetic change and complex tumor microenvironments has been incorporated into this model. Kinsler and Vogelstein have classified the genetic changes as those that monitor growth (“gatekeepers”), assisted by genes that indirectly suppress neoplasia by regulating genomic stability (“caretakers”). It is well-recognized that tumors can recruit normal cells (e.g., endothelial cells) and induce changes in their microenvironments that are conducive to growth (10, 11). Thus they also recognize that enabling genes, which they call “landscapers,” may affect nontarget cells (12, 13). In their elegant definition of the genetic hallmarks of cancer, Hanahan and Weinberg (14) enumerated seven acquired capabilities that include mutations that affect co-opted normal cells, such as the production of vascular endothelial growth factor to promote angiogenesis (10). Despite the emphasis on identifying oncogenic mutations, Hanahan and Weinberg conclude by acknowledging that cancer cells do not exist in isolation and that tumors are complex collaborations between multiple cell types. Hence, their conclusion that normal cells within tumors are not idle bystanders.

9:32


Barcellos-Hoff Indeed, as discussed later, normal cells are active participants that shape the frequency and features of tumors. If the development of cancer is not limited to genetic missteps, then rather than asking how cells become cancers, a more precise question may be how do multicellular tissues become tumors?

NORMAL CELLS EXERT EPIGENETIC CONTROL ON NEOPLASTIC BEHAVIOR Quantitative studies in rodent models demonstrate that the number of cells initiated following either physical or chemical carcinogen exposure, far exceeds the number of tumors that develop in vivo (15–17). What then prevents us all from developing cancer, rather than the 1/8 incidence that spans 70+ years and the production of 1018 cells? A variety of studies suggest that expansion of an initiated population is actively opposed/suppressed by normal cells. This effect can be clearly shown in cell culture by following the frequency of “initiation” using the morphological and behavioral benchmark of neoplastic behavior called transformation. In a series of thoughtful studies, Bauer and colleagues established that the ability of radiation, chemicals and virus to transform human and rodent fibroblasts is actively mediated by the nontransformed cells in a culture [reviewed in (18)]. Bauer and colleagues found that nontransformed cells induce the selective ablation of transformed cells via apoptosis triggered in part by cytokines induced by nontransformed neighboring cells (19). If this control system acts in vivo as efficiently as it does in vitro, tumor formation should require the establishment of resistance mechanisms directed against intercellular induction of apoptosis. Indeed when cells from established tumors were tested for inhibition by normal cells in culture, they failed to be influenced (20). Studies by Terzaghi-Howe demonstrate the impact of normal cells on epithelial cell transformation. In a process called normalization, conditioned medium from normal tracheal epithelial cells induced highly malignant rat tracheal carcinoma cells to undergo dramatic changes in morphology accompanied by a loss of anchorage independent growth (21). Differential mRNA display of the malignant cells exposed to conditioned media from normal cells identified three classes of response that mediated reversion: cell adhesion, signal transduction and gene transcription and translation (22). Cell adhesion in particular is critical to the ability of cells to behave in a


P2: GCO/LOV

Tally: FTK


PP133-300576

April 10, 2001

9:32


Epigenetics, the Microenvironment and Cancer tissue-specific fashion and has recently been identified as a modulator of the neoplastic behavior of human breast cells (23, 24). Conversely, the therapeutic benefit of α-interferon in chronic myeloid leukemia is due in part to the re-establishment of cell-adhesion signals (25). The well-characterized effects and consequences of ionizing radiation in biological systems offer another example. Ionizing radiation is an established carcinogen in many tissues of both humans and animals. It has a well-defined physical basis for action, a statistical probability of total and specific events and well-studied mutagenic capacity, all of which are generally thought to be due to irreversible damage at the time of the irradiation. Studies of the carcinogenic potential of ionizing radiation have commonly focused on the initial DNA damage, which, if improperly repaired, can result in mutations or chromosome damage, some of which may lead to neoplastic transformation, others to cell death. Consequently, the nucleus is usually considered to be the major target of ionizing radiation damage. If ionizing radiation damages individual cells, one might argue that radiation response is the sum of individual cell responses. The hypothesis that “radiation induces DNA damage, DNA damage induces mutations that cause cancer” is challenged by a variety of experimental findings [reviewed in (26)]. Little and colleagues showed that irradiated cells transmit genetic instability to their progeny by a nonmutational mechanism, and that cytoplasmic irradiation can result in increased mutations. Genomic instability is a phenomenon in which cells exhibit a heritable high frequency of genetic or chromosomal aberrations. Recent evidence indicates that cells surviving irradiation produce progeny exhibiting genomic instability at a frequency that is too high to be due to conventional mutational changes (27). In addition, it has recently been recognized that neighbors of irradiated cells respond as if they too were exposed to radiation as evidenced by the induction of so-called stress proteins. This response has been shown to involve cell-cell contact (28) and soluble signals (29). These observations lead to the conclusion that the action of radiation as a carcinogen has many effects that cannot be readily explained by its ability to disrupt bonds within DNA. In reality, our understanding of the mechanistic processes involved in initiation, transformation or maintenance of the transformed phenotype over many generations of cell replication is incomplete. When cell number and culture conditions, including the presence of nontransformed cells, are manip-

215 ulated, transformation is clearly as much a matter of selection as of initiation (30). Such data indicate that transformation, reflecting the cell culture equivalent of initiation, may be a more complex event than previously thought (31); therefore, one may question whether we really understand the nature of the ‘initiating’ events in carcinogenesis. The forces that inhibit or stimulate expression of tumorigenic potential are probably more critical in determining cancer frequency than initiation, if initiation is a mutational event (32, 33).

