|
|
|
Vol. 07 Issue 3, Early Fall 2002
Elena N. Shmidt and Alexandre Yu. Nikitin
Department of Biomedical Sciences, Cornell University
Breast cancer is the most common malignancy among women, affecting approximately one in nine females in Western countries in their lifetimes. In the United States, breast cancer is the leading cause of death among women between the ages 40 and 55 years. (ACS, 2000).
Modeling human cancer in laboratory animals provides unique opportunities for studying the nature of the disease, as well as for testing new diagnostic tools, drugs, therapeutic approaches, and strategies for prevention and control. In the past decades, a large variety of mouse models for breast cancer has been developed. However, until recently, most of the mouse models poorly reflected human disease and were not defined genetically.
Unlike humans, many naturally occurring breast cancers in mice are caused by tumor viruses, such as Mouse Mammary Tumor retro-Virus (MMTV). (See The Ribbon, Volume 5, Number 3, early Fall 2000.) These mammary neoplasias, as well as those induced by exposure to chemical carcinogens and ionizing radiation, were used until the late 1990's (Gould, 1995). Unfortunately, morphology and biological characteristics of these tumors are quite distinct from those in humans. For example, the majority of human breast tumors have a pronounced stromal reaction and frequently metastasize (spread). In contrast, mouse MMTV-induced tumors usually have very little stroma and metastasize rarely.
An alternative approach for the modeling of mammary cancer is based on preparing transplantable tumors (Heppner, 2000). In this approach, tumors arising in humans are serially transplanted into immunodeficient mice. These models allow the study of biological behaviors of human neoplasia, and are frequently used for studying new therapeutic agents. However, these models do not allow study of the organism's immune responses to the tumor. Because such responses are among well-established factors affecting tumor formation, their omission may render transplantable experiments less biologically relevant.
This difficult situation with faithful mouse models for breast cancer has changed dramatically during the past two decades due to two main reasons.
First, remarkable progress has been made in understanding genetic mechanisms of cancer. In particular, it has become clear that genes involved in carcinogenesis can be of two main types: oncogenes and tumor susceptibility genes (Vogelstein, 1993; Weinberg, 1995). Normal products of proto-oncogenes usually play roles in promoting cell proliferation and in the prevention of differentiation and programmed cell death (apoptosis). This group also includes genes that increase genetic instability and favor mutagenesis. "Gain of function" mutations, amplification (increased copy number per cell) or overexpression (increased amount of product) of known oncogenes are associated with cancer. Examples of oncogenes involved in breast cancer include neu/erbB2/HER2, ras and myc. In contrast to oncogenes, tumor susceptibility genes either accomplish negative control of cell proliferation and promote differentiation (tumor suppressors), or positively control genetic stability and are directly involved in repair of genetic material. Usually such genes have "loss of function" mutations in tumors. Among the best known genes of this group are retinoblastoma susceptibility gene (Rb), p53, and breast cancer susceptibility genes (BRCA) 1 and 2.
Second, rapid advances were made in mouse embryology and genetics. It has become possible to construct genetically accurate mouse models of human cancer by insertion, deletion, or mutation of cancer-relevant genes in the mouse genome. Thereby, the role of individual human oncogenes and tumor susceptibility genes in cancer can be tested directly.
Specifically, several approaches have been developed for in vivo study of gene functions. In transgenic technology (Jaenisch, 1988), an artificial DNA sequence - transgene - is engineered in a test tube and introduced into a fertilized mouse egg (Fig. 1). Foreign DNA of the transgene then randomly integrates into a chromosome and becomes a part of the mouse genome. Any transgene consists of two main genetic elements: a coding sequence for a gene of interest and a promoter that drives an expression of this gene in cells of a host animal. Some promoters are ubiquitous and direct transgene expression in every cell type of the animal. Others are more specific. For example, in targeting transgene expression to the mammary gland MMTV and WAP promoters are frequently used (Cardiff, 1993). The MMTV promoter is derived from the MMT virus known to infect mouse mammary epithelial cells, and, as discussed above, induce mammary tumors. The WAP promoter is a part of the whey acidic protein gene, which is synthesized in normal mammary epithelial cells during the last half of pregnancy and lactation period. As a result, expression of a transgene under the control of this promoter depends on the physiological status of the animal.