HOW DO MULTICELLULAR TISSUES BECOME TUMORS? Specialized microenvironments, composed of insoluble extracellular matrix and soluble growth factors play a pivotal role in normal tissue development and function (1). That such interactions can efficiently suppress the expression of the neoplastic phenotype has been shown in a variety of models (17, 34, 35). Perhaps the best recognized are the experiments discussed by Pierce in which carcinoma cells are induced to ‘normalize’ by virtue of their placement within developing embryos (2). Despite the presence of genetic sequence alterations, these cells respond to the overwhelming influence of the microenvironment and their neighbors to behave appropriately. Pierce likened this effect to the process of differentiation that occurs in normal tissues. Pierce also proposed the corollary that carcinogenesis is a caricature of normal development. In abnormal tissues, the influence of normal cells on neoplastic behavior can be compromised since tissue pathology arises from fundamental disruption of orchestrated communication between cells and among different cell types. Examples of abnormal stromal/epithelial interactions that enhance the ability of cells to express the neoplastic phenotype are discussed later. That experimental manipulation of the microenvironment, rather than the target cells, stimulates tumorigenesis is further evidence that neoplastic potential is highly responsive to tissue factors. Schor and colleagues postulated that in some familial breast cancers the inherited mutation may disrupt stromal/epithelial interactions, which then provide a stimulus for initiated cells to exhibit more aggressive neoplastic behaviors (36, 37). This hypothesis arose from the identification of fetal-like migratory characteristics in tumor fibroblasts. Surprisingly, skin fibroblasts also showed this phenotype in familial


P2: GCO/LOV

Tally: FTK


PP133-300576

April 10, 2001

216 but not spontaneous breast cancer patients (38). In addition, the normal appearing tissue next to breast cancer also displays such changes (39). The frequency of first-degree relatives who exhibit the altered skin fibroblast phenotype is consistent with an inherited trait (40). Thus the increased probability of cancer occurrence in these individuals is postulated to be due not to the probability of acquiring mutations in the epithelium, but to an increased potential for their establishment due to the presence of an abnormal stroma. Experimental manipulations also support the crucial role of stromal interactions. Mammary preneoplastic nodules produce more tumors when the tissue is completely disaggregated before transplantation, suggesting that neoplastic behavior is suppressed by the remnants of tissue architecture occurring in tissue fragments (41). When adult mouse mammary epithelium is combined with salivary gland mesenchyme, it not only undergoes ductal branching patterns typical of salivary gland (42), but also shows a greater incidence of tumors (43). A role for an abnormal stroma early in neoplastic progression has also been suggested in hematopoeitic malignancies resulting from misregulation of adhesive properties by diseased or genetically aberrant stromas (44). Wounding can act as a promoter, apparently by creating a favorable microenvironment for proliferation, a prerequisite for wound repair [reviewed in (45)]. Experiments in the 1960s by Fisher and colleagues showed that tumors metastasize preferentially to wound sites in parabiotic animals injected with invasive tumor cells (46). Experiments performed with Rous sarcoma virus showed that tumors formed preferentially at sites of injections or at distant wounds (47). Several transgenic oncogene models show preferential tumorigenesis at wound sites (48). Experimental animal models have shown that carcinogenesis enhanced by the activated stroma induced by wounding (47), overexpression of platelet-derived growth factor (11), or misregulation of stromelysin (49, 50). Several studies support the view that radiation exposure compromises tissue integrity by altering the flow of information among cells (51). Our studies in mammary gland showed that irradiated tissue undergoes rapid remodeling of the microenvironment characterized by changes in extracellular matrix and activation of latent transforming growth factor β1 (TGF-β1)2 (52, 53). We established that 2

Abbreviations: transforming growth factor-β1 (TGF-β).

9:32


Barcellos-Hoff these events were functionally related by showing that microenvironment remodeling was blocked by treating animals with TGF-β neutralizing antibodies before irradiation (54). These experiments provided functional confirmation of the hypothesis that radiation induces TGF-β activation and implicate TGF-β as a mediator of tissue response to ionizing radiation. Aspects of radiation-induced remodeling parallel events in dermal wound healing (55), thus the effects of ionizing radiation on tissue microenvironment may be similar to effects of other agents that elicit an activated stroma that in turn fosters neoplastic behavior. But understanding whether such responses are a net benefit to the reestablishment of homeostasis or are detrimental and contribute to radiation late effects is complicated by multicellular interactions. If the unit of function is taken into account, i.e., tissue, it becomes evident that many of these events are likely directed to the good of the whole, rather than the part, i.e., the cell. Based on the known carcinogenic risk of radiation exposure, the dependence of cells on extracellular signaling and the rapid remodeling observed in irradiated tissue, we asked whether microenvironment remodeling contributes to radiogenic carcinogenesis (56). We evaluated the effect of an irradiated mammary stroma on the neoplastic potential of unirradiated, p53-mutant COMMA-D mammary epithelial cells transplanted to mammary fat pads previously cleared of epithelium in syngeneic hosts (57). Tumor incidence was increased four-fold when animals were irradiated prior to transplantation compared with sham-irradiated hosts. Furthermore, tumors were significantly larger and arose more quickly in fat pads in irradiated hosts. Since tumors formed only in fat pads on the irradiated side of hemi-body irradiated animals, the influence of the irradiated tissue dominated over systemic effects. These data indicate that radiation-induced changes in the stromal microenvironment can contribute to neoplastic progression in vivo. We postulated that radiation-induced microenvironments are evidence of an additional class of carcinogenic action, distinct from those leading to mutations or proliferation (56). Greenberger and colleagues proposed a model of indirect γ -irradiation leukemogenesis based on co-cultures of heavily irradiated bone marrow stromal cell lines that selectively bound M-CSF receptor positive unirradiated hematopoietic progenitor cells resulting in selection of tumorigenic subclones [reviewed in (58)]. Additional evidence that radiation effects on stroma alter


P2: GCO/LOV

Tally: FTK


PP133-300576

Epigenetics, the Microenvironment and Cancer the behavior of neoplastic cells comes from studies of tumor bed effect, in which stroma that is heavily irradiated before tumor transplantation inhibits tumor growth but fosters metastatic behavior (59). In one sense, these experiments suggest that radiationinduced bystander phenomena exist in vivo; that is, products of irradiated cells can significantly alter the phenotype of unirradiated cells. Such studies support the conclusion that radiation has global and persistent consequences on stromal function, consequences that in turn can influence the expression of neoplastic potential. Recent studies by Bemis and Schedin suggest that even physiological changes in the extracellular matrix of mammary gland can modulate the neoplastic behavior of breast cancer cells (60). This principle is also evident in chemical carcinogenesis. Hodges and colleagues showed that cultured carcinogen-treated stroma recombined with normal bladder epithelium produce neoplastic changes in epithelial morphology (61). Zarbl and colleagues found that Hras1 gene mutations in mammary tumors from N-nitroso-N-methylurea treated rats arose from cells with preexisting Hras1 mutations that had occurred during early development (62). Thus, although clearly mutagenic in its own right, N-nitroso-N-methylurea exposure apparently led to the expansion and neoplastic progression of Hras 1mutation containing populations. Carcinogen-induced microenvironments are not necessarily mutagenic or mitogenic per se. Rather, changes in the microenvironment may promote neoplastic behavior by disrupting normal cell functions regulated through cell-cell contact, cell-matrix interactions and growth factor signaling. Thus, if ionizing radiation induces a microenvironment that modifies restrictive interactions, then it may promote the malignant phenotype in a way that is functionally equivalent to the acquisition of additional mutations in the initiated cell. Alternatively, the microenvironment elicited by carcinogen exposure could create novel selective pressures that would affect the features of a developing tumor. Disruption of solid tissue interactions is a previously unrecognized activity of radiation as a carcinogen, and a novel avenue through which to explore new strategies for intervening in the neoplastic process. DISRUPTION OF MULTICELLULAR RESPONSES: TGF-β When injury, such as inflammation, UV and ionizing radiation, leaves tissue intact, what molecular