The transgenic technology allowed direct testing of numerous human oncogenes for their transforming potential in mouse mammary cells. Notably, many of those oncogenes were able to induce mammary tumors in the mouse. Such genes include mammary epithelial cell growth factors (for example, Transforming Growth Factor (TGF) a, TGF-b, heregulin, Fibroblast Growth Factors (FGF), etc.), receptors for these growth factors (TGF-b DNIIR, ErbB2, etc.), second messengers in signal transduction pathways (e.g., ras, PTEN), regulators of cell cycle (e.g. myc and cyclin D) and differentiation (e.g. WNT and NOTCH4) (Hennighausen, 2000, Balmain, 2002).
The transgenic approach is useful for studying oncogenes. However, for inactivation of tumor suppressor genes a different method is required. Such a method relies on gene targeting technology (Capecchi, 1994). In this approach, desired genetic modification is first introduced into a cloned copy of the chosen gene by recombinant DNA technology in a test tube. Then, modified DNA is transferred into cultured embryonic stem (ES) cells which are undifferentiated cells capable of multiplying and differentiating into specialized cells. In the ES cells modified DNA replaces the normal gene by homologous recombination (Fig. 2). Homologous recombination is a regular genetic event that occurs in all cell types at low frequency. Recombinant ES cells are identified and injected into mouse blastocysts (early stage of embryonic development), which are in turn surgically transferred to foster mothers. The newborn "chimeric" mice are capable of transmitting the mutant genetic locus to their offspring.
The "knock out" technology has been successfully used for inactivation of many genes. Unfortunately, many tumor susceptibility genes are vitally important and their complete inactivation results in embryonic lethality (Rb and BRCA1). Yet inactivation of some other genes, such as p53, leads to development of other tumors, for instance, lymphomas, prior to mammary tumor formation (Deng, 2001, Gusterson, 1999). Thus, limited conditional inactivation of such genes is required for studying their functions in the mammary epithelial cells.
A number of approaches for conditional inactivation of tumor susceptibility genes have been developed. In the Cre-loxP system (Rossant, 1999), gene deletion is accomplished by Cre-recombinase enzyme from bacterial virus (phage) P1. This enzyme is able to excise (delete) any DNA sequence located between flanking recombinase-specific loxP sites. Placing Cre under the control of mammary specific promoter (see transgenic technology above) allows for expression of the recombinase followed by gene deletion only in cells of the mammary gland. For such experiments, two types of genetically engineered mouse strains are generated (Fig. 3): one contains "floxed" (flanked by loxP sites) gene of interest, and the other Cre-recombinase transgene. After intercrosses between these two strains, both genetic alterations appear in all cells of the progeny. As a result, Cre-recombinase that is selectively expressed in mammary epithelial cells catalyzes excision of "floxed" tumor suppressor gene in tissue-specific manner so that only mammary cells become predisposed for the cancer development. This method was successfully applied for conditional inactivation of the BRCA1, BRCA2 and p53 tumor suppressor genes in mammary epithelial cells (Xu, 1999; Jonkers, 2001). In agreement with the known role of these genes in human breast cancer, mice with targeted deletion of these genes developed mammary carcinomas.
In addition to limiting Cre expression to the mammary gland, it is often necessary to control time and anatomical area of gene deletion. For example, for understanding of mechanisms of cancer formation, it is necessary to study the effects of gene deletion at different stages of mammary gland development. Therefore, we developed an approach that combines Cre-mediated gene inactivation with tetracycline-controlled gene expression (Utomo, 1999). The tetracycline-controlled system consists of a gene controlled by the tetracycline-inducible operator. The operator is a special DNA sequence which is activated by the tetracycline-sensitive regulator protein (tetracycline-responsive activator), which is expressed under the control of a cell-specific promoter (Baron, 2000). Administration of antibiotic tetracycline activates tetracycline activator, and subsequent expression of the target gene, for example Cre recombinase, occurs. Because, tetracycline can be applied at any time, our approach allows initiation of Cre-recombinase-mediated gene deletion by administration of tetracycline whenever it is required. Furthermore, direct injection of tetracycline allows for gene inactivation in a limited area of the mouse body. Thus direct comparison of the mammary gland with and without gene deletion is possible in the same mouse.
The majority of mouse genetic models have been prepared relatively recently, and their complete characterization is still in progress. However, even the limited available observations indicate that genetically modified mice represent an extremely promising tool for modeling human breast cancer. For example, it was demonstrated that the activation of oncogenes and inactivation of tumor susceptibility genes identified in humans initiates breast cancer in mice. Importantly, alterations of these genes in the mouse gives rise to distinct tumor types, which are morphologically alike to known human breast cancer variants with similar genetic alterations (Cardiff, 2001, our unpublished observations). Furthermore, unlike spontaneous neoplasms, genetically defined tumors are frequently metastatic, similarly to their human counterparts. Since similar genetic alterations produce comparable phenotype both in the men and in the mouse, it is reasonable to expect parallel biological responses to molecular and pharmacological agents for cancer treatment and prevention.