April 10, 2001

9:32


217 mechanisms record or ‘sense’ the damage? Intracellular mechanisms for sensing DNA damage are thought to dictate individual cell responses, such as growth delay and DNA repair. If this is true, the response of a given cell type to a given dose of radiation should be invariant since the amount of DNA damage is due to the deposited energy. However, the presence of cytokines and nature of the substratum appear to dictate whether the cell lives or dies, proliferates or stops (63, 64). Thus it seems unlikely that damage sensors operate only within cells. Furthermore, damaged cells incapable of contributing to the repair/reconstitution of tissue are frequently eliminated via apoptosis to ensure maintenance of tissue integrity (65). Is this solely based on cellular fate decisions following damage? Or, might one postulate the existence of tissue-level sensors that have evolved to register tissue damage and produce a signal that will recruit nondamaged cells to facilitate tissue recovery. The flow of information both locally between cells in tissues and distantly between organs is mediated in large part by cytokines (66). TGF-β is one such critical factor that orchestrates multicellular responses to damage via effects on proliferation, apoptosis, extracellular matrix composition, growth factor production, chemotaxis and immune function (67, 68). Although TGF-β accumulates at wound sites, healing is improved if its effects are attenuated; thus, it appears to have evolved to quickly repair, rather than restore, tissue integrity (69, 70). Such observations have lead to the notion that TGFβ is poised for an exuberant response that can also become a liability. Our studies have shown that ionizing radiation elicits rapid and persistent TGF-β activation (54). Other DNA damaging agents such as cis-platinum (71) and alkylating agents (72) also induce TGF-β activity. In the cultured fibroblast transformation model, Bauer et al. have found three distinct, but competing, roles for TGF-β (73). Although TGF-β helps maintain the transformed state, it also enables nontransformed neighbors to recognize the transformed cells, triggers an apoptosis-inducing response. If injected directly into distal sites in the Rous sarcoma model, TGF-β promotes tumor formation in a manner similar to wounding (74). The postulated action of TGFβ in sarcoma formation is consistent with the role of a tumor promoter in that it stimulates mesenchymal proliferation (75). In the experiments by Terzaghi-Howe, TGF-β produced by the differentiated normal epithelial cells


P2: GCO/LOV

Tally: FTK


PP133-300576

April 10, 2001

9:32


218 inhibited carcinogen-altered cells (21). However, if TGF-β is an epithelial growth inhibitor, it appears paradoxical that it is both elevated in and promotes the growth of breast cancer (76). The paradox arises from assuming a single effect of this pleiotropic growth factor. Normally its multicellular effects are coordinated, as in wound healing, but misregulation leads to significant alterations in tissue dynamics. These competing effects are observed in transgenic mouse models that have targeted expression of a constitutively active TGF-β. Epidermally-targeted overexpression of TGF-β inhibited the establishment of benign early skin tumors following carcinogen exposure, and likewise MMTV-promoter driven expression of TGF-β suppressed the formation of mammary tumors induced by transgenic induction of an oncogene (77). Nevertheless, in skin, continued exposure to TGF-β increased the frequency with which benign tumors converted to malignant spindle cell carcinomas (78, 79). In a similar model, genetic inactivation of TGF-β signaling by overexpression of a dominant/negativeTGF-β type II receptor in the mouse epidermis also accelerated tumor progression with a six-fold increase in the frequency of conversion from benign papillomas to carcinomas (80). A spindle cell carcinoma reflects an epithelial to mesenchymal transition that occurs during normal development but also can be induced by TGF-β to produce invasive metastatic carcinoma cells [reviewed in (81)]. Nonetheless carcinogen treatment of TGF-β1 null heterozygotes reveals that TGF-β insufficiency promotes tumorigenesis (82). Together these data suggest that TGF-β action is two-edged. TGF-β can suppress the establishment of tumorigenesis but its continued elevation, or perturbation of its signaling, promotes malignant behavior. The latter effect is seen in recent experiments demonstrating that cyclosporin A, an inhibitor of transplant rejection, promotes invasive behavior of A549 lung carcinoma cells via the induction of TGF-β (83). A most intriguing finding is that the cultured keratinocytes from TGF-β null animals have elevated genomic instability as indicated by gene amplification assay and the frequency of instability was decreased by addition of low levels of exogenous TGF-β (84). Aneuploidy, chromosome breaks, and malignant transformation of v-ras(Ha) transduced primary TGF-β1 null keratinocytes were also suppressed by exogenous TGF-β. From these studies, the authors conclude that genomic instability is a mechanism that accelerates tumor progression in tumors harboring defects in TGF-β signaling (85).

Barcellos-Hoff It is surprising that an extracellular factor regulates genome integrity, a task usually associated with internal gatekeepers like p53. At present, little is understood of the processes resulting in genomic instability and in the maintenance and transmission of the phenotype over many generations. One might ask how the absence of TGF-β contributes to, or amplifies, genomic instability. Is it due to defective growth regulation, allowing damaged cells to proliferate when they should be blocked in the cell cycle to repair the damage? Or is it possible, as suggested by the transformation studies described earlier, that TGF-β selectively impedes the survival of aberrant cells? The fact that exogenous TGF-β corrected the defect in TGF-β null cells suggests either that genomic instability is a reversible phenotype or that TGF-β regulates the detection of instability, rather than its induction. If this is true, rather than being an induced process in cells, genomic instability may reveal the loss of a multicellular pathway that effectively controls the survival of abnormal cells. CONCLUSIONS Abundant evidence indicates that normal tissues can effectively thwart the establishment of cancer cells at multiple steps and conversely, that abnormal tissues are active participants in the development of cancer. In the past two decades, a great deal of information has accumulated regarding the nature of cellular oncogenetic changes. In the next decade, further understanding of how tissue normalization forces are subjugated during tumor formation could provide the means to modify cancer progression. The promise of this approach is shown by studies showing that reestablishing appropriate interactions of human mammary cancer cells with the substratum can reverse neoplastic behavior even in the presence of grossly abnormal genetic damage (23, 24). Even more exciting is the demonstration that clinical remissions in chronic mylogenous leukemia following treatment with interferon-γ are due to reattachment of the cancer cells to bone marrow stroma (25). Thus understanding how carcinogens interfere with the ability of normal cells to restrain neoplastic behavior will provide new strategies to prevent cancer. ACKNOWLEDGMENTS The author is supported by funding from the NASA Specialized Center of Research and Training