Perhaps the main future challenge in the mouse modeling is to recapitulate complex natural settings of tumor formation. Breast cancer represents a heterogeneous group consisting of different sets of genetic alterations, histopathological types, and metastatic potential. Therefore, future experiments will be aimed at deciphering the crosstalk between individual cancer genes, and, hopefully, will help in pinpointing the most significant genetic alterations. At the same time, since breast cancer formation is also influenced by a number of non-genetic factors, such as diet and hormonal status (Alberg, 1999), mouse models shall be very useful for evaluation of the complex interplay between genetic and non-genetic alterations.
Taken together, rapid advances in molecular biology and embryology have allowed generation mouse models for breast cancer which in many respects are similar to human disease. Availability of such models will further facilitate our understanding of breast cancer, and will provide an invaluable tool for developing new rationally designed approaches for diagnosis, treatment, and prevention of this debilitating disorder.
Acknowledgements
This work was supported by grants from NIH (CA96823) and DOD (BC991016 and PC010342) to AYN. We thank Corinna Levine for critical reading of the manuscript.
References
American Cancer Society. Cancer Facts and Figures - 2000. American Cancer Society, Atlanta (2000).
Alberg, A.J. and B.B. Biesecker. Epidemiology, prevention, and early detection of breast cancer. Curr. Opin. Oncol. 11 (1998):435-441.
Balmain, A. Cancer as a complex genetic trait: tumor susceptibility in humans and mouse models. Cell 108 (2002):145-152.
Baron, U. and H. Bujard. Tet repressor-based system for regulated gene expression in eukaryotic cells: principles and advances. Methods Enzymol. 327 (2000):401-421.
Capecchi, M.R. Targeted gene replacement. Scientific Am. 270 (1994):52-59.
Cardiff, R.D. and W.J. Muller. Transgenic mouse models of mammary tumorigenesis. Cancer Surv. 16 (1993):97-113.
Cardiff, R.D. Validity of mouse mammary tumour models for human breast cancer: comparative pathology. Microsc. Res. Tech. 52 (2001):224-230.
Deng, C.X. and S.G. Brodie. Knockout mouse models and mammary tumorigenesis. Semin. Cancer Biol. 11 (2001):387-394.
Gould, M.N. Rodent models for the study of etiology, prevention and treatment of breast cancer. Semin. Cancer Biol. 6 (1995):147-152.
Gusterson, B., B. Howard, et al. Do we now have a relevant animal model for breast cancer? Breast Cancer Res. 1 (1999):2-4.
Hennighausen, L. Mouse models for breast cancer. Breast Cancer Res. 2 (2000):2-7.
Heppner, G.H., F.R. Miller, and P.M. Shekhar. Nontransgenic models of breast cancer. Breast Cancer Res. 2 (2000):331-334.
Jaenisch, R. Transgenic animals. Science 240 (1988):1468-1474.
Jonkers, J., R. Meuwissen, H. van der Gulden, H. Peterse, M. van der Valk, and A. Berns. Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer. Nature Genet. 29 (2001):418-425.
Rossant, J. and A. McMahon. "Cre"-ating mouse mutants - a meeting review on conditional mouse genetics. Genes Dev. 13 (1999):142-145.
Utomo, A.R., A.Yu. Nikitin, and W.-H. Lee. Temporal, spatial, and cell type-specific control of Cre-mediated DNA recombination in transgenic mice. Nature Biotechnol. 17 (1999):1091-1096.
Vogelstein, B. and K.W. Kinzler. The multistep nature of cancer. Trends Genet. 9 (1993):138-141.
Weinberg, R.A. The molecular basis of oncogenes and tumor suppressor genes. Ann N Y Acad. Sci. 758 (1995):331-338.
Xu, X., K.-W. Wagner, D. Larson, Z. Weaver, C. Li, T. Ried, L. Hennighausen, A. Wynshaw-Boris, and C.-X. Deng. Conditional mutation of Brca1 in mammary epithelial cells results in blunted ductal morphogenesis and tumour formation. Nature Genet. 22 (1999):37 - 43
Program on Breast Cancer and Environmental Risk Factors
Cornell University, College of Veterinary Medicine
Vet Box 31, Ithaca, NY 14853-6401
Phone: 607.254.2893; Fax: 607.254.4730
Email: breastcancer@cornell.edu
We subscribe to the HONcode principles. Verify here.