P2: GCO/LOV

Tally: FTK


PP133-300576

Epigenetics, the Microenvironment and Cancer in Radiation Health, California Breast Cancer Research Program BG00-038, and the Office of Biological and Environmental Research, Office of Energy Research of the U.S. Department of Energy.

REFERENCES 1. M. J. Bissell and M. H. Barcellos-Hoff (1987). The influence of extracellular matrix on gene expression: Is structure the message? J. Cell Sci. 8:327–343. 2. G. B. Pierce, R. Shikes, and L. M. Fink (1978). Cancer: A Problem of Developmental Biology. Prentice-Hall, Inc., Englewood Cliffs, New Jersey. 3. C. Grobstein (1967). Mechanism of organogenetic tissue interaction. Natl. Cancer Inst. Monogr. 26:279–299. 4. K. Kratochwil (1969). Organ specificity in mesenchymal induction demonstrated in the embryonic development of the mammary gland of the mouse. Dev. Biol. 20:46–71. 5. E. Fuchs and J. A. Segre (2000). Stem cells: A new lease on life. Cell 100:143–155. 6. G. Evan and T. Littlewood (1998). A matter of life and cell death. Science 281:1317–1322. 7. E. Farber (1984). The multistep nature of cancer development. Cancer Res. 44:4217–4223. 8. R. A. Weinberg (1989). Oncogenes, antioncogenes, and the molecular bases of multistep carcinogenesis. Cancer Res. 49:3713–3721. 9. S. H. Kim, K. A. Roth, A. R. Moser, and J. I. Gordon (1993). Transgenic mouse models that explore the multistep hypothesis of intestinal neoplasia. J. Cell Biol. 123:877–893. 10. A. van den Hoof (1988). Stromal involvement in malignant growth. Adv. Cancer Res. 50:159–196. 11. M. Skobe and N. E. Fusenig (1998). Tumorigenic conversion of immortal human keratinocytes through stromal cell activation. Proc. Natl. Acad. Sci. U.S.A. 95:1050–1055. 12. K. W. Kinzler and B. Vogelstein (1997).Cancer-susceptibility genes. Gatekeepers and caretakers [news; comment]. Nature 386:761, 763. 13. K. W. Kinzler and B. Vogelstein (1998). Landscaping the cancer terrain [comment]. Science 280:1036–1037. 14. D. Hanahan and R. A. Weinberg (2000). The hallmarks of cancer. Cell 100:57–50. 15. M. Terzaghi and P. Nettesheim (1979). Dynamics of neoplastic development in carcinogen-exposed tracheal mucosa. Cancer Res. 39:3004–3010. 16. R. L. Ullrich (1986). The rate of progression of radiationtransformed mammary epithelial cells is enhanced after lowdose-rate neutron irradiation. Rad. Res. 105:68–75. 17. K. Kamiya, J. Yasukawa-Barnes, J. M. Mitchen, M. N. Gould, and K. H. Clifton (1995). Evidence that carcinogenesis involves an imbalance between epigenetic high-frequency initiation and suppression of promotion. Proc. Natl. Acad. Sci. U.S.A. 92:1332–1336. 18. G. Bauer (1996). Elimination of transformed cells by normal cells: A novel concept for the control of carcinogenesis. Histol. Histopathol. 11:237–255. 19. I. Engelmann and G. Bauer (2000). How can tumor cells escape intercellular induction of apoptosis? Anticancer Res. 20:2297– 2306.

April 10, 2001

9:32


219 20. I. Engelmann, H. Eichholtz-Wirth, and G. Bauer (2000). Ex vivo tumor cell lines are resistant to intercellular induction of apoptosis and independent of exogenous survival factors. Anticancer Res. 20:2361–2370. 21. M. Terzaghi-Howe (1986). Inhibition of carcinogen-altered rat tracheal epithelial cell proliferation by normal epithelial cells in vivo. Carcinogenesis 8:145–150. 22. G.-W. Chang and M. Terzaghi-Howe (1998). Multiple changes in gene expression are associated with normal cell-induced modulation of the neoplastic phenotype. Cancer Res. 58:4445– 4452. 23. M. M. Zutter, S. A. Santoro, W. D. Staatz, and Y. L. Tsung (1995). Re-expression of the α2β 1 integrin abrogates malignant phenotype of breast carcinoma cells. Proc. Natl. Acad. Sci. U.S.A. 92:7411–7415. 24. V. M. Weaver, O. W. Petersen, F. Wang, C. A. Larabell, P. Briand, C. Damsky, and M. J. Bissell (1997). Reversion of the malignant phenotype of human breast cells in threedimensional culture and in vivo by integrin blocking antibodies. J.Cell Biol. 137:231–245. 25. R. Bhatia, J. B. McCarthy, and C. M. Verfaillie (1996). Interferon-α restores normal β1 integrin-mediated inhibition of hematopoietic progenitor proliferation by the marrow microenvironment in chronic myelogenous leukemia. Blood 87:3883–3891. 26. J. B. Little (2000). Radiation carcinogenesis. Carcinogenesis 21:397–404. 27. C. L. Limoli, M. I. Kaplan, J. Corcoran, M. Meyers, D. A. Boothman, and W. F. Morgan (1997). Chromosomal instability and its relationship to other end points of genomic instability. Cancer Res. 57:5557–5563. 28. E. I. Azzam, S. M. de Toledo, T. Gooding, and J. B. Little (1998). Intercellular communication is involved in the bystander regulation of gene expression in human cells exposed to very low fluences of α particles. Rad. Res. 150:497–504. 29. B. E. Lehnert and E. H. Goodwin (1997). Extracellular factor(s) following exposure to α particles can cause sister chromatid exchanges in normal human cells. Cancer Res. 57:2164– 2171. 30. M. Chow and H. Rubin (1999). The cellular ecology of progressive neoplastic transformation: A clonal analysis. Proc. Natl. Acad. Sci. U.S.A. 96:2093–2098. 31. J. E. Trosko, C. C. Chang, and B. V. Madhukar (1990). Modulation of intercellular communication during radiation and chemical carcinogenesis. 123:241–251. 32. I. Tomlinson and W. Bodmer (1999). Selection, the mutation rate and cancer: Ensuring that the tail does not wag the dog. Nature Med. 5:11–12. 33. I. P. Tomlinson, M. R. Novelli, and W. F. Bodmer (1996). The mutation rate and cancer. Proc. Natl. Acad. Sci. U.S.A. 93:14800–14803. 34. J. J. Decosse, C. L. Gossens, J. F. Kuzma, and D. Unsworth (1973). Breast cancer: Induction of differentiation by embryonic tissue. Science 181:1057–1058. 35. M. Cooper and H. Pinkus (1977). Intrauterine transplantation of rat basal cell carcinoma as a model for reconversion of malignant to benign growth. Cancer Res. 37:2544–2552. 36. S. L. Schor, A. M. Schor, A. Howell, and J. Haggie (1988). The possible role of abnormal fibroblasts in the phathogenesis of breast cancer. In M. A. Rich, J. C. Hager, and D. M. Lopez, (eds.), Breast Cancer: Scientific and Chemical Progress, Kluwer Academic Publishers, Boston, pp. 142–157.


P2: GCO/LOV

Tally: FTK


PP133-300576

April 10, 2001

220 37. S. L. Schor, A. M. Schor, A. Howell, and D. Crowther (1987). Hypothesis: Persistent expression of fetal phenotypic characteristics by fibroblasts is associated with an increased susceptibility to neoplastic disease. Expl. Cell Biol. 55:11–17. 38. S. L. Schor, A. M. Schor, P. Durning, and G. Rushton (1985). Skin fibroblasts obtained from cancer patients display foetallike migratory behavior on collagen gels. J. Cell Sci. 73:235–244. 39. A. M. Schor, G. Rushton, J. E. Ferguson, A. Howell, J. Redford, and S. L. Schor (1994). Phenotypic heteogeneity in breast fibroblasts: Functional anomaly in fibroblasts from histologically normal tissue adjacent to carcinoma. Int. J. Cancer 59:25–32. 40. J. A. Haggie, S. L. Schor, A. Howell, J. M. Birch, and R. A. S. Sellwood (1987). Fibroblasts from relatives of hereditary breast cancer patients display fetal-like behavior in vitro. Lancet 1:1455–1457. 41. D. Medina, F. Shepherd, and T. Gropp (1978). Enhancement of the tumorigenicity of preneoplastic mammary nodule lines by enzymatic dissociation. JNCI 60:1121–1126. 42. T. Sakakura, Y. Sakagami, and Y. Nishizuka (1979). Persistence of responsiveness of adult mouse mammary gland to induction by embryonic mesenchyme. Dev. Biol. 72:201–210. 43. T. Sakakura, Y. Sakagami, and Y. Nishizuka (1981). Accelerated mammary cancer development by fetal salivary mesenchyme isografted to adult mouse mammary epithelium. J. Natl. Cancer Inst. 66:953–959. 44. M. Y. Gordon, C. R. Dowding, G. P. Riley, J. M. Goldman, and M. F. Greaves (1987). Altered adhesive interactions with marrow stroma of haematopoietic progenitor cells in chronic myeloid leukaemia. Nature 328:342–344. 45. M. H. Sieweke and M. J. Bissell (1994). The tumor promoting effect of wounding: A possible role for TGFβ induced fibrotic stroma. Crit. Rev. Oncogenesis 5:297–311. 46. B. Fisher and E. R. Fisher (1968). Role of host and tumor calcium in metastases. Cancer Res. 28:1753–1758. 47. D. S. Dolberg, R. Holingworth, M. Hertle, and M. J. Bissell (1985). Wounding and its role in RSV-mediated tumor formation. Science 230:676–678. 48. A. C. Schuh, S. F. Keating, F. S. Moneclaro, P. K. Vogt, and M. L. Breitman (1990). Obligatory wounding requirement for tumorigenesis in v-jun transgenic mice. Nature 346:756–760. 49. A. Lochter, S. Galosy, J. Muschler, N. Freedman, Z. Werb, and M. J. Bissell (1997). Matrix metalloproteinase stomelysin-1 triggers a cascade of molecular alterations that leads to stable epithelial-to-mesenchymal conversion and premalignant phentoype in mammary epithelial cells. J. Cell Biol. 139:1861–1872. 50. N. Thomasset, A. Lochter, C. J. Sympson, L. R. Lund, D. R. Williams, O. Behrendtsen, Z. Werb, and M. J. Bissell (1998). Expression of autoactivated stromelysin-1 in mammary glands of transgenic mice leads to a reactive stroma during early development. J. Amer. Sci. 153:457–467. 51. J. E. Trosko (1998). Hierarchcal and cybernetic nature of biologic systems and their relevance to homeostatic adaptation to low-level exposures to oxidative stress-inducing agents. Environ. Health Perspect. 106:331–339. 52. M. H. Barcellos-Hoff (1993). Radiation-induced transforming growth factor β and subsequent extracellular matrix reorganization in murine mammary gland. Cancer Res. 53:3880– 3886. 53. M. H. Barcellos-Hoff, R. Derynck, M. L.-S. Tsang, and J. A.Weatherbee (1994). Transforming growth factor-β activation in irradiated murine mammary gland. J. Clin. Invest. 93:892–899.

9:32


Barcellos-Hoff 54. E. J. Ehrhart, A. Carroll, P. Segarini, M. L.-S. Tsang, and M. H. Barcellos-Hoff (1997). Latent transforming growth factor-β activation in situ: Quantitative and functional evidence following low dose irradiation. FASEB J. 11:991–1002. 55. M. H. Barcellos-Hoff (1998). How do tissues respond to damage at the cellular level? The role of cytokines in irradiated tissues. Rad. Res. 150:S109–S120. 56. M. H. Barcellos-Hoff (1998). The potential influence of radiation-induced microenvironments in neoplastic progression. J. Mam. Gland Biol. Neoplasia 3:165–175. 57. M. H. Barcellos-Hoff and S. A. Ravani (2000). Irradiated mammary gland stroma promotes the expression of tumorigenic potential by unirradiated epithelial cells. Cancer Res. 60:1254– 1260. 58. J. Greenberger, M. Epperly, A. Zeevi, K. Brunson, K. Goltry, K. Pogue-Geile, J. Bray, and L. Berry (1996). Stromal cell involvement in leukemogenesis and carcinogenesis. In Vivo 10:1– 17. 59. J. T. Leith and S. Michelson (1990). Tumor radiocurability: Relationship to intrinsic tumor eterogeneity and to the tumor bed effect. Invasion Metastasis 10:329–351. 60. L. T. Bemis and P. Schedin (2000). Reproductive state of rat mammary gland stroma modulates human breast cancer cell migration and invasion. Cancer Res. 60:3414–3418. 61. G. M. Hodges, R. M. Hicks, and G. D. Spacey (1977). Epithelialstromal interactions in normal and chemical carcinogen-treated adult bladder. Cancer Res. 37:3720–3730. 62. R. S. Cha, W. G. Thilly, and H. Zarbl (1994). N-nitroso-Nmethylurea-induced rat mammary tumors arise from cells with preexisting oncogenic Hras1 gene mutations. PNAS 91:3749– 3753. 63. A. Haimovitz-Friedman, I. Vlodavsky, A. Chaudhuri, L. Witte, and Z. Fuks (1991). Autocrine effects of fibroblast growth factor in repair of radiation damage in endothelial cells. Cancer Res. 51:2552–2558. 64. D. M. Gadbois and B. E. Lehnert (1997). Control of radiationinduced G1 arrest by cell-substratum interactions. Cancer Res. 57:1151–1156. 65. S. Kondo (1988). Altruistic cell suicide in relation to radiation hormesis. Int. J. Radiat. Biol. Phys. Chem. Med. 53:95–102. 66. C. Nathan and M. Sporn (1991). Cytokines in context. J.Cell Biol. 113:981–986. 67. A. B. Roberts, N. L. Thompson, U. Heine, C. Flanders, and M. B. Sporn (1988). Transforming growth factor-β: Possible roles in carcinogenesis. Brit. J. Cancer 57:594–600. 68. J. Massague, S. W. Blain, and R. S. Lo (2000). TGF-β signaling in growth control, cancer, and heritable disorders. Cell 103:295– 309. 69. M. W. J. Ferguson (1994). Skin wound healing: Transforming growth factor β antagonists decrease scarring and improve quality. J. Interferon Res. 14:303–304. 70. G. S. Ashcroft, X. Yang, A. B. Glick, M. Weinstein, J. L. Letterio, D. E. Mizel, M. Anzano, T. Greenwell-Wild, S. M. Wahl, C. Deng, and A. B. Roberts (1999). Mice lacking Smad3 show accelerated wound healing and an impaired local inflammatory response. Nature Cell Biol. 1:260–266. 71. T. Ohmori, J. L. Yang, J. O. Price, and C. L. Arteaga (1998). Blockade of tumor cell transforming growth factor-βs enhances cell cycle progression and sensitizes human breast carcinoma cells to cytotoxic chemotherapy. Exp. Cell Res. 245:350–359. 72. B. A. Teicher, Y. Maehara, Y. Kakeji, G. Ara, S. R. Keyes, J. Wong, and R. Herbst (1997). Reversal of in vivo drug


P2: GCO/LOV

Tally: FTK


PP133-300576

April 10, 2001

Epigenetics, the Microenvironment and Cancer

73.

74.

75.

76.

77.

78.

79.

resistance by the transforming growth factor-β inhibitor decorin. Int. J. Cancer 71:49–58. T. Haufel, ¨ S. Dormann, J. Hanusch, A. Schwieger, and G. Bauer (1999). Three distinct roles for TGF-β during intercellular induction of apoptosis: A review. Anticancer Res. 19: 105–111. M. H. Sieweke, N. L. Thompson, M. B. Sporn, and M. J. Bissell (1990). Mediation of wound-related Rous sarcoma virus tumorigenesis and TGF-β. Science 248:1656–1660. G. Furstenberger, M. Rogers, R. Schnapki, G. Bauer, P. Hofler, and F. Marks (1989). Stimulatory role of transforming growth factors in multistage skin carcinogenesis: Possible explanation for the tumor-inducing effect of wounding in initiated NMRI mouse skin. Int. J. Cancer 43:915–921. T. M. Fynan and M. Reiss (1993). Resistance to inhibition of cell growth by transforming growth factor-β and its role in oncogenesis. Crit. Rev. Oncogen. 4:493–540. D. F. Pierce, A. E. Gorska, A. Chythil, K. S. Meise, D. L. Page, R. J. Coffey, Jr., and H. L. Moses (1995). Mammary tumor suppression by transforming growth factor β1 transgene expression. Proc. Natl. Acad. Sci. U.S.A. 92:4254–4258. W. Cui, D. J. Fowlis, S. Bryson, E. Duffie, H. Ireland, A. Balmain, and R. J. Akhurst (1996). TGFβ1 inhibits the formation of benign skin tumors, but enhances progression to invasive spindle carcinomas in transgenic mice. Cell 86:531–542. G. Portella, S. A. Cumming, J. Liddell, W. Cui, H. Ireland, R. J. Akhurst, and A. Balmain (1998). Transforming growth factor β

9:32


221

80.

81. 82.

83.

84.

85.

is essential for spindle cell conversion of mouse skin carcinoma in vivo: Implications for tumor invasion. Cell Growth Differ. 9:393–404. C. Amendt, P. Schirmacher, H. Weber, and M. Blessing (1998). Expression of a dominant negative type II TGF-β receptor in mouse skin results in an increase in carcinoma incidence and an acceleration of carcinoma development. Oncogene 17:25–34. E. D. Hay (1995). An overview of epithelio-mesenchymal transformation. Acta Anat (Basel) 1:8–20. B. Tang, E. P. Bottinger, S. B. Jakowlew, K. M. Bangall, J. Mariano, M. R. Anver, J. J. Letterio, and L. M. Wakefield (1998). Transforming growth factor-β1 is a new form of tumor supressor with true haploid insufficiency. Nature Medicine 4:802–807. M. Hojo, T. Morimoto, M. Maluccio, T. Asano, K. Morimoto, M. Lagman, T. Shimbo, and M. Suthanthiran (1999). Cyclosporine induces cancer progression by a cell-autonomous mechanism. Nature 397:530–534. A. B. Glick, W. C. Weinberg, I. H. Wu, W. Quan, and S. H. Yuspa (1996). Transforming growth factor β1 suppresses genomic instability independent of a G1 arrest, p53, and Rb. Cancer Res. 56:3645–3650. A. Glick, N. Popescu, V. Alexander, H. Ueno, E. Bottinger, and S. H. Yuspa (1999). Defects in transforming growth factor-β signaling cooperate with a Ras oncogene to cause rapid aneuploidy and malignant transformation of mouse keratinocytes. Proc. Nat. Acad. of Sci. U.S.A. 96:14949–14954.